Beyond Peaceful Coexistence The Emergence of Space, Time and Quantum

www.ebook3000.
com
P1045_9781783268313_TP.indd 1 7/3/16 10:18 AM
May 2, 2013 14:6 BC: 8831 - Probability and Statistical Theory PST˙ws
This page intentionally left blank
www.ebook3000.com
ICP
P1045_9781783268313_TP.indd 2 7/3/16 10:18 AM

Published by
Imperial College Press
57 Shelton Street
Covent Garden
London WC2H 9HE
Distributed by
World Scientific Publishing Co. Pte. Ltd.
5 Toh Tuck Link, Singapore 596224
USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601
UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE
Library of Congress Cataloging-in-Publication Data

Names: Licata, Ignazio, editor.
Title: Beyond peaceful coexistence : the emergence of space, time and quantum / editor
Ignazio Licata (Institute for Scientific Methodology, Italy & School of Advanced International
Studies for Applied Theoretical and Non Linear Methodologies of Physics, Italy).
Description: London : Imperial College Press, [2016] | Singapore ;
Hackensack, NJ : Distributed by World Scientific Publishing Co. Pte. Ltd. | “2016
Identifiers: LCCN 2015049353| ISBN 9781783268313 (hc ; alk. paper) |
ISBN 178326831X (hc ; alk. paper)
Subjects: LCSH: Space and time. | Quantum theory. | Physics--Philosophy.
Classification: LCC QC173.59.S65 B49 2016 | DDC 530.11--dc23
LC record available at http://lccn.loc.gov/2015049353
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.
Copyright © 2016 by Imperial College Press

All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means,
electronic or mechanical, including photocopying, recording or any information storage and retrieval
system now known or to be invented, without written permission from the Publisher.
For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance
Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy
is not required from the publisher.
In-house Editors: Shilpa/Mary Simpson
Typeset by Stallion Press

Email: enquiries@stallionpress.com
Printed in Singapore
www.ebook3000.com
Shilpa - Beyond Peaceful Coexistence.indd 1 23/12/2015 9:12:53 AM

March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page v
Contents
Foreword ix
Gerard ’t Hooft
From Peaceful Coexistence to Co-Emergence xi

Ignazio Licata
The Algebraic Way 1

B. J. Hiley
Fermi Blobs and the Symplectic Camel: A Geometric

Picture of Quantum States 27
Maurice A. de Gosson
Space–Time in Quantum Gravity: Does Space–Time

have Quantum Properties? 45
Reiner Hedrich
Introduction to the Quantum Theory of Elementary

Cycles 93
Donatello Dolce
Observers and Reality 137

George Jaroszkiewicz
v
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page vi
vi Contents
The Stability of Physical Theories Principle 153

R. Vilela Mendes
Factory of Realities: On the Emergence of Virtual

Spatiotemporal Structures 201
Romàn R. Zapatrin
Space–Time from Topos Quantum Theory 221

Cecilia Flori
From Born Reciprocity to Reciprocal Relativity: A

Paradigm for Space–Time Physics 265
Peter Jarvis
On Non-Equilibrium Thermodynamics of Space–

Time and Quantum Gravity 287
Joakim Munkhammar
World Crystal Model of Gravity 299

Hagen Kleinert
Quantum Features of Natural Cellular Automata 307

Hans-Thomas Elze
Structurally Dynamic Cellular Networks as Models

for Planck Scale Physics and the Quantum Vacuum 327
Manfred Requardt
On a Time–Space Operator (and other

Non-Self-adjoint Operators) for Observables
in QM and QFT 371
Erasmo Recami, Michel Zamboni-Rached and Ignazio Licata
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page vii
Contents vii
Emergent Space–Time 419

George Chapline
The Idea of a Stochastic Space–Time:

Theory and Experiments 433
M. Consoli and A. Pluchino
. . .And Kronos Ate His Sons 465

Giuseppe Vitiello
The Emergence of Space–Time: Transactions

and Causal Sets 487
Ruth E. Kastner
An Adynamical, Graphical Approach to Quantum

Gravity and Unification 499
W. M. Stuckey, Michael Silberstein and Timothy McDevitt
Is Bohr’s Challenge Still Relevant? 545

Leonardo Chiatti
In and Out of the Screen. On Some New

Considerations about Localization
and Delocalization in Archaic Theory 559
Ignazio Licata
Schrödinger–Milne Big Bang — Creating a “Universe

of Threeness” 579
Geoffrey F. Chew
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page viii
viii Contents
Quantized Fields à la Clifford and Unification 615

Matej Pavšič
Non-commutative Einstein, almost Kähler–Finsler

and Quantum Deformations 661
Sergiu I. Vacaru
Index 697
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page ix
Foreword
Gerard ’t Hooft
Protons, electrons, photons, and all other basic building block of matter
are subject to the laws of quantum mechanics, laws that at first sight seem
to come from an alien world. This fact may rightfully be considered to be
one of the most profound discoveries of the 20th century. “Quantum” is
the counterpart of “Continuum”, and indeed, classical mechanics may be
regarded as the mechanics of the continuum. Already, the ancient Greeks
suspected that matter is quantized, and they called the quantum of mat-
ter “atoms”. Now, we know that also energy, angular momentum, electric
charge, and many other notions from physics come in integer quantities of
elementary units, which were called “quanta”.
Being indivisible means that these quanta must be entirely different
from their classical counterparts, and indeed, their behavior is very odd.
The most curious feature is that the presence of quanta is to be regis-
tered by entities that we call wave functions, and these wave functions may
cause quite complex interference phenomena. Wave functions of quanta can
become “entangled”, and if that is the case, new twists are to be given to
what we normally call probability distributions. They seem to defy our
common sense. Theoreticians could imagine experiments that would yield
bizarre outcomes; experimentalists took the challenge and actually did the
experiments, just to observe, indeed, that the predicted “impossible” out-
comes were what they found.
Yet, theoreticians realized that all of this bizarre physics is calling for a
more natural explanation. The world of large, “continuous” objects seems
to be orderly and simple. Why then are their most fundamental building
blocks, the quanta, so strange? Where does quantum mechanics come from?
Actually, however, one may turn the question around: is quantum mechanics
ix
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page x
x G. ’t Hooft
indeed as strange as it looks? Or could it be that this is what we should

have expected, the only conceivable result of chopping entities originally
assumed to be continuous into unbreakable pieces?
Suppose now that literally everything that we thought of as being con-
tinuous is actually composed of unbreakable elementary units, not only
matter, charge and energy, but also space and time and everything else.
In that case, our world would start to look more and more like a computer,
a device that chops up all its data into fundamental units that can be pro-
cessed. This would be a majestic alley towards a fundamental unified theory
of nature. Every aspect of our universe could be framed with bits and bytes
of data. Everything we thought we knew about our universe would actually
be “emergent”: simplified descriptions, using simplified notations, of our
incomplete knowledge. Today, we have learned to use the real number sys-
tem to describe the continua that we think we are observing. When we buy
a pound of sugar, we forget that sugar is composed of grains, and although
these grains could be broken up to yield tinier grains, the ultimate quantum
of sugar is the sucrose molecule. So the real numbers we use to express the
number of sucrose molecules in terms of pounds are not quite right; indeed,
real numbers are manmade inventions, which we should use with care.
How should we understand the transition from discrete to continuous,
and from continuous to discrete? Whence this stupefying “quantum logic”?
How stupefying actually is this logic? If you do not know exactly how a
continuum is to be split up in quanta, should quantum mechanics not be
the only tool available? Quantum logic is not just bizarre, it is also beautiful
in its mathematical perfection. If you come to think of it, we could not
have wished for a better tool for doing calculations. Combining it with
other physical principles such as special relativity opens up a grandiose
mathematical scheme called quantum field theory, allowing us to describe
the most elementary constituents of matter in a synthesis of simplicity
and complexity. Numerous experiments have shown that this really does
describe the world we live in. So, quantum mechanics is here to stay, but
we still have to learn how it emerged from simple logical principles. The
quantum world should not be looked upon as alien, but one where straight
and natural logic rules.
Back to peaceful coexistence of all its inhabitants. If only we knew how.
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xi
From Peaceful Coexistence

to Co-Emergence
Ignazio Licata
ISEM, Institute for Scientific Methodology, 90146 Palermo
and
School of Advanced
International Studies on Applied Theoretical
and Non Linear Methodologies of Physics, 70132 Bari
Ignazio.licata@ejtp.info
It may be that a real synthesis of quantum and relativity theories requires

not just technical developments but radical conceptual renewal
J. S. Bell
A physical theory consists of a formal structure and one or more interpre-

tations. The latter can come out from cultural and cognitive tension going
far beyond any sound operational pact between theoretical constructs and
empirical data. We have no reason to doubt about the consistency and
efficacy of syntaxes if properly used in the right range. The formal side
of Physics has grown in a strongly connected and stratified way through
an almost autopoietic, self-dual procedure (let’s think of the extraordinary
success of the gauge theories), whereas the foundational debate is still blus-
tering about the two pillars of such monumental construction. The general
relativity (GR) and the quantum mechanics (QM), which still appear to be
greatly incompatible and stopped in a limited peaceful coexistence between
local causality in space–time and quantum non-locality [1]. The formidable
challenges waiting for us beyond the Standard Model seem to require a new
semantic consistency [2] within the two theories, so as to build a new way
to look at them, to work and to relate them.
On the side of QM, the huge work in optics laboratories on non-locality
and entanglement in the last decade has stimulated a renewed interest
xi
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xii
xii I. Licata
for the foundational debate and the historical stratifications of Quantum

Physics [3, 4]; the authority of the Copenhagen Interpretation has eroded
more and more, and also consequently the meaning of the Bell Inequal-
ity has been questioned. It is clear that a renewed comprehension of QM
is not just a philosophical “passing fancy”, but a physics’ necessity, and
confirms that what we call “quantum mechanics” today has the reasons of
its probabilism elsewhere. In particular, it is no more possible to consider
the research on quantum foundations as an isolated “conceptual planet”, it
must include the structural connections with its ripest filiation, the quan-
tum field theory (QFT) and the new challenges of quantum gravity (QG).
The conceptual compromises, as the various forms of subquantal medium,
have to deal with more radical and constraining proposals.
Also, space–time has undergone revitalizing cures due to both the dif-
ferent GR variants, and, above all, the QG exigencies. The GR and its
classical variations are well set within the metric elasticity and thermody-
namics of vacuum, which thus describes gravity on large scales and low
energies as arising from the statistical behavior of microscopic degrees of
freedom. In this regard, we have to remember the pioneering and revo-
lutionary work by A. Sacharov on the ST metric elasticity [5, 6]. At this
point, the temptation becomes very strong for additional hypotheses on
the “Eddington rubber-sheet” able to include the quantum stochastic fea-
tures; actually, some moderately successful attempts have been made. Once
again, things are not so simple, indeed. The recent models of emergent ST,
based on a top-down approach, i.e. zooming from a classical GR view to
smaller and smaller spatial scales, well characterize a mesoscopic range, but
they let the cosmological questions and the authentic microscopic domain
of QG unsolved. This is where the most interesting theories have maybe
emerged. The Covariant Loop QG has realized the “traditional” program
of the GR quantization (“start with a classical theory, then quantize”)
and showed clearly the conceptual atomism/relational dialectics guiding
Physics, whereas other approaches like Quantum Graphity have offered the
first models of geometrogenesis [7–9]. Thus, from the microscopic viewpoint,
the classical ST appears as a very rich structure intertwined with causal
relations; it emerges according to a pre-geometric fabric of reality, like in
the It from Bit by Wheeler and the Holomovement by Bohm [10, 11].
At this point, the question that spontaneously comes out is: where is QM
situated? Is it something fundamental and irreducible, a point of no return
with respect to classical physics, as it is thought to be since 1927 (not
without some authoritative dissidents!), the expression of an “undivided
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xiii
From Peaceful Coexistence to Co-Emergence xiii
oneness” or can we consider it “emergent” in some way? The quantum

events are measured in the ST, but the last one is a structure of relation,
fixed by the causal relations between interactions, which are an exquisitely
quantum fruit. On the other hand, the QM structure has always allowed
deterministic/stochastic models, and we know that contextuality is a fea-
ture of complex systems [12]. These considerations seem to suggest to us
that if the ST is not a fundamental structure, then it is neither the QM
and viceversa.
An extremely powerful guiding principle, even in the plurality of its
formulation, is the holographic principle [13–16]. Physicists know well that
when difficulties arise, they can always count on thermodynamics. In addi-
tion, as Boltzmann teaches us, “if something heats up, it has microstruc-
ture”! that is where — just like for classical space–time — the idea of an
emergent nature of QM has developed, which proposes again the Einstein
vs Bohr debate in a new way [17].
It is not by chance if the most radical challenge to QM comes from Ger-
ard ’t Hooft, one of the researchers who “closed” the standard model and
gave start to the road towards QG [18]. ’t Hooft had the idea, simple and
charming, to put forward again the question of the hidden variables, shifting
it at Planck scale. In other words, it is no more a matter of looking for hid-
den variables inside the wave function, neither to consider it as the formal
expression of an undivided non-locality nor the systemic approximation of
a “superfluid” behavior. ’t Hooft idea is that of a dynamical sub-quantum
level able to express a non-banal vacuum that is characterized by discrete,
local and deterministic cells whose collective behaviors — describable as
cellular automata (CA) in a Hilbert space — make exactly the quantum
level! Thus, the wave function is not to be interpreted, it is just a statistical
facing here, and neither do we have to find any corresponding “element of
the physical reality”, which is situated, on the contrary, at the “ontologi-
cal” level of fundamental cells. God plays dice by using a discrete form of
determinism (radical quantization approach or prequantization):
Quantum mechanics could well relate to micro-physics the same way
thermodynamics relates to molecular physics: it is formally correct, but it
may well be possible to devise deterministic laws at the micro scale. Why
not? The mathematical nature of quantum mechanics does not forbid this,
provided that one carefully eliminates the apparent no-go theorems associ-
ated to the Bell inequalities. There are ways to re-define particles and fields
such that no blatant contradiction arises. One must assume that all macro-
scopic phenomena, such as particle positions, momenta, spins, and energies,
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xiv
xiv I. Licata
relate to microscopic variables in the same way thermodynamic concepts

such as entropy and temperature relate to local, mechanical variables. The
outcome of these considerations is that particles and their properties are
not, or not entirely, real in the ontological sense. The only realities in this
theory are the things that happen at the Planck scale. The things we call
particles are chaotic oscillations of these Planckian quantities. [Gerard’t
Hooft, Does God Play Dice, Physics World, December 2005]
Between the levels of observables and the fundamental one of the beables,
there are the changeables, dynamical variables changing by means of the
degrees of freedom which get lost in passing from a level to another during
the creation of structures; so, it is useless searching for a one-to-one cor-
respondence between the two ranges. The Planck level and the quantum
one are connected by a mapping which the quantum states emerge from —
non-locality included — as classes of equivalence, i.e. a plurality of vacuum
states can give rise to the same emergences. In this way, we make the most of
one of the well-known lessons in the physics of nonlinear systems, in accor-
dance with dissipative QFT as well as the scale usually considered for QG.
It must be said that, at this stage, the CA model is nothing more than
a black-box, waiting for more cogent physical hypotheses. A very promising
line of research, which overturns, in a sense, the traditional relationships
between classical and quantum physics, consists in associating a nonlinear
oscillator whose coupling ways can generate the known forms of matter
and energy to each cell of such singular Rubik–Planck cube, with a peculiar
care for the subtle problems arising in QFT and the Standard Model [19].
Basically, the QM and TQC laws are the low energy consequences emerg-
ing from a high-energy chaos. If the world is a great cellular automaton,
which rules and particles meta-rule on changeable guide it? What about
cosmological questions, the start-problem?
We have already said that the holographic principle can be conjugated in
different ways. We could say it is a principle in search of a theory. Generally,
the mapping is between two dynamics — one on a screen and the other on a
bulk, and some lines of research are really ambitious about emergence both
of gravity and QM from a holographic screen as entropy reservoir [20]. More
radical approaches, as for both syntax and semantics, that are closer to the
Bohm explicate order or the von Weizsäcker Ur Materie are also possible.
There the setup is purely algebraic, non-locality is fundamental and the
concept itself of “dynamics” is replaced by a logics of distinctions — a
parameter of explication — between background information and actualized
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xv
From Peaceful Coexistence to Co-Emergence xv
information. We should rather speak about manifestation theories as U.

Mohrhoff points out:
The manifestation of the world consists in a transition from a condition

of complete indefiniteness and indistinguishability to a condition of com-
plete or maximal definiteness and distinguishability, and what occurs in the
course of this transition-what is not completely definite or distinguishable
can only be described in terms of probability distributions over what is com-
pletely definite and distinguishable (. . .) The reason why local explanations
do not work may be the same as the reason why the manifestation of the
spatiotemporal world cannot be explained by processes that connect events
within the space–time arena. The manifestation of the world is the nonlocal
event par excellence. Instead of being an event in space–time, it is, depend-
ing on one’s point of view, either “outside” of space–time (i.e. not localized
at all) or coextensive with space–time (i.e. completely delocalized). It is the
process by which Being enters into reflexive relations and matter and space
come into being as a result (. . .) [21]
We will not dwell on the cultural and paradigmatic differences hidden

by a shared formal acceptance of the status quo of theoretical physics. The
t’Hooft vision belongs to a line of thought going from Democritus to Turing
and Einstein, whereas the “generative” approach to the space–time matter
problem seems to bypass the so-called “dogma motion”, the very roots of
the “becoming” in physics, a problem that D. Bohm had already evidenced
clearly and that has peeped out many times in cosmology and in QG as
pre-space–time structures [22]. We seem that these approaches — currently
well concealed by any possibility of experimental verification — get some
merits as well as difficulties. The merit is that they both overcome the
dichotomy quantizing geometry/geometrizing quantum by means of radical
proposals suggesting to look at Universe as an information network. The
problem is already known with geometry [23] and lies in the fact that it is
difficult to imagine information without any material correlate. Thus, just
like the Planck cells, pre-geometries, generative algebras, and any kind of
informational causet need new and strong ideas.
Another question that fully falls under this scenario regards the size
and nature of the space–time “atoms”. The history of Planck’s Physical
Units of Nature is charming and worthy to be mentioned shortly. As the
experimental evidences of observables’ discretization were collected — such
as the electrical charge — the idea of a quantization of the world at a
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xvi
xvi I. Licata
“fundamental” level grew more and more. In 1881, G. Stoney was the first
one to propose his “Natural Units” and to use the word electron. For a
long time, the question of the units of nature was not connected to the
general space–time structure, as we generally do today, but to the extended
structural particle models (Lorentz based his interpretation of Relativity
on these ones; M. Abraham, A. Sommerfield). Planck’s units came in with
his Quantum Theory as a necessary corollary. To be more precise, Planck
thought that the units that would have taken his name were the most radical
innovative aspect of his work. Let us remember that he was a “reluctant
revolutionary” about quanta (so he was defined by Helge Kragh in the
famous commemorative paper on Physics World, December 2000) and he
was fundamentally sceptic on the developments of Quantum Mechanics.
About quanta, he writes they were: “an act of despair . . . I was ready to
sacrifice any of my previous convictions about physics.” The tune is quite
different when he describes the natural units: “These necessarily retain their
meaning for all times and for all civilizations, even extraterrestrial and non-
human ones, and can therefore be designated as ‘natural units’.” Although
they arrived before QM and GR, most physicists think that Planck scale is
the natural scale of QG, the one where the gravitational interaction has the
same strength as the other ones, the one which defines the causal scheme
of the space–time continuum and under which it is impossible to localize a
physical event.
An elegant and historical reflection about the meaning of Planck’s units
can be found in the work by Treder [24]. If, at the beginning, the theoretical
physicists attention was centered on the compatibility with a Rubik cube
world at Planck scale with Relativity (or, anyway, with a Relativity!) and
on non-commutativity [25], today the question seems to be shifted to their
activity and especially to “why” the space and time bricks are made of
these units. This brings back to the problem of the epistemological meaning
of “natural constant” [26]. Just within the holographic approach, there
are interesting attempts to connect the Planck constant with a Boltzmann
“counting” of states [27].
After all, the word “space–time” evokes both a “theatre of coordinates”
and specific aspects of the Einstein theory. Today, we have good reasons
to think that at microscopic level, such conceptions have to undergo a
change whose nature we cannot see clearly yet, but from which the rel-
ativistic theatre emerges as a “coarse grain” description. At that level,
space–time becomes something that the matter sees as metrics. All that
makes the notion of “elementary physical event” and its intrinsic spatiality
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xvii
From Peaceful Coexistence to Co-Emergence xvii
and temporality come to the foreground again. Any relativistic pact fails
there and everything requires a redefinition, as Fotini Markopoulou reminds
us: Our faulty assumption is that space is real. We enter here the nobody’s
land from the reformulation of the isomorphism between SU(2) and SO(3)
in qbits tone to the amplituhedron, there is a strong exigency of a generation
mechanism starting from an archaic precursor of time and space [28]. Other
aspects emerge from some versions of the holographic principle and seem
to confirm the intuition of some semi-classical geometrodynamic theories
about a deep connection between entanglement and ST production [29].
Moreover, is time a definite object or a process? Despite most of Planck
scale supporters, there exists a chronon approach firmly rooted in experi-
mental reasons. Is it more correct to describe the Caldirola chronon (based
on classical electron radius) and the Finkelstein chronon (based on analy-
sis of high energy scattering process; TOTEM Collaboration, 2012: for pp
processes, the value is 10−25 s) as scale-time instead of a “minimum time”.
The question is not just “how big a chronon is?” We should ask how many
chronons coexist [30]. Actually, the problem is inescapable also in the Rubik
on Planck scale-like models: how do we define the “transition time” from a
state to another one of the cell?
The original meaning of “peaceful coexistence” in A. Shimony sense (the
collapse postulate and non-locality) appears to be as the tip of the iceberg
on more radical and structural problems crossing the whole Physics. What
is a “particle”? How to introduce non-locality in QFT? Are the “realistic”
interpretations only FAPP or something more? Does some radical form of
discretization (Planck scale on other kinds of chronon) exist? Does space–
time emerge from a net of a sort of primeval “quantum events”? What is
its relation with information theory? What level QM as it is can be placed?
Does an unexpected third way between determinism and randomness exist?
Such questions have deep implications on the concept itself of “physical
theory”.
In this volume, I invited the authors to reflect on the sea around the
tip of the iceberg. The chapters might be ideally grouped according to
the approaches presented geometrodynamics, emergent, discrete, algebraic,
geometric, topological ones, the new interpretative spectrum of Quantum
theories at the border and beyond Copenhagen, the discrete time theories,
the timeless approaches, the ‘super-fluid’ pictures of space–time. Obviously,
there are wide and fecund superposition zones. We know that physics does
not develop like a pyramid, but like a totality of theories more and more
strongly connected [31].
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xviii
xviii I. Licata
Acknowledgment
A challenge like this is quite demanding. I feel a deep gratitude towards all
the authors who has risen to the challenge, and also for those who could not
be present. I learnt something from each of them, and that is how physics
makes life beautiful. Victor Weisskopf used to say that when you are down
you hate to think that Mozart and Quantum physics do exist. A part of the
history of the volume started from me and David Finkelstein exchanging
mails on Parmenides and Heraclitus. I am deeply and happily indebted to
B. Hiley and I owe special thanks to L. Chiatti, G. Vitiello and T. Elze. The
volume also owes a lot to the spirit of DICE conferences. A big thanks to
G. ’t Hooft for his kindness, but also for the clarity as well as the deepness
of the words in his Preface. The cover artwork is based on the installation
Multiverse project 2015, a gift of the artist Teresa Iaria.
Finally, the project could not have taken shape without my precious
editors at the Imperial College: Alice Owen, Tasha Cruz and Mary Simpson.
References
1. A. Shimony, Metaphysical problems in the foundations of quantum mechan-
ics, Int. Phil. Quarterly 18(1), 3–17 (1978).
2. C.F.F. von Weizsäcker, The Structure of Physics, Springer, 2006.
3. A. Valentini and G. Bacciagaluppi, Quantum Theory at the Crossroads:
Reconsidering the 1927 Solvay Conference, Cambridge University Press,
Cambridge, 2009.
4. O. Jr. Freire, The Quantum Dissidents. Rebuilding the Foundations of Quan-
tum Mechanics (1950–1990), Springer, 2015; L. Gilder, The Age of Entan-
glement: When Quantum Physics Was Reborn, Knopf, 2008.
5. A. Sacharov, Vacuum quantum fluctuations in curved space and the theory
of gravitation, Doklady Akad. Nauk S. S. R. 177, 70–71 (1987).
6. B.L. Hu, Can space-time be a condensate? Int. J. Theor. Phys. 44(10), 1785–
1806 (2005); E. Di Casola, S. Liberati, and S. Sonego, Between quantum and
classical gravity: Is there a mesoscopic space-time? Found. Phys. 45(2), 171–
176 (2015).
7. T. Padmanabhan, Lessons from classical gravity about the quantum structure
of space-time. J. Phys. Conf. Ser. 306, 012001 (2011).
8. F. Vidotto, Atomism and Relationalism as Guiding Principles for Quantum
Gravity, arXiv:1309.1403 [physics.hist-ph], 2013.
9. F. Markopoulou, The Computing Space-time, in How the World Computes,
S.B. Cooper, A. Dawar, B. Löwe, Springer, 472–484 (2012); see also Space
does not exist, so time can, arXiv:0909.1861 [gr-qc], 2009.
10. J.A. Wheeler, Information, physics, quantum: the search for links, Proc. III
Int. Symp. Found. Quantum Mechanics, Tokyo, 354–368, 1989.
www.ebook3000.com
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xix
From Peaceful Coexistence to Co-Emergence xix
11. D. Bohm, Wholeness and the Implicate Order, Routledge, 1980.

12. Y. Orlov, Quantum — Type Coherence as a Combination of Symmetry and
Semantics, arXiv:quant-ph/9705049, 1997; K. A. Kitto, Contextualised gen-
eral systems theory, Systems 2(4), 541–565 (2014); see also: W. McHarris,
Chaos and the Quantum: how nonlinear effects can explain certain quantum
paradoxes, J. Phys. Conf. Ser. 306, 012050 (2011).
13. G. ’t Hooft, Dimensional Reduction in Quantum Gravity, Salamfest arXiv:gr-
qc/9310026, 1993.
14. L. Susskind, The world as a hologram, J. Math. Phys. 36(11), 6377–6639
(1995).
15. R. Bousso, The holographic principle, Rev. Mod. Phys. 74(3), 825–874 (2002).
16. P.C.W. Davies, The Implications of a Cosmological Information Bound for
Complexity, Quantum Information and the Nature of Physical Law, in Ran-
domness and Complexity, From Leibniz to Chaitin, ed. C. Calude, World
Scientific, 69–87 (2007).
17. An agile historical survey, in addition to the already cited [3], can be found
in: R.P. Crease, A.S. Goldhaber, The Quantum Moment: How Planck, Bohr,
Einstein, and Heisenberg Taught us to Love Uncertainty, W.W. Norton &
Company, 2014; M. Kumar, Quantum: Einstein, Bohr, and the Great Debate
about the Nature of Reality, W.W. Norton & Company, NY, 2011.
18. G. ’t Hooft, For a Short Introduction: The Fate of the Quantum,
arXiv:1308.1007 [quant-ph], 2013. For a recent and detailed analysis of the AC
model see: G. ’t Hooft, The Cellular Automaton Interpretation of Quantum
Mechanics. A View on the Quantum Nature of our Universe, Compulsory or
Impossible? arXiv:1405.1548 [quant-ph], 2015.
19. The literature grows exponentially, for an introduction: Blasone, M., Jizba,
P., Vitiello, G., Dissipation and quantization, Phys. Lett A 286(3/4), 205–
210 (2001); H.T. Elze, Deterministic models of quantum fields, J. Phys.:
Conf. Ser. 33, 399–404 (2006); P. Jizba, F. Scardigli, M. Blasone, and G.
Vitiello, ’t Hooft, Quantization for interacting systems, J. Phys.: Conf. Ser.
343 012110 (2012). For a naı̈ve approach to pre-quantization see also: Licata,
I. Minkowski space-time and Dirac vacuum as ultrareferential fundamental
frame, Hadr. J. 14(3), 225–250 (1991); Licata, I. The Big Computer, Com-
plexity and Computability in Physical Universe, in Determinism, Holism, and
Complexity, Pellegrini, C., Cerrai, P., Freguglia, P., Benci, V., and Israel,
G. Kluwer Academic/Plenum Press, NY, pp. 117–123, 2003; Winterberg,
F. Planck mass plasma vacuum conjecture, Z. Naturforsch. 58a, 231–267
(2003); Wintenberg, F. Relativistic quantum mechanics as a consequence of
the Planck mass plasma conjecture, Int. Jour. Theor. Phys. 46(12), 3294–
3311 (2007).
20. D. Acosta, P. Fernandez de Cordoba, J.M. Isidro, J.L.G. Santander, A holo-
graphic map of action onto entropy, J. Phys.: Conf. Ser. 361, 012027 (2012);
J. Munkhammar, Canonical relational quantum mechanics from information
theory, EJTP 8(25), 93–108 (2011).
March 8, 2016 8:8 Beyond Peaceful Coexistence 9in x 6in b2362-fm page xx
xx I. Licata
21. U. Mohrhoff, Why the Wave Function, of All Things? arXiv:1411.1145 [quant-
ph], 2014; U. Mohrhoff, Quantum mechanics and the manifestation of the
world, Quantum Stud. Math. Found. 1(3), 195–202 (2014).
22. E. Anderson, Problem of time in quantum gravity, Annalen Phys. 524,
757–786 (2012); C. Kiefer, Conceptual Problems in Quantum Gravity and
Quantum Cosmology, ISRN Mathematical Physics, 509316 (2013); I. Licata,
A Note on The Origin of Time in Archaic Universe, NeuroQuantology, 12, 1
(2014).
23. D. Meschini, M. Letho, Is empty space-time a physical thing? Found. Phys.
36(8), 11931216 (2006).
24. H.J. Treder, On the Physical Meaning of Planck’s Natural Units, in The
Meaning of Quantum Gravity, eds. H.H. von Borzeszkowski, H.J. Treder,
Springer, pp. 114–123 (1988).
25. H. Snyder, Quantized space-time, Phys. Rev. D 38–41 (1947); A. Schild, Dis-
crete space-time and integral Lorentz transformations, Canad. J. Math. 1,
29–47 (1949); P.E. Gibbs, The Small Scale Structure of Space-Time: A Biblio-
graphical Review, arXiv:hep-th/9506171, 1996; A.I. Vialtsev, Discrete Space-
Time, Nauka, Moscow, 1965. [in Russian]
26. M.J. Duff, L.B. Okun, and G. Veneziano, Trialogue on the number of funda-
mental constants, JHEP 0203 (2002).
27. J.Y. Ng, Space-time foam: From entropy and holography to infinite statistics
and non-locality, Entropy 10, 441–461 (2008).
28. F. Markopoulou, Space Does Not Exist, So Time Can, arXiv:0909.1861 [gr-
qc], 2009; N. ArkaniHamed, J. Trnka, The Amplituhedron, arXiv:1312.2007,
2013.
29. B. Swingle, B. Constructing Holographic Space-Times using Entanglement
Renormalization, arXiv:1209.3304 [hep-th], 2009; On the semi-classical side
see: G. Resconi, I. Licata, D. Fiscaletti, Unification of quantum and grav-
ity by non classical information entropy space, Entropy 15, 3602–3619
(2013). Suggestive the transPlamckian geometry proposed in : Winterberg, F.
Teichmüller space interpretation of quantum mechanics, Ann. Found. Louis
de Broglie 38, 129–137 (2013).
30. G. Jaroszkiewicz, Principles of Discrete Time Mechanics, Cambridge Univer-
sity Press, 2014.
31. I. Licata, Methexis, Mimesis and Self-Duality: Theoretical Physics as Formal
Systems, Versus, 118, 119–140 (2014).
www.ebook3000.com
March 8, 2016 8:7 Beyond Peaceful Coexistence 9in x 6in b2362-ch01 page 1
Chapter 1
The Algebraic Way

B. J. Hiley
Physics Department, UCL and TPRU, Birkbeck,
University of London, Malet Street, London WC1E 7HX, UK
b.hiley@bbk.ac.uk
In this chapter, we examine in detail the non-commutative symplectic

algebra underlying quantum dynamics. By using this algebra, we show
that it contains both the Weyl–von Neumann and the Moyal quantum
algebras. The latter contains the Wigner distribution as the kernel of the
density matrix. The underlying non-commutative geometry can be pro-
jected into either of two Abelian spaces, so-called ‘shadow phase spaces’.
One of these is the phase space of Bohmian mechanics, showing that it
is a fragment of the basic underlying algebra. The algebraic approach is
much richer, giving rise to two fundamental dynamical time development
equations which reduce to the Liouville equation and the Hamilton–
Jacobi equation in the classical limit. They also include the Schrödinger
equation and its wave-function, showing that these features are a partial
aspect of the more general non-commutative structure. We discuss briefly
the properties of this more general mathematical background from which
the non-commutative symplectic algebra emerges.
1. Introduction
The basic principle of the algebraic approach is to avoid starting with a
specific Hilbert space scheme and rather to emphasize that the primary
objects of the theory are the fields (or the observables) considered as purely
algebraic quantities, together with their linear combinations, products, and
limits in the appropriate topology ([1]).
In order to understand the motivation for “The Algebraic Way”, we
need to recall the origins of quantum theory. History tells us that the first
pioneering papers to develop a mathematical approach to quantum phe-
nomena were those of Born et al. [2–4]. Their attempts to accommodate
1
2 B. J. Hiley
the Ritz–Rydberg combination principle, an empirical discovery in atomic

spectra, into a dynamical theory forced the classical variables of Hamilto-
nian dynamics to be replaced by non-commuting analogues.
With the emerging law of non-commutative multiplication, the need
for a matrix representation of x and p was soon recognized even though
the physical meaning of such a change was unclear. To the physicist,
these matrix representations opened up a new field of unfamiliar non-
commutative algebras with which they were not very comfortable and since
the mathematics itself appeared to have no obvious physical interpretation,
the approach was eventually abandoned in favor of the Schrödinger wave
mechanics approach.
This approach gained greater impetus when particles were found to
exhibit the wave-like behavior predicted by de Broglie. These experimental
results encouraged Schrödinger [5] to look for what he called a “Hamilto-
nian undulatory mechanics” by modifying the Hamilton–Jacobi offshoot of
Hamiltonian dynamics. His motivation came from noting that while ray
optics could be explained using equations that were analogous in form to
Hamilton’s equation-of-motion for particles, the Hamilton–Jacobi theory
contained surfaces of constant action, which suggested an analogy with the
wave fronts used in the Huygens construction to explain interference phe-
nomena in light.
This exploration led Schrödinger [5] to a differential equation which
immediately produced energy levels that conformed with the Ritz–Rydberg
data. The mathematical techniques involved in solving differential equations
were well known to physicists at that time and the faith in this equation
was further reinforced with the introduction of Born’s probability postulate,
establishing the relation between wave and particle. Although this relation
was not entirely clear conceptually, it enabled the formalism to be applied
with outstanding success.
Conceptually, the wave and algebraic approaches were very different, one
being based on a very familiar wave phenomenon, the other being based on
an unfamiliar non-commutative dynamics with no obvious interpretation.
Soon Schrödinger himself showed how the two approaches were related and,
since the techniques for solving the Schrödinger equation were very familiar,
this approach became established as the way to understand the physics of
quantum phenomena.
Nevertheless, many conceptual problems remained, generating many dif-
ferent interpretations, some naive others quite bizarre, all based on the
assumption that the Schrödinger equation tells the whole story not only
www.ebook3000.com
The Algebraic Way 3
for understanding individual experimental phenomena but also in defining

what ultimately constitutes ‘reality’. However, as we will show in this paper
that, in spite of its great successes, it is only a part of the whole story. In
order to see this, we need to return to examine the details of the original
algebraic approach in some detail.
2. Representations
Before discussing these issues, I would like to briefly highlight the relevant
features of the Schrödinger approach that we will need in order to motivate
our presentation. Of course, we will start with the Schrödinger equation,
even though it is not clear exactly how it was derived from the Hamilton–
Jacobi theory:
∂ψ
i = Ĥψ with H(x, p) → Ĥ(X̂, P̂ ),
∂t
where the classical Hamiltonian H(x, p) is replaced by its operator form
Ĥ(X̂, P̂ ).
To work with the equation, we must go to a specific representation. It is
customary to use the Schrödinger representation for which
∂
X̂ → x P̂ → −i ψ → ψ(x, t),
∂x
so that we are working in configuration space (x1 , x2 , . . . , xn ).
However, this is not the only representation. We can use the p-
representation, where
∂
X̂ → i P̂ → p ψ → ψ(p, t),
∂p
so that in this case we are working in momentum space (p1 , p2 , . . . , pn ).

Again, we have the oscillator representation, where
√ √
X̂ → (a† + a)/ 2 P̂ → i(a† − a)/ 2 N̂ = a† a.
This representation enables us to work more easily with an arbitrary num-

ber of particles and is essential for quantum field theory. Of course, the
Schrödinger representation is favoured because we believe that quantum
processes actually occur in Minkowski space–time.
Although there is an abundance of mathematical representations, the
Stone–von Neumann theorem proves that all irreducible representations
4 B. J. Hiley
are unitarily equivalent. By this, we mean that if there are two unitary
representations, π1 and π2 , in their respective Hilbert spaces, H1 and H2 ,
π1 : G → U (H1 ) and π2 : G → U (H2 )
and there exists an operator A : H1 → H2 , then these representations are
equivalent iff there exists an operator A such that
Aπ1 (g) = π2 (g)A ∀g ∈ G.
Having established mathematical equivalence, we are left with the question,
“Are the representations also physically equivalent?” This, in turn, leaves
another question “Of what mathematical structures are they representa-
tions?”
3. Common Symmetries underlying both Classical

and Quantum Mechanics
It is generally believed that quantum phenomena “demand a fundamental
modification of the basic physical concepts and laws” [6]. In other words, we
require a totally different description from that used in classical physics.
However, there are some obvious similarities in the form of the dynam-
ical equations-of-motion. In classical mechanics, Hamilton’s equations-of-
motion can be written in the form,
ṗi = {pi , H} and generally f˙(xi , pi ) = {f (xi , pi ), H},
ẋi = {xi , H};
(1)
where H is the Hamiltonian and {., .} are the Poisson brackets defined by
∂f ∂g ∂f ∂g

{f, g} = − ,
i
∂xi ∂pi ∂pi ∂xi
giving the special case {xi , pj } = δij .

On the other hand, in quantum mechanics, Heisenberg’s equations-of-
motion appear in the form,
dX̂i dP̂i dF̂
i = [X̂i , Ĥ]; i = [P̂i , Ĥ] and generally i = [F̂ , Ĥ]. (2)
dt dt dt
Here, Ĥ is the classical Hamiltonian, where x and p are replaced by oper-
ators X̂ and P̂ and [., .] are the commutator brackets defined by
[F̂ , Ĝ] = F̂ Ĝ − ĜF̂ ,
giving in the special case [X̂i , P̂j ] = iδij .
www.ebook3000.com
The Algebraic Way 5
The similarity in the form of the two sets of equations is quite remark-
able, in spite of the differences in the nature of the elements involved. The
classical equations-of-motion are ordinary functions on a continuous phase
space while, in the quantum case, they are operators acting on vectors in
an abstract Hilbert space. However, there is one other striking similarity.
They are both invariant under the Heisenberg group.
The Heisenberg equations-of-motion directly use elements of the Heisen-
berg (Lie) algebra defined by the canonical commutation relations,
[X̂i , X̂j ] = [P̂i , P̂j ] = 0, [X̂i , P̂j ] = δij T̂ , [X̂i , T̂ ] = [P̂i , T̂ ] = 0.
Here, we have written T̂ = iIˆ for convenience, so that the elements
(X̂i , P̂i , T̂ ) generate the Heisenberg group, Hn .
On the other hand, the classical dynamical variables are representations
of the Heisenberg algebra in which commutators are replaced by Poisson
brackets. Thus, the Heisenberg group is not only significant in the quantum
domain but also operates in the classical domain. In fact, it plays a vital
role in radar theory [7], which is in no way a quantum phenomenon.
There is a further invariance which is more directly seen in the classical
mechanics in the dynamical equations-of-motion (1). They are are invari-
ant under transformations of the symplectic group Sp(2n) (i.e. canonical
transformations) for a 2n-dimensional phase space. These transformations
leave invariant the antisymmetric bilinear form ω(x, p; x , p ) = xp − x p.
Although one can prove this directly, it can also be thought of as arising
from the group of automorphisms of the underlying Heisenberg group.
If we write two elements of the Heisenberg group in the form,
n
n

Û = xi X̂i + pi P̂i + tT̂ , Û = xi X̂i + pi P̂i + t T̂ ,
i=1 i=1
we find
[Û , Û ] = ω(x, p; x , p )T̂ , (3)
where ω is again an antisymmetric bilinear form. The appearance of
ω(x, p; x , p ) in Eq. (3) implies that the Heisenberg group and, hence the
Heisenberg equations-of-motion, are invariant under the group of symplec-
tic transformations. In other words, the group of automorphisms of the
Heisenberg group is the symplectic group.
This means that the mathematical structure underlying both classical
and quantum dynamical behavior arises from symplectic geometry. It turns
out that in the quantum case, the symplectic geometry is non-commutative,
6 B. J. Hiley
while in the classical case, it is commutative. Although these structures are

clearly related mathematically, we still have a puzzle as to why there is
no trace of an underlying phase space in the quantum algebra and, even
if we were to find one, then how would it accommodate the Heisenberg
uncertainty principle?
4. How do we Relate the Quantum Algebra

to the Phase Space Description?
To find the role of phase space in quantum mechanics, we must put aside
any objections based on the uncertainty principle and follow some early
work of von Neumann [8]. Let us consider not only translations in x-space,
but also translations in p-space. As is well known, space translations can
be described using the Taylor expansion so that

∂
f (x + a) = exp α f (x). α ∈ R.
∂x
In the case of a translation in momentum space, we may similarly write

∂
g(p + β) = exp β g(p). β ∈ R.
∂p
By recalling the Schrödinger representation, we can write these exponentials

in operator form, namely
Û (α) = exp(iαP̂ ) and V̂ (β) = exp(iβ X̂).
We immediately see that these translations do not commute because
Û (α)V̂ (β) = eiαβ V̂ (β)Û (α).
The operators Û (α) and V̂ (β) generate the Weyl–von Neumann algebra.
4.1. Relation between the Weyl–von Neumann algebra

and Hilbert space
To make the link with the Hilbert space formalism, von Neumann intro-
duced the algebraic element
Ŝ(α, β) = e−iαβ/2 Û (α)V̂ (β). (4)
www.ebook3000.com
The Algebraic Way 7
Then for a system described by |ψ in the usual Hilbert space, we can form
the expectation value,
Sψ (α, β) = ψ|Ŝ(α, β)|ψ.
von Neumann then shows that any linear operator Â can be symbolically
written as

Â = a(α, β)Ŝ(α, β)dαdβ.
This leads to a quantum expectation value of the operator Â via

ψ|Â|ψ = a(α, β)Sψ (α, β)dαdβ,
where the kernel a(α, β) is defined by

a(α, β) = α + γ|Â|α − γe−2iβγ dγ.
In this way, we can completely reproduce the expectation values of quantum

mechanics in terms of functions of real variables (α, β). I refer the reader
to von Neumann for the details.
In the Weyl–von Neumann approach, then the operators of the quantum
formalism are replaced by differential functions on the α, β-space. However,
von Neumann made no attempt to explain the physical meaning of the space
spanned by the parameters α and β. Nevertheless, one fact emerges: The
multiplication of two of these functions, say, a(α, β) and b(α, β), must be
non-commutative in order to reproduce the results of quantum mechanics.
Suppose Â ↔ a(α, β) and B̂ ↔ b(α, β), then if Â.B̂ ↔ a(α, β) b(α, β),
von Neumann shows that

a(α, β) b(α, β) = e[i(αβ −α β)/2] a(α − α , β − β )b(α , β )dα dβ .
(5)
Not only is this star-producta non-commutative, it is also non-local. Thus,

non-locality appears as a basic feature of the α, β plane, so if we want to
replace the operators of the quantum formalism by continuous functions,
then the resulting structure must be non-local.
a This product, although first defined by von Neumann, is now known as the Moyal
star-product.
8 B. J. Hiley
4.2. Moyal’s contribution to the physical meaning

of the Weyl–von Neumann algebra
Moyal [9] arrived at exactly the same mathematical structure as von
Neumann but by starting from a very different approach. He was try-
ing to understand the nature of the statistics that is needed in quantum
mechanics, so he asked the question “How can we generalize the statistics
of random variables if these variables are non-commutative?”
With a pair of commutative random variables X, Y, one defines the
expectation values by introducing the characteristic function ei(Xt+Y s) .
Then, the expectation value of some function fX,Y (x, y) is

φX,Y (t, s) = E[ei(Xt+Y s) ] = ei(xt+ys) fX,Y (x, y)dxdy.
Moyal proposed that, in the non-commutative case, the characteristic func-

tion can be replaced by ψ|ei(αP̂ +β X̂) |ψ so that we can form the function

1
Fψ (x, p) = ψ|ei(αP̂ +β X̂) |ψei(αp+βx) dαdβ. (6)
4π
He then proposed that the average of any quantum operator Â can be found
using

ψ|Â|ψ = a(x, p)Fψ (x, p)dxdp. (7)
Note that Moyal has now introduced the two parameters x and p through
the Fourier transform (6) and since we are dealing with a single particle, it
has been assumed that these parameters are the position and momentum
of a single particle. If that were the case, then from the form of Eq. (7), we
could regard Fψ (x, p) as a probability distribution for the particle having
coordinates (x, p) and we can then regard Eq. (7) as giving the quantum
expectation value for the operator Â by averaging a(x, p) over a phase space.
There are two difficulties in making such an assumption.
1. As is well known, Fψ (x, p) is the Wigner functionb and can become neg-
ative. The assumption that Fψ (x, p) is a probability density then opens
up a debate as to the validity of the whole approach. However, we will
show that Fψ (x, p) is not a probability distribution, but the kernel of
a density matrix which is not necessarily positive definite or even real.
Thus, it is the interpretation of Fψ (x, p) being a probability distribution
b We will show this later in Sec. 4.3.
www.ebook3000.com
The Algebraic Way 9
that is not valid, not the method in which it arises, so we can follow
Feynman [10] and use Eq. (7) as a valid way to evaluate the quantum
expectation values without worrying about the appearance of negative
values of Fψ (x, p). We need to remember that we are dealing with a non-
commutative structure and not simply averaging over classical coordi-
nates.
2. As is not so well known, the parameters (x, p) are not the position and
momentum of a localised particle, but the mean values of a cell in phase
space associated with the particle. Thus, in this approach, the particle
cannot be considered as a point-like object. Rather, it is a non-local
distribution of energy, the quantum blob [11, 12]. This region, which we
associate with the particle, explains the non-local nature of the star-
product.
4.3. Relation to the Wigner distribution

We will now show the function Fψ (x, p) is, in fact, the one particle Wigner
function, the many-body generalization of which was first introduced by
Wigner [13] to discuss the thermodynamic properties of quantum systems.
First, consider the operator Ŝ(α, β) defined in Eq. (4) written in a
slightly modified form,
Ŝ (α, β) := eiαβ Û (α)V̂ (β) = eiαP̂ /2 eiβ X̂ eiαP̂ /2 .
It is not difficult to show that

ψ|Ŝ (α, β)|ψ = ψ ∗ (x − α/2)eiβx ψ(x + α/2)dx.
By taking the Fourier transform, we find

1
Fψ (x, p) = ψ ∗ (x − α/2)e−iαp ψ(x + α/2)dα, (8)
2π
which we recognize as the Wigner function. Thus, we see the Wigner func-
tion is intimately connected with the Weyl–von Neumann–Moyal algebraic
approach.
4.4. Non-commutative phase space

In order to confirm that we are dealing with a non-commutative phase
space, we will follow Moyal [9], who showed that the star-product (5) can
10 B. J. Hiley
be written in a more convenient way,

←− −→ → ←
− −
a(x, p) b(x, p) = a(x, p) exp[i( ∂ x ∂ p − ∂ x ∂ p )/2]b(x, p). (9)
It is not difficult to show that this expression when applied to x and p givesc
x p − p x = i.
Thus we see that although we are dealing with functions of ordinary real
(x, p) variables, the usual commutative inner product must be replaced by
a non-commutative product.
Once we have a non-commutative product we must distinguish between
left and right multiplication. However we find it easier to take this into
account by introducing two types of bracket, namely,
a b−b a a b+b a
{a, b}MB = and {a, b}BB = .
i 2
The first is the Moyal bracket, while the second is the Baker bracket (or
the Jordan product). Using the expression for the product (9), it is easy to
show
←− −→ → ←
− −
{a, b}MB = a(x, p) sin[( ∂ x ∂ p − ∂ x ∂ p )/2]b(x, p)
and
←− −→ → ←
− −
{a, b}BB = a(x, p) cos[( ∂ x ∂ p − ∂ x ∂ p )/2]b(x, p).
The importance of these brackets is that they become classical objects in

the limit O(2 ). The Moyal bracket becomes the Poisson bracket
{a, b}MB = {a, b}P B + O(2 ) = [∂x a∂p b − ∂p a∂x b] + O(2 ),
while the Baker bracket to the same approximation reduces to the simple
product
{a, b}BB = ab + O(2 ).
Thus we see that the non-local star-product now becomes the local inner
product used in classical mechanics. Thus in one single formalism we have
a way of dealing with both quantum and classical mechanics.d
c In the earlier sections, we have used the parameter p without giving it a physical mean-
ing. If we want to interpret it as a momentum, we must replace it by p/.

d These results form the basis of deformation quantization [14].
www.ebook3000.com
The Algebraic Way 11
5. Non-Commutative Dynamics: The Phase Space

Approach
As we have seen, an important lesson when dealing with a non-commutative
algebra is to carefully distinguish between left and right multiplication.e We
have been able to avoid this distinction by going to the Schrödinger repre-
sentation which gives a simpler algorithm that only uses left multiplication.
To exploit the full implications of the non-commutative structure, we have
to go deeper.
To define the dynamics in such a mathematical structure, we have to
consider the following two equations

−1 →
−
H(x, p) Fψ (x, p, t) = i(2π) e−iτ p ψ ∗ (x − τ /2, t) ∂ t ψ(x + τ /2, t)dτ
(10)
and

−1 ←−
Fψ (x, p, t) H(x, p) = −i(2π) e−iτ p ψ ∗ (x − τ /2, t) ∂ t ψ(x + τ /2, t)dτ.
(11)
Subtracting these two equations gives us one time development equation

expressed in terms of the Moyal bracket:
∂t Fψ = (H Fψ − Fψ H)/2i = {H, Fψ }MB . (12)
While by adding the two equations, we get another time development equa-
tion expressed in terms of the Baker bracket [15]:

←→
2{H, Fψ }BB = i(2π)−i e−iτ p [ψ ∗ (x − τ /2, t) ∂ t ψ(x + τ /2, t)]dτ, (13)
where
←→ →
− ←−
ψ∗ ∂ t ψ [ψ ∗ ∂ t ψ − ψ ∗ ∂ t ψ]
= . (14)
ψ∗ ψ ψ∗ ψ
It should be noted that we need both equations to get a complete description
of quantum mechanics. For a more detailed discussion see Zachos [16].
We have already seen that Eq. (12) leads to the classical Liouville equa-
tion in the classical limit. To see what Eq. (13) gives in the classical limit,
e More formally the mathematical structure of quantum mechanics is a bimodule.

12 B. J. Hiley
let us put ψ = ReiS into Eq. (13), expand out and then take the limit to
O(2 ). We find
{H, Fψ }BB = H · Fψ + O(2 ) = −2(∂t S)Fψ + O(2 ),
which then gives the classical Hamilton–Jacobi equation,

∂S
+ H = 0.
∂t
A related approach to the classical limit will be found in Schleich [17].
This is a very interesting result when we recall that Schrödinger actu-
ally started from the classical Hamilton–Jacobi equation in order to find
a “Hamiltonian undulatory mechanics”. One of the reasons why he was
forced to guess his equation was because he not did fully appreciate the
significance of non-commutativity.
6. Where does the Bohm Approach fit

into this Structure?
6.1. Conditional expectation values
in the Moyal approach
Since the Moyal algebra gives the correct quantum expectation values of
quantum operators by averaging over a symplectic phase space and since
the Bohm approach gives the same expectation values using what seems to
be a different phase space defined in terms of (x, p = ∇S), there surely must
be a relation between these two approaches. To bring out this relationship,
let us follow Moyal and treat Fψ (x, p) as a quasi-probability distribution.
We can then define the conditional expectation value of the momentum.
A value of this momentum can be obtained from the general relation
given by Moyal [9], namely
n
n n
ρ(x)p = p Fψ (x, p)dp = [(∂x1 − ∂x2 )ψ(x1 )ψ(x2 )]x1 =x2 =x .
2i
(15)
For n = 1, we find, by writing ψ = ReiS , that

1 ∗
p(x) = [ψ ∇ψ − (∇ψ ∗ )ψ] = ∇S(x).
2i
This is identical to the Bohm momentum defined by the relation p = ∇S,
the so-called “guidance relation”. However, in the approach we are exploring
www.ebook3000.com
here, there are no waves of any form and the notion of guiding wave is mean-
ingless. Everything that emerges is a consequence of the non-commutative
symplectic geometry.
This connection between the Bohm momentum and the conditional
expectation value of the momentum can be made even stronger. Moyal
shows that by starting from the Heisenberg equations-of-motion, the trans-
port of the momentum p(x, t) is given by

∂t (ρpk ) + ∂xi (ρpk ∂xi H) + ρ∂xk H = 0.
i
Then after some work and again writing ψ = ReiS , Moyal finds

∂ ∂S ∇2 ρ
+H − = 0.
∂xk ∂t 8mρ
If we choose H = p2 /2m + V , where
2 ∇2 ρ
p2 = (∇S)2 − 2 (∇R/2R)2 + .
4 ρ
Then,
∂S ∇2 ρ ∂S 1 1 ∇2 R
+H − = + (∇S)2 + V − = 0. (16)
∂t 8mρ ∂t 2m 2m R
Here, the right-hand side of Eq. (16) is the quantum Hamilton–Jacobi equa-
tion, the real part of the Schrödinger equation that plays a key role in
the Bohm approach [18]. But since the Moyal algebra contains the Bohm
approach, and in fact is exactly the von Neumann algebra (i.e. an algebra
upon which quantum mechanics is based) then clearly the Moyal and the
Bohm approach are simply different aspects of precisely the same mathe-
matical structure.
Full details of the above derivations can be found in the appendix of the
original Moyal paper [9]. Further details of the relation between the Moyal
and the Bohm approach can be found in Hiley [12].
6.2. Shadow manifolds

What Sec. 6.1 shows is that if we take the variable x as one axis of the phase
space, we can take p to be the other axis of the phase space. Thus, we have
constructed a phase space out of the variables (x, p). In this phase space,
14 B. J. Hiley
the time development equation is the quantum Hamilton–Jacobi equation,
∂t S(x, t) + (∇x S(x, t))2 /2m + Qx (x, t) + V (x, t) = 0. (17)
Here, the quantum potential, Qx (x, t), is given by

2
1 ∇ R(x, t)
Qx (x, t) = − .
2m R(x, t)
Thus, we can construct trajectories in this (x, p) space.
However, note that the distribution Fψ (x, p) is symmetric in x and p
so that we can also find the conditional expectation value of the position,
x(p, t), in terms of the momentum p. We will again follow Moyal and define
this value x as

ρ(p)x = xFφ (x, p)dx

= xψ ∗ (x )ψ(x )δ[x − (x + x )/2]eip(x −x ) dxdx dx ,
which in the p-representation takes the simpler form,

1
ρ(p)x = xFφ (x, p)dx = [(∂p1 − ∂p2 )φ∗ (p1 )φ(p2 )]p=p1 =p2 .
2i
Writing φ(p) = R(p)eiS(p) , we find the conditional expectation value of the
position, x(p), given the value of p is
x(p) = −∇p S(p).
Again in analogy with the previous case, we have another quantum

Hamilton–Jacobi, only this time in p-space. Thus,
∂t S(p, t) + p2 /2m + Qp (p, t) + V (−∇p S(p, t), t) = 0, (18)
where

1 ∂ 2 Rp
Qp (p, t) = − (19)
2mRp ∂p2
is the quantum potential in a second phase space constructed in terms of the
coordinates (x = −∇p S, p). An example of how this works for the case of a
particle in a potential V (x) = Ax3 will be found in Brown and Hiley [19],
where more details of the whole approach are given.
Thus, we find that there are, at least, two shadow phase spaces we can
access. Each gives a different phase space picture of the same overall alge-
braic structure, a feature that has already been recognized in the Wigner
www.ebook3000.com
approach by Leibfried et al. [20] who call these spaces shadow phase spaces,
a term Hiley [21] has also used.
These shadow spaces are an example of what Bohm calls ‘explicate
orders’ in his general notion of the implicate order [22]. In this case, the
algebraic structure defines the implicate order, while the two shadow phase
spaces are a pair of explicate orders. One should note that both equa-
tions (12) and (13) do not contain quantum potentials explicitly. They only
appear explicitly in Eqs. (17) and (18), namely at the level of conditional
expectation values. One should also note that in the classical limit p → p
and x → x, so that, in this limit, both quantum potentials vanish and we
have one unique phase space.
7. Non-Commutative Dynamics: The Algebraic

Approach
7.1. Operator equations
We can get more insight into this whole approach by returning to the
operator approach and exploiting the one-to-one relation Â ↔ a(x, p).
This means we should be able to form the operator equivalent of the two
Eqs. (12) and (13). In order to motivate this, let us return to consider how
the Schrödinger equation emerges from the Heisenberg equation for the
time development of the density operator ρ̂
dρ̂
i = [Ĥ, ρ̂] (20)
dt
Let us follow Dirac [23] and write ρ̂ = ψ̂.φ̂. Note that both ψ̂ and φ̂ are
operators, not vectors in a Hilbert space. Substituting this expression into
Eq. (20), we get
dψ̂ dφ̂
i φ̂ + iψ = (Ĥ ψ̂)φ̂ − ψ̂(φ̂Ĥ)
dt dt
Note we can actually form this equation by subtracting the following two
Schrödinger-like equations:
dψ̂
i = Ĥ ψ̂ (21)
dt
and
dφ̂
−i = φ̂Ĥ (22)
dt
16 B. J. Hiley
We say ‘Schrödinger-like’ because ψ̂ and φ̂ are elements of the opera-

tor algebra. Note the order of the operators in these two equations; in
Eq. (21), the operators act from the left, while in Eq. (22), the operators
act from the right. In fact, these equations are left and right translation
Schrödinger equations, the analogues of Eqs. (10) and (11) proposed in the
von Neumann–Moyal algebra.
Recall that to obtain Eq. (12), we subtracted Eqs. (10) and (11), so we
see that the Heisenberg equation-of-motion can be formed by subtracting
Eqs. (21) and (22). There is a clear analogy with the bra and ket vectors,
but here ψ̂ and φ̂ are taken to be elements of the non-commuting algebra,
not elements of an external abstract Hilbert space. ψ̂ and φ̂ are, in fact,
elements of a specific left and right ideal respectively that exist within the
non-commuting symplectic algebra itself. The implications of this for any
possible physical interpretation have been discussed in Hiley [24] and Hiley
and Callaghan [25].
Thus, in our approach, all the elements we use appear in the algebra
itself and there is no essential need to introduce an exterior Hilbert space,
although this alternative is available if required for ease of calculation. This
then shows clearly that the Schrödinger equation is, as Bohr [26] claimed,
merely an algorithm for calculating the outcome of given experimental sit-
uations. But unlike Bohr, we are giving attention to the algebra, in this
case, the non-commutative symplectic group algebra. It is this algebra that
provides a complete mathematical description of the quantum dynamics.
We will now bring out this algebraic structure more clearly by adopting
a change of notation, in which ‘operators’ simply become elements of the
algebra because they ‘operate’ on themselves. Thus, we will drop the ‘hats’
and write ψ̂ → ΨL and φ̂ → ΦR . Here, ΨL is an element of a suitable
left ideal and ΦR an element of some suitable right ideal defined by the
physics of the problem we are considering. These elements contain all the
information about the state of the system. Mathematically, they are central
features of the symplectic Clifford algebra [27]. Similar features appear in
the orthogonal Clifford algebra used to describe the spin and relativistic
properties of quantum phenomena [28]. A detailed discussion of how one
chooses these ideals will be found in that paper.
7.2. Left/right algebraic equations

Let us replace the density operator ρ̂ of a pure state by ρ = ΨL ΨR , where
ΨL is the left ideal in the algebra and ΨR is the right ideal. Then, the left-
www.ebook3000.com
and right-hand side equations of motion are

dΨL dΨR
i = HΨL and − i = ΨR H.
dt dt
Next, we form
→
− →
− ←− ←
−
i( ∂ t ΨL )ΨR = ( H ΨL )ΨR and − iΨL(ΨR ∂ t ) = ΨL (ΨR H ).
Now, we can subtract and add these two equations as before and obtain
the two algebraic equations,
−→ ←− →
− ←
−
i ( ∂ t ΨL )ΨR + ΨL (ΨR ∂ t ) = ( H ΨL )ΨR − ΨL (ΨR H ), (23)
−→ ←− →
− ←
−
i ( ∂ t ΨL )ΨR − ΨL (ΨR ∂ t ) = ( H ΨL )ΨR + ΨL (ΨR H ). (24)
Since we are writing ρ = ΨL ΨR , Eq. (23) can be written in the form,
i∂t ρ = [H, ρ]− . (25)
This is, in fact, just the quantum Liouville equation. Equation (24) can be
written in the form,
←
→
iΨR ∂ t ΨL = [H, ρ]+ , (26)
where we have used definition (14). This equation is simply the expression
for the conservation of energy. Thus, Eqs. (25) and (26) then are the alge-
braic equivalents of (12) and (13) and give a complete algebraic description
of a single quantum system.
7.3. Emergence of the Bohm approach through

projections
In the previous sub-section, we showed Eqs. (25) and (26) to be the defining
equations for the time development of a single quantum system in terms of
the non-commutative symplectic structure. Note once again that there is
no explicit quantum potential in these equations.
To see how these equations are related to the usual Hilbert space
approach, we first introduce a projection operator Πa = |aa| and apply it
to each equation in turn. We obtain
∂P (a)
i + [ρ, H]− a = 0,
∂t
∂S
2P (a) + [ρ, H]+ a = 0.
∂t
Here, P (a) is the probability of finding the system in the quantum state
ψ(a) which we have written in polar form ψ(a) = R(a)eiS(a) .
18 B. J. Hiley
In order to get a feel for this approach, it is useful to consider particular

p2 2
examples. Therefore, let us consider the harmonic oscillator, H = 2m + Kx
2
for its simplicity and for the fact that it is symmetric in x and p. We will
choose two specific projection operators, Πx = |xx| and Πp = |pp|.
We will begin by projecting into the x-representation using Πx = |xx|
to obtain

∂P (x) ∇x Sx
+ ∇x · P (x) = 0, (27)
∂t m
2
∂Sx 1 ∂Sx 1 ∂ 2 Rx Kx2
+ − + = 0. (28)
∂t 2m ∂x 2mRx ∂x2 2
Thus, we see that Eq. (27) is the Liouville equation which is the expres-
sion for the conservation of probability in the x-representation. Equation
(28) is the quantum Hamilton–Jacobi equation in the x-representation that
appears in Bohmian mechanics.
Let us now project into the p-representation by choosing the projection
operator Πp = |pp| to obtain

∂Pp ∇p Sp
+ ∇p · Pp = 0, (29)
∂t m
2
∂Sp p2 K ∂2R K ∂Sp
+ − + . (30)
∂t 2m 2Rp ∂p2 2 ∂p
K ∂ R
Note the appearance again of a quantum potential Qp = − 2R p ∂p2 .
Thus, we see the quantum potential becomes manifest only as a result of
the projections. Note that when the quantum potential is negligible, we
recover the classical behavior, Eqs. (28)–(30) being related by a canon-
ical transformation. Although we have illustrated these projections for
the harmonic oscillator, it follows trivially that they work for any general
Hamiltonian.
Thus, projections from the non-commutative algebraic time develop-
ment Eqs. (25) and (26) produce exactly the same results as obtained from
the two von Neumann–Moyal equations (12) and (13). Both lead to the same
pair of shadow phase spaces. Both produce the same quantum Hamilton–
Jacobi equations, namely, Eqs. (17) and (18).
www.ebook3000.com
8. Conditional Expectation Values from the Algebra

Let us now return to our original motivation, namely, that the primary
mathematical structures necessary to describe quantum phenomena are
non-commutative geometric algebras. In this chapter, we have concentrated
on the non-commutative symplectic geometry, restricting ourselves to spe-
cific examples to motivate the general method. In a series of papers, Hiley
and Callaghan [25, 28, 29] have shown how the orthogonal Clifford algebras
can be used to describe the spin and relativistic properties of quantum
systems.
These two algebraic approaches are very similar in their mathematical
structure, so there is clearly a more general structure of which these algebras
are specific examples. Indeed they are both simple examples of von Neu-
mann algebras and general methods for handling these non-commutative
algebras now exist [30].
We will be particularly interested in their relevance to non-commutative
probability theory, and in particular, the appearance of conditional expec-
tation values in these structures, which has non-commutative integration
theory at its heart [31]. We have seen the need to consider left and right
differentiation, so that the inverse of differentiation, namely, integration has
to take this two-sidedness into account. Equation (7) has been interpreted
as providing the expectation value of a(x, p) taken over Fψ (x, p), treating it
as if it were a classical probability density. When it was subsequently dis-
covered that Fψ (x, p) can become negative, alarm bells may have sounded
as has been discussed in Refs. [10, 32, 33] and more recently in Ref. [34].
Yet, in spite of these difficulties, the expectation values ψ|Â|ψ calculated
by these methods always turn out to be positive.
The explanation of these results lies in non-commutative measure the-
ory, particularly in the papers of Umegaki [35] and Jones [36]. What
Umegaki shows is that a positive definite conditional expectation value
always exists in a sub-algebra N of a type II factor von Neumann algebra M,
which is the type of algebra we are discussing in this chapter. In particu-
lar, the conditional expectation EN : M → N is defined by the relation
tr(EN y) = tr(xy) for x ∈ M and y ∈ N . The map EN is normal and has
the following properties:
EN (axb) = aEN (x)b, for x ∈ M, a, b ∈ M
EN (x∗ ) = EN (x)∗ , ∀x ∈ M
EN (x∗ )EN (x) EN (x∗ x) and EN (x∗ x) = 0 ⇒ x = 0.
20 B. J. Hiley
Since the von Neumann–Moyal algebra we are discussing here is a type

II von Neumann algebra, a trace exists and it remains to evaluate this trace
for the two possible projections from the (x, p) algebra to the two Abelian
sub-algebras, one spanned by x and the other by p. Our case is trivial since
we are considering the special case of a single particle.
One of the projections we have introduced is EP : (x, p) → (x) which
was defined by Eq. (15). A careful examination of the origins of Fψ (x, p)
shows that it is actually the kernel of the density matrix itself. This result
has already been pointed out in Hiley [21], but we will outline the argument
briefly again here.
Let us start with the density operator ρ̂ψ = |ψψ| and form ρψ (x1 , x2 ) =
ψ ∗ (x1 )ψ(x2 ) which is the kernel of the density matrix [37]. Now, let us go
to the momentum representation and write

1
ψ(x) = φ(p)eipx dp.
(2π)
Then the density kernel can be written as

1
ρψ (x1 , x2 ) = φ∗ (p1 )e−ix1 p1 φ(p2 )eix2 p2 dp1 dp2 .
2π
Now let us change co-ordinates to
X = (x1 + x2 )/2 η = x2 − x1 and P = (p1 + p2 )/2 π = p2 − p1 ,
so that the density kernel can be written in the form

1
ρψ (X, η) = φ∗ (P − π/2)eiXπ φ(P + π/2)eiηP dP dπ.
2π
Take the Fourier transform,

ρψ (X, η) = Fψ (X, P )eiηP dP
and find

1
Fψ (X, P ) = φ∗ (P − π/2)eiXπ φ(P + π/2)dπ. (31)
2π
Recalling that

1
φ∗ (P − π/2) = √ ψ ∗ (x1 )e−i(P −π/2)x1 dx1 ,
2π

∗ 1
φ (P + π/2) = √ ψ ∗ (x2 )e−i(P +π/2)x2 dx2 .
2π
www.ebook3000.com
Using these in Eq. (31), we find

1
Fψ (X, P ) = ψ ∗ (X − η/2)e−iηP ψ(X + η/2)dη
2π
which is just the expression we used in Eq. (8) with η = α.

Note in this construction that the resulting Wigner function is a function
in the (X, P ) phase space. This phase space has been constructed from a
pair of points in (x1 , x2 ) configuration space and the coordinates (X, P )
are the mean position and mean momentum of a cell in an (x, p) phase
space. Thus, the Wigner function Fψ (X, P ) is a density matrix over a cell
constructed in the underlying (x, p) classical phase space. We have kept our
arguments deliberately simple to arrive at this result. A rigorous geometric
approach that produces this result and its generalization can be found in
Cariñena et al. [38].
The first point to note is that the Wigner function is a complex density
matrix, not a probability density. This shows why it is incorrect to regard
Fψ (x, p) as a probability distribution of particle positions and momenta.
Thus, the worries about negative and complex “probabilities” are totally
unfounded [34].
The second point to note is that the Wigner approach, when applied
to a single particle, is non-local depending on a region rather than a sin-
gle point. This means we must represent the particle by a region in phase
space, namely, the “quantum blob” introduced by de Gosson [11]. How-
ever, this non-locality should not be surprising because as we have already
pointed out, the star-product is non-local. The fact that non-locality is an
essential feature of the description should again not be surprising. Indeed,
the phase space must be non-local otherwise we would be in violation of
the uncertainty principle. That the star-product must be a non-local prod-
uct has already been pointed out by Gracia–Bondia and Várilly [39, 40].
Indeed, further details of the mathematical structure lying behind some of
the results discussed in this paper will be found in these papers.
9. Conclusion
The aim of this chapter has been to show that the algebraic structure of
the quantum operators defined by von Neumann [8] and later developed
by Moyal [9] gives a more general mathematical structure in which the
usual Schrödinger representation with its wave-function provide but a par-
tial mathematical account of quantum phenomena. Elsewhere [41], we have
22 B. J. Hiley
shown that the information contained in the wave-function can be encoded

in the algebra in terms of certain ideals already contained in the algebra
itself. Hence, there is no fundamental need to postulate an external Hilbert
space, and this is in accord with the principle outlined in the above quota-
tion taken from Emch [1], namely, that the primary objects of the quantum
formalism should be purely algebraic quantities.
The geometries underlying these structures are non-commutative in gen-
eral and by concentrating on a non-commutative symplectic geometry, we
have shown that the quantum dynamics can be described either by the
elements of an abstract algebra or by functions on a generalized phase
space. The multiplication rule for combining these functions is necessar-
ily the non-commutative star-product introduced by von Neumann [8] and
Moyal [9].
Moyal’s contribution was to show how the algebra generalized classi-
cal statistics to a non-commutative statistics that emerges from a more
general non-commutative probability theory [30]. By recognizing this gen-
eralization, we have shown that the Wigner function emerges from a repre-
sentation of the kernel of the density matrix. We argue that it is therefore
incorrect to regard this kernel as a probability density. Furthermore, this
fact explains why the negative values of the Wigner function present no
difficulty.
Within this theory, we can introduce conditional expectation values from
which Bohmian mechanics emerges under the assumption that space–time
is basic. But one has an x ↔ p symmetry in the algebra so that it is possible
to define an alternative “mechanics” taking the momentum space as basic.
Thus, the Bohm approach does contain the (x, p) symmetry that Heisenberg
claimed it lacked [42]. Moreover, this symmetry produces shadow phase
spaces as used in Ref. [20]. In Bohm’s implicate order, these are what he
calls explicate orders. We have also shown how these shadows manifold
merge into a single commutative phase space in the classical limit.
We noted that the star-product is a non-local product, as does [39]. Fur-
thermore, we have shown that the kernel of the density matrix describes a
cell-like structure, rather than a point particle in phase space. Again, this
suggests that the quantum particle is represented by a region of the under-
lying non-commutative symplectic space, so that the quantum formalism
is basically non-local in a radically new way even for the single particle,
locality arising only at the classical limit.
www.ebook3000.com
Acknowledgments
I would like to thank Robert Callaghan, Maurice de Gosson, Glen Dennis,
and David Robson for their invaluable and enthusiastic discussions.
References
1. G.G. Emch, Algebraic Methods in Statistical Mechanics and Quantum Field
Theory, Wiley-Interscience, New York, 1972.
2. M. Born and P. Jordan, Zur Quantenmechanik, Z. Phys. 34, 858–888 (1925).
English translation B.L. van der Waerden, Sources of Quantum Mechanics,
North Holland, Amsterdam, 1976, pp. 277–306.
3. M. Born, W. Heisenberg and P. Jordan, Zur Quantenmechanik II, Z. Phys.
35, 557–615 (1926). English translation B.L. van der Waerden, Sources of
Quantum Mechanics, North Holland Press, Amsterdam, 1976, pp. 321–385.
4. P.A.M. Dirac, The Fundamental equations of quantum mechanics, Proc. Roy.
Soc. A 109, 642–653 (1926).
5. E. Schrödinger, Über das Verhältnis der Heisenberg–Born–Jordanschen
Quantenmechanik zu der meinen, Ann. Physik. 384, 734–756 (1926). English
translation in G. Ludwig, Wave Mechanics, Pergamon Press, 1968, pp. 94–
105.
6. L.D. Landau and E.M. Lifshitz, Quantum Mechanics (Non-relativistic The-
ory), Pergamon Press, Oxford, 1976, p. 1.
7. L. Auslander, F. Geshwind and F. Warner, Radar Waveform Design and the
Heisenberg Group, App. Comput. Harmon. Anal. 3, 350–362 (1995).
8. J. von Neumann, Die Eindeutigkeit der Schrödingerschen Operatoren, Math.
Ann. 104, 570–587 (1931).
9. J.E. Moyal, Quantum mechanics as a statistical theory, Proc. Camb. Phil.
Soc. 45, (1949), 99–123 (1972).
10. R.P. Feynman, Negative probability, in Quantum Implications: Essays in
Honour of David Bohm (eds.) B.J. Hiley and F.D. Peat, pp. 235–248, Rout-
ledge and Kegan Paul, London, 1987.
11. M.A. de Gosson, Quantum Blobs, Found. Phys. 43, 440–457 (2013).
12. B.J. Hiley, Phase space description of quantum phenomena, in quantum the-
ory, Proc. Int. Conf. Quantum Theory: Reconsideration Found. 2, 267–286
(2003).
13. E.P. Wigner, On the quantum correction for thermodynamic equilibrium,
Phys. Rev. 40, 749–759 (1932).
14. A.C. Hirshfeld and P. Henselder, Deformation quantisation in the teaching
of quantum mechanics, Am. J. Phys. 70, 537–547 (2002).
15. G.A. Baker, Jn., Formulation of quantum mechanics based on the quasi-
probability distribution induced on phase space, Phys. Rev. 109, 2198–2206
(1958).
24 B. J. Hiley
16. C. Zachos, (2002), Deformation quantization: Quantum mechanics lives and

works in phase space, Int. J. Mod. Phys. A 17, 297–316 (2002).
17. W.P. Schleich, Quantum optics in phase space. John Wiley & Sons, 2011.
18. D. Bohm and B.J. Hiley, The Undivided Universe: An Ontological Interpre-
tation of Quantum Mechanics, Routledge, London, 1993.
19. M. Brown and B.J. Hiley, Schrödinger revisited: An algebraic approach,
(2000), quant-ph/0005026.
20. D. Leibfried, T. Pfau and C. Monroe, Shadows and mirrors: Reconstructing
quantum states of atom motion, Phys. Today 51, 22–28 (1998).
21. B.J. Hiley, Phase Space Description of Quantum Phenomena, in Quantum
Theory: Reconsiderations of Foundations-2, A. Khrennikov (ed.), Växjö Uni-
versity Press, Växjö, Sweden, 2004 pp. 267–286 2010.
22. D. Bohm, Wholeness and the Implicate Order, Routledge, London, 1980.
23. P.A.M. Dirac, On the annihilation of electrons and protons, Proc. Camb.
Phil. Soc. 26, 361–375 (1930).
24. B.J. Hiley, A Note on the role of idempotents in the extended heisenberg
algebra, Proc. Int. Meeting of ANPA at Cambridge 22, 107–121 (2001).
25. B.J. Hiley and R.E. Callaghan, The Clifford algebra approach to quantum
mechanics A: The Schrödinger and Pauli particles, (2010), arXiv: 1011.4031.
26. N. Bohr, Atomic Physics and Human Knowledge, Science Editions, New
York, 1961.
27. A. Crumeyrolle, Orthogonal and Symplectic Clifford Algebras: Spinor Struc-
tures, Kluwer, Dordrecht, 1990.
28. B.J. Hiley and R.E. Callaghan, Clifford Algebras and the Dirac-Bohm Quan-
tum Hamilton–Jacobi Equation. Found. Phys. 42, 192–208 (2012).
29. B.J. Hiley and R.E. Callaghan, The Clifford Algebra Approach to Quantum
Mechanics B: The Dirac Particle and its relation to the Bohm Approach,
(2010), arXiv: 1011.4033.
30. D.E. Evans and Y. Kawahigashi, Quantum Symmetries on Operator Algebras,
Oxford Mathematical Monographs, Clarendon Press, Oxford, 1998.
31. I.E. Segal, A non-commutative extension of abstract integration, Ann. Math.
57, 401–457 (1953).
32. M.S. Bartlett, Negative probability, Math. Proc. Cam. Phil. Soc. 41, 71–73
(1945).
33. R.P. Feynman, Simulating physics with computers, Int. J. Theor. Phys. 21,
467–488 (1982).
34. H.F. Hofmann, Complex joint probabilities as expressions of reversible trans-
formations in quantum mechanics, New J. Phys. 14, 043031 (2012).
35. H. Umegaki, Conditional expectation in an operator algebra, Tohoku Math
J. 6, 358–362 (1954).
36. V.F.R. Jones, Index for sub-factors, Inven. Math. 72, 1–25 (1983).
37. M. de Gosson, Symplectic Geometry and Quantum Mechanics, Birkhäuser
Verlag, Basel, 2006.
38. J.F. Cariñena, J. Clemente-Gallardo, E. Follana, J.M. Gracia-Bondı́a, A.
Rivero and J.C. Váilly, Connes tangent groupoid and strict quantization, J.
Geom. Phys. 32, (1999) 79–96.
www.ebook3000.com
39. J.M. Gracia–Bondia and J.C. Várilly, Algebras of distributions suitable for
phase-space quantum mechanics, I, J. Math. Phys. 29, 869–879 (1988).
40. J.M. Gracia-Bondia and J.C. Várilly, Algebras of distributions suitable for
phase-space quantum mechanics, II, Topologies on the Moyal algebra, J.
Math. Phys. 29, 880–887 (1988).
41. B.J. Hiley, Process, distinction, groupoids and clifford algebras: An alterna-
tive view of the quantum formalism, in New Structures for Physics, ed. B.
Coecke, Lecture Notes in Physics, Vol. 813, Springer, 2011, pp. 705–750.
42. W. Heisenberg, Physics and Philosophy: The Revolution in Modern Science,
George Allen and Unwin, London, 1958, p. 118.
www.ebook3000.com
Chapter 2
Fermi Blobs and the Symplectic

Camel: A Geometric Picture
of Quantum States
Maurice A. de Gossona
University of Vienna,
Faculty of Mathematics, NuHAG
A-1090 Vienna, Austria
We have explained in previous work the correspondence between the

standard squeezed coherent states of quantum mechanics, and quantum
blobs, which are the smallest phase space units compatible with the
uncertainty principle of quantum mechanics and having the symplectic
group as a group of symmetries. In this work, we discuss the relation
between quantum blobs and a certain level set (which we call “Fermi
blob”) introduced by Enrico Fermi in 1930. Fermi blobs allows us to
extend our previous results not only to the excited states of the general-
ized harmonic oscillator in n dimensions, but also to arbitrary quadratic
Hamiltonians. As is the case for quantum blobs, we can evaluate Fermi
blobs using a topological notion, related to the uncertainty principle, the
symplectic capacity of a phase space set. The definition of this notion
is made possible by Gromov’s symplectic non-squeezing theorem, nick-
named the “principle of the symplectic camel”.
1. What We will Do, and How We will Do it

In this chapter, we introduce a non-trivial extension of the notion of “quan-
tum blob” we defined and studied in Refs. [1–7]. Quantum blobs are defor-
mations of the phase space ball |x|2 + |p|2 ≤ by linear (or affine) canonical
a This work has been supported by the Austrian Research Agency FWF (Project Number
P20442-N13).
27
28 M. A. de Gosson
transformations. They provide us with a coarse-graining of phase space dif-

ferent from the usual coarse graining by cubes with volume ∼ hn commonly
used in statistical mechanics. They can thus be viewed as space units of min-
imum uncertainty in one-to-one correspondence with the squeezed coherent
states of quantum mechanics, and have allowed to recover the exact ground
states of generalized harmonic oscillators, as well as the semi-classical
energy levels of quantum systems with completely integrable Hamiltonian
function, and to explain them in terms of the topological notion of symplec-
tic capacity [8,9] originating in Gromov’s [10] non-squeezing theorem (alias
“the principle of the symplectic camel”). Quantum blobs, do not, however,
allow a characterization of excited states; for instance, there is no obvious
relation between them and the Hermite functions. Why this does not work
is easy to understand: quantum blobs correspond to the states saturating
the Schrödinger–Robertson inequalities
(∆Xj )2 (∆Pj )2 ≥ ∆(Xj , Pj )2 + 14 2 , 1 ≤ j ≤ n; (1)
as is well known, the quantum states for which all these inequalities become
equalities are Gaussians, in this case, precisely the SCS who are them-
selves the ground states of generalized harmonic oscillators. As soon as one
considers the excited states, the corresponding eigenfunctions are Hermite
functions and for these, the inequalities (1) are strict. The way out is to
define new phase space objects, the “Fermi blobs” of the title of this paper.
We will show that a complete geometric picture of excited states can be
given using a gimmick introduced by the physicist Enrico Fermi in a largely
forgotten paper [11] from 1930. Fermi associates to every quantum state Ψ
a certain hypersurface gF (x, p) = 0 in phase space. The underlying idea
is actually surprisingly simple. It consists in observing that any complex
twice continuously differentiable function Ψ(x) = R(x)eiΦ(x)/ (R(x) ≥ 0
and Φ(x) real) defined on Rn satisfies the partial differential equation
2

2 2 ∇x R
(−i∇x − ∇x Φ) + Ψ = 0, (2)
R
where ∇2x is the Laplace operator in the variables x1 , . . . , xn (it is assumed

that R(x) = 0 for x in some subset of Rn ). Performing the gauge transfor-
mation −i∇x −→ −i∇x − ∇x Φ, this equation is in fact equivalent to the
trivial equation,
2

2 2 2 ∇x R
− ∇x + R = 0. (3)
R
www.ebook3000.com
Fermi Blobs and the Symplectic Camel 29
The operator,
2 2 ∇2x R
F = (−i∇x − ∇x Φ) +
g , (4)
R
appearing in the left-hand side of Eq. (2) is the quantization (in every
reasonable physical quantization scheme) of the real observable
2 ∇2x R
gF (x, p) = (p − ∇x Φ) + 2 (5)
R
and the equation gF (x, p) = 0 in general determines a hypersurface HF in
phase space R2n
x,p which Fermi ultimately identifies with the state Ψ itself.
The remarkable thing with this construction is that it shows that to an
arbitrary function Ψ it associates a Hamiltonian function of the classical
type,
1 2
H= (p − ∇x Φ) + V, (6)
2
even if Ψ is the solution of another partial (or pseudo-differential) equation.
We note that when Ψ is an eigenstate of the operator HΨ = EΨ, then
gF = 2(H − E) and HF is just the energy hypersurface H(x, p) = E.
Of course, Fermi’s analysis was very heuristic and its mathematical rigor
borders the unacceptable (at least by modern standards). Fermi’s paper has
recently been rediscovered by Benenti [12] and Benenti and Strini [13], who
study its relationship with the level sets of the Wigner transform of Ψ.
It turns out that the hypersurface gF (x, p) = 0 has a surprising prop-
erty which connects it with the notion of quantum potential familiar from
Bohm’s theory of quantum motion (Dennis et al. [14]). For simplicity,
assume that n = 1; the equation gF (x, p) = 0 is then represented by a
curve consisting of two branches γF± described by the equations,

p± = ϕ (x)± 2mQ(x) (7)
(ϕ the first x-derivative of the phase) where the “energy” Q(x) is Bohm’s
quantum potential [15–18]:
2 R (x)
Q(x) = − . (8)
2m R(x)
As we will see, when γF is a closed curve, the area of the surface ΩF it
encloses is never less than 12 h, i.e. one half of the quantum of action. Else-
where [19], we have called such surfaces (and their generalizations to higher
dimensions) “quantum blobs”, and shown that they are closely related to
30 M. A. de Gosson
the uncertainty principle. One plausible interpretation is that the quantum

force (derived from the quantum potential) prevents the “quantum blob”
ΩF from collapsing to a particle-like point. This interpretation is consis-
tent with Hiley’s observation [20] that in “. . .a process based approach, the
notion of a ‘particle’ is not a sharply defined point object, instead it is a
quasi-local invariant feature of the total process. Each individual process
may be characterized by a mean position and a mean momentum. . ..” In
fact, Eq. (7) has the following precise meaning in terms of the quantum
theory of motion. For instance, the positive branch γF+ is described by the
equation,

p+ = pB (x)+ 2mQ(x),

where pB (x) is Bohm’s momentum and the term 2mQ(x) can be viewed
as an additional momentum coming from the quantum force.
Notation 1. The points in configuration and momentum space are written

x = (x1 , . . . , xn ) and p = (p1 , . . . , pn ), respectively; in formulas x and p are
viewed as column vectors. We will also use thecollective notation z = (x, p)
0 I
for the phase space variable. The matrix J = (0 and I the n × n
−I 0
zero and identity matrices) defines the standard symplectic form on the
phase space R2n
x via the formula σ(z, z ) = Jz · z = p · x − p · x. We write
= h/2π, h being Planck’s constant. The symplectic group is denoted by
Sp(2n, R): it is the multiplicative group of all real 2n × 2n matrices S such
that σ(Sz, Sz ) = σ(z, z ) for all z, z .
2. The Quantum Oscillator

As an appetizer, we begin by considering the fiducial coherent state,
2
Ψ0 (x) = (π)−1/4 e−|x| /2
. (9)
It is the ground state of the one-dimensional isotropic harmonic oscillator

with mass and frequency equal to one:
2 2
1
2 (− ∇x + |x|2 )Ψ0 = 12 Ψ0 . (10)
The Fermi operator (2) is here:

2 2 2
F = − ∇x + |x| −
g (11)
www.ebook3000.com
and the relation g

F Ψ0 = 0 is hence equivalent to Eq. (10). The Fermi
function is
gF (x, p) = p2 + x2 − (12)
and the Fermi set ΩF is thus here the disk x2 + p2 ≤ whose area is
π = h/2.
Consider next the N th eigenstate ΨN . We have
2

1 2 d 2 1
− + x ΨN = N + ΨN
2 dx2 2
and the eigenfunction ΨN is the (unnormalized) Hermite function,

2 √
ΨN (x) = e−x /2 HN (x/ ), (13)
where
2 dN −x2
HN (x) = (−1)n ex e (14)
dxN
is the N th Hermite polynomial. Since ΨN is real, the corresponding Fermi
function is in this case
gF (x, p) = p2 + x2 − (2N + 1) (15)
and the Fermi set ΩF is the circle
p2 + x2 = (2N + 1), (16)
whose area is (2N + 1)π = (N + 12 )h. We thus see that the Fermi sets
allow a classification
√ of√ the √
states of the quantum oscillator by concentric
circles with radii , 3, 5 . . . .
In the case of an arbitrary number n of degrees of freedom (DOF), the
eigenstate ΨN is the tensor product of n Hermite functions (13) and one
finds that
gF (x, p) = |p|2 + |x|2 − (2N + 1),
hence the Fermi set is this time the ball,
|p|2 + |x|2 = (2N + 1), (17)
which has symplectic capacity (N + 12 )h. We will see that the situation is
less simple in the case of many DOFs, and that we will have to use a more
32 M. A. de Gosson
complicated notion of symplectic capacity to describe the Fermi ellipsoid.

In fact, consider the ground state of the n-dimensional oscillator,
2 2
1
2 (− ∇x + |x|2 )Ψ0 = 12 Ψ0 . (18)
It is the function,
2
Ψ0 (x) = (π)−n/4 e−|x| /2
(19)
with |x|2 = x · x.
3. Squeezed States
3.1. The Fermi function of a Gaussian
We next consider arbitrary (normalized) squeezed coherent states,
n/4
1 1
ΨX,Y (x) = (det X)1/4 exp − (X + iY )x · x , (20)
π 2
where X and Y are real symmetric n×n matrices, and X is positive definite.
Setting Φ(x) = − 21 Y x · x and R(x) = exp(− 2
1
Xx · x), we have
∇2x R(x) 1 1
∇x Φ(x) = −Y x, = − Tr X + 2 X 2 x · x, (21)
R(x)
hence the Fermi function of ΨX,Y is the quadratic form,
gF (x, p) = (p + Y x)2 + X 2 x · x − Tr X. (22)
We can rewrite this formula as

x
gF (x, p) = (x , p)MF − Tr X,
p
where MF is the symmetric matrix,
2
X +Y2 Y
MF = . (23)
Y I
A straightforward calculation shows that

T X 0
MF = S S, (24)
0 X
where S is the symplectic matrix,

X 1/2 0
S= −1/2 . (25)
X −1/2 Y X
www.ebook3000.com
It turns out — and this is really a striking fact — that the matrix (24)
is closely related to the Wigner transform,
n

1 i
W ΨX,Y (z) = e− p·y ΨX,Y x + 12 y Ψ∗X,Y x − 12 y dy (26)
2π
Rn
of the state ΨX,Y because we have

n
1 1
W ΨX,Y (z) = exp − Gz · z , (27)
π
where G is the symplectic matrix,

T X + Y X −1 Y Y X −1
G=S S= , (28)
X −1 Y X −1
(see e.g. [5, 21]). It follows from Eq. (24) that

n
1 1
W ΨX,Y (z) = e− Tr X exp − gF (S −1 D−1/2 Sz) (29)
π

X 0
with D = . In particular, when n = 1 and ΨX,Y (x) = Ψ0 (x), the
0 X
fiducial coherent state (9) we have S −1 D−1/2 S = I and Tr X = 1, hence
the formula,
1/4
1 1 1
W Ψ0 (z) = exp − MF z · z ,
π e
which was already observed by Benenti and Strini [13].
3.2. Geometric interpretation

Recall (formula (A.4) in the Appendix) that the symplectic capacity c(Ω)
of an ellipsoid M z · z ≤ 1 (M a symmetric positive definite 2n × 2n matrix)
is given by
c(Ω) = π/λσmax , (30)
where λσmax = max{λσ1 , . . . , λσn }, the λσj being the symplectic eigenvalues
of M . We denote by ΩF the ellipsoid MF z ·z ≤ Tr X bounded by the Fermi
hypersurface HF corresponding to the squeezed coherent state ΨX,Y . Let us
34 M. A. de Gosson
perform the symplectic change of variables z = Sz; in the new coordinates,

the ellipsoid ΩF is represented by the inequality,
Xx · x + Xp · p ≤ Tr X, (31)
hence c(ΩF ) equals the symplectic capacity of the ellipsoid (31). Applying
the rule above, wethus have
to find the symplectic eigenvalues of the block-
X 0
diagonal matrix ; a straightforward calculation shows that these
0 X
are just the eigenvalues ω1 , . . . , ωn of X and hence
c(ΩF ) = π Tr X/ωmax, (32)
where ωmax = max{ω1 , . . . , ωn }. In view of the trivial inequality,

n

ωmax ≤ Tr X = ωj ≤ nλωmax , (33)
j=1
it follows that we have

1 nh
h ≤ c(ΩF ) ≤ . (34)
2 2
A consequence of the inequality 12 h ≤ c(ΩF ) is that the Fermi ellipsoid
ΩF of a squeezed coherent state always contains a “quantum blob”.
Note that when all the eigenvalues ωj are equal to a number ω, then
c(ΩF ) = nh/2; in particular, when n = 1, we have c(ΩF ) = h/2 which
is exactly the action calculated along the trajectory corresponding to the
ground state. This observation leads us to the following question: what is
the precise geometric meaning of formula (32)? Let us come back to the
interpretation of the ellipsoid defined by the inequality
(31). We have seen
X 0
that the symplectic eigenvalues of the matrix are precisely the
0 X
eigenvalues ωj , 1 ≤ j ≤ n, of the positive-definite matrix X. It follows
that there exist symplectic coordinates (x , p ) in which the equation of
the ellipsoid ΩF takes the normal form,
n
n

ωj (x2 2
j + pj ) ≤ ωj , (35)
j=1 j=1
whose quantum-mechanical interpretation is clear: Dividing both sides by

two, we get the energy shell of the anisotropic harmonic oscillator in its
www.ebook3000.com
ground state. Consider now the planes P1 , P2 , . . . , Pn of conjugate coor-

dinates (x1 , p1 ), (x2 , p2 ), . . . , (xn , pn ). The intersection of the ellipsoid ΩF
with these planes are the circles,
n

C1 : ω1 (x2 2
1 + p1 ) ≤ ωj ,
j=1
n
C2 : ω2 (x2 2
2 + p2 ) ≤ ωj ,
j=1
·················
n

Cn : ωn (x2 2
n + pn ) ≤ ωj .
j=1
Formula (32) says that c(ΩF ) is precisely the area of the circle Cj with the
smallest area which corresponds to the index j such that ωj = ωmax . This is
of course perfectly in accordance with the definition of the Hofer–Zehnder
capacity cHZ (ΩF ) since all symplectic capacities agree on ellipsoids. This
leads us now to another question: Is there any way to describe topologically
Fermi’s ellipsoid in such a way that the areas of every circle Cj becomes
apparent? The problem with the standard capacity of an ellipsoid is that
it only “sees” the smallest cut of that ellipsoid by a plane of conjugate
coordinate. The way out of this difficulty lies in the use of the Ekeland–
Hofer capacities cEH
j described in Appendix. To illustrate the idea, let us
first consider the case n = 2; it is no restriction to assume ω1 ≤ ω2 . If
ω1 = ω2 , then the ellipsoid,
ω1 (x2 2 2 2
1 + p1 ) + ω2 (x2 + p2 ) ≤ ω1 + ω2 (36)
√
is the ball B 2 ( 2) whose symplectic capacity is 2π = h. Suppose now
ω1 < ω2 . Then, the Ekeland–Hofer capacities are the numbers,
π π 2π 2π
(ω1 + ω2 ), (ω1 + ω2 ), (ω1 + ω2 ), (ω1 + ω2 ) . . . (37)
ω2 ω1 ω2 ω1
and hence
π
cEH
1 (ΩF ) = c(ΩF ) = (ω1 + ω2 ).
ω2
What about cEH
2 (ΩF )? A first glance at the sequence (37) suggests that we
have
π
cEH
2 (ΩF ) = (ω1 + ω2 ),
ω1
36 M. A. de Gosson
but this is only true if ω1 < ω2 ≤ 2ω1 because if 2ω1 < ω2 , then (ω1 +
ω2 )/ω2 < (ω1 + ω2 )/ω1 so that in this case,
π
cEH
2 (ΩF ) =(ω1 + ω2 ) = cEH
1 (ΩF ).
ω2
The Ekeland–Hofer capacities thus allow a classification of the eigenstates.
4. The Fermi Function of Excited States

The squeezed coherent states can be viewed as the ground states of a
generalized harmonic oscillator, with Hamiltonian function a homogeneous
quadratic polynomial in the position and momentum coordinates:

H(x, p) = aij pi pj + bij pi xj + cij xi xj .
i,j
Such a function can always be put in the form,

1
H(z) = M z · z, (38)
2
where M is a symmetric matrix (the Hessian matrix, i.e. the matrix of
second derivatives, of H). We will assume for simplicity that M is positive-
definite; we can then always bring it into the normal form,
n
ωj
K(z) = (x2j + p2j ),
j=1
2
using a linear symplectic transformation of the coordinates (symplectic

diagonalization): There exists a symplectic matrix S (depending on M )
such that

T Λ 0
S MS = D = , (39)
0 Λ
where Ω is a diagonal matrix whose diagonal entries consist of the symplec-
tic spectrum ω1 , . . . , ωn of M . Thus, we have K(z) = H(Sz), or, equiva-
lently,
H(z) = K(S −1 z). (40)
The ground state of each one-dimensional quantum oscillator,
2

K j = ωj x2j − 2 ∂
2 ∂xj
is the solution of K j Ψ = 1 ωj Ψ, it is thus the one-dimensional fidu-
2
2
cial coherent state (π)−1/4 e−x /2 . It follows that the ground Ψ0 state
www.ebook3000.com

=
of K
j Kj is the tensor product of n such states, that is Ψ0 (x) =
−n/4 −|x|2 /2
(π) e , the fiducial coherent state (9). Returning to the ini-
tial Hamiltonian H, we note that the corresponding Weyl quantization
H satisfies, in view of Eq. (40), the symplectic covariance formula H =
S−1 where S is any of the two metaplectic operators corresponding to
SK
the symplectic matrix S. It follows that the ground state of H is given by

the formula, Ψ = SΨ0 .
The case of the excited states is treated similarly. The solutions of the
one-dimensional eigenfunction problem K j Ψ = EΨ are given by the Her-
mite functions,
2 √
ΨN (x) = e−x /2 HN (x/ ) (41)
with corresponding eigenvalues EN = (N + 12 )ωj (see Sec. 2). It follows

= EΨ are the tensor
that the solutions of the n-dimensional problem KΨ
products,
Ψ(N ) = ΨN1 ⊗ ΨN2 ⊗ · · · ⊗ ΨNn , (42)
where (N ) = (N1 , N2 , . . . , Nn ) is a sequence of non-negative integers, and

the corresponding eigenvalue is
n

E(N ) = Nj + 12 ωj . (43)

j=1
This allows us to give a geometric description of all eigenfunctions of the

generalized harmonic oscillator, corresponding to a classical Hamiltonian
(38). We have
Let Ψ be an eigenfunction of the quantum operator,

= (x, −i∇x )M (x, −i∇x )T .
H (44)
The symplectic capacity of the corresponding Fermi blob

ΩF is
n

1
c(ΩF ) = Nj + 2 h, (45)
j=1
where the numbers N1 , N2 , . . . , Nn are the non-negative

integers corresponding to the state (42) of the diagonalized
operator K = n K j .
j=1
38 M. A. de Gosson
This is easily proven using the previous results, after having put the
Hamiltonian H in normal form using a symplectic diagonalization of the
matrix M (“Williamson diagonalization”, see Ref. [5] for a review, and
applications).
Appendix A: Symplectic Capacities and Quantum

Blobs
For a review of these notions from a physical point of view, see de Gosson
and Luef [22].
A1. Symplectic Capacities

A1.1 Intrinsic symplectic capacities
An intrinsic symplectic capacity assigns a non-negative number (or +∞)
c(Ω) to every subset Ω of phase space R2n ; this assignment is subjected to
the following properties:
• Monotonicity: If Ω ⊂ Ω , then c(Ω) ≤ c(Ω );
• Symplectic invariance: If f is a canonical transformation (linear, or
not), then c(f (Ω)) = c(Ω);
• Conformality: If λ is a real number, then c(λΩ) = λ2 c(Ω); here, λΩ is
the set of all points λz when z ∈ Ω;
• Normalization: We have
c(B 2n (R)) = πR2 = c(Zj2n (R)); (A.1)
here, B 2n (R) is the phase space ball |x|2 + |p|2 ≤ R2 and Zj2n (R) the
phase space cylinder x2j + p2j ≤ R2 .
Let c be a symplectic capacity on the phase plane R2 . We have
c(Ω) = Area(Ω) when Ω is a connected and simply connected surface.
In the general case, there exist infinitely many intrinsic symplectic capac-
ities, but they all agree on phase space ellipsoids as we will see below.
The smallest symplectic capacity is denoted by cmin (“Gromov width”): By
definition, cmin (Ω) is the supremum of all numbers πR2 such that there
exists a canonical transformation such that f (B 2n (R)) ⊂ Ω. The fact that
cmin really is a symplectic capacity follows from a deep and difficult topo-
logical result, Gromov’s [10] symplectic non-squeezing theorem, alias the
principle of the symplectic camel. (For a discussion of Gromov’s theorem
from the point of view of Physics, see de Gosson [6], de Gosson and Luef
www.ebook3000.com
[22].) Another useful example is provided by the Hofer–Zehnder [8] capacity

cHZ . It has the property that it is given by the integral of the action form
pdx = p1 dx1 + · · · + pn dxn along a certain curve:

HZ
c (Ω) = pdx, (A.2)
γmin
when Ω is a compact convex set in phase space; here, γmin is the shortest
(positively oriented) Hamiltonian periodic orbit carried by the boundary
∂Ω of Ω. This formula agrees with the usual notion of area in the case
n = 1.
It turns out that all intrinsic symplectic capacities agree on phase space
ellipsoids, and are calculated as follows (see e.g. [5, 8, 22]). Let M be a
2n × 2n positive-definite matrix M and consider the ellipsoid:
ΩM,z0 : M (z − z0 )2 ≤ 1. (A.3)
Then, for every intrinsic symplectic capacity c, we have
c(ΩM,z0 ) = π/λσmax , (A.4)
where λσmax = is the largest symplectic eigenvalue of M . The symplectic
eigenvalues of a positive definite matrix are defined as follows: The matrix
JM (J the standard symplectic matrix) is equivalent to the antisymmet-
ric matrix M 1/2 JM 1/2 , hence its 2n eigenvalues are of the type ±iλσ1 , · · · ,
±iλσn where λσj > 0. The positive numbers λσ1 , . . . , λσn are called the sym-
plectic eigenvalues of the matrix M .
In particular, if X and Y are real symmetric n × n matrices, then the
symplectic capacity of the ellipsoid,
Ω(A,B) : Xx2 + Y p2 ≤ 1 (A.5)
is given by

c(Ω(A,B) ) = π/ λmax , (A.6)
where λmax is the largest eigenvalue of AB.
A1.2. Extrinsic symplectic capacities

The definition of an extrinsic symplectic capacity is similar to that of an
intrinsic capacity, but one weakens the normalization condition (A.1) by
only requiring:
• Non-triviality: c(B 2n (R)) < +∞ and c(Zj2n (R)) < +∞.

40 M. A. de Gosson
In Ref. [23], Ekeland and Hofer defined a sequence cEH EH EH

1 , c2 , . . . , ck , ...
of extrinsic symplectic capacities having the non-triviality properties,

k+n−1
cEH
k (B 2n
(R)) = πR2 , cEH 2n
k (Zj (R)) = kπR .
2
(A.7)
n
Of course, cEH
1 is an intrinsic capacity; in fact, it coincides with the Hofer–
Zehnder capacity on bounded and convex sets Ω. We have
cEH EH EH
1 (Ω) ≤ c2 (Ω) ≤ · · · ≤ ck (Ω) ≤ · · · . (A.8)
The Ekeland–Hofer capacities have the property that for each k there exists
an integer N ≥ 0 and a closed characteristic γ of ∂Ω such that

ck (Ω) = N pdx
EH (A.9)
γ
(in other words, cEH k (Ω) is a value of the action spectrum [24] of ∂Ω); this
formula shows that cEH k (Ω) is solely determined by the boundary of Ω;
therefore, the notation cEH k (∂Ω) is sometimes used in the literature. The
Ekeland–Hofer capacities cEH k allow us to classify phase space ellipsoids. In
fact, the non-decreasing sequence of numbers cEH k (ΩM ) is determined as
follows for an ellipsoid Ω : M z · z ≤ 1 (M symmetric and positive-definite):
Let (λσ1 , . . . , λσn ) be the symplectic eigenvalues of M ; then,
{cEH σ
k (Ω) : k = 1, 2, . . .} = {N πλj : j = 1, . . . , n; N = 0, 1, 2, . . .}. (A.10)
Equivalently, the increasing sequence cEH EH

1 (Ω) ≤ c2 (Ω) ≤ · · · is obtained by
writing the numbers N πλσj in increasing order with repetitions if a number
occurs more than once.
A2. Quantum Blobs

2n
By definition,
2n
√a quantum blob √ QB (z0 , S) is the image of the phase space
ball B (z0 , ) : |z − z0 | ≤ by a linear canonical transformation (iden-
tified with a symplectic matrix S). A quantum blob is thus a phase space
ellipsoid with symplectic capacity π = 12 h, but it is not true that, con-
versely, an arbitrary phase space ellipsoid with symplectic capacity 12 h is
a quantum blob. One can however show (de Gosson [3–5], de Gosson and
Luef [22]) that such an ellipsoid contains a unique quantum blob. One
proves (ibid. 9) that a quantum blob QB2n (z0 , S) is characterized by the
two following equivalent properties:
www.ebook3000.com
√
• The supremum of the set of all numbers πR2 such that the ball B 2n ( R) :
|z| ≤ R can be embedded into QB2n (z0 , S) using canonical transfor-
1
√ (linear, or not) is 2 h. Hence, 2n
mations no phase space ball with radius
R > can be “squeezed” inside QB (z0 , S) using only canonical
transformations.
It turns out (de Gosson [5]) that in the first of these conditions, one
can replace the plane of conjugate coordinates with any symplectic plane
(a symplectic plane is a 2D subspace of R2n
z on which the restriction of the
symplectic form σ is again a symplectic form). There is a natural action,
Sp(2n, R) × QB(2n, R) −→ QB(2n, R)
of the symplectic group on quantum blobs.
Appendix B: Metaplectic Group

The symplectic group Sp(2n, R) has a covering group of order two, the
metaplectic group Mp(2n, R). That group consists of unitary operators (the
metaplectic operators) acting on L2 (Rn ). There are several equivalent ways
to describe the metaplectic operators. For our purposes, the most tractable
is the following: assume that S ∈ Sp(2n, R) has the block-matrix form

A B
S= with det B = 0. (B.1)
C D
The condition det B = 0 is not very restrictive because one shows
(de Gosson [5, 25]) that every S ∈ Sp(2n, R) can be written (non-uniquely)
as the product of two symplectic matrices of the type above; moreover, the
symplectic matrices arising as Jacobian matrices of Hamiltonian flows deter-
mined by physical Hamiltonians of the type “kinetic energy plus potential”
are of this type for almost every time t. To the matrix (B.1), we associate
the following quantities (de Gosson [5, 25]):
• A quadratic form,
1 1
W (x, x ) = DB −1 x · x − B −1 x · x + B −1 Ax · x ; (B.2)
2 2
the matrices DB −1 and B −1 A aresymmetric because S is symplectic;
• The complex number ∆(W ) = im | det B −1 | where m (“Maslov index”)
is chosen in the following way: m = 0 or 2 if det B −1 > 0 and m = 1 or
3 if det B −1 < 0.
42 M. A. de Gosson
The two metaplectic operators associated to S are then given by

n/2

SΨ(x) = 2πi 1 i
∆(W ) e W (x,x ) Ψ(x )dn x . (B.3)
The fact that we have two possible choices for the Maslov index is directly
related to the fact that Mp(2n, R) is a two-fold covering group of the sym-
plectic group Sp(2n, R).
The main interest of the metaplectic group in quantization questions
comes from the two following (related) “symplectic covariance” properties:
• Let Ψ be a square integrable function (or, more generally, a tempered
distribution), and S a symplectic matrix. We have

W Ψ(S −1 z) = W (SΨ)(z), (B.4)
where S is any of the two metaplectic operators corresponding to S;
be the Weyl quantization of the symbol (= observable) H. Let
• Let H
S be a symplectic matrix. Then, the quantization of K(z) = H(Sz) is
= S−1 H
K S where S is again defined as above.
References
1. M. de Gosson, The “symplectic camel principle” and semiclassical mechanics.
J. Phys. A: Math. Gen. 35(32), 6825–6851 (2002).
2. M. de Gosson, Phase space quantization and the uncertainty principle. Phys.
Lett. A 317(5–6), 365–369 (2003).
3. M. de Gosson, The optimal pure Gaussian state canonically associated to a
Gaussian quantum state. Phys. Lett. A 330(3–4), 161–167 (2004).
4. M. de Gosson, Cellules quantiques symplectiques et fonctions de Husimi–
Wigner. Bull. Sci. Math. 129, 211–226 (2005).
5. M. de Gosson, Symplectic Geometry and Quantum Mechanics, Birkhäuser,
Basel, series “Operator Theory: Advances and Applications” (subseries:
“Advances in Partial Differential Equations”), Vol. 166 (2006).
6. M. de Gosson, The symplectic camel and the uncertainty principle: The tip
of an iceberg? Found. Phys. 99, 194–214 (2009).
7. M. de Gosson, On the use of minimum volume ellipsoids and symplectic
capacities for studying classical uncertainties for joint position–momentum
measurements, J. Stat. Mech. 11, P11005 (2010), doi: 10.1088/1742-
5468/2010/11/P11005.
8. H. Hofer and E. Zehnder, Symplectic invariants and Hamiltonian dynamics,
Birkhäuser Advanced Texts, (Birkhäuser Verlag, Swizerland) 1994.
9. L. Polterovich, The Geometry of the Group of Symplectic Diffeomorphisms,
Lectures in Mathematics, Birkhäuser, Switzerland, 2001.
10. M. Gromov, Pseudoholomorphic curves in symplectic manifolds, Invent.
Math. 82, 307–347 (1985).
www.ebook3000.com
11. E. Fermi, Rend. Lincei 11, 980 (1930); reprinted in Nuovo Cimento 7, 361
(1930).
12. G. Benenti, Gaussian wave packets in phase space: The Fermi gF function,
Am. J. Phys. 77(6), 546–551 (2009).
13. G. Benenti and G. Strini, Quantum mechanics in phase space: First order
comparison between the Wigner and the Fermi function, Eur. Phys. J. D 57,
117–121 (2010).
14. G. Dennis, M. de Gosson and B.J. Hiley, Fermi’s ansatz and Bohm’s quantum
potential, Phys. Lett. A 378(32), 2363–2366 (2014).
15. D. Bohm, Wholeness and the Implicate Order, Routledge, London, 1980.
16. D. Bohm, Hidden variables and the Implicate Order, in Quantum Implica-
tions: Essays in Honour of David Bohm, eds. Hiley B.J. and Peat F. David.
Routledge, London, 1987.
17. D.J. Bohm, B.J. Hiley. The de Broglie Pilot Wave Theory and the Further
Development of New Insights Arising Out of It, Found. Phys. 12(10), 1001–
1016 (1982).
18. P.R. Holland, The quantum theory of motion. An Account of the de Broglie-
Bohm Causal Interpretation of Quantum Mechanics. Cambridge University
Press, Cambridge, 1995.
19. M. de Gosson, Quantum Blobs. To appear in Found. Phys. 43(4), 440–477
(2013).
20. B.J. Hiley, Bohmian Non-commutative Dynamics: History and New Devel-
opments, arXiv:1303.6057v1 [quant-ph], 2013.
21. R.G. Littlejohn, The semiclassical evolution of wave packets, Phys. Rep.
138(4–5), 193–291 (1986).
22. M. de Gosson and F. Luef, Symplectic capacities and the geome-
try of uncertainty: The irruption of symplectic topology in classi-
cal and quantum mechanics, Phys. Rep. 484, 131–179 (2009), doi:
10.1016/j.physrep.2009.08.001.
23. I. Ekeland and H. Hofer: Symplectic topology and Hamiltonian dynamics, II.
Math. Zeit. 203, 553–567 (1990).
24. K. Cielibak, H. Hofer, J. Latschev and F. Schlenk: Quantitative symplectic
geometry. Recent Progress in Dynamics, MSRI Publications, NA 54 (2007);
arXiv:math/0506191v1 [math.SG].
25. M. de Gosson, The Principles of Newtonian and Quantum Mechanics: The
need for Planck’s Constant, h. With a foreword by Basil Hiley. Imperial
College Press, London 2001.
www.ebook3000.com
Chapter 3
Space–Time in Quantum Gravity:

Does Space–Time have
Quantum Properties?
Reiner Hedrich
Krofdorfer Strasse 62, 35398 Giessen, Germany
reiner.hedrich@udo.edu
The conceptual incompatibility between General Relativity and Quan-

tum Mechanics is generally seen as sufficient motivation for the develop-
ment of a theory of Quantum Gravity. If — so a typical argument goes —
Quantum Mechanics gives a universally valid basis for the description of
the dynamical behavior of all natural systems, then the gravitational field
should have quantum properties, like all other fundamental interaction
fields. And if General Relativity can be seen as an adequate description
of the classical aspects of gravity and space–time — and their mutual
relation — this leads, together with the rather convincing arguments
against semi-classical theories of gravity, to a strategy which takes a
quantization of General Relativity as the natural avenue to a theory
of Quantum Gravity. And because in General Relativity, the gravita-
tional field is represented by the space–time metric, a quantization of
the gravitational field would in some sense correspond to a quantization
of geometry. Space–time would have quantum properties.
But, this direct quantization strategy to Quantum Gravity will only
be successful if gravity is indeed a fundamental interaction. Only if it
is a fundamental interaction, the given argument is valid, and the grav-
itational field, as well as space–time, should have quantum properties.
What if gravity is instead an intrinsically classical phenomenon? Then, if
Quantum Mechanics is nevertheless fundamentally valid, gravity cannot
be a fundamental interaction; a classical and at the same time fundamen-
tal gravity is excluded by the arguments against semi-classical theories
of gravity. An intrinsically classical gravity in a quantum world would
have to be an emergent, induced or residual, macroscopic effect, caused
by a quantum substrate dominated by other interactions, not by gravity.
45
46 R. Hedrich
Then, the gravitational field (as well as space–time) would not have any
quantum properties. And then, a quantization of gravity (i.e. of General
Relativity) would lead to artifacts without any relation to nature.
The serious problems of all approaches to Quantum Gravity that
start from a direct quantization of General Relativity (e.g. non-
perturbative canonical quantization approaches like Loop Quantum
Gravity) or try to capture the quantum properties of gravity in the form
of a ‘graviton’ dynamics (e.g. Covariant Quantization, String Theory) —
together with the, meanwhile, rich spectrum of (more or less advanced)
theoretical approaches to an emergent gravity and/or space–time —
make this latter option more and more interesting for the development
of a theory of Quantum Gravity. The most advanced emergent gravity
(and space–time) scenarios are of an information-theoretical, quantum-
computational type. A paradigmatic model for the emergence of gravity
and space–time comes from the Pregeometric Quantum Causal Histories
approach.
1. Introduction: The Mutual Incompatibility

of General Relativity and Quantum
Mechanics/Quantum Field Theory
The decisive motivation for the development of a theory of Quantum Grav-
ity is generally seen in the mutual conceptual incompatibility between Gen-
eral Relativity on the one hand and Quantum Mechanics and Quantum
Field Theory on the other hand. The following three crucial points should
elucidate this situation:
(1) General Relativity, today our best theory of gravity as well as of

space–time, treats the gravitational field as a classical dynamical field,
represented by the (pseudo-) Riemannian metric of space–time. But,
according to Quantum Mechanics, dynamical fields have quantum prop-
erties. So, if Quantum Mechanics is taken to be universally valid,
it seems reasonable to assume the necessity of a (direct or indirect)
quantization of the gravitational field. An additional motivation for
the quantization of gravity comes from rather conclusive arguments
against semi-classical modifications of the Einstein field equations that
treat gravity classically and everything else quantum mechanically
[44, 45, 100–102, 147, 193].
www.ebook3000.com
Space–Time in Quantum Gravity: Does Space–Time have Quantum Properties? 47
“The right-hand side of the field equations [of general relativity]

describes matter sources, the behavior of which is governed by
quantum theory. The left-hand side of the field equations describes
gravitation as a classical field. If the right-hand side represents quan-
tized matter then the field equations as they stand are inconsistent
[166] p. 2.”
(2) In General Relativity, the gravitational field is represented by the met-

ric of space–time. Therefore, a quantization of the gravitational field
would correspond to a quantization of the metric of space–time. The
quantum dynamics of the gravitational field would correspond to a
dynamical quantum space–time. But, Quantum Field Theories presup-
pose a fixed, non-dynamical background space–time for the description
of the dynamics of quantum fields. They are conceptually inadequate for
a description of a dynamical quantum geometry. So, a quantum theory
of the gravitational field can scarcely be a Quantum Field Theory, at
least not one in the usual sense. But it is not only the dynamical charac-
ter of general relativistic space–time which makes traditional quantum
theoretical approaches problematic. The active diffeomorphism invari-
ance [61, 62] of General Relativity is fundamentally incompatible with
any fixed background space–time [58–62, 64, 133–135]. (Active diffeo-
morphisms are point transformations, in contrast to passive diffeomor-
phisms, which are coordinate transformations.)
(3) In General Relativity, time is a component of dynamical space–time. It
is dynamically involved in the interaction between matter/energy and
the space–time metric. It can be defined only locally and internally;
there is no global time. (It is again the active diffeomorphism invariance
of General Relativity that leads to the problem of time. See Sec. 5 as
well as Refs. [22, 60, 91, 152, 159, 173, 177, 197].) Quantum Mechanics,
on the other hand, treats time as a global background parameter, not
even as a physical observable represented by a quantum operator.
2. The Quantization of Gravity and the Quantum

Nature of Space–Time — The Naive View
If we assume that, fundamentally, all natural systems are quantum sys-
tems, and that gravity is a universal interaction with influence on all nat-
ural systems, then the conceptual incompatibility of Quantum Mechanics
and General Relativity leads to a severe problem for our description of
48 R. Hedrich
nature. Under these conditions, it is natural to assume that at least one

of our actually most fundamental, well-established, but mutually incom-
patible physical theories is only an approximation to a more fundamental
physical description of nature. But, what would be the most promising way
to the construction of such a more fundamental physical theory?
Taking into account the successful experiences with the implementation
of all other fundamental interactions into a quantum mechanical descrip-
tion, leading to the (at least empirically) successful Standard Model of
Quantum Field Theories, the most natural way to get to a theory of Quan-
tum Gravity seems to be a more or less direct quantization of the gravi-
tational field. If Quantum Mechanics gives a fundamental and universally
valid basis for the description of the dynamical behavior of all natural sys-
tems, the gravitational field should have quantum properties like all other
fundamental interaction fields. (Caveat: This argument is only valid if grav-
ity is a fundamental interaction. See Secs. 6 and 7.) These quantum prop-
erties of the gravitational field should be the subject of the searched-for
theory of Quantum Gravity.
The additional and completely rational assumption that General Rel-
ativity can be seen as an adequate description of the classical aspects of
gravity and space–time — and their mutual relation — leads then, together
with the arguments against semi-classical theories of gravity, to a strategy
which consists basically in a quantization of General Relativity as a natural
avenue to a theory of Quantum Gravity.
In General Relativity, the gravitational field is represented by the metric
of space–time. Gravity is identical to properties of a dynamical geometry.
Therefore, a quantization of the gravitational field would correspond to a
quantization of the metric of space–time. The quantum dynamics of the
gravitational field would correspond to a dynamical quantum space–time,
a dynamical quantum geometry. A theory of Quantum Gravity should then,
so the argument, be or lead to a description of quantum space–time.
“[. . . ] general relativity is not just a theory of gravity — in an appropriate

sense, it is also a theory of space–time itself; and hence a theory of
quantum gravity must have something to say about the quantum nature
of space and time” [35] p. 34.
What would we have to expect with regard to the quantum properties

of space–time? Again, taking into account the experiences with Quantum
Mechanics, we would probably suspect that the space–time metric should
be the expectation value of a quantum variable. On the quantum level,
www.ebook3000.com
we would probably expect quantum uncertainties and fluctuations of the

space–time metric as well as superpositions of space–time metrics. And
the experiences with Quantum Field Theories would possibly suggest some
exchange boson for gravity: the ‘graviton’. Quantum Gravity, one could
think, should under these assumptions possibly be a theory describing the
dynamics of gravitons exchanged between matter particles.
3. Quantum Space–Time — Problems with the

Naive View
But, this is certainly a much too naive picture because (i) the apparatus of
Quantum Field Theory with its fixed background space–time is incompat-
ible with General Relativity and its active diffeomorphism invariance (i.e.
background independence; see Sec. 4 for an example of the consequences of
a background dependent quantization of a background independent theory.)
and (ii) gravitons representing the gravitational field, corresponding to the
metric field, describing therefore the quantum dynamics of space–time, can
scarcely be understood as moving within a (miraculously already existing
classical) space–time (required by Quantum Field Theory).
However, one does not necessarily need to take into account the concep-
tual problems of the assumption of a graviton dynamics to see that a simple
quantization of General Relativity is probably a very questionable route to
a theory of Quantum Gravity. Quantum fluctuations of the space–time met-
ric, to be expected within the outlined naive picture of a direct quantization
of gravity and space–time, are totally sufficient to lead to serious problems:
“[. . . ] once we embark on constructing a quantum theory of gravity, we

expect some sort of quantum fluctuations in the metric, and so also in
the causal structure. But in that case, how are we to formulate a quantum
theory with a fluctuating causal structure?” [35, p. 64].
Quantum fluctuations of space–time are fluctuations of the basic causal

structure. This leads very probably to insurmountable problems for a direct
quantization of the gravitational field using the standard quantization pro-
cedures.
“The main idea [. . . ] is that fluctuations in the gravitational field imply

fluctuations in the spatiotemporal, and hence causal, structure of the
world. But it is hard to see how one can make sense of canonical
50 R. Hedrich
commutation relations and hence quantize anything in the absence of

a stable causal structure” [44, p. 22].
So, fluctuations of the space–time metric, to be expected within the naive

approach to a description of quantum space–time, are completely suffi-
cient to make clear that one cannot get over the mutual incompatibility of
General Relativity and Quantum Mechanics by simply applying standard
quantization procedures to the gravitational field. Possibly, fluctuations of
the causal structure of space–time exist. Possibly, there does not even exist
any basic causal structure in nature. Possibly, causal structure is only an
approximately valid concept or an emergent phenomenon. But, all these
imaginable possibilities and speculations about the quantum properties of
space–time will not be elucidated any further within an approach starting
from a simple standard quantization of the gravitational field, inspired by
the procedures of Quantum Field Theory. The methodological problems
posed by quantum fluctuations of the space–time metric, to be expected
within this naive approach, are insurmountable within this approach. They
lead to implications pointing beyond the context of such a simple direct
quantization of General Relativity.
At this point, it is reasonably clear that a direct quantization of General
Relativity, following the outlined strategy, will not be an adequate route to
Quantum Gravity. Such a direct quantization of General Relativity implies
the assumption of a fundamental space–time with additional quantum prop-
erties: Quantum corrections to a classical space–time. This assumption
leads in turn to problems which make a direct quantization of General Rel-
ativity within the standard strategies of Quantum Mechanics and Quantum
Field Theory conceptually impossible. This suggests that General Relativity
and Quantum Mechanics are too different to allow such a simple amalga-
mation. A theory of Quantum Gravity, a theory that gets over their mutual
incompatibility, has to be constructed in a different way. So, let us see first
(Sec. 4) what happens concretely when one tries in fact to quantize gravity
the naive way, and then (Sec. 5) if there are more sophisticated ways of
quantizing General Relativity that could lead to the discovery of any loop-
holes in the foregoing arguments against a direct quantization of gravity.
4. Covariant Quantization of General Relativity:

Graviton Dynamics
The Covariant Quantization [50,51] of General Relativity reflects the prob-
lems outlined for the naive picture of a quantum space–time in a direct
www.ebook3000.com
way. Covariant Quantization consists in the attempt to construct a Quan-

tum Field Theory of gravity, which means A Quantum Field Theory of the
metric field.
“The idea was [. . . ] to do unto the gravitational field as was done to the
electromagnetic field: quantize the gravitational field to get a particle (the
graviton) that mediates the interaction. However, just as photons require
a background metrical structure, so does the graviton” [160, p. 16].
Because Quantum Field Theories need a background space–time with fixed

metric for the definition of its operator fields, Covariant Quantization uses
a standard perturbation-theoretical approach, working with a fixed (usu-
ally Minkowski) background metric and a perturbation on this background
to be treated quantum mechanically. This leads to a Quantum Field The-
ory of the fluctuations of the metric. The properties of the corresponding
field quanta of gravity are a consequence of symmetry arguments and of the
properties of classical gravity: long-range, exclusively attractive. ‘Gravitons’
are massless and have spin 2. They represent the assumed quantum prop-
erties of space–time, and they behave according to standard Feynman rules
on a fixed background space–time. But Covariant Quantization with its
perturbation expansion of the fluctuations of the space–time metric turns
out to be non-renormalizable. This makes the theory, ultimately, irrelevant
as a fundamental description of space–time:
“It is generally agreed that this non-renormalisability renders perturba-

tively quantised Einstein gravity meaningless as a fundamental theory
because an infinite number of parameters would be required to make any
physical prediction” [132, p. 3].
The non-renormalizability of the theory can be seen as a direct consequence

of the self-interaction of the graviton, which is in turn a quantum-field-
theoretical expression of the non-linearity of classical gravity. Gravity cou-
ples to mass and, because of the mass–energy equivalence, to every form of
energy. (All other interactions couple only to their ‘charges’, not to energy.)
Therefore, the self-interaction contributions to gravity increase for decreas-
ing distances or increasing energies. So, the contribution of virtual particles
with increasing energies dominates the higher orders of the perturbation
expansion. This leads to uncontrollable divergences of the expansion.
“[. . . ] such non-renormalizable theories become pathological at short dis-

tances [. . . ] — perhaps not too surprising a result for a theory which
attempts in some sense to ‘quantize distance’ ” [44, p. 5].
52 R. Hedrich
The non-renormalizability of Covariant Quantization is indeed not much

of a surprise. The background independence of General Relativity, together
with its identification of the gravitational with the metrical field, makes a
background dependent approach to a theory of Quantum Gravity highly
questionable. Covariant Quantization tries to quantize a background inde-
pendent theory — General Relativity — by means of a background depen-
dent method. The result is a non-renormalizable construct, useless as a
fundamental theory. Consequently, gravitons have to be seen as theoretical
artifacts, resulting from a conceptually inadequate methodology.
“The failure of the perturbative approach to quantum gravity in terms

of linear fluctuations around a fixed background metric implies that the
fundamental dynamical degrees of freedom [DoF] of quantum gravity at
the Planck scale are definitively not gravitons. At this stage, we do not
yet know what they are” [118, p. 2].
So, Covariant Quantization shows explicitly that it is not possible to

get over the mutual incompatibility of General Relativity and Quantum
Mechanics/Quantum Field Theory by simply amalgamating gravity and
the quantum. The conceptual foundations of both are obviously much too
different. A Quantum Field Theory of gravity does not exist because it is
not possible to quantize a background independent theory of space–time by
means of a background dependent approach, describing a dynamics on (an
already fixed) space–time. Covariant Quantization exemplifies that it is not
possible to describe the quantum dynamics of space–time on space–time.
Nonetheless, this is exactly what String Theory [74, 98, 150, 151] tries to do
although in a more sophisticated way than Covariant Quantization. String
Theory seems to evade — obviously with more success — the problem of the
non-renormalizability of the Covariant Quantization scheme by means of a
unification of all interactions. Instead of simply describing the dynamics of
gravitons on a fixed space–time, it describes — simply — the dynamics of
one-dimensionally extended strings on a fixed space–time. So, it does not
start from a direct quantization of General Relativity, but from a quanti-
zation of the classical dynamics of a relativistic string. Gravitons turn out
to be quantum states of this string. But, also in String Theory, these gravi-
ton states move on a fixed classical space–time. All known formulations
of String Theory are background dependent; although they seem to evade
the non-renormalizability problem of Covariant Quantization, they lead to
www.ebook3000.com
various severe and — after more than four decades of development — still
unsolved problems, not to be discussed in the present context. (For a further
discussion of String Theory and its problems, see Refs. [78–81].)
5. The Space–time Picture of Loop Quantum Gravity

Loop Quantum Gravity [4–7, 9, 29, 131, 132, 169–171, 175, 176, 180–182, 186,
187, 194–196] is a much more sophisticated attempt at a direct quantiza-
tion of General Relativity than the perturbative Covariant Quantization
approach. As a Canonical Quantization approach, starting from the Hamil-
tonian formulation of General Relativity, it is intrinsically non-perturbative.
And, in particular, it is background independent. In contrast to the old
geometrodynamical Canonical Quantization approach [49,65,103,107,110],
which started from a Hamiltonian formulation of General Relativity with
the metric and the curvature of space–time as basic variables, Loop Quan-
tum Gravity starts from a Hamiltonian formulation of General Relativity
based on the Ashtekar variables [1, 2]: a spatial SU(2) connection variable
and an orthonormal triad.
The Hamiltonian formulation of General Relativity results from a split-
ting of space–time into spatial hypersurfaces and a time parameter. In the
case of the Ashtekar variables, it is a three-dimensional (3D) connection
and a time parameter. The latter is necessary for the definition of the
canonical momentum as well as for the canonical quantization procedure.
The active diffeomorphism invariance of General Relativity — the formal
expression of its general covariance [58–62, 64, 133–135], which has to be
taken into account in the transition to the quantum theory — translates
in the Hamiltonian approach into the constraints. [22, 23, 72, 73, 83]. (The
primary constraints are a direct consequence of the transition from the
Lagrangian formalism to the Hamiltonian formalism by means of a Leg-
endre transformation.) These constraints are necessary because the plain
Hamiltonian theory and its basic variables do not take into account diffeo-
morphism invariance. The corresponding phase space contains redundant
representations of physically identical space–times (as well as representa-
tions of physically impossible states — states that lie outside the ‘con-
straint surface’). The identification of equivalence classes of representations
of physically identical space–times — equivalence classes of representations
that can be transformed into each other by a diffeomorphism — (as well
as the identification of physically impossible states) has to be introduced
additionally by means of the constraints.
54 R. Hedrich
Constraints are typical for the Hamiltonian formulation of dynamics

with an unphysical surplus structure. Such an unphysical surplus structure
is, on the other hand, typical for systems with gauge freedom. In gauge
systems, it is the gauge invariance that captures unphysical redundancies
in the description of a system; in the Hamiltonian formalism, it is the
constraints that capture them. So, according to Loop Quantum Gravity,
the constraints of the Hamiltonian formulation of General Relativity can
be understood as generators of gauge transformations. The corresponding
gauge invariance is active diffeomorphism invariance.
“Because active [diffeomorphism] invariance is a gauge, the physical con-

tent of [general relativity] is expressed only by those quantities, derived
from the basic dynamical variables, which are fully independent from the
points of the manifold” [173, p. 108].
Gauge transformations, capturing transitions between unphysical redun-

dancies, are unobservable, and if one wants to keep up the predictive power
of the theory, then ‘observables’ have to be gauge-invariant. Formally, in
the Hamiltonian approach, this means that all observables have (weakly, i.e.
on the constraint surface) vanishing Poisson brackets with all (first class)
constraints. (First class constraints are constraints with vanishing Poisson
brackets with all other constraints.) In the quantum case, this translates
into: All quantum observables have to commute (weakly) with all (first
class) quantum constraints.
Already in the geometrodynamical version of the Hamiltonian formu-
lation of General Relativity, after the splitting of space–time into spatial
hypersurfaces and a time parameter, there are four constraints: the scalar
or Hamiltonian constraint and three momentum or diffeomorphism con-
straints:
“[. . . ] the momentum and Hamiltonian constraints are believed to cap-

ture the invariance of general relativity under spacelike, and timelike
diffeomorphisms respectively” [44, p. 19].
In the Ashtekar version, because of an additional redundancy connected

with the new variables, one has three additional Gauss constraints, which
generate SU(2) gauge transformations:
“In the connection and loop approaches, three additional (local) con-
straints emerge because of the freedom to choose the local triads upon
which the formulation is based” [102, p. 9].
www.ebook3000.com
Loop Quantum Gravity uses a further modification of the classical Hamil-

tonian theory: A transition from Ashtekar’s connection variables to loop
variables (Wilson loops). From this formulation of General Relativity, it
starts into the quantization procedure using the Dirac quantization method
[83, 214] for constrained Hamiltonian systems.
“[. . . ] Dirac introduced a systematic quantization program. Here, one

first ignores constraints and introduces a kinematic framework consisting
of an algebra a of quantum operators and a representation thereof on a
Hilbert space Hkin . This provides the arena for defining and solving the
quantum constraints. When equipped with a suitable inner product, the
space of solutions defines the physical Hilbert space H phy ” [5, p. 2].
Under ‘solving the constraints’, one understands — in the classical case — a

transition from a description based on the full (unconstrained) Hamiltonian
phase space, containing descriptive redundancies, to a reduced phase space
that captures only the ‘true’ (physical) DoF of the system. In the quan-
tum case, this corresponds to the transition from the full (unconstrained)
‘kinematical’ quantum mechanical Hilbert space, containing redundancies
(e.g. in form of gauge symmetries), to a reduced ‘physical’ Hilbert space
representing only the ‘true’ physical states of the system.
The Dirac quantization method consists in a quantization of the full
Hamiltonian phase space of the classical theory — canonical commutation
relations for the quantum counterparts of the classical variables, an operator
algebra and, finally, the quantum counterparts of the classical constraints
are to be defined with the intention to solve the quantum constraints after-
ward, and to identify thereby the true physical states.
“Note that, in this approach, the commutation relations are simply pos-
tulated” [190, p. 73].
An alternative to Dirac quantization would consist in solving the constraints

first, for the classical theory, and then to quantize the reduced classical
theory, which then has no constraints any more.
“To pass to the quantum theory, one can use one of the two standard
approaches: (i) find the reduced phase space of the theory representing
‘true [DoF]’ thereby eliminating the constraints classically and then con-
struct a quantum version of the resulting unconstrained theory; or (ii)
first construct quantum kinematics for the full phase space ignoring the
constraints, then find quantum operators corresponding to constraints
and finally solve quantum constraints to obtain the physical states. Loop
quantum gravity follows the second avenue [. . . ]” [7, p. 51].
56 R. Hedrich
Unfortunately, the alternative to Dirac quantization is nothing more than

a chimera:
“A distinct quantization method is the reduced phase space quantization,

where the physical phase space is constructed classically by solving the
constraints and factoring out gauge equivalence prior to quantization.
But for a theory as complicated as general relativity it seems impossible
to construct the reduced phase space” [69, p. 9]. — “Relatively little is
presently known about the structure of the reduced phase space of general
relativity” [23, p. 229].
Already at this point, one could ask: Why should it be easier to solve the
constraints in the quantum case? And indeed, solving all the quantum con-
straints and finding the physical Hilbert space, and thereby the true states
of Loop Quantum Gravity, is anything but easy: The quantized Hamilto-
nian constraint, the so-called Wheeler–DeWitt equation, is well known for
its resistance against any attempt to solve it.
However, there are already very interesting results for the kinemati-
cal Hilbert space in Loop Quantum Gravity. For the spatial hypersurfaces,
after solving only the quantum Gauss constraints, one finds a discrete,
polymer-like graph structure: according to Loop Quantum Gravity, the dis-
crete quantum substructure to the (spatial part of the) space–time contin-
uum of General Relativity.
It has to be emphasized that the discreteness of this spin network is a
result of the direct non-perturbative quantization of General Relativity, not
a feature the theory started with.
“Space itself turns out to have a discrete and combinatorial character.

Notice that this is not imposed on the theory, or assumed. It is the result
of a completely conventional quantum mechanical calculation of the spec-
trum of the physical quantities that describe the geometry of space” [176,
p. 14].
However, the discreteness of the spin network is not that of a regular cellular
arrangement or grid (like e.g. in cellular automata), but a discreteness that
requires the continuum of real numbers (like Quantum Mechanics) for its
definition.
“This discreteness of the geometry, implied by the conjunction of [general

relativity] and [quantum mechanics], is very different from the naive idea
that the world is made by discrete bits of something. It is like the dis-
creteness of the quanta of the excitations of a harmonic oscillator. A
generic state of space–time will be a continuous quantum superposition
www.ebook3000.com
of states whose geometry has discrete features, not a collection of ele-

mentary discrete objects” [173, p. 110].
This means that the discreteness of the spin network presupposes the space–
time manifold of General Relativity, although Loop Quantum Gravity tries
to discuss away the manifold after quantization.
“Let us emphasize again that the ‘discreteness’ of the spin networks does
not correspond to a naive discretisation of space. Rather, the underlying
continuum, on which the spin networks ‘float’, the spatial manifold Σ, is
still present” [132, p. 18].
The spin network structure represents the discrete eigenvalues of two geo-
metric operators one can define in Loop Quantum Gravity: the area and the
volume operator.
“[. . . ] a quantum space–time can be decomposed in a basis of states that

can be visualized as made by quanta of volume (the intersections) sepa-
rated by quanta of area (the links). More precisely, we can view a spin
network as sitting on the dual of a cellular decomposition of physical
space. The nodes of the spin network sit in the center of the 3-cells, and
their coloring determines the (quantized) 3-cell’s volume. The links of
the spin network cut the faces of the cellular decomposition, and their
color j determine the (quantized) areas of these faces [. . . ]” [171, p. 8].
And this discrete structure is a rather surprising result for the kinematical
level:
“It is somewhat surprising that an important issue such as the funda-

mental discreteness of space emerges already at the kinematical level.
One would have instead expected that is a result that emerges from the
treatment of the Hamiltonian constraint, which encodes the ‘dynamical’
features of Einstein’s theory. The discreteness thus seems to hold for
more general theories than quantum general relativity” [101, p. 194].
Up to this point, only the Gauss constraints are solved. The spin net-
works, as well as the related area and volume operators, are therefore not
diffeomorphism invariant; they do not commute with the other quantum
constraints.
“Note that the area operator is not invariant under 3D diffeomorphisms.

[. . .] It does also not commute with the Hamiltonian constraint. An area
operator that is invariant should be defined intrinsically with respect to
curvature invariants or matter fields. A concrete realization of such an
operator remains elusive” [102, p. 11].
58 R. Hedrich
The next step consists in solving the (spatial) diffeomorphism (or momen-
tum) constraints. This is realized in a transition from the spin networks to
the diffeomorphism invariant S-knots: equivalence classes of spin networks
with regard to spatial diffeomorphisms.
“Within the framework of loop quantum gravity, regarding s-knot states,

rather than spin-network states, as the genuine physical states is not an
optional move that one might be persuaded to take in response to some
analogue of the hole argument. A quantum theory which countenances
spin-network states as physical states is simply not a quantum version
of general relativity” [153, p. 378].
S-knots are abstract topological objects — excitation states of the gravita-

tional field — that do not live on a background space, but rather represent
space itself. Although the space–time manifold is required to derive the
S-knots, they are, according to Loop Quantum Gravity, the entities defin-
ing space. Every localization is a localization with regard to the S-knots.
According to Loop Quantum Gravity, space is a completely relational con-
struct defined by the S-knots.
“The spin network represent relational quantum states: They are not
located in a space. Localization must be defined in relation to them.” [173,
p. 110] — “[. . .] in quantum gravity the notion of space–time disappears
in the same manner in which the notion of trajectory disappears in the
quantum theory of a particle” [176, p. 21].
But S-knots represent only quantum space, not space–time. They are not
invariant with regard to temporal diffeomorphisms. They are not yet the
states of the true, physical Hilbert space of the theory. The necessary last
step would consist in solving the quantum Hamiltonian constraint (i.e. the
Wheeler–DeWitt equation). But, as yet, Loop Quantum Gravity has not
succeeded with this project.
“The main open problem is the correct implementation (and solution) of

the Hamiltonian constraint” [101, p. 198] — “[. . .] so far the problem of
finding physical observables in quantum gravity is still very little explored
territory [. . .]” [69, p. 47].
Some insiders do not even expect (any more) a complete solution to this
problem:
“The final step [. . .] remains to be done: The physical states of the theory
should lie in the kernel of the quantum Hamiltonian constraint operator.
Of course we do not expect to find a complete solution of the Hamiltonian
www.ebook3000.com
constraint, which would correspond to a complete solution of the theory”

[69, p. 39].
Only treatments with many simplifications exist.
“A more complete treatment would include exponentially increasing

efforts [. . .]” [69, p. 41].
Not even the definition of the quantum Hamiltonian constraint is unam-

biguous.
“[. . .] there is still a large number of poorly controlled ambiguities in the

definition of the Hamiltonian constraint” [5, p. 12].
Obviously, many researchers have meanwhile abandoned the intention of

specifying and solving the quantum Hamiltonian constraint, and, thereby,
that of completing the project of a non-perturbative canonical quantization
of General Relativity. (The actual focus is instead on a covariant variety of
the theory, based on Spin Foam models [8, p. 182].)
And there are further serious problems in Loop Quantum Gravity: One of
these, and probably the most severe, is that no phenomenologically accept-
able low-energy limit has been derived as yet.
“The main difficulties of loop quantum gravity lie in recovering low

energy phenomenology. Quantum states corresponding to the Minkowski
vacuum and its excitation have not yet been constructed, and particle
scattering amplitudes have not been computed” [178, p. 1301] — “[. . .]
there remain however, hard issues concerning whether and how classical
general relativity dominates a suitably defined low energy limit. The fact
that the theory is well defined and finite does not, so far as we know,
guarantee that the low energy limit is acceptable” [188, p. 27].
Here, one should remember that it is not a necessary requirement for a

theory of Quantum Gravity to quantize General Relativity in a conceptually
coherent way (although this seems to be a natural strategy). Rather, the
basic and indispensable requirement for such a theory is that it is able to
reproduce the phenomenology of gravity: the classical, macroscopic low-
energy case. (The next step would ideally consist in the prediction of small
deviations from General Relativity — not yet in conflict with empirical
data — that could be the target of empirical tests in the future.) Should
it not be possible to meet this basic requirement, this would be the end of
Loop Quantum Gravity.
60 R. Hedrich
“Loop quantum gravity [. . .] will fail if it turns out that the low energy
limit of quantum general relativity coupled to matter is not classical gen-
eral relativity coupled to quantum matter fields” [188, p. 32].
Although it is still unclear if Loop Quantum Gravity will finally succeed in

the reproduction of the macroscopic phenomenology of gravity, it is already
totally clear that it has radical implications in comparison to the well-
established theories of physics.
“[. . .] the theory gives up unitarity, time evolution, Poincaré invariance

at the fundamental level, and the very notion that physical objects are
localized in space and evolve in time” [178, p. 1302].
Probably, the most radical of its consequences is the problem of time. It

is already present in General Relativity, but it has more severe implica-
tions in Loop Quantum Gravity. In General Relativity, coordinate time is
not diffeomorphism invariant. According to the gauge-theoretical interpre-
tation [22, 23, 61, 62] of (the Hamiltonian formulation of) General Relativ-
ity, the Hamiltonian constraint, capturing the transition from one spatial
hypersurface to another, and therefore the dynamics of the system, can be
understood as a generator of gauge transformations, and coordinate time
is a gauge variable.
“This means that each dynamical trajectory lies in a single gauge orbit:
As the gravitational field evolves, it always stays in the same gauge orbit”
[23, p. 225].
Essentially, this is nothing more than a circumscription of the fact that, as

an implication of the diffeomorphism invariance of the theory, dynamical
transitions — in the Hamiltonian picture: generated by the Hamiltonian
constraint — do not lead to any observable consequences. So, because it
is not diffeomorphism invariant, coordinate time is unobservable in Gen-
eral Relativity. And clock time, as an observable physical quantity, is a
non-trivial function of the gravitational field, leading to such effects as the
clock paradox. There simply does not exist any observable external time
parameter in General Relativity. This is finally a consequence of general
covariance, captured in the diffeomorphism invariance of the theory.
“[General Relativity] does not describe evolution with respect to an exter-

nal time, but only relative evolution of physical variables with respect to
each other. In other words, temporal localization is relational like spatial
www.ebook3000.com
localization. This is reflected in the fact that the theory has no Hamilto-
nian (unless particular structures are added), but only a ‘Hamiltonian’
constraint” [171, p. 20].
But, in the classical case, the practical consequences of the problem of time
are limited:
“Such a weakening of the notion of time in classical [general relativity]
is rarely emphasized, because, after all, in classical physics we may dis-
regard the full dynamical structure of the dynamical theory and consider
only a single solution of its equations of motion. [. . .] A single solution
of the [general relativistic] equations of motion determines a space–time,
where a notion of proper time is associated to each timelike worldline”
[178, p. 1318].
This is different for the quantum case:

“In the quantum context, on the other hand, there is no single space–
time, as there is no trajectory for a quantum particle, and the very con-
cept of time becomes fuzzy” [178, p. 1318].
After the canonical quantization of General Relativity, there remains noth-

ing resembling a real temporal evolution of a physical system. This is, again,
because of the fact that the temporal evolution is coded into the (quantum)
Hamiltonian constraint, which generates gauge transformations. The corre-
sponding gauge symmetry reflects nothing more than a descriptive redun-
dancy of the theory, something with no observable physical counterpart. All
observables of Loop Quantum Gravity are timeless because all correspond-
ing quantum operators have to commute with the quantum Hamiltonian
constraint.
“The definition of ‘observable’ in the context of constrained systems is
given as a variable that (weakly) commutes with all the first class con-
straints. However, since one of these is the generator of time evolu-
tion (the Hamiltonian constraint), the observables must be constants of
motion” [159, p. 12].
So, the quantized Hamiltonian constraint makes Loop Quantum Gravity a

theory without time. And, apparently, here in the quantum case, nothing
can change this fact, if one is decided only to accept observable quantities
as physically relevant, in other words, if one is decided only to accept gauge-
invariant operators: Quantum observables which commute with all quantum
constraints. However, this timelessness is, ultimately, in conflict with the
world of changes we perceive.
62 R. Hedrich
One might be tempted to say: The world of changes could be a purely

macroscopic emergent phenomenon, obviously based on a timeless micro-
scopic substrate. But, here we get the next problem [126]: For an exper-
imental test of a timeless fundamental theory, it would be necessary to
calculate a low-energy limit. But, it is exactly its timelessness which makes
it impossible to derive such a low-energy limit because even the concept of
energy presupposes time.
There are different attempts at a solution of the problem of time (in
the context of Loop Quantum Gravity as well as in that of the geometrody-
namical version of the Canonical Quantization approach), which we will not
discuss here. (The main approaches are Barbour’s timeless universe [13,14];
see also, [34], Kuchar’s attempt to restore time by treating the Hamiltonian
constraint differently from the other constraints [108, 109], and Rovelli’s
relational time approach [167, 168, 173, 174, 176, 178, 179].)
6. Gravity as an Intrinsically Classical Phenomenon?

Why has not even one of the attempts to quantize gravity as yet led to a
completely successful and phenomenologically adequate theory of Quantum
Gravity, although the first of these attempts go back to the 30s and the 40s
of the 20th century? Why did all attempts to quantize gravity lead to
(more or less) serious conceptual problems? Why does gravity pose such
problems to its quantization? In other words: What is so special about
gravity? There could be a simple explanation for the problems with the
attempts to quantize gravity:
“[. . .] gravity could all in all be an intrinsically classic/large scale phe-

nomenon [. . .]” [70, p. 1].
But, if gravity is an intrinsically classical phenomenon, what about the

arguments against semi-classical theories of gravity? Those arguments pre-
suppose that gravity is a fundamental interaction. They lose their validity
if gravity is not fundamental if it does not even appear in a fundamental
quantum description of nature. Then, on the fundamental level, there is no
semi-classical hybrid dynamics that would lead to conceptual inconsisten-
cies. So, if gravity is an intrinsically classical phenomenon, it cannot be a
fundamental interaction. It has to be an induced or residual effect, caused
by a quantum substrate dominated by other interactions. This quantum
substrate would not include gravity, but would induce gravity on a higher
structural level for small energies or for long distances. So, an intrinsically
www.ebook3000.com
classical gravity has to be an emergent phenomenon that does not exist on

the fundamental level of the quantum substrate.
This emergence scenario does not only reconcile an intrinsically classical
gravity with the known arguments against semi-classical theories of gravity;
it negates at the same time any motivation for a quantization of gravity as
a means to get over the (apparent) incompatibilities between General Rela-
tivity and Quantum Mechanics. If gravity is not a fundamental interaction,
it is not to be quantized to make it compatible with Quantum Mechanics.
Resulting as a classical phenomenon from a quantum substrate, it would
already be compatible with Quantum Mechanics. And it would not only
be unnecessary to quantize gravity — rather it would ultimately be erro-
neous. A quantization of gravity would be a quantization of collective, non-
fundamental, emergent DoF. This would explain very well the problems of
all attempts to quantize gravity.
Under these conditions, the strategy for the development of a theory of
‘Quantum Gravity’ — a theory which would dispel the apparent incompati-
bility between General Relativity and Quantum Mechanics — would rather
consist in the search for an adequate quantum substrate and for a theory
that would explain how the dynamics of this quantum substrate leads to
an emergent level with an intrinsically classical gravity, having the known
phenomenology. Then, the search for a theory of ‘Quantum Gravity’ does
not mean any more a search for a theory which tries to identify the quan-
tum properties of gravity, but for a theory which identifies the quantum
substrate from which gravity emerges as a purely classical phenomenon.
7. Emergent Gravity and/or Emergent Space–Time

If gravity is an intrinsically classical, residual, or induced, emergent phe-
nomenon, without any quantum properties, what about space–time? If Gen-
eral Relativity gives an adequate description of classical gravity, the general
relativistic relation between gravity and space–time, i.e. the geometrization
of gravity, should be taken seriously, at least as long as no better reasons
make this questionable. General Relativity would have to be seen as a clas-
sical, low-energy, long-distance limit to a searched-for theory describing
the quantum substrate from which gravity and space–time results. But,
then this substrate would neither contain gravity nor would it presuppose
space–time, at least not the continuous, dynamical space–time of General
Relativity into which the gravitational field is encoded as metric field. The
space–time of General Relativity — we would have to expect — would be,
64 R. Hedrich
like gravity, an emergent phenomenon. It would not be fundamental, but

the macroscopic result of the dynamics of a ‘pregeometric’ substrate —
where ‘pregeometric’ does not necessarily mean ‘non-geometric’, but ‘pre-
general-relativistic-space–time-continuum’.
And there is already a convincing argument for the existence of discrete
microscopic DoF below the level of a continuous space–time. It comes from
the Bekenstein–Hawking entropy of black holes [17–21, 31, 207, 208]. (For
additional arguments for a microscopic discreteness, see Refs. [57, 139].)
The Covariant (or Holographic) Entropy Bound [19–21, 31–33, 148, 149],
which can be motivated within the thermodynamics of black holes, can
be seen as an indication of a finite information content of any space–time
volume: a finite number of independent DoF within a space–time region.
This is inconsistent with a continuous space–time and with the idea of fields
defined on this continuous space–time, fields that imply an infinite number
of DoF for any space–time region. The argument for a finite information
content of any finite space–time region can be read as an indication either
for a discrete space–time structure or for a finite pregeometric structure
of micro-constituents, from which space–time results. The first alternative,
that space–time has a discrete quantum substructure, i.e. that space–time
has quantum properties leading to a finite information content, finds one
of its best realizations in the spin networks at the kinematical level of Loop
Quantum Gravity. However, the limited success of the attempts to quantize
gravity and space–time makes this first alternative less probable. So, the
best explanation for the finite information content can be seen in the second
alternative; it would then have to be read as an indication of a (with regard
to its DoF) finite pregeometric micro-structure from which space–time
emerges.
But from which structure do gravity and space–time emerge? Of what
entities, interactions, or processes does the substrate consist? (Certainly it
won’t be fields because they presuppose an infinite information content as
well as a continuous space–time on which they are defined.) Does matter
(and do other quantum fields) also emerge from the substrate? Meanwhile,
there exist a lot of different, more or less (mostly less) convincing scenar-
ios that try to answer these questions (see also Refs. [47, 184]); some are
conceptually interrelated and some are completely independent. Some of
these scenarios take General Relativity as an adequate description of grav-
ity and space–time — as an effective theory for the macroscopic, low-energy
regime — keep to the general relativistic relation between gravity and
space–time, and treat them as emerging together from a pregeometric
www.ebook3000.com
substrate. Others take General Relativity as a theory with limited validity,

even for the classical, macroscopic regime — especially with regard to its
geometrization of gravity — and describe the emergence of gravity from a
substrate that already presupposes space–time. Some are pregeometric with
regard to space, but not with regard to time, which is presupposed, either as
a continuous parameter, or in the form of discrete time steps (see e.g. Ref.
[46]). Most of the scenarios presuppose the validity of Quantum Mechanics
on the substrate level, but a few try also to explain the emergence of Quan-
tum Mechanics from a (sometimes deterministic) pre-quantum substrate
[37–42, 155–158]. Here is a selection:
(a) Space–time as an expression of a spectrum

of states of pregeometric quantum systems
In the scenario of Kaplunovsky and Weinstein [99] (which does not even
mention gravity), space and its dimensionality and topology are dynam-
ical results of the formation of higher-level order parameters within the
spectrum of states constituting the low-energy regime of relatively simple
pregeometric quantum systems. The quantum system, originally pregeo-
metric, has a geometric low-energy phase. And phase transitions between
space–times of different dimensionality are to be expected under specific
dynamical conditions.
“The existence of these phases implies the possibility that finite-

temperature effects can cause dimension-changing phase transitions” [99]
p. 1893. — “[. . .] dimension can be thought of as an integer-valued order
parameter which characterizes distinct phases of a single dynamical sys-
tem” [99, p. 1895].
Fermionic DoF leads to a flat space, bosonic DoF to a rolled-up space.
“[. . .] if the system was dominated by bosonic rather than fermionic fields
then space–time would curl up instead of flattening” [99, p. 1896].
Effective gauge DoF can also appear in the low-energy regime:
“[. . .] residual interactions among the low-energy [DoF] which have the
structure of a gauge theory” [99, p. 1896].
In this model, the distinction between ‘geometric’ and ‘internal’ DoF can
be seen as a low-energy artifact that has only phenomenological relevance.
Space is finally nothing more than a fanning-out of a quantum mechanical
66 R. Hedrich
state spectrum. It is an expression of the fact that a quantum system has

a low-energy state spectrum that shows a phenomenology, which can be
interpreted best in a geometrical way.
“[. . .] the space–time continuum as an illusion of low-energy dynamics”

[99, p. 1879].
But, the model, based on standard Quantum Mechanics, presupposes an

external time parameter, which is finally incompatible with General Rela-
tivity.
“There seem to be quantum systems which start out with a well-defined

notion of time but no notion of space, and dynamically undergo a tran-
sition to a space–time phase [. . .]” [99, p. 1879].
However, meanwhile, first ideas are arising with regard to the question of
how a temporal dynamics could emerge from a timeless ‘dynamics’ [70, 71,
112]. And such a timeless ‘dynamics’ would probably even have empirically
testable consequences.
“[. . .] the invariance under Lorentz transformations is only an approxi-

mate property of the field equations [. . .]. [. . .] our theory will show aether
effects beyond second order” [70, p. 4].
(b) Space–time and gravity as emergent thermodynamic

or statistical phenomena
Jacobson [67, 92–95] (see also Refs. [140–142]) has shown that the Einstein
field equations can be derived from a generalization of the proportionality
between entropy and horizon area for black holes (Bekenstein–Hawking
entropy) under consideration of the thermodynamical relation between
heat, temperature, and entropy.
“It is difficult to resist concluding [. . .] that the horizon entropy density

proportional to area is a more primitive concept than the classical Ein-
stein equation, which appears as a thermodynamic consequence of the
interplay of entropy and causality” [95, p. 337].
For the derivation of the Einstein equations, temperature has to be inter-

preted as Unruh temperature of an accelerated observer within a local
Rindler horizon. Heat is to be interpreted as energy flow through a causal
horizon in the past, leading to a curvature of space–time, corresponding
www.ebook3000.com
to a gravitational field. But, General Relativity, derived in this way from

thermodynamics, is probably only valid under equilibrium conditions:
“[. . .] one might expect that sufficiently high frequency or large ampli-
tude disturbances of the gravitational field would no longer be described
by the Einstein equation, not because some quantum operator nature of
the metric would become relevant, but because the local equilibrium con-
ditions would fail. It is my hope that [. . .] we shall eventually reach an
understanding of the nature of ‘non-equilibrium space–time’” [92, p. 7].
The fundamental dynamics behind the causal horizon, from which the
energy flow results in Jacobson’s approach, is unobservable in principle,
and therefore unknown. Knowledge about this fundamental dynamics is
not necessary for the derivation of the Einstein equations. They are generic.
Nothing about the fundamental dynamics can be inferred from them. Indi-
cations independent from General Relativity are necessary.
(c) Gravity and/or space–time as emergent

hydrodynamic or condensed matter phenomena
Hydrodynamic and condensed matter models for emergent gravity go back
to (and are partially inspired by) Sakharov’s Induced Gravity scenario
[15, 183, 198, 209] of the 60s, which takes gravity as a residual effect of elec-
tromagnetism, induced by quantum fluctuations. According to this model,
gravity results from Quantum Electrodynamics in the same way as hydro-
dynamics results from molecular physics; the Einstein–Hilbert action of
General Relativity would be an approximate implication of the effective
action of a Quantum Field Theory.
In Hu’s model [86–90, 96, 97, 112, 138, 185], on the other hand, space–
time is taken to be a collective quantum state of many micro-constituents,
forming a macroscopic quantum coherence, comparable to a Bose–Einstein
condensate.
“In our view general relativity is the hydrodynamical (the low energy,
long wavelength) regime of a more fundamental microscopic theory of
space–time, and the metric and the connection forms are the collective
variables derived from them. At shorter wavelength or higher energies,
these collective variables will lose their meaning, much as phonon modes
cease to exist at the atomic scale” [86, p. 2].
Hu makes a crucial point explicit: He emphasizes that taking hydrodynamic

and condensed matter models for emergent gravity seriously means a major
68 R. Hedrich
change of strategy for Quantum Gravity:
“This view marks a big divide on the meaning and practice of quan-
tum gravity. In the traditional view, quantum gravity means quantizing
general relativity, and in practice, most programs under this banner focus
on quantizing the metric or the connection functions. Even though the
stated goals of finding a micro-structure of space–time is the same, the
real meaning and actual practice between these two views are fundamen-
tally different. If we view [general relativity] as hydrodynamics and the
metric or connection forms as hydrodynamic variables, quantizing them
will only give us a theory for the quantized modes of collective excitations,
such as phonons in a crystal, but not a theory of atoms or [Quantum
Electrodynamics]. [. . .] we find it more useful to find the micro-variables
than to quantize macroscopic variables” [86, p. 2].
Volovik, who developed the probably most advanced condensed matter

model for emergent gravity [199–206] (see also Refs. [66,68,96,141,192,215]),
has the same convictions with regard to the quantization of General Rela-
tivity. But, as Secs. 6–9 should motivate: This insight is not only valid for
the hydrodynamic and condensed matter approaches, but essentially for all
emergent gravity/emergent space–time scenarios.
According to Volovik’s approach, gravity and space–time could be emer-
gent phenomena resulting from excitation states of a fermionic system with
Fermi-point (i.e. a topological defect in momentum space). These systems
are claimed to belong to a universality class showing low-energy behav-
ior that reproduces the phenomenology of gravitation as well as dynamical
structures similar to those of the Standard Model of Quantum Field The-
ory. They contain chiral fermions as low-energy quasi-particles as well as
collective bosonic excitation states of the Fermi quantum liquid, and they
lead to effective gravitational and gauge fields with their corresponding
symmetries.
“The quasiparticles and collective bosons perceive the homogeneous

ground state of condensed matter as an empty space — a vacuum —
since they do not scatter on atoms comprising this vacuum state: Quasi-
particles move in a quantum liquid or in a crystal without friction just as
particles move in empty space. The inhomogeneous deformation of this
analog of the quantum vacuum is seen by the quasiparticles as the metric
field of space in which they live. It is an analog of the gravitational field”
[201, p. 3].
Unfortunately, the identification of the concrete substrate — one of the

main objectives of a theory of Quantum Gravity — is difficult within
www.ebook3000.com
Volovik’s condensed matter approach to emergent gravity. The best one can
achieve is the identification of a universality class, from which the known
low-energy phenomenology can be reproduced. But such a universality class
contains, in general, completely different dynamical systems, which all lead
to the same low-energy phenomenology. (In this sense, it is comparable to
the landscape of String Theory. See Ref. [79] and the references therein.)
In the Fermi–point model, the emergent, effective space–time is natu-
rally 4D and can have curvature, black holes, and event horizons. (Volovik’s
model leads — like Hu’s — also to a natural explanation for a small cosmo-
logical constant, as well as for the flatness of the universe. See Ref. [205].)
But, the equivalence principle and the general covariance of General Rela-
tivity are only approximately valid.
“The effective gravity may essentially differ from the fundamental gravity
even in principle. Since in the effective gravity the general covariance is
lost at high energy, the metrics which for the low-energy observers look
as equivalent, since they can be transformed to each other by coordinate
transformations, are not equivalent physically. As a result, in emergent
gravity some metrics, which are natural in general relativity, are simply
forbidden. [. . .] Some coordinate transformations in [general relativity]
are not allowed in emergent gravity; [. . .] The non-equivalence of different
metrics is especially important in the presence of the event horizon” [203,
p. 6].
Volovik’s idea is that this is not necessarily a weakness of the theory: pos-
sibly General Relativity contains theoretical artifacts without counterparts
in reality. Its diffeomorphism invariance, representing the general covari-
ance of the theory, could be such an artifact, ultimately going beyond the
empirically tested phenomenology of gravity.
Actually, it is unclear at the moment, to what extent the hydrody-
namic and condensed matter models of an emergent gravity are in conflict
with basic conceptual implications of General Relativity, e.g. what kind
of background they need, and if they necessarily need an external time
parameter or a quasi-local change rate. Could the background indepen-
dence of General Relativity, finally, be just a theoretical artifact, as some
of the emergent gravity scenarios suggest? Could, finally, gravity be emer-
gent, but space–time fundamental? — For an emergent gravity model, a
possible background dependence would at least be less problematic than
for an approach starting from a direct quantization of General Relativity
(as long as there is no conflict with known phenomenology). In the direct
70 R. Hedrich
quantization approach instead, background dependence would be a concep-

tual contradiction: a background dependent quantization of a background
independent theory. For emergent gravity, there could still be reasons to
take the background independence of General Relativity as a theoretical
artifact. But they would have to be very good reasons.
(d) Space–time as a phenomenological result

of a computational process
As the model to be presented in the next section will show, if space–time
should be an emergent information-theoretical phenomenon, some of the
problematic implications of the hydrodynamic and condensed matter mod-
els, e.g. their possible inability to achieve background independence, may be
avoided. However, many alternative scenarios of an information-theoretical
emergence of gravity and/or space–time with different substrate construc-
tions (and sometimes with their own specific problems) exist. Most pre-
suppose quantum principles [75, 85, 114–117, 216–218], but some start from
a non-quantum substrate and try not only to elucidate the emergence of
gravity and space–time, but also to reconstruct Quantum Mechanics as an
emergent phenomenon [38, 155–157].
Already in Wheeler’s agenda for a future physical theory, destined to
overcome the mutual incompatibility between General Relativity and Quan-
tum Mechanics, one can find the following recommendation:
“Translate the quantum versions of string theory and of Einstein’s

geometrodynamics from the language of continuum to the language of
bits” [212, p. 362].
This is the central idea of Wheeler’s It from bit concept [210–212]: Going
beyond space–time to a truly pregeometric substrate, constituted by pure
information. Lloyd modifies this in his Computational Universe approach
[115–117] to an It from qubit: Space–time is here to be reconstructed as an
emergent result of a background independent quantum computation.
“Because distances are derived from dynamics, without reference to

an underlying space–time manifold, the resulting theory is intrinsically
covariant and background independent” [116, p. 2].
Quantum computations are here understood as superpositions of compu-

tational histories. The transition from these superpositions to a classical
www.ebook3000.com
macroscopic space–time consists, according to Lloyd, in their decoherence.
“The visible universe that we see around us presumably corresponds to

one such decoherent history” [116, p. 21].
And the metric of space–time is a direct result of the fundamental quantum

computations.
“The information that moves through the computation effectively ‘mea-

sures’ distances in space–time in the same way that the signals passed
between members of a set of GPS satellites measure space–time” [116,
p. 7] — “[. . .] distances are quantities that are derived from the underly-
ing dynamics of quantum systems” [116, p. 2].
Because of the background independence of the substrate, emergent space–

time fulfills, as Lloyd suggests, necessarily the Einstein field equations in
their discrete form as Einstein–Regge equations.
“Since general covariance [. . .] implies Einstein’s equations, the geome-

try induced by the computational universe obeys Einstein’s equations (in
their discrete, Regge calculus form)” [116, p. 7].
The model includes the back-reaction of the emergent space–time metric

on (computational) matter in a natural way:
“The computational universe model is intrinsically a theory of quantum

matter coupled to gravity, and not a theory of either quantum matter or
quantum gravity on its own” [116, p. 13].
But, as in almost all emergent gravity/emergent space–time scenarios, the

concrete substrate dynamics, finally, remains obscure. For the Computa-
tional Universe approach, this translates into: It is unknown on which con-
crete computation our universe with its specific space–time chronogeometry
is based.
“Every quantum computation corresponds to a family of metrics, each of

which obeys the Einstein–Regge equations. But which computation cor-
responds to the universe we see around us? What is the ‘mother’ com-
putation? We do not know” [116] p. 23.
It might even be that it is a superposition of all possible quantum com-

putations from which our universe (or a multiverse to which it belongs)
results.
72 R. Hedrich
“An appealing choice of quantum computation is one which consists of a

coherent superposition of all possible quantum computations [. . .]” [116,
p. 23].
Lloyd and some other investigators [213] (see also Ref. [154]) take quantum
cellular automata to be the best candidates for a concretization of the
quantum-computational scenario. (The question how the possibly necessary
assumption of time steps for the quantum-computational substrate can be
made compatible with background independence will be discussed in the
next section.)
However, independent of the problem of the identification of the sub-
strate, the question remains: How can space–time emerge from something
so completely different from space–time: Quantum information, information
flow, or basic causal relations? How can the chronogeometry of space–time
emerge from something completely pregeometric? This is probably one of
the most fundamental questions to be posed with regard to the information-
theoretical scenarios for the emergence of space–time. The question results
from the obvious conflict of these scenarios with our intuitions about space–
time. A possible reconciliation with our intuitions comes from the Holo-
graphic Screens idea [129].
Take an acyclic network (a graph) of directed relations (‘lines’) between
elementary quantum systems (‘vertices’) without any (continuous, metrical)
space–time background. The directed relations are instantiated by flows of
quantum information between the elementary quantum systems (and can be
interpreted as causal relations). Dynamical changes occur locally in discrete
steps. There are no continuous space–time DoF on the fundamental level.
Then define screens that separate adjacent parts of the network, cutting
through some of the lines of the network. For each screen, a specific quantum
information flow capacity can be found.
The crucial idea of the Holographic Screens concept starts from an
inversion of the central implications of the Bekenstein–Hawking entropy
[17–21, 31, 207, 208]: According to Bekenstein, the entropy of a black hole
is proportional to the area of its event horizon. And, according to the
Holographic (or Covariant) Entropy Bound [19–21, 31–33, 148, 149], this
Bekenstein–Hawking entropy defines the maximum information content of
the corresponding volume. So, the maximum information (corresponding to
the number of independent DoF) contained within a space–time volume is
finite and proportional to the area of the surface of the space–time volume.
The inversion of this Holographic Bound — the core of the Holographic
Screens concept — consists now in the idea that the amount of quantum
www.ebook3000.com
information that can flow through a screen — the quantum information

flow capacity of the screen — defines the area of the screen.
“This leads us to suggest that the Bekenstein bound may be inverted

and area be defined to be a measure of the capacity of a screen for the
transmission of quantum information.” [129, p. 3].
And then, after having defined area as information flow capacity, a space–
time geometry can be established by means of a network of ‘holographic
screens’, to be defined on the (primary) network of elementary quantum
systems and their causal relations. So, the Holographic Screens idea exem-
plifies how Wheeler’s It from bit — modified to an It from qubit — could
work in principle.
8. The Paradigmatic Case for Emergent Gravity

and Emergent Space–time: Pregeometric Quantum
Causal Histories
A theoretical approach that goes (in its most general form) by the
name (Pregeometric) Quantum Causal Histories [77, 106, 120–125, 127]
(related models and approaches are known as Internal Relativity [53–
56] and Quantum Graphity [104, 105]) can be seen as the probably
most clear-cut, paradigmatic case of an attempt to construct a the-
ory of ‘Quantum Gravity’ that can explain how gravity as well as
space–time — here both have no quantum properties because they
are intrinsically classical phenomena — could emerge from a ‘prege-
ometric’ quantum substrate, presupposing for the substrate only very
simple basic constituents and dynamics. This approach is worth a
more extended discussion, not at least, because it succeeds in avoid-
ing the most prominent problems of the direct quantization approaches,
as well as those of most other emergent gravity/emergent space–time
scenarios.
“It is peculiar that the approaches that advocate that gravity is only an
effective theory (string theory, condensed matter) are based explicitly on
a space–time being present while approaches that are background inde-
pendent consider gravity to be fundamental./Here, we will advocate an
approach orthogonal to the quantum field theory-like approaches above
(we are background independent) but also orthogonal to the usual back-
ground independent approaches (there will be no fundamental [DoF] for
the gravitational field)” [124, p. 2].
74 R. Hedrich
Pregeometric Quantum Causal Histories can be seen not only as the

paradigmatic case of a pregeometric theory of Quantum Gravity but also
as a synthesis or a point of convergence of many different approaches to
a pregeometric quantum substrate — approaches that all have in com-
mon that they assume a basic (quantum) structure representing causal
relations. Quantum Causal Histories are, on the one hand, a concep-
tual extension of Sorkin’s Causal Set approach [30, 84, 162–165, 189, 191],
enriched by the Holographic Screens concept [129] and elements from
Lloyd’s Computational Universe scenario [115–117], which itself owes a lot
to Wheeler’s It from bit [212]. On the other hand, they can also be seen
as a generalization of causal spin networks and of the Spin Foam approach
[8,11,12,113,128,136,137,145,146,182], enriched by elements from Algebraic
Quantum Field Theory.
The Pregeometric Quantum Causal Histories approach is at the moment
completely speculative, but it gives at least an idea, how gravity and space–
time could emerge from a pregeometric substrate based exclusively on quan-
tum information and its flow. Its basic assumptions are as follows
— There is no continuous space–time on the substrate level. The funda-
mental level does not even contain any geometric DoF at all.
(Then, Quantum Causal Histories are necessarily background indepen-
dent — and this in a much more extensive sense than General Relativity,
which presupposes at least a fixed topology.)
— Causal order is more fundamental than properties of space–time, like
metric or topology.
— Causal relations are to be found on the substrate level in the form of
elementary causal network structures.
— Only a finite amount of information can be ascribed to a finite part of
the substrate network of causal relations.
(This assumption is motivated explicitly by the Bekenstein–Hawking
entropy and the Holographic Entropy Bound which lead to finite infor-
mation limits for finite regions. The Bekenstein–Hawking entropy is
not only taken to be relevant in the context of the emergent grav-
ity/emergent space–time approaches, but also in that of the more tradi-
tional approaches to Quantum Gravity; so, it could be reproduced under
certain conditions even by Loop Quantum Gravity [130] and by String
Theory [48, 111, 119, 143, 144].)
— Quantum Mechanics is valid on the fundamental level.
www.ebook3000.com
Quantum Causal Histories are relational networks of quantum systems with

only locally defined dynamical transitions. The basic structure is a discrete,
directed, locally finite, acyclic graph. To every vertex (i.e. elementary event)
of the graph, a finite-dimensional Hilbert space (and a matrix algebra of
operators working on this Hilbert space) is assigned. (This is one of the
most important extensions in comparison with the Causal Set approach.)
So, every vertex is a quantum system. Every (directed) line of the graph
stands for a causal relation: a connection between two elementary events;
formally, it corresponds to a quantum channel, describing the quantum evo-
lution from one Hilbert space to another. So, the graph structure becomes
a network of flows of quantum information between elementary quantum
events. Quantum Causal Histories are information processing quantum sys-
tems; they are quantum computers.
Because there are no space–time DoF on the fundamental level of
description, Quantum Causal Histories are necessarily background inde-
pendent, and therefore not in direct conceptual conflict with General Rela-
tivity. But, if this approach intends to be successful as a theory of Quantum
Gravity, it has to explain geometrogenesis; it has to explain how space–time
emerges from a pregeometric quantum substrate. This would be the first
step on the way to a reproduction of the empirically well-tested phenomeno-
logical implications of General Relativity — the most basic and indispens-
able requirement for any theory of Quantum Gravity: General Relativity is
at least to be reproduced as an effective theory for the macro-level. A fur-
ther step would possibly consist in the explicit reproduction of the Einstein
field equations as a classical, macroscopic approximation.
The basic idea with regard to the first step — geometrogenesis — is the
following: Macroscopic space–time and classical gravity do not result from
a coarse-graining of quantum-geometric DoF — those do not exist accord-
ing to the Quantum Causal Histories approach, but from the dynamics of
propagating coherent excitation states of the substrate [106].
“[. . .] instead of looking for ways to coarse-grain the quantum geometry
directly, one can first look for long-range propagating [DoF] and recon-
struct the geometry from these” [124, p. 15] — “[. . .] we will take up the
idea that the effective description of a background independent theory
can be characterized by the dynamics of coherent excitations in the fun-
damental theory and implement it by importing the method of noiseless
subsystems from quantum information theory” [124, p. 3].
Then, macroscopic space–time is necessarily dynamical because it results

from a background independent pregeometric dynamics. But, the dynamics
76 R. Hedrich
of the effective DoF on the macro-level are necessarily decoupled from the
dynamics of the substrate degrees of freedom [96]. If they would not be
decoupled, there would not be any space–time or gravity on the macro-
level because there is none on the substrate level. (However, not every
pregeometric substrate has necessarily a geometric phase.) In the same
way, causality on the macro-level, finding its expression in the macro-level
interactions, is decoupled from causality on the substrate-level. And space–
time-locality on the macro-level, if it emerges from the dynamics of coherent
excitation states, has nothing to do with ‘locality’ on the substrate graph
structure level.
“In a given graph (the fundamental theory) there will be a notion of

locality: In a graph two nodes are neighbors if they are connected by a
link. We call this microlocality. In the known background independent
theories, the dynamics is generated by moves that are local in the micro-
scopic sense. But if this is to be a good theory, there should be a notion of
classical space–time geometry that emerges from the quantum geometry.
This will give rise to another notion of locality, which we may char-
acterize as macrolocality. [. . .] they do not coincide. [. . .] the notion of
macrolocality should be defined directly from the interactions of the noise-
less subsystems that we identify with the emergent [DoF] [. . .] It is the
fundamental evolution that is non-local with respect to our space–time”
[124, p. 24f].
It is the same for time: The temporal development on the macro-level, cor-
responding to the dynamics of the coherent excitation states, is completely
decoupled from the local temporal steps on the substrate-level [126].
“[. . .] truly effective space–time means effective locality and effective time
direction that are not simply Planck scale quantum corrections on the
classical ones” [124, p. 29].
But what are these coherent, propagating excitation states, resulting from
the substrate dynamics and leading to space–time and gravity? And how do
they give rise to space–time and gravity? The answer given by the Quantum
Causal Histories approach consists in a coupling of geometrogenesis to the
genesis of matter. The idea is that the coherent excitation states result-
ing from, and at the same time dynamically decoupled from the substrate
dynamics are matter DoF. And these give rise to space–time because they
behave as if they were living in a space–time.
“We propose that it is properties of the interactions of these excitations

that we understand as space–time” [124, p. 2] — “[. . .] all we can mean by
www.ebook3000.com
a Minkowski space–time is that all coherent [DoF] and their interactions

are Poincaré invariant at the relevant scale” [124, p. 18] — “In our
approach the relationship between particles and symmetry group is exactly
reversed. It is the particles that determine structures like the light cone
and the symmetry group. We are thus proposing not to use the Poincaré
group and its representation theory in the basic setup of the theory” [55,
p. 10].
So, the genesis of matter, resulting from the substrate dynamics, implies
at the same time geometrogenesis. Both are inseparably coupled to each
other.
“In our view, matter and geometry have a more dual role. One can not
have one without the other. Both emerge from the fundamental theory
simultaneously” [55, p. 4].
Ultimately, the space–time of the Quantum Causal Histories approach is

nothing more than an implication of the behavior of matter. Space–time
is a completely relational construct, an expression of the phenomenology
of matter dynamics. And the matter DoF give rise at the same time to
gravity because the space–time they bring forth by means of their behavior
is a curved space–time. Gravity is nothing more than an expression of this
curved space–time. And gravity has, as part of macro-causality, a finite
propagation speed because the coherent excitation states of the substrate,
the matter DoF, have a finite propagation speed.
“Once the velocities of the bound objects are no longer small we have to
take into account that the change of [the state of the order parameters] is
not instantaneous. Gravity here has a finite propagation speed” [55, p. 8].
The still unproved central hypothesis of the Quantum Causal Histories

approach is that the Einstein field equations are necessarily an implication
of the dynamics of the coherent excitation states and that they can finally
be derived from the substrate dynamics.
“[. . .] the same excitations of the underlying system (characterizing the

geometrogenesis phase transition) and their interactions will be used to
define both the geometry and the energy–momentum tensor Tµυ . This
leads to the following Conjecture on the role of General Relativity:/If the
assignment of geometry and Tµυ from the same excitations and interac-
tions is done consistently, the geometry and Tµυ will not be independent
but will satisfy Einstein’s equations as identities./What is being ques-
tioned here is the separation of physical [DoF] into matter and gravita-
tional ones. In theories with a fixed background, such as quantum field
78 R. Hedrich
theory, the separation is unproblematic, since the gravitational [DoF] are

not really free and do not interact with the matter. In the classical back-
ground independent theory, general relativity, we are left with an intricate
non-linear relation between the two sets: The Einstein equations. As the
practitioners of canonical quantum gravity know well, cleanly extracting
dynamical gravitational [DoF] from the matter is fraught with difficulties.
If such a clean separation could be achieved, canonical quantum gravity
would have succeeded at least two decades ago. / The new direction uni-
fies matter and gravity in the pre-geometric phase and provides a path
towards explaining gravity rather than just quantizing it” [125, p. 19].
But, what kind of matter does emerge from the substrate of the Quantum
Causal Histories approach? And what is it that stabilizes the coherent
excitation states corresponding to matter? The proposed answer to the last
question is: topology. The idea is that the coherent excitation states can be
identified with stable topological knot structures: braids with crossings and
twists [24, 28]. These topological structures seem to be conserved by the
substrate dynamics because of topological symmetries and corresponding
topological conservation principles.
“We have shown that braidings of graph edges are unaffected by the usual
evolution moves. Any physical information contained in the braids will
propagate coherently [. . .]” [124, p. 19] — “The states are bound here,
not by fields, but by quantum topology. [. . .] the states are bound because
there are conserved quantum numbers that measure topological properties
of the states” [28, p. 2].
Interestingly, the basic properties of these stable topological structures can

be identified with the basic properties of elementary particles. So, the twist
of a braid structure can be interpreted as electromagnetic charge.
“Twist is interpreted as U (1) charge, so that a ±2π twist in a ribbon

represents charge ±e/3” [28, p. 4].
There are also topological counterparts to charge conjugation, to quark

colors, to parity, etc.
“[. . .] parity inversion [. . .] for a braid is equivalent to a left–right inver-

sion, while not affecting the handedness of any twists on the strands”
[28, p. 6].
“It is then possible that all the quantum numbers, including the geometric
labels used in loop quantum gravity, can be regarded as composites of
fundamentally topological properties” [28, p. 11].
www.ebook3000.com
And so, all particles of the Standard Model can be identified with specific
topological structures [24–28].
“The simplest non-trivial braids can be made with three ribbons and two
crossings [. . .]. It is remarkable that with a single condition, these map
to the first generation of the standard model” [28, p. 4] — “It is natural
to hypothesize then that the second generation standard model fermions
come from the next most complicated states, which have three crossings.
[. . .] it is also proposed that the gauge vector bosons of the standard model
are composite, and are represented by triplets of ribbons with no cross-
ings. Braids with three ribbons and no crossings are mapped to the bosons
of the electroweak interaction. The electroweak interactions between the
fermions and the photon and vector bosons are then described by cutting
and joining operations on 3-ribbon braids. These preserve the relevant
quantum numbers” [28, p. 8f].
Naturally, the spectrum of topological structures does not contain any

counterpart to the graviton. According to the Quantum Causal Histories
approach, there are no gravitons: Gravity is an intrinsically classical, emer-
gent phenomenon; it does not have any quantum properties or quantum
constituents.
What is still missing is a dynamical explanation that elucidates the
identification of the topological properties with the physical properties of
elementary particles. It should, finally, be possible to derive energy conser-
vation principles from the dynamics of these topological structures, which
should be translation-invariant; and this should, not at least, lead to an
explanation for particle masses.
“Ultimately such rules have to arise from the dynamics” [28, p. 7].
9. Emergent Gravity, Emergent Space–Time,

and the Search for a Theory of ‘Quantum Gravity’
As the emergent space–time/emergent gravity scenarios show, it is con-
ceptually quite possible that space–time and/or gravity are intrinsically
classical, emergent, residual or induced, macroscopic phenomena without
any quantum properties. And, if gravity should indeed be an emergent,
intrinsically classical phenomenon, it would be completely nonsensical to
try to quantize gravity. There would be no quantum properties of gravity,
no gravitons, etc. Gravity would result from a substrate without any gravi-
tational DoF. A quantization of gravity would correspond to a quantization
of collective, macroscopic DoF. A quantization of General Relativity would
80 R. Hedrich
be the quantization of an effective theory describing the dynamics of these

collective DoF. It would be as useful as a quantization of the Navier–Stokes
equation of hydrodynamics. The resulting ‘Theory of Quantum Gravity’
would be analogous to something like a Quantum Hydrodynamics, an arti-
ficial, formal quantization of a classical theory describing collective, macro-
scopic degrees of freedom, without any implications for or any clarifications
with regard to an underlying quantum substrate. It would be simply the
wrong DoF which are quantized.
Should space–time — under the assumption that it is related to gravity
in the general relativistic sense — be, in the same way, a collective expres-
sion or result of completely different non-space–time degrees of freedom,
there would be no quantum space–time, no fluctuations, no uncertainties,
no superpositions of space–time, etc. And, should space–time not be related
to gravity in the general relativistic sense, then there would not even be any
initial reason at all to assume that space–time should have any quantum
properties.
A quantization of gravity (or space–time) is only (but not necessarily) a
reasonable strategy for the construction of a theory of Quantum Gravity if
gravity is a fundamental interaction. If it is not a fundamental interaction,
the adequate strategy consists rather in a search for the substrate dynamics
from which gravity emerges. ‘Quantum Gravity’ would then be the name of
a theory describing this substrate and explaining how gravity (and space–
time) emerge(s) from this substrate.
One of the basic requirements for such a theory describing the sub-
strate dynamics is that it can reproduce the empirically well-confirmed
macroscopic phenomenology of gravity. (This is a basic requirement for all
attempts to construct a theory of Quantum Gravity, also for those who
try to quantize gravity directly. As we have seen, a direct quantization of
General Relativity does not at all imply or even guarantee the reproduction
of the phenomenology of General Relativity as a low-energy limit.)
Even better than the fulfillment of this basic requirement would be the
prediction of small deviations from General Relativity, not yet in contra-
diction to its empirically confirmed phenomenology, which could be tested
in future experiments.
So, the present situation with regard to the different attempts to con-
struct a theory of Quantum Gravity teaches us not to cling to only one
strategy, especially when this strategy meets serious problems in all its
different instantiations. Rather, we should take into account all consistent
www.ebook3000.com
conceptual possibilities, even the more radical or exotic ones. Quantum

Gravity, if we finally should succeed in the construction of a consistent and
empirically confirmable theory, could very well lead to rather unexpected
implications with regard to our view of gravity and space–time.
Acknowledgments
Research has been generously supported by the Fritz–Thyssen–Stiftung
für Wissenschaftsförderung under the project Raumzeitkonzeptionen in der
Quantengravitation. Thanks also to Brigitte Falkenburg! For an extensive
report, see Ref. [82].
References
1. A. Ashtekar, New variables for classical and quantum gravity, Phys. Rev.
Lett. 57, 2244–2247 (1986).
2. A. Ashtekar, New Hamiltonian formulation for general relativity, Phys. Rev.
D36, 1587–1602 (1987).
3. A. Ashtekar, Gravity and the quantum, New J. Phys. 7, 198 (2005), arXiv:
gr-qc/0410054.
4. A. Ashtekar, An introduction to loop quantum gravity through cosmology,
(2007), arXiv: gr-qc/0702030.
5. A. Ashtekar, Loop quantum gravity: Four recent advances and a dozen
frequently asked questions, (2007a), arXiv: 0705.2222 [gr-qc].
6. A. Ashtekar, Introduction to loop quantum gravity, (2012), arXiv:1201.4598
[gr-qc].
7. A. Ashtekar and J. Lewandowski, Background Independent Quantum Grav-
ity — A Status Report, Classical and Quantum Gravity 21, R53 (2004),
arXiv: gr-qc/0404018.
8. A. Ashtekar, M. Reuter and C. Rovelli, From general relativity to quantum
gravity, (2014), arXiv:1408.4336 [gr-qc].
9. A. Ashtekar, C. Rovelli and L. Smolin, Weaving a classical metric with
quantum threads, Phys. Rev. Lett. 69, 237–240 (1992).
10. A. Ashtekar and J. Stachel (eds.), Conceptual Problems of Quantum Gravity,
Birkhäuser, Boston, 1991.
11. J. Baez, Spin foam models, Classical and Quantum Gravity 15, 1827–1858
(1998), arXiv: gr-qc/9709052.
12. J. Baez, An introduction to spin foam models of quantum gravity and BF
theory, Lect. Notes in Phys. 543, 25–94 (2000), arXiv: gr-qc/9905087.
13. J. Barbour, The timelessness of quantum gravity I — The evidence from
the classical theory, Classical and Quantum Gravity 11, 2853–2873 (1994).
14. J. Barbour, The End of Time. The Next Revolution in Physics, Weidenfeld
& Nicolson, London, 1999.
82 R. Hedrich
15. C. Barcelo, S. Liberati and M. Visser, Analogue gravity, Living

Reviews in Relativity (Electronic Journal) 8/12 (2005), available on:
www.livingreviews.org, also: arXiv: gr-qc/0505065.
16. J.D. Barrow, P.C.W. Davies and C.L. Harper (Eds.), Science and Ultimate
Reality: Quantum Theory, Cosmology, and Complexity, Cambridge Univer-
sity Press, Cambridge, 2004.
17. J.D. Bekenstein, Black holes and entropy, Phys. Rev. D 7, 2333–2346 (1973).
18. J.D. Bekenstein, Generalized second law of thermodynamics in black hole
physics, Phys. Rev. D 9, 3292–3300 (1974).
19. J.D. Bekenstein, Universal upper bound on the entropy–to–energy ratio for
bounded systems, Phys. Rev. D 23, 287–298 (1981).
20. J.D. Bekenstein, Holographic bound from second law, (2000), arXiv: gr-
qc/0007062.
21. J.D. Bekenstein, The limits of information, Studies in History and Philoso-
phy of Modern Physics 32, 511–524 (2001), arXiv: gr-qc/0009019.
22. G. Belot and J. Earman, From Metaphysics to Physics, in J. Butterfield
and C. Pagonis (eds.), From Physics to Philosophy, Cambridge University
Press, Cambridge, 1999, pp. 166–186.
23. G. Belot and J. Earman, Pre-socratic Quantum Gravity, in C. Callender and
N. Huggett (eds.), Physics meets Philosophy at the Planck Scale. Contempo-
rary Theories of Quantum Gravity, Cambridge University Press, Cambridge,
2001, pp. 213–255.
24. S.O. Bilson–Thompson, A topological model of composite preons, (2005),
arXiv: hep-ph/0503213.
25. S. Bilson–Thompson, J. Hackett and L.H. Kauffman, Particle topology,
braids, and braided belts, (2009), arXiv: 0903.1376 [math.AT].
26. S. Bilson–Thompson, J. Hackett, L. Kauffman and L. Smolin, Particle iden-
tifications from symmetries of braided ribbon network invariants, (2008),
arXiv: 0804.0037 [hep-th].
27. S. Bilson–Thompson, J. Hackett, L. Kauffman and Y. Wan, Emergent
braided matter of quantum geometry, (2011), arXiv:1109.0080 [hep-th].
28. S.O. Bilson–Thompson, F. Markopoulou and L. Smolin, Quantum gravity
and the standard model, (2006), arXiv: hep-th/0603022.
29. S. Bilson–Thompson and D. Vaid, LQG for the Bewildered, (2014),
arXiv:1402.3586 [gr-qc].
30. L. Bombelli, J. Lee, D. Meyer and R.D. Sorkin, Space-time as a causal set,
Phys. Rev. Lett. 59, 521–524 (1987).
31. R. Bousso, The holographic principle, Rev. Modern Phys. 74, 825–874
(2002), arXiv: hep-th/0203101.
32. R. Bousso, H. Casini, Z. Fisher and J. Maldacena, Proof of a Quantum
Bousso Bound, Phys. Rev. D 90, 044002 (2014), arXiv:1404.5635 [hep-th].
33. R. Bousso, B. Freivogel and S. Leichenauer, Saturating the holographic
entropy bound, Phys. Rev. D 82, 084024 (2010), arXiv:1003.3012 [hep-th].
34. J. Butterfield, Julian barbour: The end of time — Critical notice, British
J. Philos. Sci. 53, 289–330 (2002).
www.ebook3000.com
35. J. Butterfield and C. Isham, Spacetime and the philosophical challenge of

quantum gravity, in [43], also: (2001), arXiv: gr-qc/9903072.
36. J. Butterfield and C. Pagonis (Eds.), From Physics to Philosophy, Cam-
bridge University Press, Cambridge, 1999.
37. R.T. Cahill, Process physics — From quantum foam to general relativity,
(2002), arXiv:gr-qc/0203015.
38. R.T. Cahill, Process Physics — From Information Theory to Quantum Space
and Matter, Nova Science Publishers, New York, 2005.
39. R.T. Cahill and C.M. Klinger, Pregeometric modelling of the space–time
phenomenology, Phys. Lett. A 223, 313–319 (1996), arXiv: gr-qc/9605018.
40. R.T. Cahill and C.M. Klinger, Bootstrap universe from self-referential noise,
(1997), arXiv:gr-qc/9708013.
41. R.T. Cahill and C.M. Klinger, Self-referential noise and the synthesis of 3D
space, (1998), arXiv: gr-qc/9812083.
42. R.T. Cahill and C.M. Klinger, Bootstrap universe from self-referential noise,
Prog. Phys. 2, 108–112 (2005).
43. C. Callender and N. Huggett (eds.), Physics meets Philosophy at the Planck
Scale. Contemporary Theories of Quantum Gravity, Cambridge University
44. C. Callender and N. Huggett, Introduction, in [43] (2001a).
45. C. Callender and N. Huggett, Why quantize Gravity (or any other field for
the matter)?, Philos. Sci. (Proc.), 68, S382–S394 (2001b).
46. M. Cortês and L. Smolin, The universe as a process of unique events, (2013),
arXiv:1307.6167[gr-qc].
47. K. Crowther, Appearing Out of Nowhere: The Emergence of Space-
time in Quantum Gravity, Ph.D. Thesis, University of Sydney, 2014,
arXiv:1410.0345 [physics].
48. S.R. Das and S.D. Mathur, The quantum physics of black holes — Results
from string theory, Annual Rev. Nucl. Particle Sci. 50, 153–206 (2001),
49. B.S. DeWitt, Quantum theory of gravity. I. The canonical theory, Phys.
Rev. 160, 1113–1148 (1967).
50. B.S. DeWitt, Quantum theory of gravity. II. The manifestly covariant the-
ory, Phys. Rev. 162, 1195–1239 (1967a).
51. B.S. DeWitt, Quantum theory of gravity. III. Applications of the covariant
theory, Phys. Rev. 162, 1239–1256 (1967b).
52. D. Dieks (Ed.), The Ontology of Spacetime, Elsevier, Amsterdam, 2006.
53. O. Dreyer, Background independent quantum field theory and the cosmo-
logical constant problem, (2004), arXiv: hep-th/0409048.
54. O. Dreyer, Emergent general relativity, (2006), arXiv: gr-qc/0604075.
55. O. Dreyer, Why things fall, (2007), arXiv: 0710.4350 [gr-qc].
56. O. Dreyer, Internal relativity, (2012), arXiv:1203.2641 [gr-qc].
57. O. Dreyer, The World is Discrete, (2013), arXiv:1307.6169 [gr-qc].
58. J. Earman, Why space is not a substance (at least not to the first degree),
Pac. Philos. Quart. 67, 225–244 (1986).
84 R. Hedrich
59. J. Earman, World Enough and Spacetime — Absolute Versus Relational

Theories of Space and Time, MIT Press, Cambridge, Ma., 1989.
60. J. Earman, Thoroughly modern McTaggart. Or what McTaggart would have
said if he had learned general relativity theory, Philosopher’s Imprint 2, 1–
28 (2002). Available on: http://www.philosophersimprint.org.
61. J. Earman, Two challenges to the requirement of substantive general covari-
ance, Synthese 148, 443–468 (2006).
62. J. Earman, The Implications of General Covariance for the Ontology and
Ideology of Spacetime, in D. Dieks (ed.), The Ontology of Spacetime, Else-
vier, Amsterdam, 2006, pp. 3–23.
63. J. Earman and J. Butterfield (Eds.), Handbook of the Philosophy of Science,
Vol. 2: The Philosophy of Physics, North Holland/Elsevier, Amsterdam,
2007.
64. J. Earman and J.D. Norton, What Price Spacetime Substantivalism? —
The Hole Story, British J. Philos. Sci. 38, 515–525 (1987).
65. J. Ehlers and H. Friedrich (Eds.), Canonical Gravity — From Classical to
Quantum, Springer, Berlin, 1994.
66. C. Eling, Hydrodynamics of space–time and vacuum viscosity, (2008), arXiv:
0806.3165 [hep-th].
67. C. Eling, R. Guedens and T. Jacobson, Non-equilibrium thermodynamics
of space–time, Phys. Rev. Lett. 96, 121301 (2006), arXiv: gr-qc/0602001.
68. D.R. Finkelstein, Quantum Relativity: A Synthesis of the Ideas of Einstein
and Heisenberg, Springer, Berlin, 1996.
69. M. Gaul and C. Rovelli, Loop quantum gravity and the meaning of dif-
feomorphism invariance, Lect. Notes Phys. 541, 277–324 (2000), arXiv: gr-
qc/9910079.
70. F. Girelli, S. Liberati and L. Sindoni, On the emergence of time and gravity,
(2008), arXiv:0806.4239 [gr-qc].
71. F. Girelli, S. Liberati and L. Sindoni, Is the notion of time really fundamen-
tal?, (2009), arXiv:0903.4876 [gr-qc].
72. D. Giulini, The superspace of geometrodynamics, (2009), arXiv: 0902.3923
[gr-qc].
73. J. Govaerts, The quantum geometer’s universe: Particles, interactions and
topology, (2002), arXiv: hep-th/0207276.
74. M.B. Green, J.H. Schwarz and E. Witten, Superstring Theory, 2 Vols., Cam-
bridge University Press, Cambridge, 1987.
75. L. Hardy, Quantum Gravity Computers: On the Theory of Computation
with Indefinite Causal Structure, (2007), arXiv: quant-ph/0701019.
76. J. Harvey, S. Kachru and E. Silverstein (eds.), Strings, Branes, and Grav-
ity — TASI 99, World Scientific, Singapore, 2001.
77. E. Hawkins, F. Markopoulou and H. Sahlmann, Evolution in Quantum
Causal Histories, (2003), arXiv:hep-th/0302111.
78. R. Hedrich, Superstring Theory and Empirical Testability. Available on:
http://philsci-archive.pitt.edu, Document: 608, 2002.
www.ebook3000.com
79. R. Hedrich, String Theory — From Physics to Metaphysics, Phys. Philos.

(Online Journal) 2006, 005. Available on: http://philsci-archive.pitt.edu,
Document 2709; and (2006), arXiv: physics/0604171.
80. R. Hedrich, Von der Physik zur Metaphysik — Physikalische Verein-
heitlichung und Stringansatz, Frankfurt am Main, 2007.
81. R. Hedrich, The internal and external problems of string theory — a
philosophical view, J. General Philos. Sci. 38, 261–278 (2007a). Available
on: http://philsci-archive.pitt.edu, Document: 3012; and: arXiv: physics/
0610168.
82. R. Hedrich, Raumzeitkonzeptionen in der Quantengravitation, (2011),
arXiv: 1101.1835[physics].
83. M. Henneaux and C. Teitelboim, Quantization of Gauge Systems, Princeton
University Press, Princeton, 1992.
84. J. Henson, The causal set approach to quantum gravity, (2006), arXiv: gr-
qc/0601121.
85. S.D.H. Hsu, Information, information processing and gravity, (2007), arXiv:
0704.1154 [hep-th].
86. B.L. Hu, Can Spacetime be a Condensate, Int. J. Theor. Phys. 44, 1785–
1806 (2005), arXiv: gr-qc/0503067.
87. B.L. Hu, Emergent quantum gravity: Macro/micro structures of space–time,
(2009), arXiv: 0903.0878 [gr-qc].
88. B.L. Hu and E. Verdaguer, Stochastic Gravity: A Primer with Applications,
Classical and Quantum Gravity 20, R1–R42 (2003), arXiv: gr-qc/0211090.
89. B.L. Hu and E. Verdaguer, Stochastic Gravity: Theory and Appli-
cations, Living Reviews in Relativity (Electronic Journal) 7/3 (2004),
www.livingreviews.org; also: arXiv: gr-qc/0307032.
90. B.L. Hu and E. Verdaguer, stochastic gravity: Theory and applications,
(2008), arXiv:0802.0658 [gr-qc].
91. C.J. Isham, Canonical Quantum Gravity and the Problem of Time, in L.A.
Ibort and M.A. Rodriguez (Eds.), Integrable Systems, Quantum Groups,
and Quantum Field Theory, Dordrecht, 1999, 157–288, also: arXiv: gr-
qc/9210011.
92. E. Jacobson, Thermodynamics of Spacetime: The Einstein Equation of
State, Phys. Rev. Lett. 75, 1260–1263 (1995), arXiv: gr-qc/9504004.
93. T. Jacobson, On the Nature of Black Hole Entropy, (1999), arXiv: gr-
qc/9908031.
94. T. Jacobson, Gravitation and vacuum entanglement entropy, (2012),
arXiv:1204. 6349 [gr-qc].
95. T. Jacobson and R. Parentani, Horizon Entropy, Foundations Phys. 33,
323(2003).
96. G. Jannes, On the condensed matter scheme for emergent gravity and inter-
ferometry, (2008), arXiv:0810.0613 [gr-qc].
97. G. Jannes, Emergent gravity: The BEC paradigm, (2009), arXiv: 0907.2839
[gr-qc].
98. M. Kaku, Introduction to Superstrings and M-Theory, 2nd edn., Springer,
New York, 1999.
86 R. Hedrich
99. V. Kaplunovsky and M. Weinstein, Space–Time: Arena or Illusion?, Phys.

Rev. D31, 1879–1898 (1985).
100. C. Kiefer, Probleme der quantengravitation, Philosophia Naturalis 31, 309–
327 (1994).
101. C. Kiefer, Quantum Gravity, 2nd edn., Oxford University Press, Oxford,
2007.
102. C. Kiefer, Quantum Gravity: General Introduction and Recent Develop-
ments, Annalen der Physik 15, 129–148 (2005), arXiv: gr-qc/0508120.
103. C. Kiefer, Quantum geometrodynamics: Whence, whither?, (2008), arXiv:
0812.0295 [gr-qc].
104. T. Konopka, F. Markopoulou and S. Severini, Quantum Graphity: A Model
of Emergent Locality, Phys. Rev. D 77, 104029 (2008), arXiv: 0801.0861
[hep-th].
105. T. Konopka, F. Markopoulou and L. Smolin, Quantum graphity, (2006),
arXiv: hep-th/0611197.
106. D.W. Kribs and F. Markopoulou, Geometry from quantum particles, (2005),
107. K. Kuchar, Canonical geometrodynamics and general covariance, Found.
Phys. 16, 193–208 (1986).
108. K. Kuchar, The Problem of Time in Canonical Quantization of Relativistic
Systems, in [10] pp. 141–171 (1991).
109. K. Kuchar, Time and Interpretation of Quantum Gravity, in G. Kunstat-
ter et al. (eds.), Proceedings of the 4th Canadian Conference on General
Relativity and Relativistic Astrophysics, Singapore, 1992.
110. K. Kuchar, Canonical quantum gravity, (1993), arXiv: gr-qc/9304012.
111. J.P.S. Lemos, Black holes and fundamental physics, (2005), arXiv: gr-
qc/0507101.
112. S. Liberati, F. Girelli and L. Sindoni, Analogue models for emergent gravity,
(2009), arXiv:0909.3834 [gr-qc].
113. E.R. Livine and D. Oriti, Implementing causality in the spin foam quantum
geometry, Nucl. Phys. B 663, 231–279 (2003), arXiv: gr-qc/0210064.
114. E.R. Livine and D.R. Terno, Quantum causal histories in the light of quan-
tum information, Phys. Rev. D 75, 084001 (2007), arXiv: gr-qc/0611135.
115. S. Lloyd, Universe as quantum computer, (1999), arXiv: quant-ph/9912088.
116. S. Lloyd, A theory of quantum gravity based on quantum computation,
(2005), arXiv:quant-ph/0501135.
117. S. Lloyd, Programming the Universe: A Quantum Computer Scientist takes
on the Cosmos, New York, 2007.
118. R. Loll, The emergence of spacetime, or, quantum gravity on your desktop,
(2007), arXiv:0711.0273 [gr-qc].
119. J.M. Maldacena, Black Holes in String Theory, PhD Thesis, (1996), arXiv:
hep-th/9607235.
120. F. Markopoulou, The Internal Description of a Causal Set: What the Uni-
verse looks like from the inside, Commun. Math. Phys. 211, 559 (2000),
www.ebook3000.com
121. F. Markopoulou, Quantum causal histories, Classical and Quantum Gravity

17, 2059 (2000a), arXiv: hep-th/9904009.
122. F. Markopoulou, An insider’s guide to quantum causal histories, Nucl. Phys.
88 (Proc. Suppl.), 308–313 (2000b), arXiv: hep-th/9912137.
123. F. Markopoulou, Planck–Scale Models of the Universe, in [16] pp. 550–563
(2004), arXiv: gr-qc/0210086.
124. F. Markopoulou, Towards gravity from the quantum, (2006), arXiv: hep-
th/0604120.
125. F. Markopoulou, New directions in background independent quantum grav-
ity, (2007), arXiv:gr-qc/0703097.
126. F. Markopoulou, Space does not exist, so time can, (2009), arXiv: 0909.1861
[gr-qc].
127. F. Markopoulou, The computing space–time, Lect. Comp. Sci. 7318, 472
(2012), arXiv:1201.3398 [gr-qc].
128. F. Markopoulou and L. Smolin, Causal evolution of spin networks, Nucl.
Phys. B 508, 409 (1997), arXiv: gr-qc/9702025.
129. F. Markopoulou and L. Smolin, Holography in a Quantum Spacetime,
(1999), arXiv: hep-th/9910146.
130. K. Meissner, Black hole entropy in loop quantum gravity, Classical and
Quantum Gravity 21, 5245–5252 (2004), arXiv: gr-qc/0407052.
131. H. Nicolai and K. Peeters, Loop and spin foam quantum gravity, (2006),
132. H. Nicolai, K. Peeters and M. Zamaklar, Loop quantum gravity: an out-
side view, Classical and Quantum Gravity 22, R193 (2005), arXiv: hep-
th/0501114.
133. J.D. Norton, The hole argument, in A. Fine and J. Leplin (eds.), PSA 1988,
Vol. 2, East Lansing, (1988), pp. 56–64.
134. J.D. Norton, General covariance and the foundations of general relativity,
Rep. Prog. Phys. 56, 791–858 (1993).
135. J.D. Norton, The Hole Argument, in E.N. Zalta (ed.), Stanford Encyclopedia
of Philosophy, 2004. Available on: http://plato.stanford.edu.
136. D. Oriti, Spacetime geometry from algebra: spin foam models for non-
perturbative quantum gravity, Rep. Prog. Phys. 64, 1489 (2001), arXiv:gr-
qc/0106091.
137. D. Oriti, Spin Foam Models of Quantum Spacetime, Ph.D. Thesis, (2003),
arXiv:gr-qc/0311066.
138. D. Oriti, A Quantum Field Theory of Simplicial Geometry and the Emer-
gence of Spacetime, (2006), arXiv: hep-th/0612301.
139. D. Oriti, Disappearance and emergence of space and time in quantum grav-
ity, (2013), arXiv:1302.2849 [physics].
140. T. Padmanabhan, Gravity from Spacetime Thermodynamics, (2002), arXiv:
gr-qc/0209088.
141. T. Padmanabhan, Gravity as elasticity of spacetime: A paradigm to under-
stand horizon thermodynamics and cosmological constant, Int. J. Modern
Phys. D 13, 2293–2298 (2004), arXiv: gr-qc/0408051.
88 R. Hedrich
142. T. Padmanabhan, Gravity as an emergent phenomenon: A conceptual

description, (2007), arXiv:0706.1654 [gr-qc].
143. A.W. Peet, The Bekenstein formula and string theory, Classical and Quan-
tum Gravity 15, 3291–3338 (1998), arXiv: hep-th/9712253.
144. A.W. Peet, TASI lectures on black holes in string theory, in [76], (2001),
arXiv:hep-th/0008241.
145. A. Perez, Spin foam models for quantum gravity, Classical and Quantum
Gravity 20, R43–R104 (2003), arXiv: gr-qc/0301113.
146. A. Perez, The spin-foam-representation of loop quantum gravity, (2006),
147. A. Peres and D.R. Terno, Hybrid classical–quantum dynamics, Phys. Rev.
A 63, 022101 (2001).
148. A. Pesci, From unruh temperature to generalized bousso bound, (2007),
arXiv: 0708.3729 [gr-qc].
149. A. Pesci, On the statistical-mechanical meaning of bousso bound, (2008),
arXiv: 0803.2642 [gr-qc].
150. J.G. Polchinski, String Theory. Vol. 1: An Introduction to the Bosonic
String, Cambridge University Press, Cambridge, 2000.
151. J.G. Polchinski, String Theory. Vol. 2: Superstring Theory and Beyond,
Cambridge University Press, Cambridge, 2000.
152. J.M. Pons and D.C. Salisbury, The Issue of Time in Generally Covariant
Theories and the Komar–Bergmann Approach to Observables in General
Relativity, Phys. Rev. D 71, 124012 (2005), arXiv: gr-qc/0503013.
153. O. Pooley, A Hole Revolution, or are we back where we started, Studies in
His. Phil. Modern Phys. 37, 372–380 (2006).
154. W. Poundstone, The Recursive Universe — Cosmic Complexity and the
Limits of Scientific Knowledge, William Morrow & Co., Chicago, 1985.
155. M. Requardt, Discrete Mathematics and Physics on the Planck–Scale exem-
plified by means of a Class of ‘Cellular Network Models’ and their Dynamics,
(1996), arXiv: hep-th/9605103.
156. M. Requardt, Emergence of Space–Time on the Planck–Scale Described as
an Unfolding Phase Transition within the Scheme of Dynamical Cellular
Networks and Random Graphs, (1996a), arXiv:hep-th/9610055.
157. M. Requardt, Let’s call it nonlocal quantum physics, (2000), arXiv: gr-qc/
0006063.
158. M. Requardt, Quantum Theory as emergent from an undulatory translocal
Sub-Quantum Level, (2012), arXiv:1205.1619 [quant-ph].
159. D. Rickles, Interpreting Quantum Gravity. Available on: http://philsci-
archive.pitt.edu, Document: 2407, 2005.
160. D. Rickles and S. French, Quantum Gravity Meets Structuralism: Inter-
weaving Relations in the Foundations of Physics, in [161] pp. 1–39 (2006).
161. D. Rickles, S. French, and J. Saatsi (Eds.), The Structural Foundations of
Quantum Gravity, Clarendon Press, Oxford, 2006.
162. D.P. Rideout, Dynamics of Causal Sets, (2002), arXiv: gr-qc/0212064.
163. D.P. Rideout and R. D. Sorkin, A classical sequential growth dynamics for
causal sets, Phys. Rev. D 61, 024002 (2000), arXiv: gr-qc/9904062.
www.ebook3000.com
164. D.P. Rideout and R.D. Sorkin, Evidence for a continuum limit in causal set
dynamics, Phys. Rev. D 63, 104011 (2001), arXiv: gr-qc/0003117.
165. D. Rideout and P. Wallden, Emergence of spatial structure from causal sets,
(2009), arXiv:0905.0017 [gr-qc].
166. P.J. Riggs, Spacetime or quantum particles: The ontology of quantum grav-
ity, in P.J. Riggs (ed.), Natural Kinds, Laws of Nature and Scientific Method-
ology, Kluwer, Dordrecht, 1996.
167. C. Rovelli, Time in quantum gravity: an hypothesis, Phys. Rev. D 43, 442–
456 (1991).
168. C. Rovelli, Is there Incompatibility between the Way Time is treated in
General Relativity and in Standard Quantum Mechanics, in A. Ashtekar and
J. Stachel (Eds.), Conceptual Problems of Quantum Gravity, Birkhäuser,
Boston, 1991, 126–140.
169. C. Rovelli, Loop Representation in Quantum Gravity: The Transform
Approach, in [10] pp. 427–439 (1991b).
170. C. Rovelli, Loop Quantum Gravity, Living Rev. Relat. (Electronic Journal).
Available on: www.livingreviews.org, (1997), arXiv: gr-qc/9710008.
171. C. Rovelli, Strings, Loops, and the Others: A Critical Survey on the Present
Approaches to Quantum Gravity, in N. Dadhich and J. Narlikar (eds.),
Gravitation and Relativity: At the Turn of the Millenium, Poona University
Press, Poona, 1998, pp. 281–331, also: arXiv: gr-qc/9803024.
172. C. Rovelli, Notes for a Brief History of Quantum Gravity, (2000), arXiv:
gr-qc/0006061.
173. C. Rovelli, Quantum Spacetime: What do we know?, in [43]; (2001), also:
174. C. Rovelli, Partial Observables, Physical Review D65, 124013 (2002),
175. C. Rovelli, A Dialog on quantum gravity, Int. J. Modern Phys. 12, 1 (2003),
176. C. Rovelli, Quantum Gravity, Cambridge, 2004.
177. C. Rovelli, Unfinished Revolution, (2006), arXiv: gr-qc/0604045.
178. C. Rovelli, Quantum Gravity, in [63], 2007.
179. C. Rovelli, Forget Time, (2009), arXiv: 0903.3832 [gr-qc].
180. C. Rovelli, A new look at loop quantum gravity, Classical and Quantum
Gravity 28, 114005 (2011), arXiv:1004.1780 [gr-qc].
181. C. Rovelli, Loop quantum gravity: The first twenty five years, Classical and
Quantum Gravity 28, 153002 (2011a), arXiv:1012.4707 [gr-qc].
182. C. Rovelli, Zakopane lectures on loop gravity, (2011b), arXiv:1102.3660 [gr-
qc].
183. A.D. Sakharov, Vacuum quantum fluctuations in curved space and the the-
ory of gravitation, General Relativity and Gravitation 32, 365–367 (2000)
(Reprint; Original: Doklady Akademii Nauk SSSR 177, 70–71 (1967)/Soviet
Physics Doklady 12, 1040–1041) (1968).
184. L. Sindoni, Emergent Models for Gravity: an Overview of Microscopic Mod-
els, (2011), arXiv:1110.0686 [gr-qc].
90 R. Hedrich
185. L. Sindoni, F. Girelli and S. Liberati, Emergent Gravitational Dynamics in

Bose–Einstein Condensates, (2009), arXiv: 0909.5391 [gr-qc].
186. L. Smolin, Nonperturbative Quantum Gravity via the Loop Representation,
in [10] pp. 440–489 (1991).
187. L. Smolin, Three Roads to Quantum Gravity, Weidenfeld & Nicolson, Lon-
don, 2000.
188. L. Smolin, How far are we from the quantum theory of gravity?, (2003),
189. R.D. Sorkin, Causal sets: Discrete gravity, (2003), arXiv: gr-qc/0309009.
190. J. Stachel, Structure, individuality and quantum gravity, in [161] pp. 53–82
(2006), also: arXiv: gr-qc/0507078.
191. S. Surya, Causal Set Topology, (2007), arXiv: 0712.1648 [gr-qc].
192. M.O. Tahim, R.R. Landim and C.A.S. Almeida, Spacetime as a deformable
solid, (2007), arXiv:0705.4120 [gr-qc].
193. D.R. Terno, Inconsistency of quantum–classical dynamics, and what it
implies, Found. Phys. 36, 102–111 (2006), arXiv: quant-ph/0402092.
194. T. Thiemann, Introduction to Modern Canonical Quantum General Relativ-
ity, (2001), arXiv:gr-qc/0110034.
195. T. Thiemann, Lectures on loop quantum gravity, (2002), arXiv: gr-
qc/0210094.
196. T. Thiemann, Loop quantum gravity: An inside view, (2006), arXiv: hep-
th/0608210.
197. W. Unruh and R. Wald, Time and the interpretation of canonical quantum
gravity, Phys. Rev. D 40, 2598–2614 (1989).
198. M. Visser, Sakharov’s Induced Gravity: a Modern Perspective, Modern
Phys. Lett. A 17, 977–992 (2002), arXiv: gr-qc/0204062.
199. G.E. Volovik, Links between gravity and dynamics of quantum liquids,
(2000), arXiv:gr-qc/0004049.
200. G.E. Volovik, Superfluid analogies of cosmological phenomena, Phys. Rep.
352,195–348 (2001), arXiv: gr-qc/0005091.
201. G.E. Volovik, The Universe in a Helium Droplet, Clarendon Press, Oxford,
2003.
202. G.E. Volovik, From quantum hydrodynamics to quantum gravity, (2006),
203. G.E. Volovik, Fermi–Point scenario for emergent gravity, (2007), arXiv:
0709.1258 [gr-qc].
204. G.E. Volovik, Emergent physics: Fermi point scenario, (2008), arXiv:
0801.0724 [gr-qc].
205. G.E. Volovik, From analogue models to gravitating vacuum, (2011),
arXiv:1111. 1155 [gr-qc].
206. G.E. Volovik, The Superfluid Universe, in K.H. Bennemann and J.B. Ket-
terson, (Eds.), Novel Superfluids, International Series of Monographs on
Physics, Vol. 1, 156, 570–618 (2013), arXiv:1004.0597 [cond-mat].
207. R.M. Wald, Quantum Field Theory on Curved Spacetime and Black Hole
Thermodynamics, University of Chicago Press, Chicago, 1994.
www.ebook3000.com
208. R.M. Wald, The Thermodynamics of Black Holes, Living Reviews in Rela-
tivity (Electronic Journal) 4/6 (2001). Available at: www.livingreviews.org;
also arXiv:gr-qc/9912119.
209. S.C.E. Weinfurtner, Emergent Spacetimes, PhD Thesis, (2007), arXiv:
0711.4416 [gr-qc].
210. J.A. Wheeler, Frontiers of Time, in N. Toraldo di Francia (Ed.), Problems
in the Foundations of Physics. Proceedings of the International School of
Physics ‘EnricoFermi’, Course 72, Amsterdam, 1979.
211. J.A. Wheeler, Law without Law, in J.A. Wheeler and W.H. Zurek (eds.),
Quantum Theory and Measurement, Princeton, N.J., 1983.
212. J.A. Wheeler, Information, Physics, Quantum: The Search for Links, in
Proceedings 3rd International Symposium on the Foundation of Quantum
Mechanics, Physical Society of Japan, Tokyo, 1989, pp. 354–368; also in
W.H. Zurek (Ed.), Complexity, Entropy and the Physics of Information,
Westview Press, New York, 1990, pp. 3–28.
213. S. Wolfram, A New Kind of Science Wolfram Media, Champaign, Ill., 2002.
214. C. Wüthrich, Raiders of the lost space–time, (2014), arXiv:1405.5552
[physics].
215. S.-C. Zhang, To see a World in a Grain of Sand, (2002), arXiv: hep-
th/0210162.
216. P.A. Zizzi, Quantum computation toward quantum gravity, General Rela-
tivity and Gravitation 33, 1305–1318 (2001), arXiv: gr-qc/0008049.
217. P.A. Zizzi, Computability at the Planck Scale, (2004), arXiv: gr-qc/0412076.
218. P.A. Zizzi, A minimal model for quantum gravity, Modern Phys. Lett. A 20,
645–653, 2005, arXiv: gr-qc/0409069.
www.ebook3000.com
Chapter 4
Introduction to the Quantum

Theory of Elementary Cycles
Donatello Dolce
The University of Camerino, Piazza Cavour 19F,
62023 Camerino, Italy
donatello.dolce@unicam.it
Elementary Cycles Theory (ECT) is a novel exact formulation of

quantum-relativistic mechanics. Here, we present an introduction to its
basic quantum aspects. On the one hand, Newton’s law of inertia states
that every isolated particle has persistent motion, i.e. constant energy
and momentum. On the other hand, undulatory mechanics associates,
by means of the Planck constant, a recurrence in time and space to
the energy and the momentum of an elementary particle, respectively.
Paraphrasing these two fundamental principles of modern physics, ECT
postulates that every elementary constituent of nature (every elemen-
tary particle) is characterized by persistent intrinsic periodicity (as long
it does not interact) whose space–time duration determines its kinemati-
cal state (energy and momentum). In other words, undulatory mechanics
is imposed as constraint “overdetermining” relativistic mechanics, with
fundamental motivations on Einstein’s proposal of unification of quan-
tum and relativistic theories. Every free particle is a (de Broglie) “peri-
odic phenomenon” which can also be regarded as a reference clock and
every system is decomposable in modulated elementary cycles. Indeed,
ECT introduces a cyclic nature to the ordinary relativistic space–time
coordinates. The resulting classical-relativistic mechanics turns out to be
fully consistent with relativity and, at the same time, reproduces exactly
all the fundamental aspects of ordinary quantum-relativistic mechanics
(without any explicit quantisation). Relativity only fixes the differen-
tial structure of space–time without giving any prescription about the
boundary of space–time, and the constraint of covariant periodicity (or
similar relativistic boundary conditions) is allowed by the variational
principle for relativistic theories. The constraint of intrinsic periodicity
enforces the local nature of relativistic space–time and the wave-particle
duality. Besides such unified description of relativistic and quantum
93
94 D. Dolce
dynamics, ECT allows for a geometrodynamical formulation of gauge

interaction which, similarly to gravity, is represented by modulations of
space–time periodicities and brings novel elements to address fundamen-
tal open problems of modern physics.
1. Introduction
Elementary Cycles Theory (ECT) is a formulation of Quantum Mechanics
(QM), Gauge interaction, and other aspects of modern physics defined in
recent peer-reviewed papers [1–13]. It is based on the empirical fact that, as
noticed by Louis de Broglie about the wave-particle duality [14,15], elemen-
tary particles, i.e. the elementary constituents of nature, have recurrences
(periodicities) in time and space determined by their energy and momen-
tum, through the Planck constant h = 2π. This implies that every system
in nature can be consistently described in terms of elementary space–time
cycles. Paraphrasing Newton’s first law (a free particle has constant energy–
momentum if viewed from an inertial frame) and de Broglie wave-particle
duality originally formulated in his PhD thesis [14,15] in terms of “periodic
phenomena” (duality between energy–momentum and space–time quantum
recurrence), we will see that classical-relativistic physics can be quantized
by postulating that
Every free elementary particle, when observed from an inertial frame,

is an intrinsically “periodic phenomenon” whose persistent spatial and
temporal recurrence is determined through the Planck constant by its
persistent momentum and energy, respectively.
These recurrences must be imposed as quantization conditions, i.e.

covariant Periodic Boundary Conditions (PBCs), “overdetermining” the
relativistic dynamics — a sort of relativistic generalization of the quanti-
zation of particle in a box in which the Boundary Conditions (BCs) quan-
tize the system. As we will describe, the resulting — cyclic — relativistic
mechanics are formally equivalent to QM in both the canonical and the
Feynman formulations [1–13].
In ECT, every particle is described as an elementary clock [7,9], in anal-
ogy with de Broglie’s conjecture of the internal clock of particles; see also
the de Broglie “periodic phenomenon” [16], and (for Dirac particles) the zit-
terbewegung proposed by Schrödinger. A similar description of elementary
particles as clocks has also been recently adopted by Roger Penrose in his
book “Cycles of Time: An Extraordinary New View of the Universe” [17]
www.ebook3000.com
Introduction to the Quantum Theory of Elementary Cycles 95
and indirectly investigated by Cattillon and Müller experimental groups

[18, 19].
Among the original motivations of ECT, we have ’t Hooft’s Cellular
Automata [20–25]. ’t Hooft has proven that the there is a close relation-
ship between the cyclic temporal dynamics of a particle moving on a circle
(“continuous cogwheel model”, the analogous of a particle in a periodic time
box) and the time evolution of the quantum harmonic oscillator. Besides
this, ECT has remarkable historical justifications in the ideas of some of the
founding fathers of QM (de Broglie, Bohr, Planck, Einstein, Schrödinger,
Feynman, Fermi, Sommerfeld, Dirac, Klein, Kaluza, etc.). Notably, it seems
to realize fundamental aspects of Einstein’s original proposal to unify quan-
tum and relativistic mechanics by means of relativistic constraints (corre-
sponding to the space–time covariant PBCs in the ECT) “overdetermining”
the relativistic differential equations. Furthermore, it can be regarded as
an evolution (generalization) of the Bohr–Sommerfeld quantization or the
WKB method to solve quantum problems.
ECT has important confirmations in modern theoretical physics, such
as a geometrodynamical description of gauge invariance [1, 5, 7, 8] in deep
analogy with general relativity and a mathematical demonstration of the
so-called Anti-de Sitter/Conformal Field Theory (AdS/CFT) correspon-
dence [6, 8, 13]. Its validity has also been successfully tested in condensed
matter physics in which it has allowed for a novel, intuitive derivation of
superconductivity and related phenomenology from first principles of QM,
as well as of the electric properties of carbon nanotubes [2, 26, 27]. By con-
struction, ECT does not involve hidden variables (thus, ECT formulation
of QM is not limited by Bell’s or similar no-go theorems) and suggests that
QM could emerge as a statistical description of the deterministic ultra-
fast cyclic dynamics associated to particles’ internal clocks. The idea was
awarded by the IV prize in the 2011 FQXi contest with the essay “Clock-
work quantum universe” [9]. The cyclic character of space–time proposed
in ECT offers the possibility of a relational, emergent description of the
time flow (with some analogies to the ideas of Kastner and Rovelli [3, 28]).
The author’s intention is to write the chapter in such a way that it
can be adopted as an introduction to QM by students and non-experts on
the subject. Indeed, ECT offers a consistent, simple (“but not simpler”
Einstein) formulation of QM, based on intuitive, natural physical principles:
Every system in nature is described by elementary cycles. The chapter is
structured in the following way: We will report de Broglie’s undulatory
mechanics by introducing a relativistic formalism; we will define ECT by
96 D. Dolce
means of the postulate of intrinsic periodicity reporting some historical

motivations for such an assumption (de Broglie, Einstein, Penrose, ’t Hooft,
etc.); we will derive the axioms of QM and the Feynman path integral
from our assumption of intrinsic periodicity; we will give examples of its
application in the solution of textbook problems of QM; finally, we will
mention some applications in modern physics and conceptual implications.
2. Overview
2.1. Basic formalism of undulatory mechanics
and space–time phase harmony
ECT is based on the undulatory mechanics and the related concept of
phase harmony. Here, we essentially report wave-particle duality as origi-
nally described in de Broglie’s PhD thesis [14, 15]. It is however convenient
to introduce a covariant notation constituting the basic formalism of ECT.
On the one hand, according to undulatory mechanics, the Planck con-
stant relates the energy E and the momentum p to a time periodicity T and
spatial wave-length (spatial periodicity) λ, also known as de Broglie–Planck
relations
h h
T = , λi = , (i = 1, 2, 3). (1)
E pi
On the other hand, relativity relates, through the speed of light c, the
mass M to the rest energy E(0), according to the universally known rela-
tion E(0) = M c2 . Hence, the combination of undulatory mechanics and
relativity implies that to the mass is associated a rest time periodicity, also
known as Compton periodicity,
h
TC = . (2)
M c2
In relativity, the energy–momentum of a particle in a given reference
frame is a derivable from its mass by means of the Lorentz transformations

E
M → pµ = , −
p , (3)
c
where E = γM c2 and = {p1 , p2 , p3 }.

p = M cγ β
Similarly, the temporal and spatial recurrences can be written in a con-
travariant (space-like, tangent) four-vector

λ
µ
T = T, . (4)
c
www.ebook3000.com
We will address it as space–time periodicity (four-periodicity). In analogy

with the four-momentum, as noticed by de Broglie, the space–time peri-
odicity is a derivable from the Compton periodicity by means of Lorentz
transformations: TC → T µ such that
· λ.
TC c = γT c − γ β (5)
µ
For the same reason, the phase of relativistic (matter) waves e−ipµ x / is
an invariant: M c2 τ = pµ xµ = const; τ is the proper-time and xµ = {ct, x}.
By substituting TC in the Compton relation (2) we obtain the invariant
relation, also known as de Broglie’s phase harmony relation
M c2 TC = γM c2 T − M cγ β
· λ = ET − p · λ = cpµ T µ = h. (6)
The energy–momentum and space–time periodicity of a particle in a generic

reference frame — or the mass and the Compton period for a rest particle —
are therefore determined by the covariant relation
M c2 TC = h → cpµ T µ = h. (7)
The phase harmony implies that the momentum–energy and the space–
time periodicity are dual quantities — they are two faces of the same coin.
This is the basic meaning of undulatory mechanics and of the wave-particle
duality expressed by (1). For instance, as it can be easily checked dividing
by 2 , the relativistic dispersion relation of the energy–momentum can be
equivalently expressed in terms of a relativistic dispersion relation of the
space–time periodicity
3
1 1 1 1 c2
M 2 c4 = pµ pµ = E 2 −
p 2 c2 ←→ = = − . (8)
Tc2 T µ Tµ T 2 i=1 (λi )2
This means that the temporal period T is modulated under transformations

of reference frames according to the relativistic Doppler effect.
2.2. Postulate of elementary space–time cycles

Classical-relativistic mechanics is based on simple postulates, such as New-
ton’s laws and Einstein’s principle of equivalence, with universally accepted
physical meaning. The other pillar of modern physics, i.e. QM, has an
axiomatic formulation (the axioms derived from EC in this paper) whose
physical meaning is still largely debated by experts on the field. Further-
more, in extreme physical systems, it is difficult to conciliate together rel-
ativity and QM. It is not a case that some of the most eminent physicists
98 D. Dolce
(e.g. Einstein, Feynman, or more recently ’t Hooft, Weinberg, Wilczek) have

expressed the necessity of a deeper compression of QM.
A fascinating physical formulation of QM emerges by combining
together Newton’s law of inertia and the undulatory mechanics of elemen-
tary particles described above. From these it immediately follows that ele-
mentary particles of persistent energy and momentum, i.e. free particles,
must have persistent recurrences in time and space. Newton’s law of inertia
describes an ideal case. In everyday life, we are used to objects that do not
have uniform rectilinear motion because of interactions (local variations of
energy–momentum), and that come to rest because of friction (see concept
of inertia in Aristotle). Similarly, as we will discuss, the intrinsic periodicity
proposed in ECT (implicit in the wave-particle duality) is an ideal condition
typical of pure quantum systems: e.g. free relativistic particles, high ener-
gies, small distances, or very low temperatures. In everyday life, of course,
we do not experience a perfectly periodic world because ordinary objects are
composed by a large number of elementary particles (even neglecting inter-
actions, a set of persistent “periodic phenomena” forms an ergodic system),
because these recurrences are typically very small [9, 13] (the Planck con-
stant is so “small” that it can be approximated to zero), or there are inter-
actions implying local modulations of the space–time recurrences [1, 7, 8]
(thus a very complex and chaotic evolution for a system of interacting ele-
mentary cycles), and because the thermal noise implies chaotic interactions
(Brownian motion) which tend to destroy the de Broglie–Planck periodicity
(perfect coherence) of pure quantum systems [2,26,27]. As we will argue, the
ideal case of elementary particle’s perfect recurrences describes the quantum
world with all its peculiar phenomena whereas the classical limit is obtained
when these recurrences can be neglected in the effective description.
ECT, in addition to the ordinary postulates of classical–relativistic
mechanics, introduces one simple postulate of intrinsic periodicity from
which the whole construction of the theory is derived. In particular, in
ECT, it is not necessary to introduce the axioms of QM; these are inferred
from the postulate. Intrinsic periodicity represents the quantization con-
dition for the classical–relativistic mechanics and no further quantisation
conditions are required. The postulate of intrinsic periodicity (elementary
cycles) in ECT can be enunciated in three equivalent ways:
• An elementary particle of mass M is an intrinsically “periodic phe-

nomenon” of Compton periodicity TC = h/M c2 .
www.ebook3000.com
• An elementary particle of mass M is a reference relativistic clock of

Compton periodicity TC = h/M c2 .
• An elementary particle of mass M is a vibrating string of fundamental
Compton periodicity TC = h/M c2 .
This also represents the definition of “elementary particle” in ECT. An ele-

mentary particle is the ontological entity described by a fundamental peri-
odic phenomenon. In terms of the wave-particle duality, i.e. the relationship
between Compton periodicity and rest mass allowed by the Planck constant,
the classical-relativistic counterpart of this postulate is that a particle at
rest has a mass determined by the Compton periodicity, M = h/TC c2 .
According to the covariant description of undulatory mechanics given
above, the postulate of elementary cycles can be generalized to isolated
particles viewed from a generic inertial reference frames
• A free elementary particle of four-momentum pµ , observed from an iner-

tial reference frame, is an intrinsic “periodic phenomenon” / reference
clock / vibrating-string of persistent de Broglie–Planck space–time peri-
odicity T µ , according to the phase harmony relation cpµ T µ = h.
The classical-relativistic counterpart of this principle is that every isolated

relativistic particle has constant energy–momentum. That is, according to
Newton, every isolated relativistic particle moves in uniform rectilinear
motion (as long as it does not interact). In fact, by means of the Planck
constant, a periodic phenomenon with persistent space–time periodicity T µ
has constant energy–momentum pµ , see phase harmony relation (7) or (1).
2.2.1. “Periodic phenomena”

The first enunciation of the postulate of intrinsic periodicity pays tribute to
de Broglie. In his seminal PhD thesis [14,15] he in fact stated the hypothesis
of wave-particle duality in this form:
“To each isolated parcel of energy [elementary particle] with a proper

mass M , one may associate a periodic phenomenon of [Compton] peri-
odicity TC = h/M c2 . The [Compton] periodicity is to be measured, of
course, in the rest frame of the particle. This hypothesis is the basis of
our theory: it is worth as much, like all hypotheses, as can be deduced
from its consequences” de Broglie (1924).
100 D. Dolce
The postulate of elementary cycles enforces the wave-particle duality.

It brings de Broglie’s hypothesis of wave-particle duality to its extreme
consequences, as relativity brings to the extreme consequences the hypoth-
esis that the speed of light is constant. Indeed, these recurrences will be
imposed as constraints to “overdetermine” relativistic dynamics. We will
see how much can be deduced from it — we will find a formal equivalence
with ordinary quantum-relativistic mechanics.
2.2.2. Reference clocks

The second way to postulate the principle of intrinsic periodicity in ECT is
indirectly suggested by de Broglie. He pointed out that the phase harmony
described above (wave-particle duality), which associates to every parti-
cle a periodic phenomenon, implies that every particle can be regarded as
having an “internal clock” of rest periodicity TC . The fact that a periodic
phenomenon of persistent periodicity, i.e. an elementary particle, can be
regarded as an elementary clock follows clearly from Einstein’s definition
of relativistic clock [29]:
“By a clock we understand anything characterised by a phenomenon
passing periodically through identical phases so that we must assume,
by the principle of sufficient reason, that all that happens in a given
period is identical with all that happens in an arbitrary period” Einstein
(1910).
Such a description of particles as elementary clocks has been, e.g.

recently adopted by Penrose which, in analogy to our postulate, wrote
“for there is a clear sense in which any individual (stable) [isolated and at
rest] massive particle plays a role as a virtually perfect clock. [. . . ] In
other words, any stable massive particle behaves as a very precise quan-
tum clock, which ticks away with [Compton periodicity]” Penrose (2011).
Massless particles have infinite Compton periodicity. This corresponds

to say that massless particles can have arbitrary large space–time periodic-
ities, that is, arbitrary small energy–momentum — in ECT they define the
emphatically non-compact space–time typical of the ordinary interpreta-
tion of relativity [1, 7, 8]. Hence, mass is crucial for a consistent description
of time flow in ECT. We say that massless particles, due to their infinite
Compton periodicity, have “frozen” internal clocks. As noted by Penrose:
“Massless particles (e.g. photons or gravitons) alone cannot be used to
make a clock, because their frequencies would have to be zero; a photon
www.ebook3000.com
would take until eternity before its internal ‘clock’ gets even to its first
‘tick’ ! To put this another way, it would appear that rest-mass is a nec-
essary ingredient for the building of a clock, so if eventually there is little
around which has any rest-mass, the capacity for making measurements
of the passage of time would be lost” Penrose (2011).
That is to say, ordinary relativistic space–time — Minkowskian space

and time are intimately mixed — by itself does not provide sufficient ele-
ments to understand the flow of time in physical systems, e.g. see the prob-
lem of the arrow of time (this has brought some authors to reject the notion
of time in physics, e.g. Rovelli or Barbour). Nevertheless, to have a consis-
tent description of the relativistic time flow it is important to consider that
some elementary particles acquire mass, i.e. a finite Compton periodicity,
otherwise all the particles would be on the light-cone with “frozen” inter-
nal clocks — it is necessary that nature defines clocks, through particles’
masses, in addition to the Minkowskian time coordinate.
2.2.3. Vibrating strings

The term “periodic phenomenon” used in the first enunciation of the pos-
tulate may seem to be vague — or it may be erroneously confused to
the “wave” of the semi-classical formulation of QM. Its meaning is speci-
fied by the third enunciation of the postulate: an intrinsic “periodic phe-
nomenon” of given periodicity is a vibrating string, i.e. the set of all the
harmonic eigenmodes allowed by a fundamental period, Fig. 1. In general,
by means of discrete Fourier transform, a periodic phenomenon of finite
period (e.g. a periodic function) can be represented as a sum of harmonic
eigenmodes. The difference with ordinary undulatory mechanics is thus
that we impose intrinsic periodicity as a constraint to “overdetermine” the
relativistic dynamics of the free elementary particle. A free particle, and
its quantum excitations, will be described as vibrations associated to the
space–time periodicities.
In the semi-classical formulation of QM or in field theory, such a
“periodic phenomenon” associated to a particle is typically limited to
the monochromatic wave of corresponding angular frequency ω = 2π/T
and angular wave-number ki = 2π/λi . This wave is then typically quan-
tized by imposing commutation relations in order to obtain the quantum
behavior of a free particle whose (normally ordered) energy quantization
is E = nω (i.e. the energy spectrum of a normally ordered quantum har-
monic oscillator). This is essentially the so-called second quantization of
102 D. Dolce
Fig. 1. Harmonic vibrational modes of a string of fundamental period T . The

state of a vibrating string can be represented as a point in a Hilbert space. The
decomposition of a vibrating string into its vibrations in distinct harmonics is
given by the projection of the point onto the coordinate axes in the Hilbert
space. The harmonic frequency spectrum fn = nf = n/T of a vibrating string is
described by a corresponding “frequency” operator defined in the corresponding
Hilbert space.
ordinary Quantum Field Theory (QFT). In ECT, the most general peri-
odic phenomenon of periodicity T = 2π/ω is not a simple monochromatic
wave, but it is represented by all possible harmonics, i.e. wave components
with discretized angular frequencies ωn = nω, Fig. 1. For instance, we may
note that this, through the Planck constant (without introducing any fur-
ther quantization condition), provides the same energy spectrum of the free
particle En = ωn = nω prescribed by ordinary (normally ordered) QFT.
In general, these higher harmonics describe the quantum excitations of a
particles (e.g. multiparticle states) whereas the fundamental harmonic is
the ordinary de Broglie matter wave of ordinary undulatory mechanics or
field theory — the negative vibrational modes correspond to antimatter
(“anti-particles”). Note that persistent periodicity is limited to free, iso-
lated quantum-relativistic particle (principle of inertia). Interactions, i.e.
local variations of the local energy–momentum of the particle, imply (phase
harmony) corresponding local modulations of space–time periodicity and
thus local deformations of the corresponding harmonic set — similar to a
non-homogeneous string.
The classical-particle description is obtained from the non-relativistic
limit of this pure “periodic phenomenon.” As we will see, in the non-
relativistic limit, the rest energy E(0) = M c2 can be neglected (infinite
www.ebook3000.com
energy gap). Thus, in the non-relativistic limit the Compton periodicity

TC tends to zero. It can be neglected as well, and the quantization associ-
ated to the “periodic phenomenon” is lost. So, a free non-relativistic particle
is effectively described by continuous energy and momentum spectra. For
interacting non-relativistic particles (e.g. bounded in a potential), the BCs
quantizing the systems, and the corresponding harmonic sets, turn out to
be determined by the geometry or by the potential (e.g. as for a particle
in a box or the atomic orbitals). Intrinsic periodicity in this case corre-
sponds to the requirement that only closed orbits (with integer number of
recurrences) are allowed.
2.2.4. Einstein’s “overdetermination” of relativistic

mechanics and “supercausality”
As already mentioned, in ETC, the constraint of intrinsic periodicity is
imposed to quantize relativistic dynamics. Interesting enough, the idea to
“overdetermine” relativistic differential equations with covariant — rela-
tivistic — constraints in order to obtain a unified description of relativistic
and quantum dynamics was originally proposed by Einstein. As reported
by Pais, Einstein was convinced that “it is necessary to start from clas-
sical [relativistic] field theories [e.g. undulatory mechanics] and ask that
quantum laws emerge as [from] constraints imposed to them” [30]. Einstein
wrote [31] [translated by the author]
“For sure; [in order to describe quantum states] we must just overdeter-
mine the variables of the [relativistic] field [i.e. the undulatory mechanics
of a particle] by means of constraints. [...] The [relativistic] dynamics of
the particles would be overdetermined in such a way that the initial
conditions would be subject to restrictive constraints” Einstein (1923).
Then he added some requirements for these constraints, which are actu-
ally satisfied by the constraint of intrinsic periodicity in ECT, such as
covariance and compatibility with electromagnetism and gravity.
Einstein considered the example of the atomic orbitals. Although this
case concerns interaction it is easy to see that the atomic orbitals can be
interpreted in terms of intrinsic periodicity. As we will argue rigorously in
Sec. 4.4, similar to Bohr’s description, the atomic orbitals are the general-
ization to interactions of the harmonics of a (non-homogeneous) vibrating
string. In fact, in the atomic orbitals the wave-function of the electron is
constrained to be periodic — integer number of space–time recurrences —
104 D. Dolce
in the Coulomb potential. The frequency spectrum, and thus the Bohr spec-
trum (or the quantization of the angular momentum), is determined by the
condition that the wave-function is closed in its space–time orbits similar
to the harmonics of a vibrating string. In other words, the atomic orbitals
can be thought of as the vibrational modes of a locally deformed (according
to the Coulomb potential) vibrating string.
The term “overdetermination” used by Einstein means that these con-
straints, encoding quantum dynamics into relativistic mechanics, must be
added to (or must replace) the ordinary initial (and final) conditions of
ordinary classical–relativistic theories, (i.e. the stationary BCs of classical
theories). For instance, for a free classical-relativistic particle, in order to
satisfy the variational principle at the boundary, it is typically assumed
null variations at the extremal times of its evolution (note that this yields a
tautology in classical mechanics: How can the particle possibly know at the
beginning of its evolution the final time at which its trajectory will have null
variation?). This also fixes the energy of the classical trajectory. In ECT,
for every given initial time and energy of the classical particle we impose
the corresponding PBCs according to the phase harmony (here the tautol-
ogy associated to the stationary BCs is solved). This “overdetermination”
quantizes the system, see, for instance, the simple case of the Black-Body
radiation in Sec. 3.1.1.
Einstein’s proposal of “overdetermination” indeed allows us to introduce
some advanced mathematical aspects of ECT — though these are not rele-
vant for the pedagogical scope of this paper. From a mathematical point of
view, ECT introduces covariant PBCs to the relativistic space–time coordi-
nates. These PBCs “overdetermining” the relativistic differential equations
satisfy the variational principle of the corresponding relativistic actions —
exactly as the stationary BCs of ordinary classical and quantum theories.
The compatibility with the variational principle guarantees that these con-
straints preserve all the fundamental properties of a relativistic theory, e.g.
the covariance and causality. This also implies that new — relativistic —
phenomena emerge along with the purely relativistic ones. ECT shows that
the manifestation of these relativistic cyclic dynamics is indistinguishable
from ordinary quantum-relativistic dynamics in all its fundamental phe-
nomenology.
Note that in this chapter we will only consider PBCs as we will not
discuss spin, but different kinds of BCs (satisfying the variational princi-
ple) are allowed by different kinds of relativistic dynamics — e.g. Dirac
dynamics. We have seen that in the free case the space–time period T µ
www.ebook3000.com
transforms in a contravariant way. Indeed, it is easy to show that these

PBCs imposed as constraint to a bosonic relativistic theory satisfy the
variational principle [7, 13]. Actually, ECT has full compatibility with both
special and general relativity. This is because relativity only fixes the dif-
ferential structure of space–time without concerning about the boundaries.
For instance, once the rate of a clock is fixed in a given reference system
(e.g. the rest frame), relativity only prescribes how this rate varies in the
other reference frames. In physics as in mathematics, a well formulated
problem needs differential equations and BCs, but relativity is not able
to answer to the simple question: where is the boundary of space–time?
It only fixes, through the metric, the differential structure of space–time,
i.e. how the periodicities of the clock varies. Actually, special and gen-
eral relativity was originally conceived by Einstein in terms of relativistic
clock modulations. This analogy between particles, i.e. elementary space–
time cycles and relativistic clock, testimonies the full mathematical com-
patibility of ECT with special and general relativity. This also allows us
to understand causality in ECT — which in the case of static time peri-
odicity would be manifestly violated as, for instance, in Closed Timelike
Curves theory. The essential point to bear in mind here is that the space–
time periodicity of a particle varies in a controvariant way with respect its
energy–momentum. That is, they are determined by the phase harmony
condition (de Broglie–Planck relation). The energy propagates according
to the retarded relativistic potential so that, when it is absorbed by a par-
ticle during interaction, the periodicity (i.e. together with the energy) of
the particle is modulated in a local, retarded way [3, 7–9, 13]. We will give
some more detail about this when we discuss the time flow in ECT, Sec. 5.3.
Actually, by using Einstein’s terminology [30, 31], we can say that the con-
straint of intrinsic periodicity yields a “supercausality”. Indeed, in ECT
the local nature of space–time is enforced, together with the wave-particle
duality.
2.2.5. Elementary space–time cycles

In ECT, relativistic elementary particles are described as embedded in
cyclic space–time dimensions (more in general compact space–time dimen-
sions with different type of topologies) whose local space–time periodicity
is determined, by means of Lorentz transformations, by the Compton peri-
odicity, i.e. by the rest energy (mass). In this way, undulatory mechanics —
and, as we will see, QM — are directly encoded into the relativistic
106 D. Dolce
mechanics. Since every system in nature is totally composed by elementary

particles and every elementary particle is a periodic phenomenon (with
modulations determined by interactions), it follows that every system in
nature can be described by means of elementary space–time cycles. In
ECT, this possibility is realized by the fact that, according to the pos-
tulate of intrinsic periodicity, the space–time coordinates are treated as
(dimensional) angular variables. ECT defines relativistic differential prob-
lems by adding intrinsic covariant BCs to the relativistic space–time dynam-
ics describing every single particle. In undulatory mechanics, this is implicit
in the fact that space–time coordinates always appear in phasors or waves
functions. An angular variable is a variable which appears as an argument
of a periodic phenomenon, for instance, in a wave or a phasor, e−iEt/ : t
here is an angular (or periodic) variable with period T = h/E. That is,
every system in nature (set of particles) is described by a set of “periodic
phenomena” (e.g. phasors or wave-functions) of the space–time dimensions,
thus every physical system in nature can be described by a set (one for every
particle) of space–time periodic (angular) coordinates. In the case of PBCs,
this can be equivalently stated for a generic reference frame as follows: In
elementary particles, the space–time dimensions have a relativistic cyclic
nature, with local space–time periodicities T µ .
Finally, we mention another formal consequence of the ECT. By gen-
eralizing the above arguments to a rest particle (Compton clock), it fol-
lows that the proper-time parameter of every elementary particle has a
cyclic (compact) character of duration TC . Similarly to the cyclic (com-
pact) space–time dimensions, this proper-time parameter has PBCs (or
other kind of BCs allowed by the variational principle for relativistic the-
ories). For the consistency of this description it is crucial to notice that
massless particles, having infinite Compton periodicity, are associated to a
proper-time parameter with infinite compactification length. Hence, mass-
less particles with arbitrary low energy and momentum turns out to have
infinite compactification lengths for the space–time coordinates. This infi-
nite space–time associated to massless particles represents the reference
(non-compact) space–time, analogous to the (emphatically non-compact)
space–time of ordinary relativity. This guarantees a consistent relational
description of physics. Interactions, i.e. local and retarded modulations
of space–time periodicity, are therefore described by local and retarded
deformation of the space–time compactification lengths. The correspond-
ing geometrodynamics of the space–time boundary or of the metric actually
describe gauge or gravitational interactions, respectively [1, 3–13].
www.ebook3000.com
3. Derivation of QM
In ETC, the postulate of intrinsic periodicity represents the quantization
condition of ECT from which it is possible to derive the fundamental axioms
of QM and the Dirac quantization. We shall also describe how the purely
relativistic (cyclic) dynamics associated to the ECT leads to the ordinary
Feynman path integral and vice versa. The correspondence to QM will
be first introduced for isolated particles [13] (persistent periodicity), then
generalized to interacting particles [7, 8] (local modulations of periodicity).
For the sake of simplicity, we will essentially consider spinless particles.
3.1. Derivation of the canonical formulation

of QM
3.1.1. Elucidative example: The Black-Body radiation
To introduce the idea of quantization by means of constraint of intrinsic
periodicity, we shall consider the Black-Body radiation [10–13]. We know
from Planck that to any elementary component (mode) of periodicity T ,
i.e. of fundamental angular frequency, ω = 2π/T of the — classical —
electromagnetic field has a quantized energy spectrum En = nω = nh/T .
In ECT, every component of the — classical — electromagnetic field is
“overdetermined” by imposing its fundamental periodicity T as constraint.
The Planck energy spectrum is therefore interpreted as the harmonic spec-
trum of a massless periodic phenomenon (vibrating string) of fundamental
time periodicity T = h/E. According to the postulate of ECT, such a
component of the electromagnetic field must be described as a intrinsic
periodic phenomenon of temporal periodicity T , i.e. as a vibrating string
of fundamental period T . Indeed, a string vibrating with such a fundamen-
tal periodicity has discrete frequency spectrum fn = nf = n/T , Fig. 1,
which in turns, multiplying by Planck constant, yields the ordinary Planck
spectrum
h
En = hfn = n = nE. (9)
T
Hence, in ETC, the Planck quantization is described as the “quantiza-
tion” (discretization) of the harmonics frequency spectrum of a string con-
strained to vibrate (i.e. PBCs) with corresponding fundamental period
T = h/E. Similarly, the constraint of spatial periodicity λi = h/pi of the
108 D. Dolce
corresponding electromagnetic component implies the quantized harmonic

spectrum for the momentum
h
p(n)i = npi = n . (10)
λi
The system, being massless (i.e. infinite Compton periodicity), has pro-
portional temporal and spatial components of the four-periodicity T =
|λ|/c, see (8), so that we get the massless dispersion relation E = |
p|c and
the related massless dispersion relation of the energy spectrum
nh nhc
En = = pn |.
= c| (11)
T |λ|
By “overdetermining” the massless relativistic dynamics with intrinsic peri-
odicity we have obtained the ordinary Planck’s description of the Black-
Body radiation which, as well-known, avoids the ultra-violet catastrophe.
3.1.2. Derivation of the axiom of the states

According to the postulate of intrinsic periodicity, in ECT avery elementary
(quantum) particle, i.e. the basic constituents of nature, is represented as
a “periodic phenomenon”/vibrating string/elementary clock described by
a function Φ(x, t) of the (cyclic) space–time coordinates; or by Φ(xµ ) or,
suppressing the Lorentz index, by Φ(x) in relativistic notation. Moreover,
such a free elementary particle of persistent four-momentum pµ , observed
from an inertial reference frame, is postulated to have persistent space–
time periodicity T µ , according to the phase harmony relation cpµ T µ = h,
related to its Compton periodicity TC by Lorentz transformations. This
means that the function Φ(xµ ) representing an elementary particle must
satisfy the following constraint of periodicity, i.e. PBCs,
Φ(xµ ) = Φ(xµ + T µ ). (12)
By means of discrete Fourier transform, i.e. in analogy with a vibrating
string of space–time periodicity T µ , the free relativistic particle is therefore
represented as an harmonic system
i
Φ(x) = cn e− En t φn (x), (13)
n
where cn are Fourier coefficients, En = nh/T is the energy spectrum asso-

ciated to the harmonic frequency spectrum fn = n/T of fundamental time
periodicity T . Similarly, a free relativistic particle has harmonic momen-
tum spectrum p(n)i = nh/λi resulting from the spatial components λi of
www.ebook3000.com
the space–time periodicity T µ . Respectively, the temporal and spatial har-

monics are
i i
φn (t) = e− En t , φn (x) = e pn ·x . (14)
Note, that these are actually the quantized spectra prescribed by ordi-
nary QFT (after normal ordering) for a free relativistic particle. Indeed, the
dispersion relations (8) implies that the energy and the temporal period for
2
a free particle varies with the momentum as E 2 (p) = T h2 (p) = p2 c2 + M 2 c4
so that the dispersion relation of the energy spectrum is
h
En (
p) = nωn (
p) = n = n p2 c2 + M 2 c4 . (15)
T (
p)
Note also that both the energy and momentum spectra are described
by the same “quantum” number n — with n ∈ Z, the negative vibrational
modes correspond to antimatter (“antiparticles”), which for neutral bosons
is indistinguishable from ordinary matter (so we can assume n ∈ N in most
of the cases investigated in this paper).a This is because the temporal and
spatial components of the space–time periodicity T µ are not independent:
they are the Lorentz projections (5) of a single fundamental periodicity,
i.e. the Compton periodicity (topology S1 ). Moreover, these spectra are
harmonic as T µ is persistent (global) for free particles.
From a historical point of view, the mathematical concept of Hilbert
space was actually introduced to describe harmonic systems, as described in
Fig. 1. Indeed, the harmonic modes φn (x) form a complete and orthogonal
set which can be used to define the basis of the Hilbert space H associated to
an elementary particle Φ. Thus, to every harmonic φn (x) of an elementary
particle is associated a basis vector in the Hilbert space
φn (x) → |n (or |φn ) (16)
such that
φn (x)
x|n = √ , (17)
Vλ
a Indeed, as also known from ordinary QFT, a relativistic particles has positive and
negative energy modes associated to the quantization of the positive and negative energy
branches resulting from the square root of the relativistic dispersion relation. For neutral
bosonic particles, the negative and positives modes are indistinguishable as particles and
antiparticles coincide. For fermionic particles, these negative modes describes holes in the
Dirac sea. This aspect has been discussed in detail in terms of ’t Hooft cellular automata
[1, 7, 8, 27, 32].
110 D. Dolce
where Vλ is the volume defined by a single spatial period (wave-length)

λ = T c. The completeness relation of the harmonics φn (x) defines the
inner product of the corresponding Hilbert space H
3
d x ∗
n|n = φ (x)φn (x) = δn,n . (18)
Vλ n
V
λ
The base vectors |n and the inner product | define the Hilbert space
associated to our free elementary particle. We may equivalently consider an
integration volume that is associated to an arbitrary large integer number
of periods N λ (with N ∈ N). The corresponding normalization is given by
substituting Vλ → VN λ in (Eqs. (17) and (18)). It is however convenient to
normalize over an infinite number of periods, i.e. over an infinite integration
volume, as in ordinary QFT. In this case, the substitution is Vλ → 2π:
3
d x ∗
n|n = φ (x)φn (x) = δn,n (19)
2π n
and
φn (x)
x|n = √ . (20)
2π
Thus, the elementary space–time cycles associated to an elementary particle
naturally defines a corresponding Hilbert space H.
At this point, it is straightforward to see that the elementary particle is
represented by a vector in the corresponding Hilbert space, i.e. by a Hilbert
state. In fact, Φ(x), see (13), is the superposition of all the possible harmonic
modes allowed by the condition of intrinsic periodicity. Therefore, in the
Hilbert space H, the free relativistic particle is described by a corresponding
Hilbert state denoted by |Φ:

Φ(x) → |Φ = cn |n. (21)
n
This Hilbert description can be easily generalized to interacting par-

ticles, see Sec. 3.1.8. More generally, every isolated system S in nature
can be decomposed in terms of a finite number Npt of elementary par-
ticles Φi with i = 1, 2, . . . , Npt (and their reciprocal interactions). Every
elementary particle Φi of the system can be represented as a state in the
corresponding Hilbert space Hi . This means that the system is represented
by a state |ΦS = |Φ1 , Φ2 , . . . , ΦNpt defined in the Hilbert space resulting
from the tensor product of all the Hilbert spaces of the elementary particles
www.ebook3000.com
HS = H1 ⊗H2 ⊗· · ·⊗HNpt . Indeed, a harmonic system characterized by more

independent fundamental periodicities is described by the tensor product
of the corresponding Hilbert spaces (Fock space). For example, we will also
see that the atomic orbitals are described by the product of the deformed
periodicity in space–time associated to the Compton clock of the electron
with Hilbert basis |n and the spherical periodicity (i.e. the vibrations of a
spheric membrane) with Hilbert basis |l, m, so that the Hilbert basis of the
atomic orbitals is given by the corresponding tensor product |n ⊗ |l, m.
In this way, from the postulate of intrinsic periodicity, we have inferred
the axiom of the state of ordinary QM:
Axiom I: To every system S is associated a Hilbert space HS and S is

represented by the Hilbert state |ΦS .
3.1.3. Derivation of the axiom of the observables

In classical mechanics, to the harmonic frequency spectrum fn = nf = n/T
of a classical string vibrating with fundamental periodic T a frequency
operator Hf in the corresponding Hilbert space can be associated such
that (13), when applied to the basis element |n (the nth harmonic), returns
the corresponding frequency eigenvalue: Hf |n ≡ fn |n.
Similar to the case of the “quantized” harmonic frequencies of a vibrat-
ing string, to the nth harmonic φn of our vibrating string (free elementary
relativistic particle) is associated a quantized energy En = hfn = nh/T and
momentum p(n)i = nh/λi . In the Hilbert space notation, this corresponds
to define two operators H and P. They return the corresponding energy or
momentum eigenvalues when applied to the corresponding harmonic:
H|n ≡ En |n , P|n ≡ pn |n. (22)
In other words, in ECT, the Hamiltonian and momentum operators

encode the quantized spectra of the elementary space–time vibrations asso-
ciated to the elementary particle. These are the fundamental operators of
QM. The other operators can be constructed from them. It is possible to
show that the condition of periodicity (PBCs) guarantees that these Hilbert
operators are Hermitians (self-adjoint). Thus, generalizing to a system of
particles, we have the axiom of the observables:
Axiom IIa: Every physical observable of the system S corresponds a linear

self-adjoint operator on the Hilbert space A.
112 D. Dolce
Similarly, by generalizing the fact that the only possible vibrational

modes allowed by the fundamental periodicity to a string correspond to
the “quantized” frequencies fn = nf , we find the third axiom of QM:
Axiom IIb: All the possible results of the measurement of the observable
A of S are the eigenvalues An of the operator.
3.1.4. Derivation of the axiom of the motion

The Schrödinger equation is the analogous in the Hilbert space notation,
of the equation describing the time evolution of a vibrating string. It is
straightforward to check that every harmonic — which is actually a wave —
associated to an isolated periodic phenomenon, see (13), satisfies the fol-
lowing (first-order) differential equation
∂
i φn (t) = En φn (t). (23)
∂t
In ECT, an elementary particle (i.e. a periodic phenomenon) is not how-

ever described by a simple wave. In general, it is the superposition of all
the possible harmonics allowed by its fundamental space–time periodicity.
The evolution of such a “periodic phenomenon” with all its harmonics is
described by the Hilbert analogous of the time evolution Eq. (23). In partic-
ular, it is easy to check that the time evolution of such a persistent “periodic
phenomenon”, representing a free elementary particle, is actually given by
the ordinary time independent Schrödinger equation
∂
i |Φ = H|Φ. (24)
∂t
The spatial evolution is derived in terms of the momentum operator in

∂
a similar way: −i ∂x i |Φ = Pi |Φ. As we will see, the time dependent
Schrödinger equation refers to the interacting case (local modulations of

periodicity). Thus, we have derived the axiom of the motion:
Axiom III: The time evolution of a quantum system S, denoted by ΦS ,

with Hamiltonian operator HS is described by the differential equation on
the corresponding Hilbert space HS :
∂
i |ΦS = HS |ΦS . (25)
∂t
www.ebook3000.com
3.1.5. Derivation of the axiom of the measurement

Now we give a more accurate physical meaning of the ontological entity
describing an elementary particle, that we have addressed intrinsic “periodic
phenomenon”, internal clock or vibrating string, and represented by Φ(x).
We will find that in ECT, QM emerges as the statistical description of the
ultra-fast, cyclic relativistic dynamics associated with the elementary par-
ticles, due to the relatively poor resolution in time of modern timekeepers.
The time periodicity T of an elementary particle viewed in a generic iner-
tial reference frame is always faster than the Compton periodicity T ≤ TC .
This follows from the fact that the time period is determined by the energy
T = h/E whereas the Compton (rest) period is determined by the rest
energy (mass) TC = h/M c2 , and in relativity E ≥ M c2 . Furthermore,
the heavier the mass of the particle, the faster the related periodicity.
For instance, even considering simple quantum electrodynamics (QED),
i.e. systems of electrons interacting electromagnetically, the upper limit of
the periodicities involved are determined by the Compton periodicity of
the electron — the lightest particle of the Standard Model (expect neu-
trinos in which there are experimental evidences of the very slow “oscilla-
tions” associated to their tiny masses). The Compton periodicity of an
electron is extremely small TCe ∼ 10−21 s. This is the time taken by
light to travel across the electron Compton length ΛC ∼ 10−12 m: i.e.
TC = λC /c. For comparison, the best resolution of modern timekeepers
is only ∆Texp ∼ 10−17 s, whereas that of the Caesium atomic clock is
∆TCs ∼ 10−10 s — not to mention the time resolution at the Planck and
de Broglie epoch. With such a poor resolution in time the direct observa-
tion of the ultra-fast cyclic dynamics of elementary particle is impossible.
Hence, only a statistical description of these ultra-fast periodic phenom-
ena is possible.b This is similar to the observation of a rolling die without
a slow motion camera: the outcomes can only be described statistically.
ECT proves that the statistical description of the ultra-fast cyclic dynam-
ics associated to elementary particles formally corresponds — is formally
b The difference of magnitude between the Cesium atomic clocks and the electron internal
clock ∆TCCs /TCelectron ∼ 1011 is of the order of that between a solar year and the age
of the universe∆TUniverse /TSeasons ∼ 10−11 . Thus, trying to predict the outcomes of a
quantum system is like trying to predict some annual dynamics having data, say, of the
stock markets or of the earth climate, with uncertainty in time of the order of the age of
the universe. Due to the poor resolution in time the only possible way to describes these
annual dynamics would be statistical.
114 D. Dolce
equivalent — to QM. The representation of elementary particles in the

Hilbert space notation actually corresponds to give a statistical description
of its cyclic dynamics, i.e. to give up with a “deterministic” description.
In order to derive the Born rule from the postulate of elementary cycles
it is necessary to consider that the assumption of ultra-fast intrinsic peri-
odicity for an elementary particle corresponds to describe the dynamics
of a particle moving very fast on a circle (elementary space–time cycles).
It has been proven by ’t Hooft that there is a close relationship between
a particle moving on a circle and the temporal evolution of the quantum
harmonic oscillator of the same period — the time evolution of a “periodic
phenomenon.” This can be intuitively understood by thinking of a rolling
die — a die is essentially a ’t Hooft’s cellular automata with six sites. If
the die rolls too fast and at two observations (e.g. through a stroboscopic
light, see Elze’s stroboscopic quantization [33]) we observe, say, the same
faces, then we can only say that the die evolution is that of a “periodic
phenomenon” described by the superposition of evolutions with frequen-
cies fn = nf , where T = 1/f is the time between the two observations.
Let us now consider the analogy to the electric current, [34]. In an
electric current, we have electrons moving very slowly in a circuit, which
can be thought of like a particles on a circle — large period T . Due to
the large number of electrons in a conductor, we typically give a statistical
description of the current by introducing a wave-function Φ(x, t) satisfying
the continuity equation
∂ρ
+ ∇ · j = 0. (26)
∂t
We can assume that the charge density, i.e. the density of electrons, is
given by ρ = |Φ|2 ; j is the current density. Of course, we can generalize
this description to neutral particles. In this case, ρ denotes the density of
particles. It describes the probability to find particles in a given section
of the circuit. Furthermore, we can generalize this statistical description
even to the case of a single particle in the circuit/circle, if it moves too
fast with respect to our timekeeper (i.e. too small period T ). In this case,
similarly to a rolling die, we cannot determine its motion in detail (“deter-
ministically”). We can only give a statistical description. Hence, the “peri-
odic phenomenon” associated to a single elementary particle in ETC is
described, at a statistical level, in analogy to the current of a single particle
moving very fast on a circle (circuit). Indeed, in ECT, Φ(x) is described
by a wave equation so that it satisfies a continuity equation (26) — as the
observation of a particle implies that it must be stopped on a detector here
www.ebook3000.com
we can assume a non-relativistic continuity equation [34]. Note that as the

particle has a cyclic motion of PBCs with unknown small period T , it can
be written as the superposition of harmonic eigenstates, through discrete
Fourier transform. Finally, we also have an important condition for Φ(x),
i.e. for the periodic phenomenon associated to an elementary particle, as we
have a single particle on a circle (circuit), the integral of the particle density
ρ = |Φ|2 over the whole circuit must be equal to one, or, by integrating over
an infinite number of period (i.e. by using a different normalisation of the

wave-function), we have the Born condition dx3 |Φ(x, t)|2 = n |cA 2
n | = 1.
This describes the Born rule for our “periodic phenomenon” associated to
an elementary particle.
Thus, we have derived the axiom of the measurement from the postulate
of intrinsic periodicity:
Axiom IVa: For a system S represented in the Hilbert space HS by |ΦS =

A A A
n cn |φn , where |φn are eigenstates of an observable A (i.e. eigenvectors
of HS ), the probability to measure the eigenvalue An of the observable A is
A2
given by P[An ] = |cA 2
n | , such that | n |cn | ≡ 1.
Moreover, since we are describing essentially a classical system we must
assume that “immediately after” a measurement our “periodic phenomena”
are in the state in which they have been observed.
Axiom IVb: For a system S represented in the Hilbert space HS , if the

result of the measure of the observable A was An , “immediately after” this
measure the state is in |φA
n .
As the canonical formulation of QM is defined by all these axioms and we
have derived them from the postulate of intrinsic periodicity of elementary
particles, we conclude that ECT is formally equivalent to ordinary QM, see
below for the generalization to the interacting case, and further evidences.
3.1.6. Derivation of the commutation relations: Dirac

quantization rule
To derive the ordinary commutation relations of QM from the postu-
late of intrinsic periodicity, we evaluate the “expectation value” of an
observable ∇F
(x) for a generic elementary periodic phenomenon, that is,
Φ|∇F
(x)|Φ — the following demonstration is analogous to the demon-
stration used by Feynman [35] to prove that the path integral formulation of
QM is equivalent to the canonical formulation of QM. This corresponds to
weight the given observable by the Fourier coefficients cn of the harmonic
116 D. Dolce
system describing the elementary particle (the statistical meaning of the

expatiation value is given by the Born rule obtained above). For conve-
nience here we use the normalization over a single period, but the result
can be easily generalized to large or infinite number of periods. By means
of integration by parts and by using the definition of the momentum oper-
ators, we find
3
i d x i i
Φ|∇F (x)|Φ =
c∗m cn e− pm ·x ( pn )e pn ·x
pm F (x) − F (x)
Vλ n,m
V
λ

i i
− [Φ∗ (x)F (x)Φ(x)]Vλ = Φ|PF(x)
− F(x)P|Φ.

Vλ
(27)
In this derivation, it is very important to note the crucial role of the
assumption of spatial periodicity: here xi is a periodic variable of period
λi . This guarantees that the boundary term of (27), obtained from the
integration by parts vanishes. That is, it guarantees that for arbitrary states
Φ we obtain
∂F (x)
[F (x), Pi ] = F (x)Pi − Pi F (x) = i . (28)
∂xi
As can be easily seen by choosing F (x) = xj , this yields the ordinary
commutation relations of QM
[xj , Pi ] = iδi,j . (29)
This remarkable result shows that the commutation relations of ordinary
QM (and thus the Heisenberg uncertainty relation, see below) are implicit
in the assumption of intrinsic periodicity of ECT — vice versa the Dirac
quantization rule, in which commutation relations are imposed to quantize
a system, encode a condition of intrinsic periodicity (commutation relations
are typical of angular variables).
Similar commutation relations can be extended to the other observables.
Thus, we find the Dirac quantization rule of the canonical QM:
Dirac quantization rule: If the commutation relation of the physical
observables A and B of a system of classical particle is described by the
Poisson brackets {A, B}P , then the related system S [system of periodic
phenomena] is described by the commutation relations
[A, B] = i{A, B}P , (30)
where A and B are the corresponding operators of the Hilbert space HS .
www.ebook3000.com
3.1.7. Heisenberg uncertainty relation

Even though the Heisenberg uncertainty relation is implicit in — is a direct
mathematical consequence of — the commutation relations derived above,
it can also be heuristically inferred directly from the assumption of intrinsic
periodicity. We have to consider that the phase of a “periodic phenomenon”
is defined by modulo factors 2π’s, and that in ECT the space–time coordi-
nates that are treated have periodic (angular) variables of period T µ [36].
We only consider, for the sake of simplicity, the fundamental harmonic
of a periodic phenomenon spatial evolution, i.e. the monochromatic wave
i i
e− pi x of spatial periodicity λi = T i c = h/pi in the ith spatial direction,
and the phase invariance 2πn with n = 1 (this gives the most stringent
uncertainty relation). Thus, in the observation of such a “periodic phe-
nomenon”, the phase invariance 2π implies a simultaneous uncertainty in
the determination of its momentum or position during its cyclic evolution:
i i i i
i i i i
(xi +∆xi )
e− pi x = e− (pi x +2π)
= e− (pi +∆pi )x = e− p , (31)
where ∆pi = h/xi and ∆xi = h/pi . According to the postulate of ele-
mentary cycles, this evolution is characterized by an intrinsic periodicity:
0 ≤ (xi mod + λi ) ≤ λi , where pi λi = 2π. Hence, such ambiguity is gov-
erned by the ordinary Heisenberg relation (the factor 1/2 can be obtained
by considering the invariance of |Φ| according to the Born rule and spin-
statistics)
(2π)2 (2π)2
∆pi ∆xi = ≥ = h. (32)
pi xi pi λi
3.1.8. Interacting quantum systems

To generalize the formal correspondence between ECT and QM to the case
of interactions, it is sufficient to consider that in undulatory mechanics
the momentum–energy and the space–time periodicity are two faces of the
same coin [7]. Interactions are local variations of the energy–momentum.
That is, if we denote by pµ the four-momentum of a free relativistic par-
ticle, when interaction is switched on, the resulting four-momentum in a
given point x = X is modified to pµ (x)|x=X = Λνµ (x)|x=X pµ . The func-
tion Λνµ (x) describes, point by point, how the four-momentum varies with
respect to the free case when the corresponding interaction scheme or poten-
tial is assumed — indeed this implies, through the phase harmony relation,
that the local space–time dimensions of our periodic phenomena are locally
deformed during interaction, and thus a geometrodynamical description of
118 D. Dolce
interactions is similar to general relativity [7]. Contrarily to the free case

in which the periodicities are persistent (Newton’s first law), interactions
correspond to local modulations of the space–time periods T µ (X) such that
in X the phase harmony (de Broglie–Planck relations) is locally satisfied

pµ (x)dxµ = 2π. (33)

T µ (X)
In analogy with modulated signals, the time and spatial evolutions of

a “periodic phenomenon” are thus described by the modulations of all the
harmonics allowed by the local PBCs
i
R t
Φ(x, t ) = cn e− t En (x)dt φn (x, t ),
n
i
R x
φn (x , t) = e−
x n
p
(y)·d
y
φn (x , t). (34)
Such modulated harmonics form locally a complete, orthogonal set. In every

point, they define the basis of a local Hilbert space with corresponding local
inner product denoted by | X . The local energy and momentum spectra
En (x) and pn (x) can be used to define local Hamiltonian and momen-
tum operators H (x) and P (x). In analogy with the classical (persistent)
four-momentum operator pµ (x), we define the (persistent, i.e. global) four-
momentum operator of the free case Pµ = {H, −P}. The (local) Hamilto-
nian and momentum operator of the interacting system is inferred from the
free case by means of the local transformation Pµ (x)|x=X = Λνµ (x)|x=X Pµ
[7]. The time evolution of modulated elementary cycles is therefore given
by the ordinary Schrödinger equation of an interacting system with local
Hamiltonian H (X):
∂
i |Φ = H (X)|Φ. (35)
∂t
An interacting system is characterized in every point of its cyclic evo-
lution by a locally modulated spatial period. Thus, by generalizing the
demonstration of the free case, we have local commutation relations
[x, P ] = i. (36)
In this way, the exact correspondence of ECT to QM (axioms of QM and

Dirac quantization rule) has been generalized to interactions.
In conclusion, as all the axioms of the canonical formulation of QM, as
well as the Dirac quantization rule (commutation relation and Heisenberg
www.ebook3000.com
Fig. 2. Some periodic classical paths (light gray lines), i.e. a small selection (n =
−7, −6, . . . , 6, 7) of the infinite set of degenerate paths characterized by different
winding numbers, between two points of a cyclic geometry. In analogy with this
picture, the evolution of a periodic phenomenon is given by the interference of
all the classical periodic paths on the cyclic space–time geometry of ETC, and it
turns out to exactly described by the ordinary Feynman path integral of QM.
relation) have been derived directly from the assumption of intrinsic peri-
odicity, we actually have found an equivalence between ETC and QM. Now
we will extend this equivalence to the Feynman path integral.
3.2. Derivation of the Feynman formulation

The evolution of “periodic phenomena”, similar to the evolution between
two points on a cylindric geometry, Fig. 2, is characterized by an infinite
number of degenerate classical paths, i.e. by the sum of all the possible
paths with different winding numbers associated to the cyclic space–time
geometry of ETC. For instance, this can be easily seen in the free case by
−inωt

using the Poisson summation: n cn e = 2π n ĉn δ(ωt + 2πn ) =

T n ĉn δ(t + n T ) where n, n ∈ Z (see footnote a), the period of the
(angular) coordinate t is T = 2π/ω, ĉn are the “discrete” Fourier transform
coefficients of cn . The space–time evolution of a free relativisitic particle is
i
Φ(x , x ) = cn e− n[E(t −t )−p·(x −x )]
n

= 2π ĉn δ(E(t − t ) − p · (x − x ) + 2πn). (37)
n
The sum over the Dirac deltas describes the interference among the classical
paths of winding numbers n of the cyclic evolution associated to the per-
sistent periodic phenomenon, i.e. the infinite set of classical paths among
two fixed points on a (space–time) cylinder. These are represented by the
blue lines in Figs. 2 and 3 — this degeneracy of classical paths provides an
alternative interpretation of the Heisenberg relation.
Let us now formally derive the Feynman path integral from the postulate
of intrinsic periodicity (without any further condition). We have seen that
120 D. Dolce
(a)
(b) (c)
Fig. 3. Here, we imagine to open the cylinder of Fig. 2 on a cartesian plane.

The periodic classical paths (light gray lines) of a cyclic evolution are equivalent
to variations (bold dashed and dotted lines) around a given path (black line)
between two points on a cartesian plane. Indeed, due to the unitarity of the
periodic paths — e.g. the dotted and dashed paths of (b) — these can be cut and
translated by periods and combined in order to form paths with the same initial
and final points, which in turn can be interpreted as variations around a given
path — (c). The iteration of this procedure shows graphically — (a) — that
the classical evolution of a periodic phenomenon is described by the ordinary
Feynman path integral. The interference between these classical paths becomes
more destructive for final (or initial) points aways from the classical path of the
corresponding classical particle.
the space–time evolution of a modulated “periodic phenomenon” is charac-

terized in every point by a local Hilbert space with its local inner product.
i
The time evolution operator in the free case was U (t , t ) = e− H(t −t ) .
According to Sec. 3.1.4, in the interacting case, the time evolution operator
is the Hilbert analogous of a modulated wave
R t
i
H (x)dt
U (t , t ) = e− t . (38)
This operator is unitary (and Markovian). Hence, the time evolution can
be sliced in N → ∞ elementary time evolutions of infinitesimal duration :
N
−1
U (t , t ) = lim U (t + tm+1 , t + tm − ), (39)
N →∞
m=0
www.ebook3000.com
where the infinitesimal time evolution is

i
U (t + tm+1 , t + tm − ) = e− H (x,tm ) .
For the sake of simplicity, we assume the normalization over an infi-

nite number of spatial periods — it is sufficient that the integration region
corresponds to an integer number of periods and completely contains the
interaction region. In perfect analogy with Feynman’s derivation, we plug
the local completeness relation of the modulated harmonic set φn of a peri-
odic phenomenon in between the elementary time evolutions. By consid-
ering the local definitions of the Hamiltonian and momentum operators in
the Hilbert space, we obtain
N −1

3
U (x , x ) = lim d xm
N →∞
m=0
U (x , xN −1 )U (xN −1 , xN −2 ) . . . U (x1 , x ).

(40)
The infinitesimal space–time evolutions of the “periodic phenomenon” are

i
U (xm+1 , xm ) = Φ̂|e− (H (xm )∆tm −P (xm )·∆xm ) |Φ̂ (41)

with |Φ̂ = n |n, ∆tm = tm − tm−1 and ∆xm = xm − xm−1 . In this way,
we have derived, without any further assumption than intrinsic periodicity,
the ordinary Feynman path integral in phase space
N −1 N −1
i
3
Z = U (x , x ) = lim d xm Φ̂|e− (H ∆tm −P ·∆xm ) |Φ̂.
N →∞
m=0 m=0
(42)
The phase of the modulated “periodic phenomenon” defines the
Lagrangian of the interaction
L = P · ẋ − H . (43)
Note that by construction, according to the definition of the local Hamil-

tonian and momentum operators given in the previous paragraph, this
Lagrangian is formally the Lagrangian of the corresponding interacting clas-
sical particle written in terms of operators. The corresponding action is

t t
S= L= (P · ẋ − H )dt. (44)
t t
122 D. Dolce
This yields the ordinary Feynman path integral in Lagrangian notation

i
Z = D3 xe S . (45)
This proves that the classical evolution of modulated elementary space–time

cycles is equivalently described by the ordinary Feynman path integral.
The equivalence between the Feynman path integral and the cyclic
dynamics of ECT is straightforward if we again make use of the Poisson
summation. This also clearly shows the physical meaning of our result.
The idea is illustrated of on Figs. 2 and 3 in analogy with a simple cyclic
geometry. We consider the free case, but the result can be generalized to
interactions.
A relativistic free particle has harmonic energy and momentum spectra:
H|φn = nE|φn and P|n = np|n. Thus, similar to (37), the elementary
space–time evolution of the Feynman path integral can be written, by means
of the Poisson summation, as
i
i
U (xm+1 , xm ) = Φ|e− (H∆tm −P·∆xm ) |Φ = e− n(E∆tm −p·∆xm )
n

= 2π δ(E∆tm − p · ∆xm + 2πnm ). (46)
nm

By substituting in (40) and using the Dirac delta property d3 xm δ(xm+1 −
xm )δ(xm − xm−1 ) = δ(xm+1 − xm−1 ) we explicitly see, in agreement with
(37), that the Feynman path integral is given by the sum over all the pos-
sible periodic paths with persistent periodicity T µ :
N −1 N −1

Zf ree = lim d3 xm 2π δ(E∆tm − p · ∆xm + 2πnm )
N →∞
m=0 m=0 nm

= 2π δ(E(t − t ) − p · (x − x ) + 2πn ). (47)
n
The demonstration in this form (i.e. based on infinitesimal space–time

paths) can be generalized to interaction. The Feynman path integral of
an interacting system is interpretable as the sum (integral) over all the
possible periodic paths with locally modulated periodicity, i.e. with all the
possible windings associated to the locally modulated periodicity T µ (x) of
the interaction scheme — local deformations of the elementary space–time
cycles. In the more powerful Lagrangian formalism of ECT [1, 7, 8, 13], the
elementary particle is actually described as the solution of a relativistic wave
www.ebook3000.com
(bosonic) action with PBCs. The least action principle in this case shows
that this classical action is minimized by infinite degenerate solutions, i.e.
the periodic paths of different windings. Thus, the Feynman path integral
can be equivalently described as interference of classical paths, without giving
up with the classical variational principle.
This remarkable result clearly reveals that the quantum evolution of a
particle, as described by the Feynman path integral, is given by the inter-
ference of all the periodic classical paths associated to its local de-Broglie–
Planck (space–time) periodicity. Note that once the local space–time period
T µ (x) (the shape of our deformed cyclic space–time geometry) is assigned,
the corresponding local (classical) energy–momentum pµ (X), is assigned
as well. In turn, the classical particle path (the path of the corresponding
classical particle) is assigned. If the final point x at which the Feynman
path integral is evaluated moves away from this classical path, the inter-
ference of these periodic paths becomes less and less constructive denoting
a lower probability to find the particle in that point, in agreement with
Feynman’s interpretation. The maximal probability is along the path of
the corresponding classical particle, on which all the periodic paths have
constructive interference.
3.3. Derivation of the Bohr–Sommerfeld quantization

and WKB method
The constraint of persistent or modulated space–time periodicity is the
quantization condition in ECT for the free or interacting case, respectively.
In the free case persistent periodicity T µ , i.e. the PBCs (12), implies
i µ i µ
Φ(x) = cn e− p(n)µ x ⇒ e− p(n)µ T c = e−i2πn . (48)
n
The temporal and spatial components of this relation yield the (normally
ordered) energy and momentum spectra for the free relativistic particle
(e.g. similarly to a particle in a spatial and temporal box)
i h
e− En T = e−i2πn → En T = 2πn → En = n ,
T
i h
e pn ·λ = e−i2πn → pn · λ = 2πn → pin = n . (49)
λi
Interaction implies that in every space–time point X on which we are
evaluating the particle a locally modulated space–time periodicity T µ (X) is
124 D. Dolce
associated. To derive the quantization condition, we must consider that the

generic solution is locally modulated, as in (34). In this case, the constraint
of locally modulated periodicity (space–time geometrodynamics) implies [7]
RX i
H
pnµ dxµ
i
p(n)µ (x)dxµ
Φ(X) = cn e − ⇒ e− T µ (X) = e−i2πn . (50)
n
In this case, the deformed energy and momentum spectra are thus given by

En dt = 2πn, pn · dx = 2πn. (51)

T (X) λ(X)
This essentially means that the “periodic phenomenon” must have closed
space–time orbits, in analogy, for instance, with Bohr orbitals. ETC is there-
fore a relativistic generalization (an evolution) of the Bohr–Sommerfeld
quantization and of the WKB method.
It must be noted that the variational principle generally allows for more
general BCs. For instance, the relativistic bosonic action allows for Dirichlet
and Neumann BCs or twisted PBCs at the ends of compact space–time
dimensions. By assuming anti-periodicity in time, i.e. Φ(x, t) = −Φ(x, t+T )
(e.g. Fermi–Dirac statistics), the resulting quantization of the energy is

− i En T −i2πn 1 1
e = −e → En T = 2π n + → En = n + hf.
2 2
(52)
This solution contains the vacuum energy E0 = hf /2.
More generally, the twist of an angle 2πα in the periodicity implies the
Morse term α in the Bohr–Sommerfeld quantization condition

p(n)µ dxµ = 2π(n + α). (53)

λ(X)
A twist factor determines the vacuum energy E0 = αhf in the spectrum.

Again, the Morse term and thus the vacuum energy can be determined
by studying the BCs of the system. For instance, for particles bounded
in a potential it is determined by the BCs at the spatial infinite or at
the center of symmetry of the potential (e.g. as in the harmonic potential
or Coulomb potential). Note that as also originally proposed by Casimir,
the modern technique to calculate the Casimir effect, manifestation of the
vacuum energy is actually based on the BCs [37, 38].
www.ebook3000.com
4. Basic Applications
Here, we will briefly report the basic procedures to solve textbook problems
of QM within the formalism of ECT.
4.1. Non-relativistic free particle

A relativistic free particle of energy E is a “periodic phenomenon” con-
strained to have a persistent time periodicity T = h/E. In analogy with
a particle in a periodic time box or a vibrating string, it has a harmonic
energy spectrum En = nE = nh/T . We assume n ∈ N — the negative
vibrational modes describe antimatter. As already inferred in Sec. 3.1.2,
according to the controvariance of T µ , (8), the time period T is frame
dependent. The resulting dispersion relation of the energy spectrum (15)
is equivalent to that prescribed by the second quantization (after normal
ordering) for bosonic particles. Similarly, the spatial periodicity λ implies
the harmonic momentum spectrum p(n)i = nh/λi . The Compton period-
icity, fixed by the mass according to the Compton relation TC = h/M c2 ,
determines, through Lorentz transformations, the space–time periodicity
and thus the quantized energy–momentum in an inertial reference frame.
As is well known from ordinary relativity, in the non-relativistic limit,
i.e. |
p| M/c, the rest energy of a particle can be omitted as it forms an
p2
infinite energy gap: E = M 2 c4 + p2 c2 ∼ M c2 + 2M , so that the dispersion
2
p
relation for a free non-relativistic particle is Eclass (
p) = 2M . Equivalently,
the Compton periodicity, i.e. the undulatory counterpart of the mass, can be
neglected as it tends to zero in the non-relativistic limit, TC → 0. The effect
of the Compton periodicity, i.e. the quantization of the particle dynamics,
can be therefore neglected. This implies that a periodic phenomenon, i.e.
an elementary particle, in the non-relativistic limit has continuous energy
and momentum spectra. The quantization associated to the Compton clock
is lost and we have ordinary non-relativistic physics.
Furthermore, it is possible to show that in ETC, |Φ(x)| is localized
within the Compton length along the classical particle path [13]. Thus, in
the non-relativistic limit (λC → 0) we get a point-like description of the
particle path, i.e. the ordinary classical path. In general, in ECT, the parti-
cle description is obtained in the limit of very small temporal periods — e.g.
electromagnetic modes at very high frequency (Black-Body UV region) can
be described by photons. ECT correctly describes the particle counterpart
of massless fields (e.g. the electromagnetic radiation) and the undulatory
behaviour of massive particles, enforcing the wave-particle duality.
126 D. Dolce
4.2. One-dimensional Schrödinger problems:

closed orbits
In non-relativistic quantum problems, the intrinsic space–time periodicity
of a particle, i.e. the BCs quantizing the system, is determined by the
potential V (x) rather than by the Compton periodicity. In this way, the
solution of non-relativistic problems such as the particle in a box, potential
well, the Dirac delta potentials, tunnel effect, etc., is essentially identical to
the ordinary ones: they in fact are easily solved by imposing BCs.
Similar to a vibrating string, the energy and momentum spectra are
determined by the requirement that, as prescribed by the postulate of
intrinsic periodicity in ECT, the particle must have closed space–time orbits
along the phase-space of the potential. The condition of closed orbits is
given in (51) or, by assuming a Morse factor, by (53). In short, in non-
relativistic problems, the “overdetermination” of the system is provided by
the potential. We have to calculate the duration of a orbit in the phase-
space of the potential. Only those orbits in which the wave-length enters
an integer number of times are allowed for the quantum system. Thus, the
solution of simple non-relativistic Schrödinger problems is possible in per-
fect analogy to the Bohr–Sommerfeld quantization or to the WKB method.
Once the energy spectrum is known, the harmonics constituting the wave-
packet of the system are derived by solving the equations of motion in order
to determine the local modulations of periodicity from the potential V (x).
As already said, in the Hilbert space notation, the evolution of our periodic
phenomenon in a potential V (x) is given by the Schrödinger equation
2
∂ P
i |Φ(x, t) = HClass (x)|Φ(x, t) = + V (x) |Φ(x, t). (54)
∂t 2M
Below, we will see some examples of the validity of this method. We will
obtain the same results reported in textbooks.
4.3. Quantum harmonic oscillator, quantum

anharmonic oscillator, linear potential,
and second quantization
The quantum harmonic oscillator, V (x) = 12 mω 2 x2 , can be easily solved by
considering the isochronism of the pendulum: every space–time orbit has
the same fundamental period T = 1/f = 2π/ω. This corresponds to the
case of a “periodic phenomenon” with persistent time periodicity. We are in
the case of an homogeneous string vibrating in time (49). Thus, the energy
www.ebook3000.com
spectrum of the quantum harmonic oscillator is harmonic En = (n + α)hf

where we have assumed a generic Morse factor. According to (54), the
spatial component of the nth harmonic of this “periodic phenomenon” is
given by the same differential equation of the ordinary description

2 ∂ 2 φn (x) M f 1
− + φn (x) = hf n + φn (x), (55)
2M ∂ 2 x 4π 2
where we have assumed α = 1/2 in order to have vanishing value (BCs) of

the harmonics at the spatial infinite. See below for the Coulomb potential.
The anharmonic potential is the harmonic potential plus a quartic term
4 4
x /l with 1 and l = h/M f . In this case, the requirement of close
orbits yields a correction ∆En = 34 (2n2 + 2n) to the quantum harmonic
oscillator energy spectrum. For the linear potential V (x) = mgx (gravita-
tional potential for small distances), the quantized energies satisfying closed
space–time orbits turn out to be En = 12 [3π(n + 1/4)]2/3 (M g 2 ).
In short, in all the possible textbook quantum problems, we obtain the
correct solutions known in literature.
4.3.1. Creation and annihilation operators: Derivation

of the second quantization
The solution of the quantum harmonic oscillator can be used to introduce
the Ladder operators in ETC, i.e. the creation and annihilation opera-
tors of QFT. They
are defined as functions
of the position and momentum
Mω i Mω i
operators: a† = 2 (x + Mω P) and a = 2 (x − Mω P). In ECT, the
†
commutation relation [a , a] = 1, similarly to ordinary QFT, is a direct
mathematical implication of the ordinary commutation relations [x, P] = i
already derived from the postulate of intrinsic periodicity. From this, it is
possible to build the Fock space of QFT. This represents the link between
ECT and the second quantization, i.e. the quantization method of QFT
based on commuting ladder operators of the quantum harmonic oscillator
corresponding to every fundamental (frequency) mode of a classical field.
4.4. Three-dimensional Schrödinger

problems, Coulomb potential, quantum
numbers and tensor product
In spherical problems, beside the space–time periodicity considered so far,
it is necessary to consider the spherical periodicity as additional constraint
128 D. Dolce
“overdetermining” the dynamics. Indeed, similar to the space–time peri-

odicity of a periodic phenomenon (topology S1 determined by the single
fundamental Compton periodicity), spherical problems are described by
two spherical angles θ and ϕ of (static) periodicities π and 2π, respec-
tively (topology S2 ). As is well known from ordinary QM, it is sufficient to
impose this spherical periodicity as constraint (without any further quan-
tization condition) to obtain the correct quantization of the corresponding
physical conjugate variables, i.e. the quantization of the angular momen-
tum of the system. We have seen that the space–time periodicities, Lorentz
projections of the single Compton periodicity, imply a decomposition in
harmonics φn (|x|, t) denoted by the quantum number {n} (one quantum
number for the topology S1 ). Similarly, the periodicity of the two spheric
angles implies additional harmonic sets described by two corresponding
quantum numbers {m, l} (two quantum numbers for the two fundamental
periodicities of S2 ). They are essentially the harmonics of a vibrating spher-
ical membrane Ylm (θ, ϕ). These form a complete, orthogonal set defining
a corresponding Hilbert space Hsphere . The corresponding solutions of the
Schrödinger equation are the spherical harmonics. The generic solution of
a spherical problem is thus described by the three quantum numbers, cor-
responding to the harmonic eigenmodes φn (|x|, t)Ylm (θ, ϕ). For instance,
to describe the 3D quantum harmonic oscillator we have to substitute the
space–time harmonic solution φn (|x|, t) obtained above, (55).
Note that in ECT the quantization by means of the constraint of intrin-
sic space–time periodicity is the space–time equivalent of the universally
accepted and tested quantization of the angular momentum in terms of the
intrinsic periodicity of spherical angles — or the relativistic generalization
of a particle in a box. Indeed, no further quantization condition for the
angular momentum is required except spherical periodicity (or its defor-
mation). In other words, in ECT, the space–time coordinates are treated
as (non-independent) angular variables for the quantization of the energy
momentum exactly as the spheric angles in the quantization of the angular
momentum. This is another confirmation of the consistence of ECT.
Another typical example of Schrödinger problem with spherical symme-
try is the Hydrogen atom. The requirement of closed space–time orbits in a
Coulomb potential (notice that it is not necessary to assume circular orbits
contrarily to Bohr’s original description) leads to the atomic energy levels
En = − 13.6eV
n2 . The requirement of closed spherical orbits yields the quanti-
zation of the angular momentum in terms of the quantum numbers {m, l}.
Thus, the atomic orbitals can be regarded as the combination of vibrational
www.ebook3000.com
modes of a periodic phenomenon associated to intrinsic space–time period-

icity modulated by the Coulomb potential, and the spherical harmonics
associated to the spherical symmetry. Note that a careful application of the
Bohr–Sommerfeld quantization similar to the method proposed here yields
a correct description of more advanced atomic problems such as atoms with
more electrons, the Zeeman effect, and so on [39].
The composition of the space–time periodicity (topology S1 ) and the
spherical periodicity (topology S2 ) are described by the tensor prod-
uct of the Hilbert spaces associated to these fundamental periodicities:
HS = H ⊗ Hsphere (S1 ⊗ S2 ). The resulting Hilbert basis defined by
φn (|x|, t)Ylm (θ, ϕ) is |n, m, l = |n ⊗ |m, l (three quantum numbers for
the three fundamental periodicities of S1 ⊗ S2 ). This example shows that in
ECT a different quantum number is associated to every fundamental peri-
odicity. The composition of two elementary particles, i.e. two distinguished
space–time periodic phenomena, is thus given by |Φ1 , Φ2 = |Φ1 ⊗ |Φ2
with corresponding Hilbert basis |n1 , n2 . This leads to the Fock space.
4.5. The role of the temperature: Dirac quantization

for magnetic monopoles, superconductivity
and graphene
We shortly describe the role of the temperature in ECT [4, 13, 27]. Since
the temperature implies random collisions among particles (thermal noise),
a quantum system at finite temperature is characterized by chaotic (Pois-
sonian) decay of the “complete coherence” of the particles with a charac-
teristic thermal time β = /kB T , i.e. a dumping e−ω/kB T = e−β/T of
the cyclic behaviour of periodicity T of the system, kB being the Boltz-
mann constant and T being the temperature. In other words, while the
Minkowskian periodicity T of QM tends to form perfect coherent states
(“periodic phenomena”), the Euclidean periodicity β describes a thermal
dumping associated to the thermal noise which tends to destroy the perfect
recurrences of the pure quantum systems. Thus, if T β, the system can
autocorrelate and give rise to pure quantum phenomena whereas in the
opposite limit the thermal noise breaks the quantum recurrence before it
can give rise to autocorrelation, leading to the classical behavior (e.g. ordi-
nary electric resistance). The dumping factor e−ωn /kB T = e−nβ/T can be
also regarded as the probability to populate the nth vibrational mode of the
system. Thus, at very low temperature T β only the fundamental mode
n = 1 will be populated, whereas at high temperature many vibrational
130 D. Dolce
modes must be considered, so that the spectrum can be approximated by

a continuum (e.g. as in the classical limit of the Bohr atom).
In ECT, gauge interactions, such as the electromagnetic interaction,
can be directly inferred (without postulating it) from the space–time
geometrodynamics in deep analogy with gravitational interaction in gen-
eral relativity. For the scope of this chapter, it is nevertheless sufficient to
introduce electromagnetic interaction by assuming the minimal substitu-
tion pµ (x) = pµ − eAµ (x) where Aµ is the electromagneitc potential. With
this substitution in the modulated periodic phenomenon, e.g. in (50), in
analogy with the Bohr–Sommerfeld quantization, the condition of intrinsic
periodicity leads directly to the Dirac quantization for monopoles (and a
link to Dirac strings)
H
ie µ
e T µ (X) Aµ dx = e−i2πn → Aµ dxµ = 2πn. (56)
T (X)
This result is at the base of the derivation of superconductivity and

its fundamental phenomenology in ECT, directly from first principles of
QM rather than from empirical models [2] with interesting insights on high
temperature superconductivity. Indeed, if we consider the gauge invariance
e
Φ(x, t) = U(x, t)Φ(x, t) where U(x, t) = e−i c θ(x,t) , the condition of intrinsic
periodicity in this case implies that the Goldstone θ(x, t) is periodic and
defined with modulo factors 2πn:
e e
θ(x, t) = θ(x, t + T ) + 2πn. (57)
c c
From the Stokes theorem and by considering a contour on an electric con-
ductor in which the field is in pure gauge Aµ = ∂µ θ, we find that the
magnetic flux through the area SΣ is quantized

B · dS
= A · dx = ·dx = n hc . (58)
e
SΣ Σ Σ
As the magnetic flux is quantized, the current cannot smoothly decay,

so that we have superconductivity [40]. Similarly, it is possible to derive
the other effects characterizing superconductivity such as the Meissner
effect, the Josephson effect, the Little–Parks effect and energy gap opening
[2, 26, 27].
Remarkably, the cyclic dynamics characterizing ECT, the relativistic
modulation of periodicity and the related dispersion relation of the energy
spectrum, the description of antimatter as negative vibrational modes,
etc., can be directly tested in carbon nanotubes. Actually, we have found
www.ebook3000.com
[2,26,27] that ECT provides an elegant and simple new technique to derive
the essential electronic properties of carbon nanotubes or similar graphene
systems.
5. Implications in Modern Physics

Even though in this chapter we have exclusively limited our considerations
to basic aspects of quantum theory typically reported in textbooks, it must
be noted that ECT has found important applications and confirmations in
advanced aspects of modern theoretical physics.
5.1. Gauge interactions from space–time

geometrodynamics
In ECT, both gravitational and gauge interactions are derived from space–
time geometrodynamics, in particular of the local deformations of the met-
ric and of the local rotations of the boundary, respectively [7]. Elementary
particles can be regarded as relativistic reference clocks and the variations
of energy–momentum during interactions are in fact equivalently described
by local and retarded modulations of their de Broglie–Planck periods T µ .
In analogy with general relativity, these modulations of clock rates in dif-
ferent space–time points are encoded in a corresponding curved space–time
metric. The intrinsic periodicity of elementary particles, also implicitly in
undulatory mechanics, is realized by defining the corresponding relativis-
tic action in compact space–time dimensions with covariant PBCs — since
T µ determines the kinematical state and is fixed by the boundary of the
action, we find that interactions are encoded in the geometrodynamics of
the boundaries, providing a fundamental link to Holography [1, 7, 8]. This
implies that local rotations of the space–time boundary can generate local
modulations of Broglie–Planck period, and in turn particular kind of inter-
actions. It turns out that these particular interactions formally correspond
to ordinary gauge interactions, e.g. electromagnetism. Again, the constraint
of periodicity represents the quantization condition “overdetermining” the
gauge dynamics. The result is a formal equivalence to ordinary QED [7].
5.2. Correspondence with extra-dimensional theories

ECT inherits fundamental aspects of modern theories [8]. The cyclic (or,
more in general, compact) world-line parameter associated to the Comp-
ton periodicity of a particle, from a mathematical point of view plays a
132 D. Dolce
role very similar to the cyclic (or compact) extra-dimension of the Kaluza–
Klein theory — the geometrodynamical description of gauge interactions
in this analogy reproduces Kaluza’s miracle — and of the cyclic (com-
pact) world-sheet parameter of closed string theory (open string theory).
To highlight these correspondences, the world-line parameter is also named
“virtual extra dimension” in ECT. The combination of the correspondence
of ECT, on the one hand to classical extra-dimensional theories, and on the
other hand to ordinary QM (Feynman path integral), leads to an intuitive
formal derivation of Maldacena’s conjecture (also known as AdS/CFT or
gauge/gravity duality) by means of simple semi-classical arguments [6, 8].
5.3. Time cycles and time flow

ECT introduces a cyclic character to the ordinary relativistic time — a
cyclic nature of time has been supported by notable philosophers of the
past and present. To understand the concept of time in this approach it is
necessary to distinguish between the time coordinate and the time flow.
Every particle, i.e. every elementary constituent of nature, is described
as a reference clock whose rate is determined by the particle’s mass [19].
The time coordinate has a cyclic (angular) nature for elementary particles,
with the exception of massless particles (photons and gravitons) which have
“frozen” Compton clock — ECT must not be confused with cyclic cosmol-
ogy: roughly speaking, in ETC every elementary particle can be regarded
as a ultra-fast cyclic universe. This radically new description of relativis-
tic time demands for a radical reconsideration of QM paradoxes. Indeed,
from such a cyclic description of relativistic time originates the formal cor-
respondence to ordinary QM, i.e. it allows to encode QM directly into
relativistic dynamics. Though elementary particles are “periodic phenom-
ena”, we must remember that interactions correspond to local and retarded
modulations of the periodicities of these elementary clocks. This leads to
time ordering (an interaction is an event from which it a past and a future
can be established), causality (since the periodicity is locally determined by
the energy, it changes according to the retarded potentials), chaotic evolu-
tions (a system of many periodic phenomena is an ergodic system, if these
interact the system will have chaotic evolution described by the statistical
laws and thus by thermodynamics). The arrow of time emerges in a rela-
tional/statistical/thermodynamics way from the combination of the tricks
of all these elementary clocks constituting a macroscopic system.
www.ebook3000.com
6. Conclusions
Where is the boundary of relativistic space–time? This question cannot be
answered by simply invoking relativistic theory. A fascinating answer to this
question is however suggested by the combination of Newton’s law of inertia
and undulatory mechanics. As also suggested by de Broglie, it follows that
every elementary particle in nature is an intrinsic “periodic phenomenon”,
i.e. every system in nature can be described in terms of elementary space–
time cycles. We have shown that, following the line already pointed out
by Einstein to unify relativistic and QM, the condition of periodicity can
be imposed as constraint to “overdetermine” relativistic mechanics in such
a way that QM emerges as novel relativistic phenomenon, without any
explicit quantisation condition. The resulting cyclic dynamics are formally
equivalent to ordinary QM in its main formulation (canonical and Feynman
formulations). In this chapter we have reported, in a pedagogical way, the
main quantum aspect of the ECT, giving a step by step demonstration of
some of the main evidences, as well as practical applications, of this unified
description of physics.
References
1. D. Dolce, Elementary space–time cycles, Europhys. Lett. 102, 31002 (2013).
2. D. Dolce and A. Perali, The role of quantum recurrence in superconductivity,
carbon nanotubes and related gauge symmetry breaking, Found. Phys. 44,
905–922 (2014).
3. D. Dolce, Elementary cycles of time, EPJ Web Conf. 58, 01018 (2013).
4. D. Dolce, Intrinsic periodicity: The forgotten lesson of quantum mechanics,
J. Phys.: Conf. Ser. 442, 012048 (2013).
5. D. Dolce, On the intrinsically cyclic nature of space–time in elementary par-
ticles, J. Phys.: Conf. Ser. 343, 012031 (2012).
6. D. Dolce, AdS/CFT as classical to quantum correspondence in a Virtual
Extra Dimension, PoS, ICHEP2012, 478 (2013).
7. D. Dolce, Gauge interaction as periodicity modulation, Annals Phys. 327,
1562–1592 (2012).
8. D. Dolce, Classical geometry to quantum behavior correspondence in a virtual
extra dimension, Annals Phys. 327, 2354–2387 (2012).
9. D. Dolce, Clockwork quantum universe, IV Prize, QFXi contest (2011).
10. D. Dolce, de Broglie deterministic dice and emerging relativistic quantum
mechanics, J. Phys.: Conf. Ser. 306, 10 (2011).
11. D. Dolce, Deterministic quantization by dynamical boundary conditions, AIP
Conf. Proc. 1246, 178–181 (2010).
12. D. Dolce, Quantum mechanics from periodic dynamics: the bosonic case, AIP
Conf. Proc. 1232, 222–227 (2010).
134 D. Dolce
13. D. Dolce, Compact time and determinism for bosons: Foundations, Found.
Phys. 41, 178–203 (2011).
14. L. d. Broglie, A tentative theory of light quanta, Phil. Mag. 47, 446 (1924).
15. L. d. Broglie, Recherches sur la théorie des quanta, Ann. Phys. 3, 22 (1925).
16. R. Ferber, A missing link: What is behind de Broglie’s “periodic phe-
nomenon”?, Found. Phys. Lett. 9(6), 575–586 (1996).
17. R. Penrose, Cycles of Time. An Extraordinary View of The Universe, chapter
2.3. Knopf, New York, 2011.
18. P. Catillon et al., A search for the de Broglie particle internal clock by means
of electron channeling, Found. Phys. 38, 659–664 (July, 2008).
19. H. Müller et al., A clock directly linking time to a particle’s mass, Science.
339(6119), 554–557 (2013).
20. G. t. Hooft, The cellular automaton interpretation of quantum mechanics.
A view on the quantum nature of our universe, compulsory or impossible?
(2014), arXiv:1405.1548.
21. G. ’t Hooft. Entangled quantum states in a local deterministic theory, (2009),
arXiv:0908.3408.
22. G. ’t Hooft, Emergent quantum mechanics and emergent symmetries, AIP
Conf. Proc. 957, 154–163 (2007).
23. G. ’t Hooft, The mathematical basis for deterministic quantum mechanics,
J. Phys.: Conf. Ser. 67, 15 (2007).
24. G.’t Hooft, Determinism in free bosons, Int. J. Theor. Phys. 42, 355–361
(2003).
25. G.’t Hooft, TransPlanckian particles and the quantization of time, Class.
Quant. Grav. 16, 395–405 (1999).
26. D. Dolce and A. Perali, On the compton clock and the undulatory nature of
particle mass in graphene systems, EPJ Plus. 140, 41 (2015).
27. D. Dolce and A. Perali, Testing elementary cycles interpretation of quantum
mechanics in superconductivity and graphene physics, to be published on J.
Phys.: Conf. Ser. 626(1), 012062 (2015).
28. R. Kastner, de broglie waves as the “bridge of becoming” between quantum
theory and relativity, Found. Sci. 18(1), 1–9 (2013).
29. A. Einstein, Principe de relativité, Arch. Sci. Phys. Nat. 29(4), 5–28 (1910).
30. A. Pais, Subtle is the Lord: The science and the life of Albert Einstein. Oxford
University Press, 1982.
31. A. Einstein, Bietet die Feldtheorie Möglichkeiten zur Lösung des Quanten-
problems?, S.B. Press. Aked. Wiss. 33, 359–364 (1923).
32. D. Dolce, Unified description of Relativity and Quantum Mechanics from
Elementary Cycles Theory, 12, IYL15-34 (2015) 17–74.
33. H.-T. Elze, Emergent discrete time and quantization: Relativistic particle
with extradimensions, Phys. Lett. A310, 110–118 (2003).
34. B. Rosenstein and L. P. Horwitz, probability current versus charge current
of a relativistic particle, J. Phys. A18, 2115–2121 (1985).
35. R.P. Feynman, Spate-time approach to non-relativistic quantum mechanics,
Rev. Mod. Phys. 20, 367–387 (1948).
www.ebook3000.com
36. H.B. Nielsen and M. Ninomiya, Intrinsic periodicity of time and non-maximal
entropy of universe, Int. J. Mod. Phys. A21, 5151–5162 (2006).
37. J. Cugnon, The casimir effect and the vacuum energy: Duality in the physical
interpretation, Few-Body Syst. 53(1–2), 181–188 (2012).
38. R.L. Jaffe, The Casimir effect and the quantum vacuum, Phys. Rev. D72,
021301 (2005).
39. E.A. Solovév, Classical approach in atomic physics, Eur. Phys. J. D 65(3),
331–351 (2011).
40. S. Weinberg, Quantum mechanics without state vectors, (2014), arXiv:1405.
3483.
www.ebook3000.com
Chapter 5
Observers and Reality

George Jaroszkiewicz
School of Mathematical Sciences, University of Nottingham, Nottingham, UK
1. Introduction
In any theory or discussion about physics, the assumed relationship between
observers and physical reality is fundamental to the interpretation of what
is going on: this relationship underpins the difference between classical
mechanics (CM) and quantum mechanics (QM). If we want to under-
stand the physical universe, it is important to understand this relation-
ship: otherwise, we may be engaging in metaphysics, a scientifically fruitless
activity.
In this context, the quantum relativist John Wheeler said [1]:
“Stronger than the anthropic principle is what I might call the participa-
tory principle. According to it we could not even imagine a universe that
did not somewhere and for some stretch of time contain observers because
the very building materials of the universe are these acts of observer-
participancy. You wouldn’t have the stuff out of which to build the uni-
verse otherwise. This participatory principle takes for its foundation the
absolutely central point of the quantum: No elementary phenomenon is
a phenomenon until it is an observed (or registered) phenomenon.”
In this chapter, we extend the participatory principle to incorporate con-

textuality. By contextuality we mean the circumstances by which the
truth value of a proposition could be established. In mathematics, con-
text includes the axioms relative to which mathematicians prove theorems;
in quantum physics, context includes the apparatus used to confirm pre-
dictions.
Following Wheeler’s participatory principle, we review an approach to
QM centered on the observer and their apparatus rather than systems under
observation (SUOs). To illustrate our approach, we discuss the monitored
137
138 G. Jaroszkiewicz
double-slit (DS) experiment with a quantum register approach tailored to

encode experimental contextuality.
2. Basic Concepts and Terms

We first review some basic concepts and terms.
Physical reality
This is the arena of space, time, and matter that human observers believe
they inhabit and in which are found the SUOs that they observe. The
divergences of quantum field theory (QFT) hint that the block universe
model of reality, that of a four-dimensional (4D) Lorentzian manifold in
which SUOs are embedded, is too simplistic.
Observers
These are complex systems that can observe SUOs and other observers.
Observers have a sense of time, memories, and purpose.
Primary observers
These are observers relative to whom specific truth values and the contexts
of generalized propositions are meaningful. We should not ask a primary
observer to observe themselves because this is self-referential. If O0 is a
primary observer observing SUO S0 , then it is meaningful to discuss the
combination (O0 , S0 ) as a new SUO S1 from the perspective of some new
primary observer O1 . Everett formulated his relative state approach to QM
by asserting that the laws of physics L0 used by O0 to describe S0 had to
be the same as the laws L1 used by O1 to discuss S1 [2].
Generalized propositions
These are propositions of the form (P, CP |O, CO ), where P is a standard
proposition (in mathematics, P is usually a theorem whilst in physics P
is usually a measurement), CP is the relative internal context, O is the
primary observer associated with the proposition and CO is the relative
external context.
Relative internal context
This is whatever the primary observer needs to validate a proposition: math-
ematicians need axioms, physicists need apparatus. Truth values cannot be
obtained without relative internal context, so this gives a method for iden-
tifying this form of context: if we changed it, could we still validate the
proposition?
www.ebook3000.com
Relative external context

This is whatever is needed to identify the primary observer O. It is classical
(i.e. certain from O’s point of view) information held by O, such as O’s
frame of reference and the times and places over which the proposition
is validated. In physics, the time at which a proposition is validated is
especially crucial. For instance, it would have been impossible to measure
the temperature in the early universe before recombination. Of course, such
a temperature is extrapolated from data obtained in the current epoch on
the basis of current knowledge of the laws of physics as we know them.
Our point is that proper science requires this to be properly documented
in generalized propositions. Suppose Dirac was right to suspect that the
constants of physics changed over time [3]?
Another reason for caution here is that counterfactuality is inadmis-
sible in QM: as Wheeler warns us in the above quote, ‘No elementary
phenomenon is a phenomenon until it is an observed (or registered) phe-
nomenon.’
Heisenberg cuts
A Heisenberg cut is a hypothetical line between the worlds of CM and QM.
According to Heisenberg [4],
“The dividing line between the system to be observed and the measuring
apparatus is immediately defined by the nature of the problem but it obvi-
ously signifies no discontinuity of the physical process. For this reason
there must, within certain limits, exist complete freedom in choosing the
position of the dividing line.”
Such cuts are contextual according to Heisenberg, which we agree with,

but there is a problem with the above: we cannot identify SUOs without
apparatus. The implications of Wheeler’s participatory principle is that
apparatus defines SUOs. This has implications for the way scientists use
language. It is metaphysics to talk about a photon traveling from a source
to a detector as if it were a particle. If it did, which path did it take?
In our approach, there is a natural separation between relative internal
context and relative external context and so there is a cut of sorts there. In
our notation for a generalized proposition (P, CP |O, CO ), the vertical bar
represents this cut.
Contextual completeness
A generalized proposition is contextually complete if it can be validated,
i.e. assigned a relative truth value 0 or 1. In mathematics, it is customary
to neglect relative external context: there is a hidden assumption that all

mathematicians would use the given axioms in the same way and arrive at
the same validation value of a theorem. On the other hand, in quantum
physics, both internal and external relative context must be specified: it
makes no physical sense not to identify the observer or the means by which
they can validate propositions.
This gives rise to an amusing analogy between mathematicians and
quantum physicists on the one hand and elementary particles on the other.
Different mathematicians can prove the same theorem, so mathematicians
are like bosons. On the other hand, only one observer can detect a given
quantum event, so physicists are like fermions and obey an observational
exclusion principle.
Contextual incompleteness
A contextually incomplete generalized proposition is one such that relative
internal context and/or relative external context is missing. Such proposi-
tions are denoted (P, ∅|O, CO ), (P, CP |O, ∅), (P, ∅|O, ∅) or even (P, ∅|∅, ∅),
where ∅ denotes the empty set, i.e. total lack of information. Examples
of contextually incomplete propositions are to be found throughout most
scientific papers, which can lead to confusion and fruitless cross-purpose
debate.
Validation
If a generalized proposition can be validated, then it can have a truth value
0 or 1 relative to the primary observer. We write
V(P, CP |O, CO ) = ±1, (1)
where V is the validation map. Not all generalized propositions can be vali-
dated: there is by definition no way of validating a metaphysical proposition.
Physical propositions
These are generalized propositions in physics that can be validated, i.e.
tested by scientists and therefore having truth values 0 (false) or 1 (true).
Metaphysical propositions
These are generalized propositions of the form (P, ∅|∅, ∅), i.e. propositions
making no reference to any validation mechanism or primary observer. Such
www.ebook3000.com
propositions cannot be validated by definition. An archetypical metaphys-

ical proposition is
(“There is a universal wavefunction for the universe”, ∅|∅, ∅).
Finke and Leshke criticized quantum cosmology essentially on these
grounds [5].
Mathematical propositions
These are generalized propositions of the form (P, A|O, ∅), where A is the
system of axioms relative to which O, the primary observer (a mathemati-
cian in this case) can establish the validity of proposition P . The axioms
are to the mathematician what apparatus is to the physicist. Mathemati-
cians generally do not discuss themselves, so relative external context is
usually missing in mathematical discussions. Significantly, constructivists
and intuitionists would like to add relative external context to mathematics,
an acknowledgment of the role of the observer even in mathematics.
Classical propositions
These are generalized propositions of the form (P, ∅|O, CO ), where an exo-
physical or endophysical observer O attached to some frame of reference
described by CO makes a proposition, such as Newton’s laws of motion.
One of the principles of CM is that such a proposition is true or else false
independently of any apparatus: therefore, relative internal context does
not matter. In particular, if P is about the assumed properties of an SUO,
these are asserted to ‘exist’ in CM regardless of observation.
This underlines what we wrote above: the structure of propositions
made in physics is directly influenced by the view taken of the relation-
ship between observers and physical reality.
Quantum propositions
These are generalized propositions of the form (P, CP |O, CO ). When in 1925
Heisenberg formulated his approach to QM he was motivated directly by
the idea that relative internal context does matter, in contrast to the view
taken in CM. Indeed, he went so far as to assert that “... the ‘orbit’ comes
into being only when we observe it ” [6].
According to his own principles, Heisenberg would have no right to
assert that something that was unobserved did not exist. The notion of
existence is contextual in QM. We would add contextuality and amend his
assertion to “... the ‘orbit’ exists as far as we are concerned, if we have
observed it.”
In other words, empirical truth is contextual.
3. Particles
Viewed in the right context, an object such as a football, planet or galaxy
can be discussed as a particle with classical properties such as color and size.
Although photons are often discussed as particles, there does not seem to be
any context where a photon can be thought of as having a geometrical size
and structure. What then does it mean to say that photons are particles?
With Wheeler’s participatory principle in mind, our answer is that par-
ticles such as photons and electrons are meaningful only in terms of sig-
nals received in apparatus. According to Brown, Feynman attempted to
construct a description of electrodynamics from such a signal perspective
during the course of his doctorate [7]. He changed his views about that
possibility subsequently when he found that in order to account for the
Lamb shift he had to use virtual photon contributions involving quantum
fluctuations in the vacuum outside the detectors, rather than the detectors
themselves. We shall comment on this later in this chapter, as it has a
bearing on the description of physical reality that we propose.
4. Questions, Answers, and Laboratories

We define a binary question as one that has a yes or no answer. The vali-
dation of any generalized proposition (P, CP |O, CO ) can always be phrased
in terms of a binary question QP of the form
QP ≡ Is it true that {V(P, CP |O, CO ) = 1}? (2)
This can be done even for metaphysical propositions (in which case the
answer to QP is no).
The basic activity of an observer is to obtain answers to physical ques-
tions, i.e. establish the truth status of generalized propositions using appa-
ratus. It is our critical assumption that all physical questions can be asked
in terms of countable collections of binary questions referred to as regis-
ters. When the rules of QM are applied to these registers they are called
quantum registers.
It can be assumed that experimentalists know the context of their exper-
iments, that is, what the yes–no signals in their apparatus mean. This con-
textual knowledge is then combined with outcome frequencies to validate
propositions about the properties of SUOs. That is not the same thing as
www.ebook3000.com
asserting that those SUOs ‘have’ those properties. The Kochen–Specker the-
orem warns us not to think that quantum states ‘have’ pre-existing values
that are measured by observers [8].
Our use of quantum registers to represent time-dependent signal states
of apparatus [9] allows us to model time-dependent apparatus and the archi-
tecture of important experiments such as delayed choice [10], quantum era-
sure [11] and the Hardy paradox [12].
5. Observers and Experiments

We consider now the observer concept in more depth. We can with con-
fidence assert that all observers exist within the universe, i.e. there are
no truly exophysical observers. In practice, real observers generally go to
very great lengths to approximate exophysics: they shield off portions of
the universe from outside influence and perform their experiments there. In
other words, they very carefully define their relative external and internal
contexts. Once they have created an effective Heisenberg cut based on that
division, they can then safely perform their delicate experiments on the
relative internal side of that line, which we shall refer to as the laboratory,
denoted by Λ. The observers stand on the other side of the line and imagine
that they are exophysical. There are as many different laboratories as there
are observers.
A laboratory need not be localized: an astrophysicist observing light
from a distant galaxy shields off or filters out radiation from other galaxies,
so in such a case the laboratory is certainly non-local. The phenomenon of
galactic lensing is another example that shows that a laboratory can indeed
have cosmological extent. A laboratory is a conceptual device as well as a
physical structure such as the Large Hadron Collider.
The observer has a sense of time which need not be regarded as con-
tinuous. What matters are stages. A stage is analogous to a hyperplane
of simultaneity in relativity: any events assigned to the same stage cannot
have any causal connections between them. However, the events in a given
stage need not be simultaneous relative to the laboratory time in Λ. They
could even be relatively time-like, as long as there is no physical interaction
between them: adequate shielding has to be provided.
What matters is that for all intents and purposes, the observer prepares
and collects information in a discrete succession of stages relative to their
own sense of time, associated with Λ. For example, a Stern–Gerlach exper-
iment consists of many repetitions of a run consisting of two stages: state
Fig. 1. The architecture of the Stern–Gerlach experiment: vertical dotted lines

are stages, circles are ESDs.
preparation occurs at stage 0 and outcome detection occurs at stage 1. In

diagrams, stages are represented by dotted lines and labelled by integers.
An elementary signal detector (ESD) is a hypothetical or actual oppor-
tunity for information input or extraction over a run of an experiment. ESDs
are represented by circles in diagrams and are labelled with two indices:
one index labels the stage and the other distinguishes the ESDs within
that stage. The Stern–Gerlach experiment has the architecture of Fig. 1,
where dotted vertical lines indicate successive stages, indexed by an integer
increasing from left to right to denote the direction of laboratory time.
Such a diagram has a mathematical representation, for which we use
the notation and terminology given in Ref. [12]. At stage n, the ith ESD
Eni has four potential basis states denoted |0, n)i , |1, n)i , |2, n)i , and |3, n)i
respectively. The last state, |3, n)i , is the (information) void state of that
ESD, representing the absence of any physical apparatus corresponding to
that ESD, relative to the given laboratory, at that stage. The existence of
this state in the formalism is required in order to allow for the possibility
of the creation or destruction of apparatus. The absence of a detector can
be an empirical fact and therefore is as much part of the informational
context that an observer has about their laboratory as the presence of
actual detector would be. The void state is in fact the natural state of
ESDs: physicists often require vast resources to lift an ESD from its void
state to any of the other three states, which are physically present in the
laboratory. The information void state is not the same thing as a ground
state or vacuum in relativistic quantum field theory (RQFT).
The other three basis states represent real apparatus created by the
observer in the laboratory. The states |0, n)i and |1, n)i represent the ground
state and signal states respectively of a working detector, whilst |2, n)i
represents the faulty or decommissioned state of that detector.
www.ebook3000.com
At stage n, the void state of a completely empty laboratory is repre-

sented by the infinite tensor product |Ω, n) ≡ i ⊗|3, n)i , which includes
all possible, imaginable ESDs. The infinite cardinality of the set of all such
ESDs is not a problem because an absence of a detector, whilst it is real
information, carries no dynamical content or cost to the observer.
Suppose now at stage n the apparatus in Λ consists of r ESDs each in
its ground state. The total state of the apparatus at that stage is denoted
by the contextual ground state or contextual vacuum:
|0, n) ≡ C1n C2n . . . Crn |Ω, n), (3)
where Cin denotes a construction operator, representing the creation in the

laboratory of the ith signal detector in its ground state. If the ith ESD
is lifted into its signal state, we represent that by an application of the
signal operator Ai+ n to the ground state. The signal operators satisfy the
commutation properties
[Ain , Aj+
n ] = δij (4)
and Ain effectively annihilates the ground state.

We show now how the Stern–Gerlach experiment is represented in our
notation, and then we will discuss the DS and monitored DS experiments.
6. The Stern–Gerlach Experiment

With reference to Fig. 1, we start at initial stage 0 with the source of the
beam creating the initial signal state
|Ψ, 0) ≡ A1+
0 |0, 0). (5)
Since the process involves electrons, we can use conservation of electric

charge to assert that the evolution of the labstate (the state of the appara-
tus, relative to the observer) is given by

|Ψ, 0) → |Ψ, 1) ≡ αA1+1 + βA1
2+
|0, 1), (6)
where |α|2 + |β|2 = 1. There are four outcome probabilities measurable at
stage 1:
P 00 ≡ (Ψ, 1|P11 P21 |Ψ, 1) = 0,

P 10 ≡ (Ψ, 1|P̄11 P21 |Ψ, 1) = |α|2 ,
P 01 ≡ (Ψ, 1|P11 P̄21 |Ψ, 1) = |β|2 , (7)
11
P ≡ (Ψ, 1|P̄11 P̄21 |Ψ, 1) = 0,
where Pin and P̄in are appropriate projection operators at stage n for ground
and signal states of the ith ESD respectively.
The formalism allows for two unusual possibilities: P 00 is the proba-
bility that no signal is received by either the up or down channels of the
experiment, whilst P 11 is the probability that a signal is seen in both.
These two possibilities are ruled out by charge conservation in this exper-
iment, but would in general be relevant to a complete description of an
arbitrary quantum process with the same architecture. Our formalism is a
half-way house between quantum wave-mechanics, which preserves particle
number, and QFT, which allows for variable particle number. In general,
there is no law that requires conservation of total signal, except if, for
example, electric charge or some equivalent conserved quantum number is
involved [9].
7. The DS Experiment
The architecture of the DS experiment is represented in Fig. 2, where now
the two ESDs at stage 1 are not tested for outcome but allowed to pass
quantum information onto stage 2, at which stage an array consisting of r
Fig. 2. The architecture of the DS experiment.
www.ebook3000.com
ESDs is used to extract information. As part of the calibration process, the

parameters of the experiment would be so organized that each of the two
ESDs at stage 1 could act as a source of amplitude on the ESDs at stage 2.
The general rule in this case is given by
r

Ai+
1 |0, 1) → U ji Aj+
2 |0, 2), i = 1, 2, (8)
j=1

with the complex transmission coefficients U ji satisfying the conditions
r
r
∗
|U ji |2 = 1, i = 1, 2, U j1 U j2 = 0. (9)
j=1 j=1
Quantum properties emerge when both ESDs at stage 1 act as simul-

taneous sources and their amplitudes on the stage 3 ESDs are superposed.
Then the evolution of the labstate is given by

|Ψ, 0) → |Ψ, 1) ≡ αA1+ 1 + βA1
2+
|0, 1)
r

→ |Ψ, 2) = αU j1 + βU j2 Aj+ 2 |0, 2). (10)
j=1
i
The outcome probability P at the jth detector on the screen is given by
P j ≡ (Ψ, 3|P̄j |Ψ, 3) = |αU j1 + βU j2 |2 (11)
which shows quantum interference terms and satisfies the total probability
conservation rule
r
P j = 1. (12)
j=1
8. The Monitored DS Experiment

In this variant of the DS experiment, an additional detector, denoted by
the circle labeled 0, 2 in Fig. 3, is used to trigger if a particle has passed
through slit #1 at stage 1 (according to the traditional interpretation of
the results). In this case, we have the dynamical rule
r
r

j+
A1+
1 |0, 1) → U j1 A0+
2 A2 |0, 2), A2+
1 |0, 1) → U j2 Aj+
2 |0, 2). (13)
j=1 j=1
Note that slit #1 now triggers ESD E20 simultaneously with one of the other
ESDs in stage 2, whereas slit #2 does not trigger E20 . Therefore, if E20 is
Fig. 3. The architecture of the monitored DS experiment, where one slit is mon-
itored for the passage of a signal.
found in its signal state, the interpretation would be that a “particle” had
passed through slit #1 and not slit #2.
The final state in this case is given as before by a superposition of
contributions from both slits, i.e.
r
j+

|Ψ , 2) = αU j1 A0+
2 + βU
j2
A2 |0, 2). (14)
j=1
This time, however, we find the outcome probability P j at the jth detector
at stage 2 given by
P j ≡ (Ψ , 3|P̄j |Ψ , 3) = |α|2 |U j1 |2 + |β|2 |U j2 |2 , (15)
i.e. with no interference terms. In this case total probability is also con-
served.
In this approach, the formalism naturally accounts for the absence of
quantum interference terms in the DS experiment if any of the slits are
monitored. Technically, this is explained by the change in relative internal
context: by placing a monitor on slit #1, the dynamics maps the signal
www.ebook3000.com
from that slit into a quantum register state (in the final quantum register)
that is orthogonal to the signal mapped from slit #2. If no monitoring takes
place, then the dynamics maps into two non-orthogonal vectors in the final
quantum register.
What is satisfactory about this account of the disappearance of inter-
ference terms is that the discussion is generic, i.e. it is independent of the
specific details of the amplitude coefficients {U2ji }, all that is needed being
the semi-unitaritya rules (9).
The approach taken here should allow for a comprehensive treat-
ment of large scale, time-dependent networks of ESDs. This includes par-
tial information extraction whilst the process is still running, using the
faulty/decommissioned state |2, n) of ESDs to represent detectors from
which information has been extracted irreversibly.
It was recognized a long time ago that QM could be regarded in terms
of binary questions and projection operators. Our approach is essentially
that with as much emphasis on contextuality and avoidance of metaphysics
as possible.
9. The Contextuality of Space and Time

A feature of our approach is that there is an appeal to the discreteness found
in real physics: although space and time appear continuous ordinarily, all
experiments run on discrete lines. Given that all apparatus is made up of
atoms and molecules, then there can be no spatially continuous measure-
ments and, given that information extraction and processing takes finite
time, there can be no truly continuous-time measurements either. Studies
of the quantum Zeno effect that refer to continuous time do not faithfully
model what happens in the laboratory but work with an effective concept
of time analogous to the concept of temperature in statistical mechanics.
This begs the question: in what way are space and time real? Do they
have an existence independent of observers? According to General Relativ-
ity and RQFT, empty space–time has dynamical properties, but Wheeler’s
participatory principle requires us to add the observers into the discussion.
According to Schwinger [13],
“The mathematical machinery of quantum mechanics is a symbolic
expression of the laws of atomic measurement, abstracted from the
a We use the terminology semi-unitary to remind us of the fact that in the DS experiment.
the map from stage 1 to stage 2 effectively increases the dimension of the Hilbert space,
so the mapping is not unitary in the usual sense.
specific properties of individual techniques of measurement. In particular,

the space–time manifold that is the background of any quantum-
mechanical description is an idealization of the function of a measure-
ment apparatus to define a macroscopic frame of reference.”
Our concept of contextual ground state or vacuum goes some way toward
realizing Schwinger’s ideas.
This brings us to the singularities of QFT. These arise precisely because
it has generally been assumed that relative internal context and relative
external context have to be modeled by the same idealized space–time
structure, a 4D Lorentzian manifold. If we accept Schwinger’s point of view
that space–time is an abstraction, then perhaps this is the place to start
remodelling our approach to physics. Quantum principles dictate that we
take seriously only those aspects of reality that we can observe. A primary
observer can be reasonably confident about their microscopic view of space
and time, because that is part of their relative external context, which is
classical and its structure can be validated. But can any observer be jus-
tified in using the same approximation in the modeling of relative internal
context?
Our intuition is that the appearance of infinities on the relative internal
context side of the generalized proposition cut is an indicator that cur-
rent modeling in RQFT is conceptually inadequate. The great challenge is
to find a formalism that keeps relative external context looking much as
it does now with a description of relative internal context that is finite.
Feynman abandoned his initial empiricist formulation of QM and papered
over the divergences with clever tricks. Dirac too was preoccupied with the
divergences, but was never satisfies with the way that Feynman and others
used renormalization to avoid the issues [14]. The detector formalism we
have outlined seems a reasonable place to start rethinking the problems: as
observers we will always be dealing with a finite or possibly countable set of
detectors with finite properties, rather than with a continuum of idealized
space–time coordinates.
Acknowledgment
I am extremely grateful to Prof. Ignazio Licata for giving me this opportu-
nity to present my ideas.
References
1. J. A. Wheeler, From the big bang to the big crunch, Cosmic Search Magazine,
1(4), 1979, Interview with J. A. Wheeler.
www.ebook3000.com
2. H. Everett, ‘Relative State’ formulation of quantum mechanics, Rev. Mod.

Phys. 29(3), 454–462 (1957).
3. P.A.M. Dirac, A new basis for cosmology, Proc. Roy. Soc (London) A,
165(921), 199–208 (1938).
4. W. Heisenberg, Questions of principle in modern physics, In Philosophic
Problems in Nuclear Science, Faber and Faber, London, 1952.
5. H. Fink and H. Leschke, Is the universe a quantum system? Found. Phys.,
13(4), 345–356 (2000).
6. W. Heisenberg, Uber den anschaulichen inhalt der quanten theoretischen
kinematik und mechanik, Zeits. Physik 43, 172–198 (1927). Reprinted
English translation: “The Physical Content of Quantum Kinematics and
Mechanics” in Quantum Theory of Measurement, ed. by J.A. Wheeler and
W.H. Zurek, Princeton University Press, N.J., 1983.
7. L.M. Brown, Feynman’s Thesis, A New Approach to Quantum Theory, World
Scientific, 2005.
8. S. Kochen and E. Specker, The problem of hidden variables in quantum
mechanics, J. Mathematics and Mechanics, 17, 59–87 (1967).
9. G. Jaroszkiewicz, Quantized detector networks: A review of recent develop-
ments, Int. J. Modern Phys. B 22(3), 123–188 (2008).
10. Y. Kim, R. Yu, S. Kulik, Y. Shih, and M. Scully, A delayed choice quantum
eraser, Phys. Rev. Lett. 84, 1–5 (2000).
11. S.P. Walborn, M.O. Terra Cunha, S. Pádua, and C.H. Monken, Double-slit
quantum eraser, Phys. Rev. A 65 033818 1–033818 6 (2002).
12. G. Jaroszkiewicz, Towards a dynamical theory of observation, Proc. Roy.
Soc. A 466(2124), 3715–3739 (2010).
13. J. Schwinger, Spin, statistics and the TCP theorem, Proc. N. A. S., 44,
223–228 (1958).
14. P.A.M. Dirac, The Principles of Quantum Mechanics, Oxford University
Press, London, 1958.
www.ebook3000.com
Chapter 6
The Stability of Physical

Theories Principle
R. Vilela Mendes∗
Centro de Matemática e Aplicações Fundamentais, Univ. Lisboa,
Av. Gama Pinto 2, 1649-003 Lisboa, Portugal
rvilela.mendes@gmail.com
1. Introduction: Physical Models and Structural

Stability
When models are constructed for the natural world, it is reasonable to
expect that only those properties of the models that are robust have a
chance to be observed. Models or theories being approximations to the
natural world, it is unlikely that properties that are too sensitive to small
changes (that is, that depend in a critical manner on particular values
of the parameters) will be well described in the model. If a fine tuning
of the parameters is needed to reproduce some natural phenomenon, then
the model is basically unsound and its other predictions are expected to
be unreliable. For this reason, a good methodological point of view, in
the construction of physical theories, consists in focusing on the robust
properties of the models or, equivalently, to consider only models which are
stable, in the sense that they do not change, in a qualitative manner, when
some parameter changes. This is what will be called the stability of physical
theories principle (SPTP).
The stable-model point of view had a large impact in the field of
nonlinear dynamics, where it led to the rigorous notion of structural sta-
bility [1, 2]. As already pointed out by Flato [3] and Faddeev [4], the same
∗ Also at Instituto de Plasmas e Fusão Nuclear — IST, rvmendes@fc.ul.pt.
153
154 R. Vilela Mendes
pattern seems to occur in the fundamental theories of Nature. In fact, the

two physical revolutions of the last century, namely, the passage from non-
relativistic to relativistic and from classical to quantum mechanics, may be
interpreted as transitions from two unstable theories to two stable ones.
Because a theory is a mathematical model for the natural world, stabil-
ity of a theory is stability of its mathematical structure. A mathematical
structure is said to be stable (or rigid ) for a class of deformations, if any
deformation in this class leads to an equivalent (isomorphic) structure. The
idea of stability of the structures provides a guiding principle to test either
the validity or the need for generalization of a physical theory. Namely, if
the mathematical structure of a given theory is not stable, one should try
to deform it until one falls into a stable one, which has a good chance of
being a generalization of wider validity.
When a mathematical structure is deformed, the deformation depends
on a certain number of parameters. Typically, if one starts from an unsta-
ble theory Tα0 , that corresponds to a particular value α0 of the parameter
α, α0 will be an isolated point, in the sense that for any other value α
of the parameter in a neighborhood of α0 , the theory Tα is not equiva-
lent to Tα0 . Conversely, a stable theory would be one for which α0 has a
neighborhood of theories all of them equivalent to Tα0 . Therefore, when one
deforms an unstable theory and falls into a stable one, the exact value of
the deformation parameter that corresponds to the actual physical theory
cannot be obtained from deformation theory because, from this point of
view, all values for which the theory is stable are equivalent. The defor-
mation parameters are therefore the natural fundamental constants that
have to be obtained from experiment. In this sense, deformation theory not
only is the theory of stable theories, it is also the theory that identifies the
fundamental constants.
The construction of physical theories operates at several distinct struc-
tural levels and, at each level, distinct mathematical structures are involved.
Therefore, the application of the ideas of stability and deformation to the
distinct structural levels requires a precise formulation of deformation the-
ory in several mathematical disciplines. Analyzing the existing physical the-
ories, one identifies a hierarchy of structural levels. In the first, which one
may call the logical level, are the basic hypothesis about what is observable
and what is not, what kind of questions can be settled by experiment and
how these questions are interrelated. At this level, one finds the distinction
between classical and quantum physics. In the literature dedicated to the
foundations of science, one finds, at times, some confusion concerning what
www.ebook3000.com
The SPTP 155
distinguishes classical from quantum mechanics. For example, one finds the
statement that classical mechanics is deterministic whereas nondetermin-
ism is the hallmark of quantum physics. In fact, quantum mechanics is as
deterministic as classical mechanics, in the sense that the Schrödinger equa-
tion is as deterministic as Hamilton’s equations. Determinism is a property
of the equations that define the time evolution and therefore it is a dynam-
ical question, not a question concerning the logical structure of the theory.
What happens in quantum theory is that, as in any logical structure, there
are questions that can be raised and questions that cannot. As Feshbach
and Weisskopf [5] said: “If you make a silly question, you obtain a silly
answer ”.
At the second level, which may be called the kinematical level, one
defines what are the observable quantities (the observables) and what are
the relations between them. At this level, one also defines what are the
mathematical quantities that in the theory correspond to each one of the
experimental apparatus. Finally, in the third level, called the dynamics, one
includes all the hypothesis relating to time evolution of the physical systems
and their interactions. The three levels of the theoretical structure define
a hierarchy of hypothesis. Hence, with one logic several kinematics may be
used and many different dynamics may be associated to each kinematics.
The hypothesis of the theory include a certain number of manipulation rules
which are needed to predict the results that are to be expected from the
experiments. These results (in general numbers) are then compared with
the corresponding results obtained in the experiments. This comparison
establishes the agreement or disagreement between the theoretical predic-
tions and the experimental results. Note that it is only at this stage that
the theory (a mathematical entity) establishes its contact with the physical
world. In particular, it is not essential and sometimes not even desirable for
all the entities in the model to have a direct physical interpretation. The
“external” physical world may contain many variables to which we have
no direct access, or that we do not care about, when we restrict ourselves
to a certain set of experiments and apparatus. Likewise, the mathematical
model may have parameters and internal entities which have no direct rela-
tion to external observable quantities. The only criterion of validity of the
theory is the agreement of its output (that is, the measurable predictions)
with the experimentally observed quantities. It is only at this level that
the theory, a mathematical entity, comes into contact with what is called
“reality”, whatever it means. One should also bear in mind the nature of
Fig. 1. The hierarchy of hypothesis in the construction of physical theories.
this precarious contact and never be misled into confusing the model with
the object that is being modeled.
As suggested in Fig. 1,a the evolution of the theoretical models oper-
ates by loops, with the signal of the theory-experiment comparison being
fed back into the model, leading to changes in the dynamics which lead
to new predictions, which are compared once more, etc. If after a number
of such steps a reasonable agreement is not obtained, one may be led to
broaden the scope of the feedback loop, that is, one might be led to change
the kinematical or even the logical structure of the theory. The scientific
revolutions that led from Galilean to Lorentzian mechanics and from clas-
sical to quantum mechanics are examples of a change of the kinematics and
a change of the logics.
The separation between theoretical construction and experimental ver-
ification is however not so clear-cut as one might be led to believe from
the discussion above. The experimental results, which serve as a control for
the theoretical framework, are never pure empirical data in the sense that
when experiments are designed to test a theoretical model, they are them-
selves contaminated by the prejudices of the theory. The following remark
by Misner, Thorne and Wheeler [6] is particularly relevant:
“All the laws and theories of physics have this deep and subtle charac-
ter, that they both define the concepts they use and make statements about
these concepts. Contrariwise, the absence of some body of theory, law and
principle deprives one of the means properly to define or even use concepts.
a Adapted from unpublished lecture notes of Prof. F. A. Matsen at University of Texas
(Austin).
www.ebook3000.com
The SPTP 157
Any forward step in human knowledge is truly creative in this sense: that
theory, concept, law and method of measurement — forever inseparable —
are born into the world in union”
The structuring effect of the theory is an important instrument in the
interpretation of the experimental data. On the other hand, prejudices are
thereby introduced in the analysis which may lead to neglecting some infor-
mation contained in data for which there is as yet no theoretical interpre-
tation.
Concerning the SPTP which is the main concern in this chapter, one
sees that to be able to discuss stability issues at all levels of the theoretical
construction one has to identify the nature of the mathematical framework
that is relevant at each one of the levels. For the structural stability of
nonlinear dynamics, the needed mathematical framework is the theory of
stable vector fields and differentiable maps. To discuss stability of the kine-
matical level, one notices that after the definition of a certain number of
observables, the structure of kinematics is the structure of the algebra of
these observables. For the logical level because logical questions may some-
times also be framed in an algebraic setting, the mathematical framework
is also an algebraic one. Note however, that to frame the logical issues in
algebraic form some choice of observables is in general needed and the dis-
cussion of stability is no longer a purely logical question. It would be more
appropriate to consider the lattice of propositions and discuss the stability
issue in the framework of lattice theory. However, as far as I know, there is
not yet a well developed deformation theory for lattices. Therefore, for the
time being, it seems appropriate to discuss the stability issues both for the
kinematical and the logical levels using algebraic tools.
The fact that semi-simple algebras are deformation-stable led Segal [7]
to propose in 1951 that, in its evolution, physical theories would tend to
be framed in terms of such algebras. However, the stability principle is
more general than the simplicity criterion because not all stable algebras
are semi-simple [8] and, for example, dynamical stability issues are not
necessarily algebraic. Nevertheless, the algebraic simplicity principle is a
powerful one, which led to interesting developments (see Finkelstein and
collaborators [9–13]).
Section 2 contains a short review of the stabilizing deformations that
lead from Galilean to relativistic dynamics and from classical to quantum
mechanics. Also discussed is the finite versus infinite dimensional issue when
dealing with algebraic deformation questions.
Section 3 examines the stability of the algebra that is obtained by com-

bining the algebras of relativistic and quantum mechanics, that is, the
Heisenberg–Poincaré algebra. One finds that the combined algebra of rel-
ativistic quantum mechanics is not stable and its stabilization by a defor-
mation forces the introduction of two length parameters, one of which
will probably have the status of a new fundamental constant. In the new
deformed algebra, the space–time coordinates no longer commute and, at
the scale where the effects of a non-zero fundamental length may be felt,
the geometry of space–time is necessarily a non-commutative geometry.
The consequences of this non-commutativity of the space–time coordinates,
their geometric aspects and experimental tests have been discussed in sev-
eral publications. The main results are summarized and some new conse-
quences are explored.
Section 4 describes structural stability of maps, its use in nonlinear
dynamics as well as the possible relevance to universality and critical phe-
nomena.
Finally, Appendix A is a review of structural stability in dynamical
systems theory, which is the field where the importance of stable theories
was first emphasized and Appendix B contains a summary of results on
deformation theory of algebras. The mathematical results contained in these
appendices, which are spread over many texts, are included here to provide
a first working knowledge on deformation tools for the reader interested in
pursuing stability explorations in his domain.
2. From Galilean to Relativistic Dynamics and From

Classical to Quantum Mechanics
Within the deformation theory of algebras, the transitions from Galilean
to relativistic and from classical to quantum mechanics may be interpreted
as the stabilizing deformations of two unstable theories.
The Lie algebra of the homogeneous Galilean group, the kinematical
group of non-relativistic mechanics, is:
[Ji , Jj ] = iijk Jk , (1)

[Ji , Kj ] = iijk Kk , (2)
[Ki , Kj ] = 0, (3)
where the angular momenta Ji are the generators of rotations and the
boosts Ki are the generators of velocity transformations. The second
www.ebook3000.com
The SPTP 159
cohomology group (Appendix B) does not vanish because, for example,

φ1 (Ki , Kj ) = iijk Jk and φ1 = 0 for all other arguments, is a 2-cocycle
that is not a 2-coboundary. The deformation
1
[Ki , Kj ] = −i ijk Jk , (4)
c2
leads to the Lorentz algebra which, being semi-simple, has vanishing second
cohomology group and is stable. The deformation parameter 1c (the inverse
of the speed of light) is a fundamental constant.
For the deformation leading from classical to quantum mechanics, recall
that the phase space of classical mechanics is a symplectic manifold W =
(T ∗ M, ω) where T ∗ M is the cotangent bundle over configuration space M
and ω is a symplectic form. In local (Darboux) coordinates {pi , qi } the
symplectic form is

dω = dpi ∧ dqi
The Poisson bracket gives a Lie algebra structure to the C ∞ -functions on

W , namely,
∂f ∂g ∂f ∂g
{f, g} = − (5)
i
∂qi ∂p i ∂p i ∂qi
in local coordinates.
The transition to quantum mechanics is now regarded as a deformation
of this Poisson algebra [14]. Let, for example, T ∗ M = R2n . Then ω =
i j
i i+n
1≤i,j≤2n ωij dx ∧ dx = 1≤i≤n dx ∧ dx .
Consider the following bidifferential operator

P r (f, g) = ω i1 j1 . . . ω ir jr ∂i1 . . . ∂ir f ∂j1 . . . ∂jr g, (6)
i1 ...ir ,j1 ...jr
where P 1 (f, g) is the Poisson bracket. P 3 (f, g) is a non-trivial 2-cocycle and,

barring obstructions, one expects the existence of non-trivial deformations
of the Poisson algebra.
Existence of non-trivial deformations has been proved in a very general
context [15–18]. They always exist if W is finite-dimensional and for a flat
Poisson manifold they are all equivalent to the Moyal bracket [19]

2 3
[f, g]M = sin P (f, g) = {f, g} − P (f, g) + · · · . (7)
2 4.3!
1
Moreover, [f, g]M = i (f ∗ g − g ∗ f ) where f ∗ g is an associative star-
product

f ∗ g = exp i P (f, g) . (8)
2
Correspondence with quantum mechanics formulated in Hilbert space is
obtained by the Weyl quantization prescription. Let f (p, q) be a function
in phase space and f its Fourier transform. Then, if to the function f we
associate the Hilbert space operator

xi Qi + yi Pi
Ω(f ) = f(xi , yi ) exp −i dxi dyi

∂
with Qi Ψ = xi Ψ and Pi Ψ = −i ∂x i
Ψ , one finds
[Ω(f ), Ω(g)] = −iΩ([f, g]M ).
In the left-hand side is the usual commutator of Hilbert space operators.

Therefore, quantum mechanics may be described either by associating self-
adjoint operators in Hilbert space to the observables or, equivalently, by
staying in the classical setting of phase space functions but deforming their
product to a ∗ product and the Poisson bracket to the Moyal bracket.
The quantization-by-deformation program initiated in Ref. [14] was later
on considerably extended to general Poisson manifolds which are not nec-
essarily sympletic manifolds [20–22]. One of the main results states that
there is a canonical correspondence between deformations of an algebra A
of C ∞ functions on a Poisson manifold M and formal Poisson structures
(πt = tπ1 + t2 π2 + · · · ) on A [23]. Furthermore, an explicit deformation
formula is provided for M = Rn and the product of the deformed algebra is
a star product, that is, in ∗ = Σtn Bn the Bn ’s are bidifferential operators.
There is a basic difference in the deformations leading from non-
relativistic to relativistic and from classical to quantum mechanics. In the
first case, one deals with the deformation of a finite-dimensional algebra
and, in the second, with the more complex case of deformation of an infinite-
dimensional algebra of functions. With the benefit of hindsight, one may
simplify the presentation by using for classical mechanics, instead of the
Poisson algebra in phase space, a formulation in Hilbert space. Then the
transition appears in both cases as a deformation of a finite-dimensional
Lie algebra. This not only simplifies the presentation but is the appro-
priate setting for further analysis of the stability of relativistic quantum
mechanics.
www.ebook3000.com
The SPTP 161
A description of classical mechanics by operators in Hilbert space was

proposed soon after the discovery of quantum mechanics by Koopman [24]
and von Neumann [25]. A constant energy surface ΩE in the phase space of
N particles carries an invariant measure µE , which is the restriction of the
Liouville measure d3N xd3N p to ΩE . In the space of square-integrable func-
tions L2 (ΩE , µE ), the Hamiltonian flow Tt induces an unitary operator by
(Ut f )(w) = f (Tt w), (9)
where w ∈ ΩE and f ∈ L2 (ΩE , µE ). Unitarity is a consequence of the

invariance of the measure, that is µ(Tt−1 F ) = µ(F ) for a measurable set
F ∈ ΩE .
In the Hilbert space L2 (ΩE , µE ), classical mechanics has an operator
formulation. The time evolution is induced by the unitary operator Ut as
in quantum mechanics and the observables are smooth functions on ΩE ,
which act as multiplicative operators in L2 (ΩE , µE ).
Considered as multiplicative operators in Hilbert space, the functions
of coordinates and momenta are an infinite-dimensional abelian algebra.
However, in the Hilbert space formulation we need not consider explicitly
the infinite-dimensional algebra because the full content of the theory is
obtained by selecting a finite set of paired observables (pi , xi ) or (pi , yi =
eixi ) and defining its transformation properties under Ut and its algebraic
properties which, in classical mechanics, are
[pi , xj ] = [pi , pj ] = [xi , xj ] = [pi , yj ] = 0. (10)
The transition to quantum mechanics is now done by the replacement of

this Abelian algebra by the Heisenberg algebra
[pi , pj ] = [xi , xj ] = 0, (11)

[xi , pj ] = iδij , (12)
where is the identity operator, a trivial center of the algebra of observ-

ables. The infinite-dimensional Moyal algebra is therefore replaced by the
simpler finite-dimensional Heisenberg algebra. The role of the Heisenberg
algebra, in the context of deformation theory has however to be discussed
carefully. Consider the one-dimensional (1D) case of a classical abelian alge-
bra [x, p] = 0. This Abelian algebra is clearly not stable and in its neigh-
borhood there is the algebra
[x, p] = ix (13)

or the Heisenberg algebra

[x, p] = i (14)
which is the central extension of the abelian algebra. Algebra (13) is a stable
algebra. Indeed, the only stable algebra in two dimensions is isomorphic to
[26]
[Y, X1 ] = X1 (15)
but the Heisenberg algebra itself is not stable.
There are two ways of looking at the instability of the Heisenberg alge-
bra. First, if we consider it as a tridimensional algebra, [X2 , X3 ] = X1 (all
the other commutators being zero), the complete structure of its neighbor-
hood, in the space of Lie algebra laws, is known [27]. Namely, the Heisenberg
algebra is a contraction of any algebra of the same dimension that carries a
linear contact form. Conversely, any perturbation of the Heisenberg algebra
supports a linear contact form. For example, from the Lie algebra of SO(3),
[X1 , X2 ] = X3 , [X2 , X3 ] = X1 , [X3 , X1 ] = X2
which is semi-simple and therefore stable, with the following linear change
of coordinates
√ √
Y1 = X1 , Y2 = X2 , Y3 = X3 ,
one obtains
[Y1 , Y2 ] = Y3 , [Y2 , Y3 ] = Y1 , [Y3 , Y1 ] = Y2
and in the → 0 limit one obtains the Heisenberg algebra.
We could also have considered the Heisenberg algebra as a 2D algebra
with a trivial center. That is, we restrict our deformations to those that
preserve the zero commutator of X1 with the other two elements. Consider
in this case the deformation
[X2 , X3 ] = X1 + αX2 + βX3 .
With the linear change of variables
Y2 = αX2 + X1 + βX3 , Y3 = α−1 X3 ,
we now fall on the stable 2D algebra (15), [Y2 , Y3 ] = Y2 .
We conclude in both cases that the Heisenberg algebra is unstable and
has a stable algebra in its neighborhood. Therefore it would seem, at first
sight, that the Hilbert space construction leads to conclusions different
www.ebook3000.com
The SPTP 163
from the phase space construction described before, which interprets the
transition from classical to quantum mechanics as a deformation from an
unstable Poisson algebra to the stable Moyal–Vey algebra. A simple rea-
soning shows however that this is not the case and that the constructions
are indeed equivalent and they are both the transition from an unstable
classical algebra to a stable quantum algebra. The apparent difference is
merely an artifact of the singling out of x as the observable, when in fact
the observables are all the smooth functions of x (and p). Consider the
explicit representation
d
p= , x = x.
i dx
The physical content of the theory will be the same if instead of the coor-
dinate x we consider any linear or nonlinear function of x. In particular,
considering y = exp(ix), one obtains the algebra
[p, y] = y
which is isomorphic to the stable 2D algebra (15). Hence, the Heisenberg

algebra is equivalent, through a nonlinear coordinate transformation that
preserves the physical content, to a stable algebra. In this sense, the tran-
sition from classical to quantum mechanics is again seen to be a stabiliz-
ing deformation of an unstable algebra. The main reason why the coor-
dinate choice leading to the Heisenberg algebra is physically convenient
is that the observable p has then a simple interpretation as the gener-
ator of translations in x. This example also shows that, when selecting
a finite subset of observables rather than an infinite-dimensional space
of functions, the notion of linear equivalence of algebras, in the sense
of (B.5), is not sufficient for the stability analysis and one should also
consider nonlinear transformations preserving the physical content of the
theory.
In both the Galilean and the Poisson algebra cases, the deformed alge-
bras are all equivalent for non-zero values of c12 and of . This means that
although we could have derived relativistic and quantum mechanics purely
from the stability of their algebras, the exact values of the deformation
parameters cannot be obtained from algebraic considerations. The defor-
mation parameters are therefore the natural fundamental constants to be
obtained from experiment. It is in this sense that deformation theory not
only is the theory of stable theories, it also is the theory that identifies the
fundamental constants.
3. Stabilizing the Heisenberg–Poincaré algebra

In Sec. 2, both the transition from Galilean to Lorentzian and the transition
from classical to quantum mechanics are cast as deformations of finite-
dimensional Lie algebras of operators in Hilbert space. A trivial point in
this construction, which however has non-trivial consequences, is the fact
that, to have both constructions in a finite-dimensional algebra setting,
it is essential to include the coordinates as basic operators in the defining
(kinematical) algebra of relativistic quantum mechanics. The full algebra of
relativistic quantum mechanics will contain the Lorentz algebra (1, 2, 4), the
Heisenberg algebra for the momenta and space–time coordinates (Pµ , xν )
in Minkowski space and also the commutators that define the vector nature
(under the Lorentz group) of Pµ and xν . Defining
Mij = ijk Jk , M0i = Ki
and measuring velocities and actions in units of c and (that is c = = 1),

one obtains
[Mµν , Mρσ ] = i(Mµσ ηνρ + Mνρ ηµσ − Mνσ ηµρ − Mµρ ηνσ ), (16)
[Mµν , Pλ ] = i(Pµ ηνλ − Pν ηµλ ), (17)
[Mµν , xλ ] = i(xµ ηνλ − xν ηµλ ), (18)
[Pµ , Pν ] = 0, (19)
[xµ , xν ] = 0, (20)
[Pµ , xν ] = iηµν (21)
with ηµν = (1, −1, −1, −1). This algebra, the Heisenberg–Poincaré algebra,
is the algebra of relativistic quantum mechanics 0 = {Mµν , Pµ , xµ , }.
We know that the Lorentz algebra, {Mµν }, being semi-simple, is stable
and that each one of the 2D Heisenberg algebras {Pµ , xµ } is also stable in
the nonlinear sense discussed in Sec. 2. When the algebras are combined
through the covariance commutators (17–18), the natural question to ask
is whether the whole algebra is stable or whether there are any non-trivial
deformations.
The answer is that the algebra 0 = {Mµν , Pµ , xµ , } defined by
Eqs. (16)–(21) is not stable [28]. This is shown by exhibiting a two-
parameter deformation of 0 to a simple algebra which itself is stable.
To understand the role of the deformation parameters consider first the
Poincaré subalgebra P = {Mµν , Pµ }. It is well known that already this
www.ebook3000.com
The SPTP 165
subalgebra is not stable and may be deformed [3] [29] to the stable simple
algebras of the De Sitter groups O(4, 1) or O(3, 2). Writing
1
Pµ = Mµ4 (22)
R
the commutation relations [Mµν , Mρσ ] and [Mµν , Pλ ] are the same as before,
that is, (16)–(17), and [Pµ , Pν ] becomes
4
[Pµ , Pν ] = −i Mµν . (23)
R2
Equations (16), (17), and (23), all together, are the algebra
[Mab , Mcd ] = i(−Mbd ηac − Mac ηbd + Mbc ηad + Mad ηbc ) (24)
of the five-dimensional pseudo-orthogonal group with metric
ηaa = (1, −1, −1, −1, 4), 4 = ±1.
That is, the Poincaré algebra deforms to the stable algebras of O(3, 2) or
O(4, 1), according to the sign of 4 .
This instability of the Poincaré algebra is well understood. It simply
means that flat space is an isolated point in the set of arbitrarily curved
spaces. Faddeev [4] points out that the Einstein theory of gravity may also
be considered as a deformation in a stable direction. This theory is based
on curved pseudo Riemann manifolds. Therefore, in the set of Riemann
spaces, Minkowski space is a kind of degeneracy whereas a generic Riemann
manifold is stable in the sense that in its neighborhood all spaces are curved.
However, as long as the Poincaré group is used as the kinematical group
of the tangent space to the space–time manifold, and not as a group of
motions in the manifold itself, it is perfectly consistent to take R → ∞ and
this deformation would be removed.
For the full algebra, 0 = {Mµν , Pµ , xµ , }, the situation is more inter-
esting. In this case, the stabilizing deformation [28] is obtained by setting
1
Pµ = Mµ4 , (25)
R
xµ =
Mµ5 , (26)

= M45 , (27)
R
to obtain
4
[Pµ , Pν ] = −i Mµν , (28)
R2
[xµ , xν ] = −i5
2 Mµν , (29)
[Pµ , xν ] = iηµν , (30)
4
[Pµ , ] = −i 2 xµ , (31)
R
[xµ , ] = i5
2 Pµ (32)
with [Mµν , Mρσ ], [Mµν , Pλ ] and [Mµν , xλ ] being the same as before.
The stable algebra ,R to which 0 has been deformed is the algebra
of the six-dimensional pseudo-orthogonal group with metric
ηaa = (1, −1, −1, −1, 4, 5 ), 4 , 5 = ±1.
In addition to the signs 4 and 5 , two deformation parameters, R and

,
with dimensions of length, characterize this stabilizing deformation. R, asso-
ciated to the non-commutativity of the generators of translations, must be
related to the local curvature. Therefore, because the curvature is not a
constant, R cannot have the status of a fundamental constant. However,
the other constant
might be a fundamental length, a new fundamental
physical constant.
As in the case of the Poincaré algebra discussed above, if one is mostly
concerned with the algebra of observables in the tangent space, one may
take the limit R → ∞ and finally obtain
[Mµν , Mρσ ] = i(Mµσ ηνρ + Mνρ ηµσ − Mνσ ηµρ − Mµρ ηνσ ), (33)
[Mµν , Pλ ] = i(Pµ ηνλ − Pν ηµλ ), (34)
[Mµν , xλ ] = i(xµ ηνλ − xν ηµλ ), (35)
[Pµ , Pν ] = 0, (36)
2
[xµ , xν ] = −i5
Mµν , (37)
[Pµ , xν ] = iηµν , (38)
[Pµ , ] = 0, (39)
2
[xµ , ] = i5
Pµ , (40)
[Mµν , ] = 0, (41)
as the stable algebra of relativistic quantum mechanics. The main features

are the non-commutativity of the xµ coordinates and the fact that , previ-
ously a trivial center of the Heisenberg algebra, becomes now a non-trivial
www.ebook3000.com
The SPTP 167
operator. These are however, the minimal changes that seem to be required
if stability of the algebra of observables (in the tangent space) is a good
guiding principle. Two constants define this deformation. One is
, the
fundamental length, the other the sign of 5 . The “tangent space” algebra
(33–41) is the kinematical algebra appropriate for microphysics. However,
for physics in the large, it should be the full stable algebra (16–18, 28–32)
to play a role. In the last part of this section, I will discuss two important
1
roles that the non-vanishing of R may play for the physical construction.
However, for the most part, the emphasis here will be in the tangent space
limit R → ∞.
The stabilization of the Heisenberg–Poincaré algebra has been further
studied and extended in Refs. [30–32]. The idea of modifying the algebra
of the space–time components xµ in such a way that they become non-
commuting operators had already appeared several times in the physical
literature. Rather than being motivated (or forced) by stability consider-
ations, the aim of those proposals was to endow space–time with a dis-
crete structure, to be able, for example, to construct quantum fields free
of ultraviolet divergences. Sometimes, they simply postulated a non-zero
commutator, others they were guided by the formulation of field theory in
curved spaces. Although the algebra arrived at in Ref. [28], Eqs. (33)–(41),
is so simple and appears in such a natural way in the context of defor-
mation theory, it seems that, strangely, it differed in some way or another
from the past proposals. In one scheme, for example, the coordinates were
assumed to be the generators of rotations in a five-dimensional space with
constant negative curvature. This possibility was proposed long ago by Sny-
der [33,34] and the consequences of formulating field theories in such spaces
have been extensively studied by Kadishevsky and collaborators [35, 36].
The coordinate commutation relations [xµ , xν ] are identical to (37), how-
ever, because of the representation chosen for the momentum operators, the
Heisenberg algebra is different and, in particular, [Pµ , xν ] has non-diagonal
terms. Banai [37] also proposed a specific non-zero commutator which only
operates between time and space coordinates, breaking Lorentz invariance.
Many other discussions exist concerning the emergence and the role of dis-
crete or quantum space–time, which however, in general, do not specify a
complete operator algebra [38–51].
Note that there other ways to deform the algebra 0 to the simple
algebra of the pseudo-orthogonal group in six dimensions. They correspond
to different physical identifications of the generators Mµ4 , Mµ5 , and M45 .
For example, putting

1
Pµ = (Mµ4 + Mµ5 ), (42)
R

xµ = (Mµ4 − Mµ5 ), (43)

2

= M45 (44)
R
and 4 = −5 = 1, the coordinates and momenta are now commuting
variables and the changes occur only in the Heisenberg algebra and the
nature of , namely

[Pµ , xν ] = i Mµν + ηµν , (45)

R

[Pµ , ] = −i Pµ , (46)
R

[xµ , ] = i xµ . (47)
R
However this identification of the physical observables in the deformed alge-
bra does not seem so natural as the previous one. In particular Eq. (45)
implies a radical departure from the Heisenberg algebra and the fundamen-
tal length scale is tied up to the large scale of the manifold curvature radius,

in the sense that, if we take R → ∞ , the whole deformation vanishes.
The ,∞ algebra (33)–(41) has a simple representation by differential
operators in a five-dimensional space with coordinates (ξ0 , ξ1 , ξ2 , ξ3 , ξ4 )
∂
Pµ = i µ + iDPµ , (48)
∂ξ

∂ ∂
Mµν = i ξµ ν − ξν µ + Σµν , (49)
∂ξ ∂ξ

∂ 4 ∂
xµ = ξµ + i
ξµ 4 − 5 ξ +
Σµ4 , (50)
∂ξ ∂ξ µ
∂
= 1 + i
4 + i
Dξ4 . (51)
∂ξ
The set (Σµν , Σµ4 ) is an internal spin operator for the groups O(4, 1) (if
5 = −1) or O(3, 2) (if 5 = +1) and DPµ and Dξ4 are derivations operating
in the space where (Σµν , Σµ4 ) acts. In this representation, the deformation
has a simple interpretation. The space–time coordinates, xµ , in addition to
www.ebook3000.com
The SPTP 169
the usual (continuous spectrum) component, have a small angular momen-

tum component corresponding to a rotation (or hyperbolic rotation) in the
extra dimension. And the center of the Heisenberg algebra picks up a small
momentum in the extra dimension.
The algebra (33)–(41) is seen to be the algebra of the pseudo-Euclidean
groups E(1, 4) or E(2, 3), depending on whether 5 is −1 or +1. For the
construction of quantum fields it might be convenient to use this repre-
sentation. Note however, that only the Poincaré part of E(1, 4) or E(2, 3)
corresponds to symmetry operations and only this part has to be imple-
mented by unitary operators.
Physical consequences of the non-commutative space–time structure
implied by the ,∞ algebra have been explored in a series of publica-
tions [52–57]. Depending on the sign of 5 , the time (5 = +1) or one space
variable (5 = −1) will have discrete spectrum. In any case
, a new funda-
mental constant, sets a natural scale for time and length. If
is of the order
of Planck’s length, observation of most of the effects worked out in the cited
references will be beyond present experimental capabilities. However, if
is much larger than Planck´s length (for example, of order 10−27 −10−26
seconds) the effects might already be observable in the laboratory or in
astrophysical observations. I refer the reader to the references above, for a
detailed analysis of the experimental predictions and just add here a few
remarks. Some of the most noteworthy effects arise from the modification of
the phase space volume and from interference effects. In addition, the sim-
ple fact that the space–time coordinates do not commute already implies
that many notions currently used in the analysis of laboratory experiments
become ill-defined. For example, because the space and the time coordi-
nates cannot be simultaneously diagonalized, speed can only be defined in
terms of expectation values,
1 d
vψi = ψt , xi ψt , (52)
ψt , ψt
dt
ψ being a state with a small dispersion of momentum around a central

value p. This would imply a deviation from c (= 1) of the “effective speed”
of massless particles of order [55]
2
∆vψ = −35
2 p0 . (53)
The deviation would be negative for 5 = +1 (

a fundamental time) or
positive for 5 = −1 (
a fundamental length). In any case, such deviation
should not be confused with a modification of the value of the fundamental

constant c.
Most of the consequences worked out in Refs. [52–56] are rather conser-
vative in the sense that they simply explore the non-vanishing of the right-
hand side of the commutators of previously commuting variables. Deeper
consequences are to be expected from the radical change from a commu-
tative to a non-commutative space–time geometry. The new geometry was
studied in Ref. [58].
For this non-commutative geometry, the differential algebra may be
defined either by duality from the derivations of the algebra or from the
triple (H, π(U ), D), where U is the enveloping algebra of ,∞ , to which
a unit and, for later convenience, the inverse of , are added.
U = {xµ , Mµν , pµ , , −1 , 1}, (54)
π(U ) is a representation of the U algebra in the Hilbert space H and D is

the Dirac operator, the commutator with the Dirac operator being used to
generate the one-forms. In a general non-commutative framework [59, 60],
it is not always possible to use the derivations of the algebra to construct
by duality the differential forms. Many algebras have no derivations at all.
However, when the algebra has enough derivations it is useful to consider
them [61,62] because the correspondence of the non-commutative geometry
notions to the classical ones becomes very clear. One considers the set V of
derivations with basis {∂µ , ∂4 } defined as followsb
∂µ (xν ) = ηµν ,
∂4 (xµ ) = −5
pµ ,
∂σ (Mµν ) = ησµ pν − ησν pµ , (55)
∂µ (pν ) = ∂µ () = ∂µ (1) = 0,
∂4 (Mµν ) = ∂4 (pµ ) = ∂4 () = ∂4 (1) = 0.
In the commutative (
= 0) case, a basis for one-forms is obtained, by
duality, from the set {∂µ }. In the
= 0 case, the set of derivations {∂µ , ∂4 }
is the minimal set that contains the usual ∂µ ’s, is maximal abelian and is
action closed on the coordinate operators in the sense that the action of ∂µ
on xν leads to the operator associated to ∂4 and conversely.
The operators that are associated to the physical coordinates are just the
four xµ , µ ∈ (0, 1, 2, 3). However, an additional degree of freedom appears
b Note that the definition of ∂4 here, is slightly different from the one in [58].
www.ebook3000.com
The SPTP 171
in the set of derivations. This is not a conjectured extra dimension but sim-
ply a mathematical consequence of the algebraic structure of ,∞ which,
in turn, was a consequence of the stabilizing deformation of relativistic
quantum mechanics. No extra dimension appears in the set of physical
coordinates because it does not correspond to any operator in ,∞ . How-
ever, the derivations in V introduce, by duality, an additional degree of
freedom in the exterior algebra. Therefore, all quantum fields that are Lie
algebra-valued connections will pick up some additional components. These
additional components, in quantum fields that are connections, are a conse-
quence of the length parameter
which does not depend on its magnitude,
but only on
being = 0.
The Dirac operator [58] is
D = iγ a ∂a (56)
with ∂a = (∂µ , ∂4 ) and the γ’s being a basis for the Clifford algebras C(3, 2)
or C(4, 1)
0 1 2 3 4 5

γ0 , γ1 , γ2 , γ3 , γ4 = γ 5 5 = +1 , (57)
γ , γ , γ , γ , γ = iγ 5 = −1
How to construct quantum, scalar, spinor and gauge fields, as operators in

U , has been described in Ref. [58]. In particular, the role of the additional
dimension in the exterior algebra on gauge interactions has been emphasized
(see also [56]). Here, another potential interesting consequence for spinor
fields will be described. Because

pµ , e 2 ν {
k xν ,−1 }
= −kµ e 2 ν {
i i
k xν ,−1 }
+ + (58)
a spinor field is written

i ν −1 i ν −1
ψ = d4 kδ(k 2 − m2 ){bk uk e− 2 kν {x , }+ + d∗k vk e 2 kν {x , }+ },
(59)
ψ ∈ U : Dψ − mψ = 0. (60)
For a massless field, the (extended) Dirac equation is

Dψ = iγ a ∂a ψ = iγ µ ∂µ + iγ 4 ∂4 ψ = 0. (61)
Write
ψ = e 2 ν{
i
k xν ,−1 }
+ u(k).
From
∂µ e 2 ν {
i
k xν ,−1 } i
kν {xν ,−1 }
+ = ik e 2 +
µ
, (62)
∂4 e 2 ν {
k xν ,−1 }
e 2 ν{
i
µ 1 2 i
k xν ,−1 }
+ = −i
k p + k +
5 µ
2
one obtains, using (62) and (58)

µ 5 1 2
−γ kµ − γ
k u (k) = 0 5 = +1,
2
(63)
1
−γ µ kµ + iγ 5
k 2 u (k) = 0 5 = −1.
2
Let 5 = −1. Iterating (63)

2
2 2 2
k − k u (k) = 0. (64)
4
This equation has two solutions, the massless solution (k 2 = 0) and another
one, of large mass (
being small)
4
k2 = . (65)

2
For 5 = +1, the large |k 2 | solution is tachyonic. The solutions of the
extended Dirac equation for k 2 = 0 are the usual ones and for k 2 = 42 , in
the rest frame and the Weyl (chiral) basis

a 2
Positive energy m0 = ,
−ia
(66)
a 2
Negative energy m0 = −
ia
the solutions of non-zero momentum being obtained by the application of

a proper Lorentz transformation. a is an arbitrary two-vector.
So far and in Refs. [52–56], consequences were explored of the (
=
0, 1/R → 0) case. However, as pointed out by several authors [63–66], even
a very small non-vanishing of the right-hand side of the commutator [Pµ , Pν ]
may have striking consequences on the nature of the representations of the
algebra, which instead of Poincaré, becomes de Sitter or anti-de Sitter.
Another interesting possibility, still unexplored, would be to promote
the right-hand side of the commutator [Pµ , Pν ], which in (28) is written
−i R
42 Mµν , to a space–time dependent field Cµν (x), from which a theory of
gravity as a deformation might be constructed.
www.ebook3000.com
The SPTP 173
Finally, note that when using algebraic stability to study the kinemat-
ical algebras, the primary results so far have concerned the nature of one-
particle states. If, instead, one is concerned with two-particle effects (or
aggregates) it is probably the deformation theory of bialgebras that comes
into play. The suggestion is that the stability theory of bialgebras might
provide useful information on the nature of the stable interactions.
4. Stability, Universality and Critical Phenomena

4.1. Bifurcations and universality
Many families of differential equations and discrete-time mappings depend-
ing on one parameter µ exhibit, when µ varies, a cascade of successive
period-doubling bifurcations of stable periodic orbits [67, 68]. A typical
example is the quadratic map x → 1 − µx2 . As µ approaches the value
µ∞ = 1.40155 from below, the ratio
µn − µn−1
,
µn+1 − µn
tends to δ = 4.669, . . . , µn being the value at which the 2n -cycle is born.
Similarly, the size of the domains in phase space associated to the successive
cycles (for example, the separation of two points in the orbits that contain
the critical point at x = 0) also scales to a constant λ = 0.399 . . . . [68].
The universality of these constants is associated to the existence of a
fixed point for the Feigenbaum functional equation [69–72].
1
− ψ ◦ ψ(−λx) = ψ(x).
λ
The values δ = 4.669 . . . and λ = 0.399 . . . depend on the quadratic nature
of the critical point. Other critical points also lead to scaling behavior but
with different constants [73]. However, the fact that the above constants
are the ones that are actually found in so many one-parameter systems and
also on experimental results [74] clearly seems to be a manifestation of the
fact that, as discussed in Appendix A, the quadratic map is the only stable
one-dimensional map.
For higher dimensions, however, we might have stable sequences of
higher order bifurcations corresponding to fixed point solutions of the
functional equation
1
− ψ ◦ · · · ◦ ψ (−λx) = ψ(x).
λ
n
Bifurcation sequences of period-tripling, period-quadrupling, etc. have

been studied for complex mappings [75–77]. Consider a family f (z, µ) of
quadratic mappings of C 1 into C 1 depending on a complex parameter µ.
In the complex µ plan there is a domain U0 of parameter values for which
there is a stable fixed point. The boundary of the U0 domain consists of
the parameter values for which the map derivative at the fixed point lies on
(1) (2)
the unit circle. Touching U0 there are two smaller domains U3 and U3
corresponding to the values of µ for which there is a stable period-3 orbit.
(1) (2)
The contact points√of the domain U0 with U3 and U3 are the cubic roots
3 (1) (2)
of the unit − 12 ± i 2 . Then, adjoining each of the domains U3 and U3 ,
there are two domains corresponding to stable period-9 orbits and so on.
Choosing parameter values µ to follow the successive contact points of all
these domains one obtains a period-tripling bifurcation sequence. The cor-
responding (complex) universal constant is δ (1,2) (3) = 4.600 · · · ± i8.981 . . ..
A similar scheme operates for other n-tuplings for which the complex uni-
versal constants have also been computed [76].
A complex C 1 → C 1 mapping may be regarded as a real R2 → R2 map-
ping and sequences of n-tuplings might therefore also be expected in real
mappings as a two-parameter effect. Structural stability imposes however
some restrictions on the observability of this phenomenon. Let us write the

quadratic z → z = 1 − µz 2 complex mapping as a real C 2 → C 2 mapping.
With µ = α + iβ and z = x + iy, one obtains

x = 1 − α(x2 − y 2 ) + 2βxy,

y = −β(x2 − y 2 ) − 2αxy.
This map however has at x = y = 0 a singularity of the Σ2 -type which

is stable only for real maps of dimension four and above. Therefore, on
the basis of the stability principle, for physical systems described by real
maps, one should expect the n-tupling effect (with n > 2) to be generic only
for phenomena which are not reducible to an effective dynamics below four
dimensions. Conversely, the observation of such higher n-tuplings in actual
complex physical systems may be a guide for the dimensional requirements
of their mathematical models.
4.2. Universality in phase transitions

The renormalization group analysis [78, 79] provides a great deal of infor-
mation on continuous phase transitions. At a continuous phase transi-
tion point, the correlation length diverges, the dynamics is dominated by
www.ebook3000.com
The SPTP 175
long-range collective effects and one expects the physics of the problem to
be insensitive to scale transformations.
In configuration (or real)-space renormalization, for a system defined on
a lattice, one replaces, at each step, all the degrees of freedom contained in
(n+1)
a block by a single block variable. Therefore, the block variable (σi ) at
(n)
step n + 1 is a function of the block variables of the preceding step (σi ).
(n+1) (n)
σi = f (σk ). (67)
The function f may be a smooth function and is normalized in such a way

that the mean-square value of the block variables is preserved at all renor-
malization steps. Each time a blocking is performed, the lattice parameter
changes from a to ba. Therefore to keep the same nominal lattice spacing,
lengths are at each step scaled down by a factor b−1 . The effective Hamil-
tonian H(σ (n+1) ) of the renormalized system is obtained by summing over
the variables of the preceding step, namely
1 (n+1) (n+1) 1 (n) (n)
e−H (σi )
= e−H (σk )
, (68)
Z (n+1) (n) (n+1)
Z (n)
f (σk )=σi
(n)
where the sum in the right-hand side is over all the configurations of the σk
(n+1)
variables that lead to the specified σi . The temperature dependence is
included in the effective Hamiltonian. In the first step, we have
(0) 1 (0)
H(0) (σi ) = H(σi ), (69)
kT
H being the temperature-independent usual Hamiltonian. However, after
the renormalization, the effective Hamiltonians obtained from Eq. (68) will
in general have a much more complicated dependence on the temperature
and on the other variables. However, they will be functions of the same
(0)
variables as H(0) (σi ) and furthermore assumed to be smooth functions.
Here, I will be mostly concerned with the dependence on temperature
and on a parameter which, for definiteness, is assumed to play the same
B (0)
role as an external magnetic field coupled by a term kT i σi in H(0) .
Hence

(0) (0) (n) B
H (σi ) = H T, , . . . . (70)
T
The dots stand for other variables like the spin–spin coupling strengths, etc.
At high temperatures, variables become independent and the correlation

length vanishes. On the other hand, using a sufficiently high-temperature
for the starting point of the renormalization, the correlation length at step
n is ξ (n) = bξn , it tends to zero as n → ∞ and one expects it to be driven
to the high-temperature fixed point.
On the other hand, close enough to T = 0 all variables are near their
ground-state values and the block averaging, resulting from the renormal-
ization, will make the block variables increasingly more uniform. Therefore,
starting from a sufficiently small temperature the system is driven by renor-
malization towards the low-temperature fixed point.
Consider now a system that has only one phase transition. Then,
between the functions that are attracted to the high-temperature fixed point
and those that are attracted to the low-temperature fixed point, there is,
in the space of smooth functions, those that are attracted to neither one.
These functions are said to lie in the critical surface and, at least some of
them, correspond to effective Hamiltonians for phase transition points at
distinct values of the physical parameters.
To make the connection with the structural stability scenario, note that
this is typically a codimension-one framework (Appendix A). Therefore,
the critical surface may be taken to be a codimension-one subset Sc in
the space of all smooth functions. The missing dimension is precisely the
direction taken by the renormalization transformation when it drives nearby
functions either to the low or the high-temperature limits. This is the precise
physical meaning of the codimension of the critical surface, as defined here.
It should not be confused with the number of relevant directions, because if
there is a renormalization group fixed point in the critical surface, some of
the directions associated to eigenvalues greater than one may point along
the critical surface.
Of course, not every function in Sc may be reached from any other by a
renormalization transformation. This is understandable because the finite
codimension subsets in the space of all smooth functions are defined by
R-equivalence, that is, by arbitrary diffeomorphisms and the renormaliza-
tion transformation is just a particular type of change of variables. Also, as
defined, the critical surface may contain the effective Hamiltonians of many
different physical systems. For each particular system, the renormalization
group generates an orbit (not necessarily dense) in the critical surface. Note
also that, instead of the critical surface containing the effective Hamiltoni-
ans, we may consider a space of (Helmholtz) free energy functions.
www.ebook3000.com
The SPTP 177
So far, this a very general framework which depends only on the exis-
tence of the low and high-temperature limits and one phase transition.
A further assumption of the renormalization group analysis is the exis-
tence, in the critical surface, of quasi-homogeneous functions. A function is
quasi-homogeneous [80] of degree d with indices y1 , . . . , yn if for any b > 0
we have
f (by1 x1 , . . . , byn xn ) = bd f (x1 , . . . , xn ). (71)
For the effective Hamiltonians the assumption is that there is a fixed point
in the critical surface and the corresponding result for the free energy per
unit mass is a relation of the type of Eq. (71). Actually, even at the fixed
point, the transformation of the free energy is slightly more complicated,
namely,
f (by1 x1 , . . . , byn xn ) = bd {f (x1 , . . . , xn ) − g(x1 , . . . , xn )},
where the function g is needed to satisfy the normalization conditions. How-

ever, for the purpose of computation of the critical exponents the first term
is considered to be sufficient (for a discussion see Ref. [79]).
For a continuous phase transition, physical intuition, derived from the
divergence of the correlation length, indeed suggests the existence of a scale-
independent point. Nevertheless, the actual existence of a renormalization
fixed point in the critical surface is an assumption and more complex (peri-
odic or chaotic) behaviors are possible. Note also that, at the fixed point, the
effective Hamiltonian that is obtained by the action of the renormalization
group may not correspond to any particularly interesting set of parameters.
The physical phase transition points are all over the critical surface. How-
ever, because the critical exponents are preserved along renormalization
group orbits, they may be computed at the fixed point.
Let, in Eq. (71), x1 be the reduced temperature
|T − Tc |
x1 = t =
Tc
and x2 the magnetic field
x2 = B.
Then y1 and y2 are the temperature and magnetic indices (or eigenvalues)
and Eq. (71) becomes Widom’s [81, 82] scaling hypothesis
f (byt t, byB B) = bd f (t, B). (72)

All critical exponents may be computed from the two numbers yt and yB
−ν
[79]. ξ ∼ |T − Tc |
−α
−1 |T − Tc | d
cB ∼ α −1 ; α=2− ,
Tc yt
d − yB
m ∼ (Tc − T )β ; B = 0; β = ,
yt
−α
−1 |T − Tc | d
cB ∼ α −1 ; α=2− ,
Tc yt
1 yB
m ∼ B δ ; T = Tc ; δ = ,
d − yB
1
G(2) (r) ∼ ; T = Tc ; B = 0; η = d + 2 − 2yB ,
γ d−2+η
−ν 1
ξ ∼ |T − Tc | ; B = 0; ν = .
yt
For each pair (yt , yB ) of renormalization group eigenvalues, one has a set of
critical exponents, which apply to a class of different physical systems. Each
set of values (yt , yB ) defines a universality class. This provides an apprecia-
ble unification in our knowledge of critical phenomena and understanding
the mechanism, through which very different physical systems may have the
same critical exponents, was the great achievement of the renormalization
group analysis. However, the renormalization group is powerless in deter-
mining the pair (yt , yB ) or in finding out how many universality classes
there is.
We now turn to structural stability considerations. One imposes, as a
hypothesis, that the critical surface is a structurally stable codimension-one
family of functions. From the table in Appendix A, one knows that there
is only one stable family of codimension-one. This family contains all the
functions that are R-equivalent to the canonical form A2 . The canonical
forms listed in the table of Appendix A are defined up to a Morse function
in the other variables. Hence, for two variables, one has
fα (x1 , x2 ) = x21 + x32 + αx2 . (73)
The last term is the unfolding that vanishes (α = 0) on the critical surface.
By R-equivalence, one generates all kinds of complex functions in the criti-
cal surface. However, the canonical form is already all one needs because it is
www.ebook3000.com
The SPTP 179
a quasi-homogeneous function. Hence, for the stable codimension-one fam-

ily, the existence of a quasi-homogeneous point is not a separate assumption.
Note that the canonical form in Eq. (73) is only appropriate at the fixed
point. Because the other functions in the critical surface are obtained from
this one by arbitrary diffeomorphisms, there is no simple relation between
the canonical form at the quasi-homogeneous point and the free energy
at other physical phase transition points. Therefore, the canonical form is
only appropriate to derive renormalization group invariants like the critical
exponents and nothing else. Note also that it is only at the fixed point
that the extra functional dimension, pointing towards the high and low
temperature limits, is generated by αx2 .
To apply the canonical form to derive the critical exponents we still
have to identify the variables x1 and x2 . Referring back to Eq. (70), we
conclude that the natural identification of even and odd variables is not t
and B, but t and Bt . Then
B
x1 = t and x2 = .
t
Therefore, from

B B
f0 t, = b−d f0 byt t, byB −yt
t t
and Eq. (73), one obtains
2yt = d,
3(yB − yt ) = d,
d
that is, yt = 2 and yB = 56 d. Then,
1 4 2 2
α = 0; β = ; γ = ; δ = 5; η = 2 − d; ν = . (74)
3 3 3 d
These values, obtained from the structural stability of the critical surface,
are indeed close to the experimental values for three-dimensional (3D) phys-
ical systems undergoing continuous phase transitions.
The similarity of the measured critical exponents for many different
experimental systems and in particular the proximity of their values to sim-
ple rational numbers has intrigued many authors. Cardy [83], for example,
uses the fact that, by letting the length rescaling factor depend continuously
on position, scale invariance is generalized to conformal invariance. Then
the critical exponents are restricted to rational numbers which, by trial,
may be identified with particular models. However no unique or strongly

preferred result is obtained.
Here, by using a structural stability hypothesis, in the codimension-one
setting, a unique result is obtained. To require structural stability, that is,
to require that the physical laws are not too sensitive to the precise values
of the couplings, is perhaps a natural requirement, at least for phenomena
that do not seem to depend on the detailed properties of the system but only
on a general scaling behavior. In obtaining the result (74), an important
role is also played by the identification of t and Bt as the variables in the
quasi-homogeneous free energy at the fixed point. This however seems a
natural choice in view of Eq. (70). Is the result (74) an accident, or is it
appropriate to use structural stability in this context? I leave to the reader
to decide.
Appendix A: Structural Stability in Dynamical

Systems Theory
A.1. Structural stability of phase portraits
Let (M, Ut ) be a dynamical system. M is the state space and Ut (with t ∈
K=R, Z, R+ or Z + ) the time evolution operator. For each initial condition
x0 ∈ M , the set {Ut x0 : t ∈ K} is an orbit of the dynamical system. The
set of all orbits is called the phase portrait P of the system.
The problem of structural stability in the theory of dynamical systems
is, in qualitative terms, the following: “If the dynamical system (M, Ut ) with
phase portrait P is perturbed to a slightly different system (M, Ut ) , is the
new phase portrait P also a small perturbation of P ? That is, is the new
system equivalent to the first? (equivalent in a sense to be specified later)”.
The perturbation of the dynamical system may be, for example, a small
change in the numerical parameters of the evolution operator.
Structural stability is a question of great physical importance because,
even if (M, Ut ) is an accurate model for a physical system, the results
obtained by the study of this model are, in practice, never applied to the
actual (M, Ut ) model of the real world but to a nearby system because the
parameters of the system, being obtained experimentally, are only known
approximately. Therefore, underlying all attempts to describe natural phe-
nomena is the assumption that the structures in Nature enjoy some stability,
otherwise we could hardly think of the possibility to describe them in an
experimentally reproducible way. Hence, the only qualitative properties of
a family of dynamical systems which are physically relevant are those that
are preserved under perturbations.
www.ebook3000.com
The SPTP 181
To discuss structural stability we have to specify what are the allowed

deformations of the systems we are concerned with (that is, what is the
topology given to the set of dynamical systems) and what is the equiva-
lence relation that decides when the perturbed system is equivalent to the
unperturbed one. In the classical theory of dynamical systems, the evolution
operator Ut is either a discrete power of a mapping T : M → M (discrete
time) or the flow induced by a vector field X in M (continuous time). The
topology is in both cases the C r -topology. Two maps are C r -close when
their values and the values of their derivatives up to order r are close at
every point. An -neighborhood of a map f in the C r -topology is the set of
all C r -maps which together with their derivatives up to order r differ from
f less than .
The equivalence relation is topological conjugacy for maps and topologi-
cal equivalence for flows. Two maps T1 and T2 are C 0 -conjugate if there is a
homeomorphism h such that h◦T1 = T2 ◦h. Two vector fields X1 and X2 are
C 0 -equivalent if there is a homeomorphism h which takes the orbits of X1
to orbits of X2 , preserving senses but not necessarily the time parametriza-
tion. This is because, for example, we allow the periods of closed orbits to
be different. Note also that the most relevant notion of equivalence is topo-
logical (C 0 ) equivalence, not for example C 1 -equivalence or C 1 -conjugacy.
This latter equivalence would be too restrictive because it would impose
invariance of the eigenvalues of the linear part of the dynamics at periodic
points.
A map f (or vector field X) is structurally stable if it has an
-neighborhood topologically conjugate (or topologically equivalent) to f
(to X). We may however not be concerned with the transients of the dynam-
ics. Therefore, we may consider stability restricted to the main part of
the orbit structure, that is to the non-wandering set Ω. A point is non-
wandering if, for any neighborhood U of x, there is an integer n such that
f n U ∩ U = ∅. Then, Ω-structural stability is structural stability restricted
to the non-wandering set. That is, given a small perturbation the perturbed
system has a non-wandering set Ω and there is a surjective map Ω → Ω
sending orbits to orbits.
For general dynamical systems, the notions of structural stability, hyper-
bolicity and transversality are closely related. Some of the strongest results
proved so far are:
Theorem ([84]). A C 1 -diffeomorphism is C 1 -structurally stable if and only

if it satisfies Axiom A and the strong transversality condition.
Theorem ([85]). A C 1 -diffeomorphism is C 1 -Ω-structurally stable if and

only if it satisfies Axiom A and the no-cycle condition.
The meaning of the terms used in these theorems is as follows:
An Axiom A dynamical system is a map (or flow) such that
(1) The non-wandering set Ω is compact and hyperbolic.
(2) The fixed points and periodic orbits are dense in Ω.
A set is hyperbolic when there is a continuous splitting T M |Ω = V + +
V of the tangent bundle restricted to Ω such that (Df )V ± ⊂ V ± and
−
Df ±n |V ∓ ≤ cθ−n , n ≥ 0 for some c > 0 , θ > 1.

The stable and unstable manifolds of a point are

Wxs = y ∈ M : lim d(f n x, f n y) = 0 ,
n→∞

Wx = y ∈ M : lim d(f n x, f n y) = 0 .
u
n→−∞
A dynamical system f satisfies the strong transversality condition if, for

each y ∈ M there are stable and unstable manifolds through y such that
Ty M = Ty Wxu + Ty Wxs (+X),
where X is added in the flow case if y is not a fixed point.
For Ω-stability, strong transversality is replaced by the no-cycle condi-
tion. The non-wandering set Ω of an Axiom A system f is a finite union
Ω = Ω1 ∪ · · · ∪ Ωn of disjoint closed invariant sets called basic sets, such
that f is topologically transitive on each Ωi . Topological transitivity means
that there is an x with dense orbit in Ωi . One writes
Ωi Ωj if WΩs i ∩ WΩuj = ∅.
The no-cycle condition means that one cannot find distinct Ωi1 , . . . , Ωip
(p > 1) such that
Ωi1 Ωi2 · · · Ωip Ωi1 .
The stability results quoted above are difficult mathematical theo-
rems. However, the relation between structural stability, hyperbolicity and
transversality is fairly intuitive and was the object of an old conjecture
([86]). Structural stability means that the nature of the system does not
change for small perturbations and, for example, a periodic point must be
hyperbolic if it remains of the same nature for small perturbations. On the
other hand, transversality means that the stable and unstable manifolds
that are the organizers of the dynamics must be in general position.
www.ebook3000.com
The SPTP 183
The generic nature of hyperbolicity and transversality might suggest

that almost all systems are structurally stable in the sense that structurally
stable systems are dense in the set of all smooth systems. Actually, this has
been proven not to be true [87, 88] except for low-dimensional cases [89].
Structural stability as it has been defined is a property that concerns the
topological properties of the dynamical system. Another notion of stability
was proposed ([90]) that deals with the invariant measure ρ of the system
under a small random perturbation. A small random perturbation is added
to the system because for a large class of noisy systems the invariant mea-
sure is unique whereas in general a deterministic system has many invariant
measures. The measure is the solution of the Fokker–Planck equation
∂t ρ = −∇(ρX) + ρ,
where X is the deterministic vector field and the diffusion coefficient is

a small quantity. Two functions ρ and ρ are equivalent if there are diffeo-

morphisms α and β of M and R such that ρ ◦ α = β ◦ ρ. Then two vector

fields X and X are −equivalent in this sense when the corresponding solu-

tions ρX,
and ρX ,
are equivalent. A vector field X is -stable if it has an
-equivalent neighborhood. It is called stable if it is -stable for arbitrarily
small > 0.
Both structural stability and (measure) stability in Zeeman’s sense are
designed for general dynamical systems and leave out whole classes of
physical interest. For example, they are not suitable for Hamiltonian sys-
tems which are all structurally unstable. This is because the perturbations
allowed in the C r -topology do not preserve any constants of motion or
symmetries that the dynamical system may have.
No system with regular first integrals may be structurally stable, in the
general sense, because the property of having no regular first integrals is
C 1 generic [91,92]. To define a structural stability concept for Hamiltonian
systems, we must exclude non-Hamiltonian perturbations. Restricting the
perturbations to the space χH of Hamiltonian vector fields, and using the
C r -topology in this space, we may define stability of the phase portrait in
the same way as before. This notion of stability is very strong and it seems
more appropriate to require only stability of the phase portrait on a single
energy surface under small perturbations of the Hamiltonian and the energy.
Because generically an Hamiltonian system restricted to an energy surface
has no other first integrals, the conflict with general structural stability
would seem to be avoided. However, the problem with the several definitions
that have been proposed so far [93] is that they do not apply to generic
Hamiltonian systems [94].
In addition to stability of the phase portraits, there are two other notions
of dynamical stability which are reviewed in next two subsections. They are
of importance for the applications described in Sec. 4.
A.2. Stability of smooth mappings and stable

dynamical families
The preceding subsection was concerned with the stability of the phase
portrait of a dynamical system, that is, the stability of the realization in
phase space of a dynamical law. Given two equivalent phase portraits, one
may in fact say that one is dealing with the same dynamics as seen in
two reference frames, related by a continuous change of coordinates. This
subsection deals not with stability of the phase portrait but with stability
of the type of dynamical law. This will be clear after the definition of
equivalence and stability of smooth mappings.
Being mostly concerned with local properties of maps between smooth
manifolds M and N one may, by a choice of local charts, reduce the problem
to Rn → Rp maps. Two smooth maps f1 , f2 : Rn → Rp are equivalent
when there are diffeomorphisms g : Rn → Rn and h : Rp → Rp such that
f1 = h−1 ◦ f2 ◦ g. A mapping f is stable when there is a neighborhood where
all mappings are equivalent to f . Neighborhoods in the space of mappings
are defined by

α α1 αn
Uf (k, ) = g : max ∂ (f − g)∂x · · · ∂x < , α = α1 + · · · + αn
α≤k
the derivatives being taken up to order r for the C r -topology.

When dealing with maps between different spaces, Rn and Rp , the equiv-
alence relation means that different choices of coordinate systems in the
source and the target spaces are allowed. If however one identifies the source
and the target space, as in a map f : Rn → Rn defining a discrete time
dynamical system and the diffeomorphisms h and g are distinct, different
dynamics are in fact obtained. Two equivalent maps in the above sense may
generate very different phase portraits. The set

{f : f = h−1 ◦ f ◦ g} (A.1)
for all possible difeomorphisms h and g represents not a single dynami-

cal system but a family of related systems. We know that in Nature we
www.ebook3000.com
The SPTP 185
sometimes have to deal with phenomena that depend on a certain number

of control parameters, which may indeed induce very different dynamical
behavior, phase transitions, etc., but which nevertheless we want to iden-
tify with different conditions of the same physical system. The action of the
diffeomorphisms in (A.1) gives then a precise and general meaning to the
notion of change of parameters in a stable family of dynamical systems. This
contains the usual notion of change of parameters in many classical exam-
ples. For example for mappings of the unit interval x → fµ (x) = 1 − µx2 ,
1
fµ and fµ are related by h(x) = x and g(x) = (µ /µ) 2 x.
If the phase portrait is not preserved, what are the features of the
dynamics that are preserved under this equivalence relation? That is, what
are the invariant properties that characterize the dynamical systems fam-
ily defined by (A.1). The most significant ones are the singularities of the
mappings. f is said to have a singularity or critical point at x if the rank
of the derivative map Df at x is less than the maximum possible value (n
for Rn → Rn mappings). The kind of dynamical properties that are con-
trolled by the critical points are universality in the approach to bifurcation
accumulation points [73] and bifurcation patterns.
For the singular points of smooth mappings, one uses Boardman’s nota-
tion Σi1 ,···,ik . A point is said to belong to Σi1 if the dimension of the kernel of
Df is i1 . The full notation is defined recursively by considering the kernels
of the restriction of Df to Σi1 , etc. That is, Σi1 ,...,ik = Σik (Df | Σi1 ,...,ik−1 ).
Actually, this characterization of the Boardman symbol Σi1 ,...,ik is correct
only if these sets are submanifolds, which is the case for the stable maps
that concern us here. That is, for stable maps the Boardman sets coincide
with Thom’s singularity sets.
The stable maps for low dimensions have been fully classified [80,95,96].
They are characterized in terms of germs and unfoldings. A smooth germ at
the point x is an equivalence class of maps which coincide when restricted
to some neighborhood of x. Given a germ f0 : (Rn , 0) → (Rn , 0) in the
neighborhood of zero, an r-parameter unfolding of f0 is the germ F : (Rr ×
Rn , 0) → (Rr × Rn , 0) given by F (u, x) = (u, f (u, x)) with f (0, x) = f0 (x).
Therefore, an unfolding is a (r + n)-dimensional map, the first r compo-
nents being the identity map and the other n a deformation of the original
f0 map.
A classification of stable Rn → Rn maps for n ≤ 4 is listed below in
terms of equivalence of its germ at any point to a standard form. Let f
be a stable map; then its germ at any point is equivalent to one of the
following:
n=1
0

(x1 = x1 ),
1,0

(x1 = x21 ).
n=2
0

(x1 = x1 ; x2 = x2 ),
1,0

(x1 = x1 ; x2 = x22 ),
1,1,0

(x1 = x1 ; x2 = x32 + x1 x2 ).
n=3
0

(x1 = x1 ; x2 = x2 ; x3 = x3 ),
1,0

(x1 = x1 ; x2 = x2 ; x3 = x23 ),
1,1,0

(x1 = x1 ; x2 = x2 ; x3 = x33 + x1 x3 ),
1,1,1,0

(x1 = x1 ; x2 = x2 ; x3 = x43 + x1 x3 + x2 x23 ).
n=4
0

(x1 = x1 ; x2 = x2 ; x3 = x3 ; x4 = x4 ),
1,0

(x1 = x1 ; x2 = x2 ; x3 = x3 ; x4 = x24 ),
1,1,0

(x1 = x1 ; x2 = x2 ; x3 = x3 ; x4 = x34 + x1 x4 ),
1,1,1,0

(x1 = x1 ; x2 = x2 ; x3 = x3 ; x4 = x44 + x1 x4 + x2 x24 ),
1,1,1,1,0

(x1 = x1 ; x2 = x2 ; x3 = x3 ; x4 = x54 + x1 x4 + x2 x24 + x3 x34 ),
2,0

(x1 = x1 ; x2 = x2 ; x3 = x3 x4 ; x4 = x23 + x24 + x1 x3 + x2 x4 ),
2,0

(x1 = x1 ; x2 = x2 ; x3 = x3 x4 ; x4 = x23 − x24 + x1 x3 + x2 x4 ).
www.ebook3000.com
The SPTP 187
On the left of each standard form is the Boardman symbol corresponding

to the singularity set to which the singular point belongs. Note that in all
cases the standard forms for singularities of type Σi are written as (n − i)-
parameter unfoldings of i-dimensional maps.
For a stable Rn → Rp map, the singularity set Σi1 ,...,ik is a smooth
submanifold of codimension
(p − n + i1 )µ(i1 , . . . , ik ) − (i1 − i2 )µ(i2 , . . . , ik ) − · · · − (ik − ik−1 )µ(ik ),
where µ(is , . . . , ik ) denotes the number of sequences of integers (js , . . . , jk )

satisfying js ≥ js+1 ≥ · · · ≥ jk ≥ 0 with ir ≥ jr for all s ≤ r ≤ k and
js > 0. In particular, for the equidimensional case (n = p), Σi has codi-
mension i2 . That is why, in the list above, singularities of the type Σ2 only
appear for n ≥ 4.
The (Boardman) singularity symbols Σi1 ,...,ik are equivalence invariants,
that is, they are invariant under a change of parameters (in the sense defined
above), and therefore, they are a robust characterization of the dynamical
system families. Note, however that, for example, the last two stable R4 →
R4 maps listed above have the same Σ2,0 symbol but are not equivalent.
Hence, the classification of singularities in ΣI classes is not complete.
For low dimensions, stable maps are dense in the space of all Rn → Rn
maps. However, for n ≥ 9 this is no longer true.
In discussing the stability of critical properties of dynamical system
families through the stability of smooth mappings one is directly concerned
with discrete time dynamics. This is not a serious limitation because in a
continuous time system one may always consider the intersections of the
orbits with some transversal surface in phase space. Conversely, for a dis-
crete dynamical system defined in K ⊂ Rn there is [97] a continuous time
system in R2n+1 for which K is a global section.
A.3. Stable dynamical families with degeneracies

Here we are concerned with properties of smooth functions f : Rn → R.
The structural stability conditions for functions are given by Morse theory:
(i) f is stable if and only if the critical points are non-degenerate (non-
vanishing Hessian) and distinct.
(ii) If f is stable, local coordinates (x1 , . . . , xn ) may be defined in such a
way that in the neighborhood of each point − →
x the function may be
written either as
f (−
→
x ) = x1
or
f (−
→
x ) = x21 + · · · + x2k − x2k+1 − · · · − x2n
(iii) Stable functions on a compact manifold are everywhere dense in the

space of all smooth functions.
Hence, in the space of all functions, stable functions are generic and
the non-stable functions form a codimension-one hypersurface, that is,
a submanifold defined by one equation. This hypersurface is called the
bifurcation set. The bifurcation set is the union of the hypersurface of
functions having degenerate critical points and the hypersurface of func-
tions with coinciding critical values. The bifurcation set divides the func-
tion space into components. When in the previous subsection we spoke
of the notion of change of parameters in a stable family of dynamics, as
induced by the diffeomorphisms h and g, this operates solely inside one of
the components of the space of functions. However we may have a more
general situation. Consider, for example, a one-parameter family. This is
represented by a curve in function space. If the intersection of this curve
with the bifurcation hypersurface is transversal then the intersection is sta-
ble in the sense that it cannot be destroyed by a small variation of the
one-parameter family. For a neighboring family the intersection will occur
for a slightly different value of the parameter and the point of intersection
itself is slightly different. However, the intersection cannot be removed by
small perturbations and the situation is qualitatively the same for all the
neighboring families. An example is
ft (x) = x3 − tx
which has a degenerate critical point at t = 0 which cannot be removed from

the family by small perturbations. We therefore reach the notion of stable
dynamical family with degeneracies. Such families represent the stable ways
to connect two non-equivalent classes of functions. This is the reason why
they might be relevant to the theory of phase transitions as illustrated in
Sec. 4.
To classify the possible classes of stable parametrized families, the notion
of universal unfolding plays an essential role. For the space of function
germs En one uses in general a notion of equivalence finer than the one
defined for general maps. Two function germs f, g : Rn → R are said to be
right-equivalent if there is a diffeomorphism germ h and a constant c such
www.ebook3000.com
The SPTP 189
that
g(x) = f ◦ h(x) + c.
The action of all possible diffeomorphisms h acting on a function f defines

an orbit of a smooth action. The codimension of the orbit is the number of
independent functional directions missing from the orbit. The codimension
is obtained by finding the quotient of the functional space with the tangent
space to the orbit. The functional space to consider is Mn , the ideal of
germs vanishing
∂f at the origin, and the tangentspace is the Jacobian ideal
∂f
∆(f ) = g1 ∂x1 + · · · + gn ∂xn : g1 , . . . , gn ∈ En . Then
Mn
cod(f ) = dim .
∆(f )
Whenever the codimension of f is finite the construction of a stable family
of dynamics fα (x) based on f is straightforward. A basis {u1 , . . . , ul } is
Mn
found for ∆(f ) and
fα (x) = f (x) + α1 u1 (x) + · · · + αl ul (x). (A.2)
This unfolding of the function f is called universal because any other

unfolding may be induced from it by a smooth change of parameters and
the number l of unfolding directions is as small as possible. If the function
f is stable the unfolding coincides with the function itself. A family of func-
tion germs is structurally stable if any small perturbation is equivalent to
it, as an unfolding. (Equivalence for two unfoldings means that they may
be obtained from each other by a smooth change of parameters.) Hence, a
universal unfolding of a germ of finite codimension is structurally stable.
The unfolding (A.2) is linear in the parameters α and for finite codi-
mension this construction characterizes all possible parametrized functional
families. A useful result is the splitting lemma which states that if the rank
of the second differential (the Hessian) of f at a singularity is r then f is
right equivalent to
g(x1 , . . . , xn−r ) ± x2n−r+1 ± · · · ± x2n .
The splitting lemma reduces the effective number of variables to n − r and

the classification of possible classes for f depends only on the classification
of g. n−r is called the corank. A list of all the classes of universal unfoldings
for codimension ≤ 5 is included here. By the splitting lemma, in each case,
we may add an arbitrary quadratic (Morse) function on the other variables.
A more extensive list may be found in Ref. [80].
The symbols Ak , Dk , and E6 are used because of the relation of these

singularities to the crystallographic groups with the same symbols. Ak and
Dk correspond to two infinite series with germs g(x1 , . . . , xn−r ) equivalent
to xk+1 and x2 y + y k−1 .
When using the stable unfoldings to model natural phenomena, the
first and most important number to be concerned with is the codimension
(of the germ g), because degenerate singularities are irremovable only in
the case of a family depending on parameters. In particular, a singularity
of codimension c is irremovable only if the number of parameters is ≥ c.
Conversely, if for some process, there are l relevant parameters then all
classes up to codimension l should be considered.
Symbol Corank,
Codimension
A2 x3 + αx 1,1
A3 ±x4 + α1 x2 + α2 x 1,2
A4 x5 + α1 x3 + α2 x2 + α3 x 1,3
A5 ±x + α1 x4 + α2 x3 + α3 x2 + α4 x
6
1,4
A6 x + α1 x5 + α2 x4 + α3 x3 + α4 x2 + α5 x
7
1,5
D4 x3 − xy 2 + α1 x2 + α2 x + α3 y 2,3
D4 x3 + xy 2 + α1 x2 + α2 x + α3 y 2,3
D5 ±(x y + y 4 ) + α1 x2 + α2 y 2 + α3 x + α4 y
2
2,4
D6 x5 − xy 2 + α1 y 3 + α2 x2 + α3 y 2 + α4 x + α5 y 2,5
D6 x5 + xy 2 + α1 y 3 + α2 x2 + α3 y 2 + α4 x + α5 y 2,5
E6 ±(x3 + y 4 ) + α1 xy 2 + α2 y 2 + α3 xy + α4 x + α5 y 2,5
Appendix B: Algebraic Deformation Theory:

Basic Notions
Deformation theory, as the study of continuous families of mathemati-
cal structures, already implicit in the work of Riemann [98], traces its
modern origins to the work of Fröhlicher–Nijenhuis [99] and Kodaira–
Spencer [100] on deformations of complex manifolds and of Gerstenhaber
[103] and Nijenhuis–Richardson [102] on the deformations of associative and
Lie algebras. So far, it is the deformation theory of algebras that seems to
play the main role on physical applications.
B.1. Deformation of Lie algebras

In physics, it is useful to have an explicit representation of the deformation
parameters, because they may play the role of fundamental constants in
the deformed stable theories. I will therefore, focus in the theory of formal
www.ebook3000.com
The SPTP 191
deformations of Lie algebras [104]. A formal deformation of a Lie algebra

L0 defined on a vector space V over a field K is an algebra Lt on the space
V ⊗ K[t] (where K[t] is the field of formal power series), defined by
∞

[A, B]t = [A, B]0 + φi (A, B)ti (B.1)
i=1
with A, B, φi (A, B) ∈ V and t ∈ K. The adjoint representation of L0 is
ρ(A)(B) = [A, B]0 . (B.2)
The (Chevalley) cohomology groups play a key role in characterizing

the stability of the Lie algebra. An n-cochain (relative to the adjoint rep-
resentation) is a multilinear, skew-symmetric mapping
V × ···× V → V
and the n-cochains form a vector space C n (ρ, V ). In particular, φi (A, B) in

Eq. (B.1) must be a 2-cochain. One also has:
• The coboundary operator

n+1
i , . . . , An+1 )
d(A1 , . . . , An+1 ) = (−1)i−1 ρ(Ai )φ(A1 , . . . , A
i=1
(B.3)
+ i , . . . , A
(−1)i+j φ([Ai , Aj ], A1 , . . . , A j , . . . , An+1 )
1≤i<j≤n+1
• A cocycle φ ∈ C n (ρ, V ) whenever dφ = 0. The set of all n-cocycles is a

vector space denoted Z n (ρ).
• A coboundary if φ ∈ d(C n−1 (ρ, V )). The set of all coboundaries is a
vector space denoted B n (ρ).
• The quotient space
Z n (ρ)
H n (ρ) =
B n (ρ)
is the n-cohomology group (relative to the ρ-representation). From (B.3),
it follows that d2 φ = 0. However, not all cocycles need to be coboundaries
and the n-cohomology groups may be non-trivial.
To illustrate the relevance of these concepts to the deformation problem,
use the deformed commutation relations (B.1) and differentiate the Jacobi
identity
[A, [B, C]t ]t + [B, [C, A]t ]t + [C, [A, B]t ]t = 0 (B.4a)
in the variable t. Then, setting t = 0, one obtains

dφ1 (A, B, C) = 0,
that is, for the deformation in (B.1) to be a Lie algebra, φ1 must be a
2-cocycle.
A deformation of L0 is said to be trivial if the algebra Lt is isomorphic to
L0 . This means that there is an invertible linear transformation Tt : V → V
such that
[A, B]t = Tt−1 [Tt A, Tt B]0 . (B.5)
If all deformations Lt are isomorphic to L0 then L0 is said to be stable
or rigid. Suppose now that the second cohomology group H 2 (ρ) is trivial.
This means that all 2-cocycles are 2-coboundaries. Then, there must be a
1-cochain γ such that φ1 = dγ. Applying the linear transformation Mt =
exp{−tγ} to the algebra Lt
[A, B]t = Mt−1 ([Mt A, Mt B]0 ).
From φ1 = dγ, one now obtains, by a simple calculation
φ1 (A, B) = φ1 (A, B) − [γ(A), B] − [A, γ(B)] + γ([A, B]) = 0.
Therefore, the power series expansion for [A, B]t begins with terms of second
order in t
[A, B]t = [A, B]0 + φ2 (A, B)t2 + · · ·
and from the Jacobi identity, as above, it follows dφ2 (A, B) = 0. Iterating
the whole process, all powers of t are successively eliminated. It means that
the limit
Tt−1 = Mt−1 Mt−1 . . .
is the transformation that establishes the equivalence of Lt and L0 . In con-
clusion, the vanishing of the second cohomology group is a sufficient condi-
tion for non-existence of non-trivial deformations, that is, it is a sufficient
condition for stability (or rigidity) of the Lie algebra. This is the content
of the “rigidity theorem” of Nijenhuis and Richardson [102]. However, the
condition is not necessary and there are rigid Lie algebras for which the
second cohomology group is non-vanishing [8, 105, 106].
Cocycles and coboundaries have a nice geometrical interpretation. The
set Ln of all possible complex n-dimensional Lie algebras is an algebraic
variety embedded in the linear space of alternating bilinear mappings in
www.ebook3000.com
The SPTP 193
3 2
C N (isomorphic to C (n −n )/2 ), the defining algebraic relations being the
Jacobi identity relations between the structure constants. In Ln , one has
two natural topologies. One is the topology induced on Ln by the open sets
in C N . the other is the Zariski topology defined by taking closed sets to be
zeros of polynomials on Ln .
The isomorphism relation (B.5) is an action of the linear group GL(n, C)
Ln × GL(n, C) → Ln : (l, T ) → T −1 ◦ l ◦ T × T, (B.6)

.
where l ∈ Ln denotes the Lie algebra law. Denoting l0 (A, B) = [A, B]0 ,
l0 will be a rigid algebra if its orbit O(l0 ) under the action of GL(n, C) is
(Zariski) open. Considering an infinitesimal transformation
T = Id + εφ,
acting on l0 (T : l0 → l0 ), a simple computation using (B.5) leads to
l0 − l0
lim (X, Y ) = dφ(X, Y ).
ε→0 ε
Therefore, the tangent space to the orbit O(l0 ) at l0 coincides with B 2 (l0 ).
On the other hand, considering a tangent line to Ln at l0
lt = l0 + εψ,
lt will satisfy Jacobi identity if and only if
d2 ψ = 0,
that is, if ψ ∈ Z 2 (l0 ), the 2-cocycle space Z 2 (l0 ) of l0 . Then we understand

why the vanishing of the second cohomology group is a sufficient condition
for rigidity of the algebra. The correspondence does not work both ways
because, in general, the algebraic variety Ln has singular points.
Semi-simple Lie algebras have a vanishing second cohomology group
[107] and are stable. More generally, for any subalgebra l of a semi-simple
Lie algebra that contains a maximal solvable algebra, one has H p (l) = 0
for all p ≥ 0 [108].
In the general case, the construction of the cohomology groups is not a
simple matter. This led to the development of different, non-cohomological,
techniques to classify the rigid Lie algebras [27, 106, 109–112]. Here, an
important role is played by the techniques of non-standard analysis. In
this context a Lie algebra law l0 is said to be rigid if any perturbation is

isomorphic to l0 . A perturbation of l0 is an algebra such that
l(A, B) ∼ l0 (A, B) (B.7)
for A, B standard or limited. The symbol ∼ means infinitesimally close.
There is a decomposition of any perturbation of l0 as follows
l = l0 + 1 φ1 + 1 2 φ2 + · · · + 1 2 . . . k φk (B.8)
which is unique up to equivalence. The φ’s are standard antisymmetric
bilinear mappings, the ’s are non-zero infinitesimals and k ≤ N .
The most useful result for the characterization of the rigid Lie algebras
is the theorem that states that if l0 is rigid there is a standard non-zero vec-
tor X such that adl0 X is diagonalizable (adl0 X(Y ) = [X, Y ]). The converse
result is not true and to classify the rigid algebras in dimension n one still
has to exclude the non-rigid ones with a diagonalizable vector. A large num-
ber is simply excluded by checking the rank of the root system and for the
rest (which is a finite number) one has to check explicitly the isomorphism
of the perturbation. For details, refer to Refs. [27, 109, 111, 112].
A related question is the strong rigidity. A finite-dimensional complex
Lie algebra is called strongly rigid if its universal enveloping algebra is rigid
as an associative algebra. Results on the strong rigidity question may be
found in Ref. [113].
B.2. Bialgebras
A bialgebra over the field K is an algebra A which, in addition to the product
m, is equipped with a coproduct : A → A ⊗ A and a counit ε : A → K
satisfying
◦ m = (m ⊗ m) ◦ τ ◦ ( ⊗ ) ; ε ◦ m = m ◦ (ε ⊗ ε)
(B.9)
◦ i = i ⊗ i; ε ◦ i = I,
where i is the unit of the algebra, I the identity map and τ a permutation
on the nearby indices. With an additional operation called the antipode
S : A → A and
m ◦ (I ⊗ S) ◦ = m ◦ (S ⊗ I) ◦ = i ◦ ε (B.10)
the bialgebra becomes an Hopf algebra. These properties have a natural
realization on (and were abstracted from) the algebra of a group G where
m (g ⊗ h) = gh ε (g) = 1
. (B.11)
(g) = g ⊗ g S (g) = g −1
www.ebook3000.com
The SPTP 195
Then, for example, the first equation in (B.9) reads
◦ m (g, h) = gh ⊗ gh = (m ⊗ m) ◦ τ ◦ ( ⊗ ) (g, h)
= (m ⊗ m) ◦ τ ◦ (g ⊗ g, h ⊗ h) = gh ⊗ gh,
where the last step follows from exchanging the second and third argument
(that is why the permutation τ is sometimes denoted (2, 3)).
Other common realizations are:
• For the algebra of functions on a group
m (f ⊗ g) (x) = f (x) g (x) ε (f ) = f (e)
(B.12)
(f ) (x, y) = f (xy) S (f ) (x) = f x−1 .
• For the tensor algebra on a vector space
m (x, y) = x ⊗ y ε (x) = 0
(B.13)
(x) = x ⊗ 1 + 1 ⊗ x S (x) = −x.
A deformation theory of bialgebras has been developed [114] and a par-
tial classification of rigid bialgebras has also been obtained [115].
References
1. A. Andronov and L. Pontryagin, Systèmes grossières, Dokl. Akad. Nauk.
SSSR 14 (1937) 247–250.
2. S. Smale, Structurally stable systems are not dense, Amer. J. Math. 88
(1966), 491—496; Differentiable dynamical systems, Bull. Am. Math. Soc.
73 (1967) 747–817.
3. M. Flato, Deformation view of physical theories, Czech.J.Phys. B32 (1982)
472–475.
4. L. D. Faddeev, On the relationship between Mathematics and Physics, Asia-
Pacific Physics News 3 (1988) 21–22 and in Cerdeira and Lundqvist Fron-
tiers in Physics, High Technology and Mathematics 238–246, World Scien-
tific, 1989.
5. H. Feshbach and V. Weisskopf, Ask a Foolish Question. . . , Physics Today,
October 1988, pp. 9–11.
6. C. W. Misner, K. S. Thorne and J. A. Wheeler, “Gravitation”, W. H. Free-
man and Company, S. Francisco, 1973, p. 71.
7. I. E. Segal, A class of operator algebras which are determined by groups,
Duke Math, Journal, 18 221–265, (1951).
8. R. W. Richardson, On the rigidity of semi-direct products of Lie algebras,
Pac. J. Math, 22 339–344, (1967).
9. J. Baugh, D. R. Finkelstein, A. Galiautdinov and H. Saller, Clifford algebra
as quantum language, J. Math. Phys., 42 1489–1500, (2001).
10. A. A. Galiautdinov and D. R. Finkelstein, Chronon corrections to the Dirac
equation, J. Math. Phys., 43 4741–4752, (2002).
11. D. R. Finkelstein, Homotopy approach to quantum gravity, Int. J. Theor.

Phys., 47 534–552, (2008).
12. D. R. Finkelstein and M. Shiri-Garakani, Finite quantum dynamics, Int. J.
Theor. Phys., 50, 1731–1751, (2011).
13. D. R. Finkelstein, Simplicial quantum dynamics, arXiv:1108.1495, Quantum
set algebra for quantum set theory, arXiv:1403.3725, Quantum field theory
in quantum set algebra, arXiv:1403.3726
14. F. Bayen, M. Flato, C. Fronsdal, C. Lichnerowicz and D. Sternheimer, Defor-
mation theory and Quantization I. Deformation of symplectic structures; II.
Physical Applications, Ann. Phys. NY., 111 61–110; 111–151, (1978).
15. J. Vey, Deformation du crochet de Poisson sur une variété symplectique,
Comm. Math. Helv., 50, 421–454, (1975).
16. M. De Wilde and P. Lecomte, Existence of star products and of formal
deformations of the Poisson-Lie algebra of arbitrary symplectic manifolds,
Lett. Math. Phys., 7, 487–496, (1983).
17. O. M. Neroslavsky and A. T. Vlassov, Sur les deformations de l’algebre des
fonctions d’une variété symplectique, C. R. Acad. Sci. Paris, 292, I, 71–73,
(1981).
18. M. Cahen and S. Gutt, Regular star-representations of Lie algebras, Lett.
Math. Phys., 6, 395–404, (1982).
19. J. Moyal, Quantum mechanics as a statistical theory, Proc. Camb. Phil.
Soc., 45, 99–124, (1949).
20. M. Kontsevich, Operads and motives in deformation quantization, Lett.
Math. Phys., 48, 35–72, (1999).
21. A. Cattaneo, B. Keller, C. Torossian and A. Bruguières, Déformation, Quan-
tification, Théories de Lie, Panoramas et Synthèses 20, Soc. Math., de
France, (2005).
22. M. Kontsevich and Y. Soibelman, Deformation Theory I, www.math.ksu.
edu/˜soibel/Book-vol1.ps.
23. M. Kontsevich, Deformation quantization of Poisson manifolds, Lett. Math.
Phys., 66, 157–216, (2003).
24. B. O. Koopman, Hamiltonian systems and transformations in Hilbert space,
Proc. Nat. Acad. Sci. USA, 17, 315–318, (1931).
25. J. von Neumann, Zur Operatoren methode in der klassischen mechanik.
Ann. Math., 33 587–642, (1932).
26. J. M. Ancochea Bermudez, in (ed. M. Hazewinkel and M. Gerstenhaber)
“Deformation Theory of Algebras and Structures and Applications” 403–
445, Kluwer, Dordrecht, 1988.
27. M. Goze, in M. Hazewinkel and M. Gerstenhaber “Deformation Theory
of Algebras and Structures and Applications” 265–355, Kluwer, Dordrecht,
1988.
28. R. Vilela Mendes, Deformations, stable theories and fundamental constants,
J. Phys. A: Math. Gen., 27 8091–8104, (1994).
29. A. O. Barut and W. E. Brittin Lectures on Theoretical Physics vol. XIII - De
Sitter and Conformal Groups and Their Applications, Colorado Associated
University Press, Boulder, 1971.
www.ebook3000.com
The SPTP 197
30. C. Chryssomalakos and E. Okon, Generalized quantum relativistic kinemat-

ics: a stability point of view, Int. J. Mod. Phys. D, 13, 2003–2034, (2004).
31. D. V. Ahluwalia-Khalilova, Minimal spatio-temporal extent of events, neu-
trinos, and the cosmological constant problem, Int. J. Mod. Phys. D, 14,
21512166, (2005).
32. D. V. Ahluwalia-Khalilova, A freely falling frame at the interface of gravita-
tional and quantum realms, Class. Quantum Gravity, 22, 1433–1450, (2005).
33. H. S. Snyder, Quantized space-time, Phys. Rev. 71, 38–41, (1947); The Elec-
tromagnetic Field in Quantized Space-Time, 72, 68–71, (1947).
34. C. N. Yang, On quantized space-time, Phys. Rev. 72 874, (1947).
35. V. G. Kadyshevsky, Fundamental length hypothesis and new concept of
gauge vector field, Nucl. Phys., B141, 477–496, (1978).
36. V. G. Kadyshevsky and D. V. Fursaev, On a generalization of the gauge
principle at high-energies, Teor. Mat. Phys., 83, 474–480, (1990) and refer-
ences therein.
37. M. Banai, Quantum relativity theory and quantum space-time, Int. J.
Theor. Phys., 23, 1043–1063, (1984).
38. A. Das, The quantized complex space-time and quantum theory of free
fields, J. Math. Phys., 7, 52–60, (1966).
39. D. Atkinson and M. B. Halpern, Non-usual topologies on space-time and
high energy scattering, J. Math. Phys., 8, 373–387, (1967).
40. S. P. Gudder, Elementary length topologies in physics, SIAM J. Appl. Math.,
16, 1011–1019, (1968).
41. D. Finkelstein, Space-time code IV, Phys. Rev., D9, 2219–2231, (1974).
42. M. Dineykhan and K. Namsrai, Quantum space-time and estimations on
the value of the fundamental length, Mod. Phys. Lett., 1, 183–190, (1986).
43. T. D. Lee, Can time be a discrete dynamical variable?, Phys. Lett., 122B,
217–220, (1983).
44. E. Prugovecki, “Stochastic quantum mechanics and quantum spacetime”,
Reidel Pub. Comp., Dordrecht, 1983.
45. D. I. Blokhintsev, Quarks in quantized space, Teor. Mat. Phys., 37, 933–937,
(1978).
46. A. Schild, Discrete space-time and integral lorentz transformations, Phys.
Rev., 73, 414–415, (1948).
47. G. Veneziano, Physics with a fundamental length, CERN preprint TH
5581/89.
48. G. ’t Hooft, in S. Deser and R. J. Finkelstein (eds.), “Themes in Contem-
porary Physics”, Vol. II, 77, World Scientific, Singapore 1990.
49. J. C. Jackson, A quantisation of time, J. Phys. A: Math. Gen., 10, 2115–
2122, (1977).
50. J. Madore, Fuzzy physics, Ann. Phys., 219, 187–19, (1992).
51. M. Maggiore, The algebraic structure of the generalized uncertainty princi-
ple, Phys. Lett., B319, 83–86, (1993).
52. R. Vilela Mendes, Quantum mechanics and non-commutative space-time,
Phys. Lett. A, 210, 232–240, (1996).
53. E. Carlen and R. Vilela Mendes, Non-commutative space–time and the
uncertainty principle, Phys. Lett. A, 290, 109–114, (2001).
54. R. Vilela Mendes, Some consequences of a non-commutative space-time

structure, Eur. Phys. J. C, 42, 445–452 (2005).
55. R. Vilela Mendes, The deformation-stability fundamental length and devi-
ations from c, Phys. Lett. A, 376, 1823–1826, (2012).
56. R. Vilela Mendes, A laboratory scale fundamental time?, Eur. Phys. J. C
72, 2239–2254, (2012).
57. R. Vilela Mendes, Stochastic calculus and processes in non-commutative
space-time, Progress in Probability, 52, 205–217, (2002).
58. R. Vilela Mendes, Geometry, stochastic calculus, and quantum fields in a
noncommutative space–time, J. Math. Phys., 41, 156–186, (2000).
59. A. Connes, Non-commutative geometry, Academic Press, 1994.
60. A. Connes and M. Marcolli, “Noncommutative Geometry, Quantum Fields
and Motives”, American Math. Society Colloquium Publications, 2007.
61. M. Dubois-Violette, Derivations et calcul differentiel non-commutatif, C. R.
Acad. Sci. Paris Ser. I, 297, 381–384, (1985).
62. M. Dubois-Violette, R. Kerner and J. Madore, Noncommutative differential
geometry of matrix algebras, J. Math. Phys., 31, 316–322, (1990) Noncom-
mutative differential geometry and new models of gauge theory, 31, 323–330,
(1990).
63. P. A. M. Dirac, A Remarkable Representation of the 3 + 2 de Sitter Group,
J. Math. Phys., 4, 901–909, (1963).
64. C. Fronsdal, Elementary particles in a curved space IV. Massless particles,
Phys. Rev. D, 12, 3819–3830, (1975).
65. M. Flato, C. Fronsdal, Composite Electrodynamics, J. Geom. Phys., 5, 37–
61, (1988)
66. M. Flato, C. Fronsdal and D. Sternheimer, Singletons, physics in AdS uni-
verse and oscillations of composite neutrinos, Lett. Math. Phys., 48, 109–119,
(1999).
67. R. M. May, Biological Populations Obeying Difference Equations : Stable
Points, Stable Cycles, and Chaos, J. Theor. Biol., 51, 511–524, (1975).
68. M. J. Feigenbaum, Quantitative universality for a class of nonlinear transfor-
mations, J. Stat. Phys., 19, 25–72 (1978); The Universal Metric Properties
of Nonlinear Transformations, J. Stat. Phys., 21, 669–706, (1979).
69. O. E. Lanford, A computer-assisted proof of the Feigenbaum conjectures,
Bull. Amer. Math. Soc., 6, 427–434, (1982).
70. J.-P. Eckmann, H. Koch and P. Wittwer, A computer-assisted proof of uni-
versality for area-preserving maps, Memoirs AMS, 47, (289), (1984).
71. M. Campanino and H. Epstein, On the existence of Feigenbaum’s fixed
point, Commun. Math. Phys. 79, 261–302, (1983).
72. E. B. Vul, Ya. G. Sinai and K. M. Khanin, Feigenbaum universality and the
thermodynamic formalism, Russ. Math. Surveys 39, 3–37, (1984).
73. R. Vilela Mendes, Critical point dependence of universality in maps of the
interval, Phys. Lett., 84A, 1–3, (1981).
74. A. Libchaber and Maurer, Local Probe in a Rayleigh-Bénard Experi-
ment in Liquid Helium, J. Phys. Lett., 39, 369–373 (1978); Une experi-
ence de Rayleigh-Bénard de géométrie réduite: Multiplication, accrochage et
démultiplication de fré quences, J. Phys. (Paris) Colloq. 41, C3, 51, (1980).
www.ebook3000.com
The SPTP 199
75. A. I. Gol’berg, Ya. G. Sinai and K. M. Khanin, Universal properties of

sequences of period-tripling bifurcations, Russ. Math. Surveys, 38, 187–188,
(1983).
76. P. Cvitanovic and J. Myrheim, Universality for period n-tuplings in complex
mappings, Phys. Lett, A94, 329–333, (1983).
77. B. Mandelbrot, On the complex mapping z → z 2 − µ for complex µ and z:
the fractal structure of its M-set and scaling, Physica, D7, 224–239, (1983).
78. K. G. Wilson and J. Kogut, The renormalization group and the expansion,
Phys. Rep., 12C, 75–199, (1974).
79. J. J. Binney, N. J. Dowrick, A. J. Fisher and M. E. J. Newman, The The-
ory of Critical Phenomena. An Introduction to the Renormalization Group,
Oxford University Press, Oxford, 1992.
80. V. I. Arnold, S. M. Gusein-Zade and A. N. Varchenko, “Singularities of
differentiable maps”, Vol. I, Birkhäuser, Boston, 1985.
81. B. Widom, Surface tension and molecular correlations near the critical
point, J. Chem. Phys., 43, 3892–3897 (1965), Equation of state in the neigh-
borhood of the critical point, 3898–3905.
82. L. P. Kadanoff, Scaling laws for Ising models near Tc, Physics, 2, 263–272,
(1966).
83. J. L. Cardy, in C. Domb and J. L. Lebowitz (eds.), Phase Transitions and
Critical Phenomena, vol. 11, 55, Academic Press, London, 1987.
84. R. Mañé, A proof of the C1 stability conjecture, Publ. Math. IHES, 66,
161–210,(1988).
85. J. Palis, On the C1 Ω-stability conjecture, Publ. Math. IHES, 66, 211–215,
(1988).
86. J. Palis and S. Smale, Structural stability theorems, Proc. A.M.S. Symp.
Pure Math., 14, 223–232, (1970).
87. M. Peixoto and C. Pugh, Structurally Stable Systems on Open Manifolds
are Never Dense, Ann. Math., 87, 423–430, (1968).
88. R. F. Williams: in Global Analysis, S. S. Chern and S. Smale (eds.) Proc.
Symp. Pure Math. 14, 341, Providence R. I., AMS, 1970.
89. M. Peixoto, Structural stability on two-dimensional manifolds, Topology, 1,
101–120, (1962).
90. E. C. Zeeman, Stability of dynamical systems, Nonlinearity, 1, 115–155,
(1988).
91. M. Peixoto, On structural stability, Ann. of Math., 69, 199–222, (1959).
92. R. C. Robinson, in S. S. Chern and S. Smale (eds.), Proceedings of the Sym-
posium in Pure Mathematics XIV, Global Analysis, 233–244, Am. Math.
Soc., Providence, R. I., 1970.
93. R. Abraham and J. E. Marsden, Foundations of Mechanics, 2nd edition,
Benjamim, Reading, Massachusetts 1978.
94. R. C. Robinson, Generic properties of conservative systems, Am. J. Math.,
92, 562–603, 897–906, (1970).
95. V. I. Arnold, Singularities of smooth mappings, Russian Math. Surveys, 23,
1–43, (1968).
96. C. G. Gibson, Singular points of smooth mappings, Pitman, London, 1979.
97. N. P. Zhidkov, Some properties of dynamical systems (Russian), Uch. Zap.

Moskovsk. Un-ta (Matematika), 163, 31–59, (1952).
98. B. Riemann, Theorie der Abel’schen Functionen, Journal für die reine und
angewandte Mathematik, 54, 101–155, (1857).
99. A. Frohlicher and A. Nijenhuis, A theorem on stability of complex struc-
tures, Proc. Nat. Acad. Sci., Wash., 43, 239–241, (1957).
100. K. Kodaira and D. C. Spencer, On the variation of almost complex struc-
tures in Algebraic Geometry and Topology, Princeton Univ. Press, 139–150,
(1957).
101. M. Gerstenhaber, On the deformation of rings and algebras, Ann. Math.,
79, 59–103, (1964) 88, 1, (1968); 99, 257, (1974).
102. A. Nijenhuis and R. W. Richardson, Cohomology and deformations in
graded Lie algebras, Bull. Am. Math. Soc., 72, 1–29, (1966); Deformations
of homomorphisms of Lie groups and Lie algebras, 73, 175–179, (1967);
Deformation of Lie algebra structures, J. Math. Mech. 17, 89–105, (1967).
103. M. Gerstenhaber and S. D. Schack, in M. Hazewinkel and M. Gerstenhaber
(eds.), Deformation Theory of Algebras and Structures and Applications 11–
264, Kluwer, Dordrecht, 1988.
104. R. Hermann, Vector Bundles in Mathematical Physics, ch. 3, W. A. Ben-
jamin, New York, 1970.
105. G. Rauch, Éffacement et Déformation, Ann. Inst. Fourier Grenoble, 22,
239–269, (1972).
106. R. Carles, Sur la structure des algèbres de Lie rigides, Ann. Inst. Fourier,
Grenoble, 33, 65–82, (1984); Sur certaines classes d’algèbres de Lie rigides,
Math. Ann., 272, 477–488, (1985); Un exemple d’algèbres de Lie résolubles
rigides, au deuxième groupe de cohomologie non-nul et pour lesquelles
l’application quadratique de D.S. Rim est injective, C. R. Acad. Sci. Paris,
300, 467–469, (1985).
107. N. Jacobson, Lie algebras, Interscience Pub., New York, 1962.
108. A. L. Onishchik and E. B. Vinberg, Lie Groups and Lie Algebras III, Ency-
clopaedia of Mathematical Sciences vol. 41, Springer, Berlin, 1994.
109. M. Goze and J. M. Ancochea Bermudez, Algèbres de Lie rigides, Indaga-
tiones Math 88, 397–415, (1985).
110. R. Carles and Y. Diakite, Sur les variétés d’algèbre de Lie de dimension ≤ 7,
Jour., Algebra, 91, 53–63, (1984).
111. M. Goze, On the Classification of Rigid Lie Algebras, J. of Algebra, 245,
68–91, (2001).
112. J. M. Ancochea Bermudez, R. Campoamor-Stursberg, L. Garcia Vergnolle
and M. Goze, Algèbres de Lie résolubles réelles algébriquement rigides,
Monatsh. Math., 152, 187–195, (2007).
113. M. Bordemann, A. Makhlouf and T. Petit, Deformation par quantification
et rigidité des algèbres enveloppantes, J. of Algebra, 285, 623–648, (2005).
114. M. Gerstenhaber and S. D. Schack, Bialgebra cohomology, deformations,
and quantum groups, Proc. Nat. Acad. Sci., 87, 478–481, (1990).
115. A. Makhlouf, Degeneration, rigidity and irreducible components of Hopf
algebras, Algebra Colloq., 12, 241–254, (2005).
www.ebook3000.com
Chapter 7
Factory of Realities:
On the Emergence of Virtual
Spatiotemporal Structures
Romàn R. Zapatrin
Informatics Dept., The State Russian Museum,
Inżenernaya 4, 191186 St.Petersburg, Russia
Roman.Zapatrin’@Gmail.com
The ubiquitous nature of modern Information Retrieval (IR) and Virtual

World give rise to new realities. To what extent are these ‘realities’ real?
Which ‘physics’ should be applied to quantitatively describe them? In
this chapter, I dwell on few examples. The first is adaptive neural net-
works, which are not networks and not neural, but still provide service
similar to classical artificial neural networks (ANNs) in extended fashion.
The second is the emergence of objects looking like Einsteinian space–
time, which describe the behavior of an Internet surfer like geodesic
motion. The third is the demonstration of nonclassical and even stronger-
than-quantum probabilities in IR, their use.
Immense operable datasets provide new operationalistic environ-
ments, which become to greater and greater extent “realities”. In this
chapter, I consider the overall IR process as an objective physical pro-
cess, representing it according to Melucci metaphor in terms of physical-
like experiments. Various semantic environments are treated as analogs
of various realities. The reader’s attention is drawn to topos approach
to physical theories, which provides a natural conceptual and technical
framework to cope with the new emerging realities.
1. Introduction
The idea to treat Information Retrieval (IR), or, in general, data and knowl-
edge proceeding as physical processes has long tradition, in particular, these
ideas gave rise to Quantum Computation. In the last decades, a research was
carried out on viewing usual (non-quantum) information deals as physical
201
202 R. Zapatrin
processes [1–3]. More than 100 years ago along with the growth of the preci-
sion of physical measurements, Quantum Mechanics was born. Now, along
with the development of World Wide Web and the emergence of immense
data corpora, new conceptual challenges arise. The amount of the existing
data together with the possibility to generate new data on-the-fly give rise to
purely theoretical inquiries whose practical goals were to boost the perfor-
mance of search engines. While the notion of “performance” in this context
remains ambiguous, efforts were made to build a general framework for IR in
the newly emerged environment. It was observed — in practice, by trials —
that the performance of search algorithms may be sometimes improved if
one follows a “wrong” probabilistic model, that is, recalculate probabilities
in discordance with Kolmogorovian laws. Similar situation takes place in
Quantum Theory, which gave rise to Quantum Probability. Furthermore,
quantum probabilistic approach was successfully applied to IR [4].
The development of World Wide Web, the emergence of massive acces-
sible data sets gave rise to a kind of new realities. Like conventional physics,
IR deals with events and, as it was recently discovered, the statistical depen-
dencies observed (that is, obtained in experiments) in IR processes may not
only be non-classical, but also demonstrate stronger-than-quantum corre-
lations. From a broader conceptual perspective, IR in its current state can
be treated as a factory producing various realities. I intentionally reduce
the picture to that extent that from the point of view of ‘end user’, modern
sophisticated physical laboratory is just a man–computer interaction. The
User affects the Environment, the Environment reacts somehow, the User,
being to this or that extent satisfied (this is to be quantified as well), refines
its query according to some underlying principles.
The above mentioned ‘underlying principles’ are based on certain pic-
tures of reality which need not be classical, or even quantum. The suggested
picture of operationalistically perceived Environment requires a consistent
realistic framework. In recent years, a formalism based on topos theory was
developed meeting this requirement. In this chapter, the appropriate math-
ematical models based on topos theory and giving rise to event structures
similar to those forming traditional space–time are overviewed.
In the meantime, attempts to build a consistent and in some sense “real-
istic” theory of Quantum Gravity faced severe problems. The reason was
that the very notion of a single, pre-defined underlying configuration space
is strange for the theory, it required a formalism dealing with varying or
emerging space–time. The problems in both IR and Quantum Gravity were
of similar kind — how to make space and time secondary, derived, rather
than fundamental entities. In order to explain the observed non-classicality,
www.ebook3000.com
Factory of Realities: On the Emergence of Virtual Spatiotemporal Structures 203
the idea of non-uniqueness of the event space within a given probabilistic

model was put forward [5]. However, the occurring “wrong” probabilities
behaved sometimes neither in quantum, nor in classical way. The advances
in quite different areas signal the need for a unified conceptual framework.
An essential breakthrough in this direction was made by introducing topos
approach to physical theories. In this chapter, possible applications of topos
theory to IR are highlighted.
2. Computational Complementarity
In this section, primitive empirical statements or propositions about
automata [1] are introduced. Such experimental statements form the basis
of the formal investigation of the corresponding logics. In particular, there
exist automata for which validation of one empirical statement makes
impossible the validation of another empirical statement and vice versa,
as it was first pointed out by Moore [6], which makes them similar to quan-
tum systems.
Thereby, one decisive feature of the setup is the intrinsic character of the
measurement process: the automaton is treated as a black box with known
description but unknown initial state. Automata experiments are conducted
by applying an input sequence and observing the output sequence.
The conventional state identification problem [6] is to obtain information
about an unknown initial state. One may think of it as choosing at random
a single automaton from an automata ensemble which differ only by their
initial state. The task then is to find out which was the initial state of the
chosen automaton.
Staying within the instrumentalist framework, we may ask if a classical
object may demonstrate elements of quantum behavior. David Finkelstein
provided a simple example of a classical automaton, which demonstrated
quantum features on the level of the structure of the set of observable
properties [1]. Like in quantum system, every measurement on its state
affected the state itself, so the logic of accessible properties of Finkelstein’s
automaton with the graph
1 x x2
x x
4 3
204 R. Zapatrin
was similar to that of a quantum particle with spin 1/2.

{1, 2, 3, 4}
t
{2, 3, 4} t {1, 3, 4} t t{1, 2, 4} t{1, 2, 3}
t
∅
In the above example, it is not necessary to input sequences containing

more than one symbol, since the non-final states are not distinguished by
the output function.
Since then a consistent theory of finite automata simulating quantum
systems with a finite number of degrees of freedom was developed [2, 7].
A more general framework was developed on both technical and opera-
tionalistic level by Aerts with co-authors, including, in particular, study of
cognition from a quantum perspective [8].
Anyway, on this step, the emerging virtual structures was on the level
of the logic of accessible properties, or, in other words, on the level of
configuration space rather than fully fledged space–time.
3. Classical Spatiotemporal Structures in IR

The basic idea of this approach, put forward in Ref. [9] and developed
in Ref. [10], is as follows. A smooth continuous manifold B is considered,
called IR space. Note that B is neither a document space, nor a query
space; instead, it has a more fundamental and unstructured nature. It may
be thought of as the set of all transmitted bits. The elements of B are all
the same; they have no structure. This is a complete analogy to the points
of space–time, or the configuration space in physical theories.
To be more specific, B is an analog to physical space–time. Both doc-
uments and queries replaced into B, providing them with both temporal
and spatial dimensions. As a consequence, the idea that a document may
www.ebook3000.com
change in time is automatically incorporated in the theory. The second con-

sequence is that the static documents, and those generated on-the-fly, are
described as entities of exactly the same nature, differing only in ‘shape’ in
our IR space–time B. Furthermore, since the points of B are uniform, they
can be treated as events.
The IR process itself is represented by trajectories in B which show the
behavior of users. When we are speaking about a huge community of users,
we no longer treat their behavior as intelligent: this gives us the right to
shift the focus of our research from the task of finding a good way to satisfy
users’ requests to the task of analyzing typical user behavior. From this
perspective, a typical user of a search environment is not more intelligent
than an elementary particle or a molecule, and we may apply the good
old principle of least action, which stems from the work of Fermat and
Euler. This principle proved its efficiency by providing simple and strongly
predictive models.
The standard IR paradigm treats the IR process as a search. That is, the
initial condition is posing a query, then, according to this or that formula,
the indexed documents are ranked. Subsequently, the results are delivered
to the user according to the ranking. But, typically, the user never makes a
single query and the process is usually progressive. After parsing the results
and considering their relevance, the user poses further queries, repeating the
process iteratively based on relevance feedback.
Building IR space is getting above these particularities. The notion of
relevance feedback is replaced by the least action:
The sequence of user’s actions is interpreted as geodesic motion. The
dynamics replace the notion of relevance, and the displacement of a user
from point t1 to point t2 is what replaces relevance feedback making it in
a sense, a relevance feedforward (Fig. 1).
Let us briefly overview how IR space is built on the basis of query log
analysis. Let us first produce the ‘flesh’ of IR space. Its elementary con-
stituent, a point, is a click: a query/HTTP request together with a body
Fig. 1. A point on IR surface together with users’ intention vector.

206 R. Zapatrin
Fig. 2. Points on clickstreams are ordered by timestamps of clicks.
Fig. 3. Clickstreams acquire metrics.
response (HTML page accessed by a result link). In order to specify a par-

ticular geodesic, we must specify the initial conditions which are the initial
position x(0) and the initial ‘intention’ ẋ(0). A typical user clickstream will
be represented as a line, whereby the points of the line x(t) are associ-
ated with the state of knowledge the user has gained from interpreting the
retrieved information until that point.
A clickstream is a progressive, ‘continuous’ sequence of user queries
and responses which have a definite start and end. The result of Step 1
is a collection of clickstreams, an ordered sequences of points as shown in
Fig. 2.
At next step, the discrete pre-space B is created. In the given collection
of clickstreams, their points are ordered and we know the distances between
them. We assume that we use certain distance between points of the threads
(any particular relevance formula can be applied). This way the clickstreams
acquire metric as shown in Fig. 3.
The next step turns the structure from spatial to spatiotemporal by
creating layers (analogs of spacelike hypersurfaces). Start with points with
label 0 (this will be a starting layer), and, using the same distance function,
www.ebook3000.com
Fig. 4. Creating transversal layers.
Fig. 5. Forming the discrete skeleton.
place them as points on a metric space. Then we pass to sequential label 1,

and form the same discrete metric space, and so on. As a result, we have
a sequence of layers labeled 0, 1, . . . , forming altogether a discrete metric
space:
The next step completes the skeleton with the ‘bones’ linking nearest
neighbors, now irrespective of the thread to which they belong as shown in
Fig. 5.
In the formalism proposed in Ref. [10], the dimension n + 1 of the future
IR space is set ad hoc (since it has a spatiotemporal structure, one dimen-
sion is reserved for the temporal parameter and n for ‘spatial’). Once n
is chosen, each layer is projected on an n-dimensional space, a foliation
is formed, labeled 1, 2, . . . together with threads, which are retained from
the first step. Finally, the resulting space is treated as IR space as shown
in Fig. 6.
208 R. Zapatrin
Fig. 6. IR space is built.
4. A Step Towards Quantum: Adaptive ANNs

ANNs have gained increasing interest, finding applications in various
domains such as business planning, medicine, engineering, geology, and
physics. ANNs have shown remarkable efficiency in solving tasks related
to forecasting, planning, control, and classification. ANNs are a natural
and intuitive field to explore, as they draw inspiration from the structure
of real biological systems. One of the basic tasks of neural networks is to
function as perceptrons, that is, to recognize signals for which we have no
structural theory — for instance, to recognize visual patterns.
We shall deal with multilayered feedforward NNs, such as
c -c -c
*

HH
c - c HHjc
* H
*
@H
c - c@HHjc

*H
@
HH
Rc
@
c j
H
that is, their nodes can be arranged in layers so that (i) no nodes in a given
layer communicate and (ii) the signals propagate only consecutively, via
layers.
www.ebook3000.com
Training and performance. Initially, one starts with a set of patterns for
which the classification is known. Usually by means of heuristic methods,
the topological structure of the network is chosen and then trained via input
of known patterns and subsequent adjustment of network parameters (tran-
sition functions). Output signals are correlated with patterns from different
classes, to be well-separated with respect to certain criterion. The most pop-
ular method to adjust transition functions is error backpropagation. Signal
propagation in the linear approximation can be viewed as a matrix multi-
plication, which reduces to a number of arithmetic operations. The more
links there are between neurons, the more computational resources are con-
sumed by the process of pattern recognition. In order for a neural network
to be faster, we should seek sparser configurations. Therefore, the criterion
for ‘good matching’ should also take performance into consideration.
From ANNs to AANNs. The core idea of the suggested approach is

to convert the primary component of ANNs — directed acyclic graphs
(DAGs) — into a more general structure, which naturally complies with
quantum parallelism. The latter inevitably means that this should be a
linear structure in order to support superpositions. In Ref. [11], the idea
to replace DAG by appropriate sets of linear matrices (associated further
with appropriate quantum variables) was suggested. The training paradigm
remains the same as it was in conventional ANNs — standard performance
criteria are used, such as Root Mean Square Error (RMSE) between tar-
gets and outputs of the neural network, the correlation (called R-value)
between the outputs and targets an so on, the are only reclassified in terms
of unitary matrices and matrix operations. The main peculiarity is that the
result of training optimization is then a matrix, or a set of matrices, rather
than a DAG, as is seen in a standard, ‘literal’ approach. This way, Adaptive
AANs (AANNs) are introduced which, strictly speaking, are not networks.
The topology of a feedforward artificial neural network, N , is described
by the template matrix of the appropriate DAG, which is formed as follows:

∗, if j → k in N
Ajk = (1)
0, otherwise,
where ∗ stands for a wildcard — any number, and the set of such numerical
matrices form an algebra [12] as it is closed under multiplication. The main
property of A is that the synaptic weights ‘follow’ it — namely, if Ajk =
0, then wjk = 0. This differs from the standard description in terms of
adjacency matrices, and the difference is that the matrix is not numeric.
210 R. Zapatrin
The simplest example is provided by a two-vertex DAG with one arrow

N = j - j (2)
whose template matrix is

∗∗
A= . (3)
0∗
In general, taking various matrices by substituting the asterisk in template
matrix (1) by numbers, we may form various products. The resulting set of
matrices is called the Rota algebra. It can be verified that this set is closed
under sums and matrix products, and thus qualifies as a closed algebra.
This description is explicated under greater detail in Ref. [11].
Rota topology. The matrix formalism deals with objects having no specific
spatiotemporal features. We call them continuously-evolving, superposed
topologies, but these topologies are yet ‘spaceless’ (while not ‘timeless,’
as we are speaking of their evolution in time, with respect to the proper
time of the experimenter). The possibility to transfer them into space–time
manifold fluctuations was studied in Quantum Gravity [11].
The key point of the spatialization procedure [11] is the following: con-
sider the full matrix algebra A as a linear space. From this perspective, the
Rota algebras are just linear subspaces of A. Having a procedure which —
starting from a subset of A — creates a topological space, providing the
capability to discuss superposed configurations of differing neural networks.
In this section, we present a procedure which — starting from a given sub-
space of A — produces a set, and endows it with a topology that can be
associated with certain acyclic directed graphs.
In brief, the spatialization procedure makes it possible to transform
matrix algebras of the form like (3) to topological spaces like (2), so that
they are endowed with the Rota topology. The emergence in this way, of
classical spaces is the result of a quantum measurement, described by the
Projection Postulate.
5. “Wrong” Probabilities: Within and Beyond

the Quantum
What do I mean by “wrong” probabilities? This is the numerical discrep-
ancy between the predictions of a model based on Kolmogorov paradigm,
that is, on the existence of a predefined sample space and the frequencies
measured within an experiment.
www.ebook3000.com
An example: Bell inequalities. Consider three random variables a, b, c

each taking values ±1. Then whatever be their distributions, the following
inequality holds:
P (a = b) + P (b = c) + P (a = c) ≥ 1. (4)
The proof is straightforward: since all a, b, c take only two values, at least
one of the above equalities hold, therefore at least one of the (always non-
negative) summands in (4) equals to 1. However, if a, b, c are the measured
projections of the spin of a quantum particle, it may happen that each sum-
mand in (4) equals to 41 . I do not dwell on this issue in detail, it gave rise to
discussions and research lasting few recent decades, see, say, Ref. [13] for a
review. The only thing to conclude for this chapter is that there is a realistic
situation, in which the standard Kolmogorovian probabilistic model is not
adequate.
Generalities. In order to test this or that model, Accardi’s statistical
invariants [14] are employed, they allow to test the applicability of Kol-
mogorovian model. Given:
• In a family of discrete maximal observables {Aα : α = 1, . . . T } (T being

(α)
finite), each observable Aα takes the finite number of values ajα labeled
by jα = 1, . . . , n
• The experimentally measurable conditional probabilities pjα ,jβ (β | α)
(β) (α)
pjα ,jβ (β | α) = P (Aβ = ajβ |Aα = ajα ). (5)
The problem is: Does there exist a probability space (Ω; F ; P ) and T mea-
(α)
surable partitions Aj of cardinality n (the number of distinct values of
each observable is assumed to be the same)
(α)
Aj , α = 1, . . . T, j = 1, . . . n,
such that for any α, β = 1, . . . T , one has

(β) (α)
(β) (β) (α) (α) P Aj ∪ Ai
P (A = aj |A = ai ) = (β) . (6)
P Aj
In general, this is a finitely decidable linear programming problem [14]. As

an example, consider the special case of three observables A, B, C, each
taking only two values a1 , a2 for A, b1 , b2 for B and c1 , c2 for the observ-
able C. The transition probability matrices for each pair of observables,
212 R. Zapatrin
being bistochastic each has only one numeric parameter, denote the appro-
priate matrices as

p 1−p
P (A | B) = P = ,
1−p p

q 1−q
P (B | C) = Q = , (7)
1−q q

r 1−r
P (C | A) = R = .
1−r r
Then these transition probabilities can be described by a Kolmogorovian

model (that is, they are produced by a single sample space) if and only if
|p + q − 1| ≤ r ≤ 1 − |p − q|. (8)
That means, when we use probabilities derived from relative frequencies,

which, in the meantime, can be calculated from other given probabilities, we
can directly test the inequality (8). If it fails, that signals that we deal with
“wrong”, non-classical probabilities. They cannot be described in terms of
a single event space and thus require a notion of a ‘varying event space’.
Non-classical probabilities in IR. In the realm of IR, the probabilities
are usually obtained from various sources, and, as it happens, they may not
admit a single event space. “Not admit” means that it is not possible to
assign a single elementary event space and a probability measure satisfying
Kolmogorovian axioms.
In practice, relative frequencies rather than probabilities are measured.
Then, the relative frequencies are thought of as probabilities and, looking
at the values of those probabilities, decisions are made. Are these decisions
good or bad? What can we change in order to enhance the performance
of search environment? The point is that for some probabilities, there are
different ways to be obtained: the value can be either
(i) Obtained directly by measuring appropriate relative frequency or

(ii) Derived from the values of other observed relative frequencies.
The latter needs an underlying probabilistic model. Suppose it is Kol-

mogorovian, so the existence of a (single) sample space Ω is required. The
events are subsets of Ω, while the points of the sample space are elementary
and independent.
www.ebook3000.com
The queries are multiplied. We are going to make evaluations based on

ratios. The standard picture is:
Retrieved
Relevant
The key problem is that the sample space Ω is not well defined. This
suggests the notion of varying, context-dependent sample space.
Melucci metaphor. Melucci metaphor is a unified view to represent sim-
plified IR environment with no reference to particular underlying logic and
rendering IR into experimental, naturalist realm. According to it [15], the
IR procedure is represented by a two-slit experiment, widely known in
physics. The IR system is thought of as a laboratory with the source, which
supplies documents according to the input query. What is described looks
intermediate between a pure Gedanken experiment and a fully fledged mea-
surement.
The documents within Melucci metaphor are treated as particles, they
may be of classical or quantum nature, or, perhaps, of some other kind. We
do specify the mechanism of producing this flux of documents-particles.
214 R. Zapatrin
What is essential, is that the number of ejected documents is supposed to

be potentially infinite, but we analyze only first N documents. Entering
a particular query means preparing the source in a particular state Q | .
From this experiment, we get the value of P (X|R), where P (X) stands for
the probability of the occurrence of a certain term X:
$
R -
*
-
:

- Check X

-
XX
H
HHX
z
X
j
H R
and from this experiment we get the value of P (X|R):
$
R
*

:

-

XX
H HXz
X
H
j
H R
-
- Check X
% -
When we are in the classical realm, there is no need to calculate P (X) due
to our Boolean belief revision (that is, the law of total probability):
P (X) = P (X|R) P (R) + P (X|R) P (R). (9)
www.ebook3000.com
But just for fun we may attempt to measure P (X) directly, removing the
relevance check:
$
R -
*
-

: -

-
XXX
H Check X
H X
z
Hj
H R
-
-
% -
and, surprisingly, discover that the result may drastically differ from (9).
Let us pass to exact numerical results. In order to evaluate the discrepancy,
Accardi statistical invariant is used:
P (X) − P (X|R)
A= . (10)
P (X|R) − P (X|R)
When the IR environment is classical, the law of total probability (9) holds,
therefore
P (X|R)P (R) + P (X|R)P (R) − P (X|R)
A=
P (X|R) − P (X|R)
P (X|R)P (R) − P (X|R)P (R)
= = P (R),
P (X|R) − P (X|R)
that is why
0≤A≤1
in classical realm. In quantum setting, this inequality may be violated.
At first sight, a violation of Accardi inequality (8) seems to provide
wrong results since a “wrong” model of the world cannot provide anything
but “wrong” results. However, the experiments were performed showing
that the violation can enhance the search effectiveness [16]. It was observed
that the terms, whose measured probabilities violates (8), are those that
increase average precision more frequently and significantly than those do
not.
So, it is conjectured that a single-event-space model may be insufficient.
As already mentioned, a theoretical support for the notion of ‘varying event
space’ is needed.
216 R. Zapatrin
6. Non-Classical Models of IR
The notion similar to that of configuration space in Classical Mechanics or
event space in Probability Theory is compelling for a theory to be realistic.
The rough reason is that the events need a room to occur. Although from
both theoretical and pragmatic points of view, one can move from real-
ism to instrumentalism. Instead of telling that a system possesses this or
that property, we say that the measurement yields this or that result. This
result we may (or may not) interpret as property. But not results are avail-
able. When we qualify a system as classical and quantum, this is not about
its ‘real’ nature. They are about the observable properties. In particular,
probabilities need not inevitably be interpreted as a reflexion of statistics,
of relative frequencies. Rather, they can be viewed as propensities, there-
fore they need not be a number between 0 and 1, they may form a more
complicated structure. This idea is developed within the topos approach to
Quantum Mechanics.
Nowadays, a topos approach to a theory of Quantum Gravity is well-
developed [17], the idea on how topos theory can be used in general to
describe theories of physics includes, in particular, a theory of Quantum
Gravity. Physical theories are formulated in a topos other than Sets, this
topos may depend on both the theory-type and the system. The notion of
theory-type is crucial, this will be further expanded from the realm of “good
old” physics to that of IR.
If a theory-type, such as classical physics, or quantum physics, or some-
thing else is applicable to a certain class of systems, then, for each system
in this class, there is a topos in which the theory is to be formulated. For
some theory-types, the topos is system-independent: for instance, classical
physics always uses the topos of sets. For other theory-types, the topos may
vary from one system to another; this is, for instance, the case for quantum
mechanics.
The main particular goal is finding, in a topos, a representation of a
certain formal language, that is, associated with the system in question.
Non-classical models of IR. The experimental observations demonstrate

violations of seemingly necessary theoretical invariants. The suggestion is to
treat this as a potential way to improve the IR effectiveness. The idea to use
quantum model was put forward by van Rijsbergen [4]. In the meantime,
there is no evidence that other inequalities, perhaps even non-quantum,
can be formulated that their violation can reveal, on the one hand, the
inconsistency between experimental observations and a particularly chosen
www.ebook3000.com
underlying model, but, on the other hand, the directions toward a further
development of models for IR. Still, the question remained why the abstract
vector spaces would be a better framework than other mathematical theo-
ries. A possible answer was provided in Ref. [4]: Hilbert spaces encompass
different models for IR, such as the probabilistic model and the Vector Space
Model. This proved the use of non-classical logic to be effective. But non-
classical does not necessarily means quantum, and a broader conceptual
background is needed to build a uniform theory. The program of building
an operationalistic framework for empirically verified theories (we may call
them physical theories, but they should be understood in a broader sense)
was successfully carried out in the last decade, see Refs. [17,18] for reviews.
Topos theory is a vast, broadly developed area of mathematics, there
is no need to define it in detail in this chapter. A very good introduction
to the subject is Ref. [19], the reviews of applications of topos theory to
physics can be found in a general context [17] and specifically for quantum
gravity in Ref. [18].
A loose definition of a topos is a category with special properties, which
make a topos “look like” the category of sets, Sets, in the sense that any
mathematical operation that can be done in set theory has an analogue
in a general topos. What is essential is that the conceptual needs for both
theoretical physics and IR are in accordance:
• Spatialization (in whatever sense) is needed in order to make theory

realistic and experimentally verifiable.
• In order to be in a position to dasein, a kind of classical snapshot of a
system is to be taken.
• To create the arena for events, a Gel’fand-like spatialization is required.
• To obtain a reasonable Gel’fand spectrum, a commutative subalgebra,
associated with a particular context, should be specified.
This scheme is common for building a probabilistic-like model in any oper-

ationalist environment. To conclude, topos paradigm provides a formalism,
which, given a physical system (or, more generally, an operationalistic envi-
ronment) associates with it a bunch of virtual context-dependent configu-
ration (or sample) spaces.
7. Concluding Remarks
We see that the experimental results over large data collections demonstrate
features of quantum behavior. In order to mimic quantum indeterminacy,
the access to complete knowledge about the system was artificially
218 R. Zapatrin
restricted. But this phenomenon is generic for IR! The point is that search
engines store some limited data about the documents rather than the doc-
uments themselves. This is a natural restriction for the access to the docu-
ments to be complete, which, in turn, could be the reason for the observed
non-classicality.
The task for IR is to range search results. The standard probability-
based approach may be too straightforward and not effective sometimes.
A more subtle evaluations like propensities are invoked. These propensi-
ties must be somehow quantified. The topos approach is flexible enough to
provide various scales for propensities. Within it, “probabilities” may not
lie with the [0, 1] interval, furthermore, it admits models where “probabili-
ties” are not totally ordered. The existence of non-comparable properties is
natural for IR. But these are practical issues, which I do not inquire here.
To conclude, I would like to formulate the main message of this chap-
ter. There are different physical theories. To evaluate them, one needs to
know what each of these theories has to do with the reality. My claim is
that nowadays we witness the creation of realities, these realities become
to greater and greater extent real in comparison with the reality. As a con-
sequence, we are now in a position to say that certain theory describes a
reality. What happens with a physical theory if it turns out that its predic-
tions are disproved by experiment? The theory is to be rejected. The reason
to reject it is that it does not describe the reality. In this chapter, I show
that nowadays we witness how a new harbor for the abandoned theories is
being built. The factory producing these realities increases its output, and,
therefore, abandoned physical theories, which were proved to be falsified by
experiment, may find the new life.
Acknowledgment
The author appreciates Cris Calude, Karl Svozil and Jozef Tkadlec for stim-
ulating discussions on quantum contextuality during my stay in Technical
University of Vienna, supported by the TU Wien Ausseninstitut and the
Institute of Theoretical Physics of the Vienna University of Technology. I
am also grateful to Petros Wallden and other participants of the Work-
ing Group Meeting ‘Foundations of Quantum Mechanics and Relativistic
Spacetime’ for COST Action MP1006, 25–26 September 2012, University
of Athens, Greece.
www.ebook3000.com
References
1. D. Finkelstein and S. Finkelstein, Computational complementarity, Int. J.
Theor. Phys. 2, 753–779 (1983).
2. A.A. Grib and R.R. Zapatrin, Automata simulating quantum logics, Int. J.
Theor. Phys. 29, 113–124 (1990).
3. R. Zapatrin, Logic programming as quantum measurement, Int. J. Theor.
Phys. 34, 1813 (1995).
4. K. van Rijsbergen, The Geometry of Information Retrieval, Cambridge
University Press, Cambridge, UK, 2004.
5. S. Robertson, On event spaces and probabilistic models in information
retrieval, J. Inform. Retrieval 8, 319 (2005).
6. E. Moore. Gedankenexperiments on sequential machines, In Automata
Studies, eds. C.E. Shannon and J. McCarthy, Princeton Univ. Press,
Princeton, pp. 129–153 (1956).
7. K. Svozil and R.R. Zapatrin, Empirical logic of finite automata: Microstate-
ments versus macrostatements, Int. J. Theor. Phys. 35, 1541 (1996).
8. D. Aerts, Quantum structure in cognition, J. Math. Psychol. 53, 314 (2009).
9. Shuming Shi, Ji-rong Wen, Qing Yu, Ruihua Song, Wei-ying Ma. Gravitation-
based model for information retrieval, In: Proceedings of the 28th Annual
International ACM SIGIR conference on Research and Development in Infor-
mation Retrieval, 488–495, ACM, New York, NY, USA, 2005.
10. R.R.Z. Daniel Sonntag, Macrodynamics of Users’ Behavior in Information
Retrieval, 2009, arXiv:0905.2501 [cs.IR].
11. I. Raptis and R. Zapatrin, Algebraic description of spacetime foam, Class.
Quantum Grav. 18, 4187 (2001).
12. G.-C. Rota, On the Foundation of Combinatorial Theory, I. The Theory of
Möbius Functions, Zeitschrift für Wahrscheinlichkeitstheorie, 2, 340 (1968).
13. A. Shimony, Bell’s Theorem, in The Stanford Encyclopedia of Philosophy, ed.
E.N. Zalta, Winter 2013 edition, 2013.
14. L. Accardi and A. Fedullo, On the statistical meaning of complex numbers
in quantum mechanics, Lett. Nuovo Cimento. 34, 161–172 (1982).
15. M. Melucci, When Index Term Probability Violates the Classical Probability
Axioms Quantum Probability can be a Necessary Theory for Information
Retrieval, 2012, arXiv:1203.2569 [cs.IR].
16. M. Melucci, Advances in Multidisciplinary Retrieval, First Information
Retrieval Facility Conference, IRFC, Vienna, Austria, pp. 46–61, May, 2010.
17. C.I. Andreas Doering, ‘What is A Thing?’: Topos Theory in the Foundations
of Physics, In Springer Lecture Notes in Physics 813, ed. B. Coecke, Springer,
Heidelberg, pp. 753–940 (2011).
18. C. Flori, Lectures on Topos Quantum Theory, 2012, arXiv:1207.1744
[math-ph].
19. R. Goldblatt, Topoi: The Categorial Analysis of Logic, North-Holland,
London, 1984.
www.ebook3000.com
Chapter 8
Space–Time from Topos

Quantum Theory
Cecilia Flori
Blacket Laboratory, Imperial College London SW72BZ, UK
cf603@imperial.ac.uk
Waikato University, Hamilton NZ
cflori@waikato.ac.nz
One of the main challenges in theoretical physics in the past 50 years

has been to define a theory of quantum gravity, i.e. a theory which
consistently combines general relativity and quantum theory in order
to define a theory of space–time itself seen as a fluctuating field. As
such, a definition of space–time is of paramount importance, but it is
precisely the attainment of such a definition which is one of the main
stumbling blocks in quantum gravity. One of the striking features of
quantum gravity is that although both general relativity and quantum
theory treat space–time as a four-dimensional (4D) manifold equipped
with a metric, quantum gravity would suggest that, at the microscopic
scale, space–time is somewhat discrete. Therefore the continuum struc-
ture of space–time suggested by the two main ingredients of quantum
gravity seems to be thrown into discussion by quantum gravity itself.
This seems quite an odd predicament, but it might suggest that perhaps
a different mathematical structure other than a smooth manifold should
model space–time. These considerations seem to shed doubts on the use
of the continuum in general in a possible theory of quantum gravity.
An alternative would be to develop a mathematical formalism for quan-
tum gravity in which no fundamental role is played by the continuum
and where a new concept of space–time, not modeled on a differentiable
manifold, will emerge. This is precisely one of the aims of the topos the-
ory approach to quantum theory and quantum gravity put forward by
Isham, Butterfield, and Doering and subsequently developed by other
authors. The aim of this article is to precisely elucidate how such an
approach gives rise to a new definition of space–time which might be
more appropriate for quantum gravity.
221
222 C. Flori
1. Introduction
One of the main challenges in theoretical physics in the past 50 years has
been to define a theory of quantum gravity, i.e. a theory which consistently
combines general relativity and quantum theory in order to define a theory
of space–time itself seen as a fluctuating field. Therefore, a definition of
space–time is of paramount importance, but it is precisely the attainment
of such a definition which is one of the main stumbling blocks in quantum
gravity.
The reason for such a difficulty is the seemingly incompatible roles of
space–time put forward by general relativity and quantum theory. In fact,
on the one hand in general relativity, although the presence of both the 4D
metric and the connection is assumed ab initio, they are both considered to
be dynamical quantities and there is no preferred foliation of space–time.
On the other hand, quantum theory assumes a fixed (with respect to
its differentiable structure and metric) space–time, required to make sense
of measurements. This is a consequence of the fact that quantum theory,
in its standard formulation, implies an instrumentalist interpretation. By
this, we mean that statements regarding quantum objects can only be made
counter-factually, i.e. after measurement.
So for example, in classical physics, the statement “the particle x has
position y” makes sense, however in quantum theory, in order to have mean-
ing, this statement should be changed to “if a measurement is performed on
the position of the particle x, then it will have a certain probability to give
outcome y”. The difference in these statements reveals the discrepancy that
exists between any classical theory and quantum theory. Although the full
implications of this will be analyzed in detail later on, for now we want to
emphasize that the concept of measurement is essential for any statement
regarding quantum systems to make sense. This in turn implies that, by
necessity, a fixed space–time background has to be assumed in which the
measurement takes place.
However, such a notion of a fixed background structure seems hard
to accommodate in a theory of quantum gravity where the varying field
is space–time itself. In fact, by adopting quantum theory as it stands, a
possible theory of quantum gravity would have to make sense of statements
of the form “if a measurement is made of property x of the space–time
field, then the outcome y will have probability z”, where the notion of
measurement requires a fixed space–time background.
www.ebook3000.com
But what can be said about the “structure” of such a space–time inher-
ent in quantum theory? To answer this question, one has to go back and
analyze how exactly quantum theories come about. This is generally done
through the process of “quantization”, by which a classical theory is ‘trans-
formed’ into a quantum theory. Now in a classical theory, in general, the
configuration space of the system is mathematically represented by a differ-
entiable manifold M, while the phase space is represented by its cotangent
bundle T ∗ M. When quantizing a classical theory, this concept of a phase
space is inherited by the quantum theory. For example, if the classical con-
figuration is Q G/H for some Lie groups G and H, then a quantization
of such a system would define the quantum states to be sections of a vec-
tor bundle over Q whose fibers carry a representation of H. By definition,
the domain of these sections would be the continuum Q. So if quantization
is so defined, and this begs the question as to why this is the case, then
the space of values of quantum states is modeled by the continuum. This
mathematical description of space–time agrees with that given by general
relativity, which models space–time by a differentiable manifold M whose
elements are interpreted as space–time points and the gravitational field is
given by the curvature tensor of the pseudo-Remannian metric on M.
Therefore, although mathematically space–time is treated in an analo-
gous way in both quantum theory and general relativity, its role in these
two theories is very different. However, when defining a theory of quan-
tum gravity, the very definition of space–time as a differentiable manifold
is thrown/put into discussion. In fact, it is believed that at microscopic
scales, space–time ceases to be continuous but acquires a discrete nature.
Therefore, the continuum structure of space–time suggested by the two
main ingredients of quantum gravity seems to be refuted by quantum grav-
ity itself. This might seem an odd predicament, but it might also suggest
that the mathematical description of space–time required for quantum grav-
ity should be radically different from the continuum picture put forward by
the two ingredient theories.
A candidate for an alternative description of space–time is given by
the topos approach. This was first introduced by Butterfield, Isham, and
Doering and subsequently developed by other people. An introduction to
this subject will be given in Section 4, for now it suffices to say that in this
approach, the notion of a space–time point is replaced by the notion of a
space–time region. Such regions should be interpreted as defining regions
which are occupied by “extended” objects.
224 C. Flori
The interesting feature is that the collection of such “regions” carry a

Heyting algebra structure which is a generalized Boolean algebra where the
law of excluded middle does not hold. This mathematical description of
space–time in terms of what is technically called a locale fits well with the
discrete notion of space–time put forward by quantum gravity.
2. A Lesson from Quantum Gravity?

Although, as mentioned in the introduction, to date there is still no agreed
upon theory of quantum gravity, it is still possible and useful to analyze
the various proposals to see if anything can be learned from them. In par-
ticular, we would like to get a feeling for how a theory of quantum gravity
should address three main topics which we believe are directly related with
the ensuing concept of space–time. These topics are (1) the use of the
continuum, (2) the role of the Planck length and (3) the relation to the
instrumentalist interpretation of quantum theory.
Use of the continuum. Both in classical physics and in quantum theory,
the continuum appears in three main areas:
1. To model configuration space.

2. As values of physical quantities.
3. As modeling the space of probabilities.
We will now analyze why the continuum is used in these areas and how it
creates problems when considered in the context of quantum gravity [1–3].
1. To model configuration space. If we consider a particle moving in three-

dimensional (3D) space, we define its configuration to be R3 . This is
a consequence of the fact that physical space is modeled by a differen-
tiable manifold. This comes as no surprise since we are used to measuring
macroscopic objects with pointers and rulers. So it seems that the choice
of a differentiable manifold to represent a classical configuration space
is a consequence of modeling physical space by a manifold. But, as dis-
cussed briefly in the introduction, such a conception of configuration
space is carried over to the quantum regime via the process of quan-
tization. Thus, by necessity, the mathematical description of quantum
theory is determined by a priori assumptions on the nature of space–
time. The question is then if such a priori assumptions are still justified
in the context of quantum gravity where now space–time is seen as a
dynamical variable.
www.ebook3000.com
2. Values of physical quantities. In classical physics, quantities are real-

valued. This is a consequence of the fact that such values are defined
in terms of measurements carried out in the physical space, i.e. point-
ers and rulers which measure quantities living in the classical physical
space. Thus, by using the continuum to model space–time, we also adopt
the view that values of quantities should be real. Similarly in quantum
theory, eigenvalues are expected to be real numbers. If we now consider
quantum gravity, space–time is no longer a smooth manifold, thus it
seems harder to justify the fact that physical quantities are real-valued.
3. Probabilities. In both classical physics and quantum theory, probabilities
are defined using the relative frequency interpretation. Namely, to obtain
a probability of a certain outcome xi , one has to repeat the measurement
a large number of times N and divide this number by the number of
times the measurement gave the desired result Ni , thus obtaining the
probability of outcome xi to be N/Ni . By necessity, this has to be a
rational number which lies between 1 (always obtain xi ) and 0 (never
obtain xi ). If one takes the number of measurements to be infinite, then
N/Ni ∈ [0, 1]. In this way, quantum probabilities take their values in the
closed interval [0, 1]. However, to make sense of such an interpretation,
one has to assume a continuum back ground space–time in which the
measurements take place. However, how can this view be reconciled in
a theory of quantum gravity where there is no such fixed background
space–time in which to carry out measurements?
There have been other notions of probabilities put forward, namely as

propensities, measures of believes and possibilities. With respect to such
interpretations, it turns out that it is harder to justify the closed interval
[0, 1] as the space of probability values. In fact, all that is really required
from such a space is that it be a partially ordered set with top and bottom
element and equipped with a semi-additive structure so that probabilities
of disjoint events can be added. However, other than this, there is no reason
why [0, 1] should be chosen.
Planck scale. We know by now that a theory of quantum gravity aims

at reconciling quantum theory with general relativity. One of the striking
features of quantum gravity is that, although both general relativity and
quantum theory treat space–time as a 4D manifold equipped with a metric,
quantum gravity would suggest that, at the microscopic scale, space–time is
somewhat discrete. For example, approaches such as loop quantum gravity,
spin foams, dynamical triangulation and causal set, all suggest that at the
226 C. Flori
fundamental level space–time is discrete. In fact, it is believed that at the

Planck scale classical space–time concepts cease to apply and a new way of
viewing space–time is needed, which is not based on the notion of a con-
tinuum. In particular, the notion of point is not regarded as fundamental
any more but is replaced by the notion of a ‘region’ which physical objects
occupy.
Therefore, the continuum structure of space–time suggested by the two
main ingredients of quantum gravity seems to be thrown into discussion
by quantum gravity itself. This seems quite an odd predicament, but it
might suggest that perhaps a different mathematical structure other than
a smooth manifold should model space–time. Clearly, if such a new math-
ematical structure is defined, the questions to answer are (i) how it relates
to the conception of space–time of both general relativity and quantum
theory? and (ii) is the manifold structure emergent?
Relation to instrumentalist interpretation. Most proponents of a the-

ory of quantum gravity seem to adopt quantum theory without changing
any of the mathematical formalism. This implies that they also inherit the
conceptual interpretation of the theory which is an instrumentalist inter-
pretation. This interpretation heavily relies on the distinction of observed
system and observer and ascribes to the process of measurement an almost
ontological status. However, such an interpretation seems to be problematic
in the context of quantum gravity. In fact, if we consider quantum grav-
ity as a theory of the entire universe, which is a closed system, then the
observer and observed system distinction cannot be applied.
Moreover, the notion of measurement in quantum theory requires a
fixed space–time background in which the measurement takes place. But
quantum gravity is a theory of space–time itself seen as a dynamical quan-
tity, hence the problem arises as to how one could measure space–time
properties? Where would this measurement take place? Failing to clearly
define measurements of space–time properties jeopardizes the entire edifice
of quantum theory, since any prediction of the theory is defined in terms of
repeated sets of measurements. Hence, the instrumentalist interpretation of
quantum theory seems to be inconsistent with a theory of quantum grav-
ity. However, if we analyze the mathematics which induces such an inter-
pretation, we quickly arrive at the conclusion that it is a consequence of
the Hilbert space formalism which ‘comes equipped’ with the Born rule for
probabilities. This reflection would suggest that maybe a theory of quantum
gravity should make do without the Hilbert space formalism of quantum
theory.
www.ebook3000.com
3. Introduction to Topos Theory

Topos theory is a very vast and deep area of mathematics, thus it would
be impossible to describe it in detail in the present chapter. A very good
introduction to the subject is given in Ref. [4], while a more advanced
exposition can be found in Ref. [5].
In this section, we will limit ourselves to giving only a very brief outline
of the subject, delineating, in particular, a few aspects of the theory which
will be used later on in this chapter.
The very hand-wavy definition of a topos is that of a category with
extra properties. So, the first thing to explain is what a category is. Roughly
speaking, a category is a collection of objects and relations between these
objects. These relations are required to satisfy certain axioms which make
the set of all such relations ‘coherent’. Given a category, it is not necessarily
the case that every two objects have a relation between them, some do
and others do not. For the ones that do, the number of relations can vary
depending on which category we are considering.
Working with category theory means adopting an external description
of mathematical objects. Normally, in set theory, we are used to an internal
description of objects, i.e. objects are defined according to a set of commu-
nal properties which all of their elements share. This description however
is not adequate for category theory, where objects are defined in terms of
the relations they have with other objects (in the same category). Thus, in
order to work with category theory, one has to perform a change in per-
spective, from internal to external. As an example, consider the definition
of the concept “element”. The internal description of an element of a set is
straightforward, it is simply given in terms of ‘belonging’ (a ∈ S). On the
other hand, externally, an element of a set S is identified with a map from
the singletona {∗} to the set itself. In fact, any map from the singleton to a
set can only pick one element of the set since its domain is only a singleton.
Hence, to each element a, a unique map fa : {∗} → S such that fa ({∗}) = a
is associated.
The formal definition of a category can be found in Ref. [6]. For the
present purpose, it suffices to think of a category as a collection of objects
together with a collection of morphisms (maps, or what we have thus far
called relations) between these objects. So for example, the collection of
all topological spaces forms a category where the objects are topological
aA singleton is a set with only one element.

228 C. Flori
spaces and the maps are continuous maps between these. Another example
is given by the collection of groups and homomorphisms between them.
Now, a topos is a category, but a very special type of category in the
sense that it has a rich internal structure which in a way makes it behave
like Sets (this is the category whose objects are sets and whose maps are
functions between sets). In particular, any mathematical operation which
can be done in set theory can be done in a general topos. Therefore, a topos
is a category for which all the categorical versions of set constructs exist
and are well defined. This is why it is said that a topos ‘looks like’ Sets.
For example, we have the topos analogues of the set-theoretic notions of
Cartesian product S × T , disjoint union S Π T , and exponential S T , the set
of all functions from T to S. Also, each topos has a terminal object, denoted
by 1. This is an object with the property that given any other object A in
the topos, there exists only one map from A to 1. In Sets, the terminal
object is the singleton 1 = {∗}.
The notions of particular importance for the purpose of doing physics
in a topos are the following [4, 5]:
Heyting algebra. In Sets, the collection of all subsets of a given set forms
a Boolean algebra which represents the internal logic of Sets. Similarly,
given a topos, there exist an internal logic derived from the collection of
all subobjects of any object in the topos. Such a logic is called a Heyting
algebra, which is a distributive lattice for which the law of excluded middle
does not necessarily hold, i.e. S ∨ ¬S ≤ 1. This represents a generalization
of the Boolean algebra in Sets for which S ∨¬S = 1. An example of Heyting
algebra is given by the collection of all open sets in a topological space.
Subobject classifier. Each topos comes equipped with an object Ω called

the subobject classifier, which represents the generalization of the set
{0, 1} {false, true} of truth-values in Sets. As the name suggests, the
subobject classifier identifies subobjects. In the case of Sets, given a set A,
we say either A ⊆ X or A X. Thus, to the proposition “A is a subset
of X”, we can ascribe either the value true (1) or false (0), respectively.
Moreover, if A ⊆ X, one can ask which points x ∈ X lie also in A. This
can be expressed mathematically using the so-called characteristic function:
χA : X → {0, 1}, which is defined as follows:

0 if x ∈
/A
χA (x) = , (1)
1 if x ∈ A
www.ebook3000.com
where, {0, 1} = {false, true}. Thus, in sets, we only have true or false
as truth values, i.e. Ω = {0, 1}. This type of truth values determines the
internal logic of set to be Boolean, i.e. S ∨ ¬S = 1.
However, in a general topos, Ω will be a more general object (not nec-
essarily a set) leading to a multivalued logic. In this setting, we obtain a
well-defined mathematical notion of what it means for an object to nearly
be a subobject of a given object and how far it is from being a subob-
ject. Thus, the role of a subobject classifier Ω in a topos is to define how
subobjects fit in a given object.
The elements of this object Ω, similarly as was the case in Sets, repre-
sent the truth values. The collection of all such truth values forms a Heyting
algebra.
For example, let as assume that a person Bob is very thirsty but only a
little hungry. We then want to define the truth values of the propositions
“Bob is thirsty” and “Bob is hungry”. In classical logic, we would obtain
true for both propositions, so that we can only infer that Bob is both thirsty
and hungry, presumably to the same degree. On the other hand, in a multi-
valued logic, otherwise know as an intuitionistic logic, the statements “Bob
is very thirsty” is given a truth value which is ‘bigger’ than the truth value
given to “Bob is hungry”. In this way we can infer that Bob is more thirsty
than hungry. Moreover, we are also able to infer how much more thirsty
he is. In this sense, intuitionistic logic reflects more accurately language.
This brings us to our next topic which deals precisely with elucidating the
intimate connection between topos theory and language.
3.1. The language of a theory of physics

A language, in its most raw definition, consists of a collection of atomic vari-
ables, and a collection of primitive operations called logical connectives,
whose role is to combine together such primitive variables transforming
them into formulas or sentences. In order to reason with a given language,
one also requires rules of inference, i.e. rules which allow you to gener-
ate other valid sentences from the given ones. However, the semantics or
meaning of the logical connectives is not given by the logical connectives
themselves, but it is defined through a so-called evaluation map, which is a
map from the set of atomic variables and sentences to a set of truth values.
Such a map enables one to determine when a formula is true and, thus,
defines its semantics. Therefore, truth values determine the meaning of log-
ical connectives and, thus, the logic that a given language will exhibit will
230 C. Flori
depend on what the set of truth values is considered to be. This definition of
a language is clearly very abstract. In order to actually use a language, one
needs to find a mathematical setting in which to represent these abstract
terms so that both elementary and compound propositions will be repre-
sented by certain mathematical objects and the set of truth values will itself
be identified with an algebra.
For example, in standard classical logic, the mathematical context used
is Sets and the algebra of truth values is the Boolean algebra {0, 1}. In
this context, propositions are represented by sets while logical connectives
are represented by the set operations of intersection, disjoint union and
compliment [4, 5, 7].
So, if we start with an abstract language L and a mathematical universe
τ with an internal algebra (or logic) A, then a representation of L in τ is
a map π from the set of primitive propositions to elements in the algebra
A. Logical connectives are then represented by operations in A. However,
it turns out that for certain mathematical universes τ , the converse is also
true, i.e. the internal logic gives rise to a language. Since we are interested
in topos theory, we will explain the above statement when τ is a topos.
In this case, an internal language L(τ ) is defined by
1. Objects in τ are typesb in L(τ ).

2. Morphisms idA : A → A in τ represent variables of type A.
3. Morphisms A → B in τ are terms of type B with a free variable of
type A.
4. Terms of type Ω are propositions. These form a Heyting algebra.
5. The operations of the Heyting algebra of Ω represent logical connectives.
Given a topos τ and a language L, we then have
π Lang
L −−→ τ ; τ −−−→ L(τ ).
This correspondence is of uttermost importance when utilizing topos

theory to construct a theory of physics. In fact, in the topos framework,
the development of a theory of physics consists of three main steps [8]:
(1) Association of a local language, L(S) to each physical system S. The

precise definition will be given later on.
b Types are particular kinds of objects, such that the primitive propositions are charac-
terized by belonging to a certain type.
www.ebook3000.com
(2) The application of the theory-type (for example, classical physics or

quantum physics) to S. This implies choosing the appropriate topos τ
within whose framework the theory, as applied to S, is to be formulated
and interpreted.
(3) Since each topos has an internal language associated with it, a theory
of physics is equivalent to finding a translation/representation of L(S)
into the internal language of that topos.
Thus, the construction of a physical theory for a system S is defined through

an interplay between a language L(S), associated to the system S, a topos
τ and a translation of the language, L(S), of the system to the internal
language L(τ ) of the topos. The choice of both the topos and the repre-
sentation will depend on the theory-type being used, i.e. if it is classical or
quantum theory.
In order to make sense of the above definition, we need to explain what
L(S) is.
The Language L(S) of a System S. The minimum set of type symbols

and formulas, which are needed for a language to be useful as a language
to talk about a physical system S, are the following:
1. The state space object and the quantity value object. These are repre-
sented in L(S) by the ground type symbols Σ and R.
2. Given a physical quantity A, it is standard practice to represent such
a quantity in terms of a function from the state space to the quan-
tity value object. Thus, we require L(S) to contain the set of function
symbols FL(S) (Σ, R) = {Σ → R}, such that each physical quantity is
Ai : Σ → R.
3. We would like to have values of physical quantities. These are defined
in L(S) as terms of type R with a free variable s of type Σ, i.e. A(s),
where (A : Σ → R) ∈ FL(S) (Σ, R).
4. The collection of truth values is represented in L(S) as the ground type
symbol Ω called the truth object.
5. Generally, when talking about a system, we talk about values of its prop-
erties, thus we deal with propositions of the form “A ∈ ∆”, meaning “the
value of the quantity A lies in the interval ∆”. Such propositions are rep-
resented in the language L(S) as terms of typec P Σ, i.e. as subobjects
c Here, P Σ represents the power set of Σ, i.e. the collection of all its subobjects.
232 C. Flori
of the state space object. In particular, the proposition A ∈ ∆ is repre-

sented by the subobject Q = {s|A(s) ∈ ∆}.
6. States are identified with terms w of type P Σ. Given a proposition Q =
{s|A(s) ∈ ∆}, which is of type P Σ with a free variable ∆ of type P R, we
want to know whether the elements in w have the property A(s) ∈ ∆,
i.e. we want to know whether the proposition Q is true, given the state
w. To this end, we need to check the assertion,
w ⊆ Q. (2)
This is a term of type Ω.

7. Any axioms added to the language have to be represented by the arrow
true : 1 → Ω which gives the maximal truth value in Ω.
To better understand how L(S) gets represented in a topos through the
map π : L(S) → τ , let us consider the case in which S is a classical system
and τ is the topos Sets. In this case, we have
1. The state space and the quantity value object are defined by π(Σ) =
{s|s is state of the system} =: S, and π(R) = R, respectively.
2. Physical quantities are represented by functions from the state space to
the reals. Thus, each physical quantity, A, is represented by a function
fA : S → R.
3. Values of physical quantities are subsets of the reals.
4. The truth object is π(Ω) = {0, 1}, whose elements are the truth values
“true” and “false”.
5. Any proposition of the form “A ∈ ∆” (“the value of the quantity A
lies in the subset ∆ ⊆ R”) is represented by a subset of the state space
S, namely that subspace for which the proposition is true. This is just
fA−1 (∆) = {s ∈ S|fA (s) ∈ ∆}. The collection of all subsets of S forms a
Boolean algebra denoted Sub(S).
6. States ψ are identified with Boolean-algebra homomorphisms ψ:
Sub(S) → {0, 1} from the Boolean algebra Sub(S) to the two-element
set {0, 1}. Each state s either lies in fA−1 ({∆}) or it does not. Equiv-
alently, given a state s, every proposition about the values of physical
quantities in that state is either true or false.
7. Any axioms added to the language have to be represented by the arrow
true : {∗} → {0, 1}.
Given the above representation of the language L(S) in Sets, one has
all the necessary ingredients to construct a classical theory of S, keeping
in mind that if any axiom is needed, then it should be represented by
www.ebook3000.com
the morphism {∗} → {0, 1}. So, for example, consider Newton’s third
law: “When one body exerts a force on a second body, the second body
simultaneously exerts a force equal in magnitude and opposite in direc-
tion on the first body.” How would this look like in the framework we
just outlined? First of all, we need to consider the state space of the
composite system which we identify with S1 × S2 where S1 and S2 are
the state spaces of the individual systems. We then need to define the
observables associated with the force that system one exerts on system
two and vice versa. These are given by the maps F1 : S1 × S2 → R and
F2 : S1 × S2 → R, respectively. Clearly, the precise definition of these
functions will depend on the type of physical systems we are consider-
ing but for our purpose, it will suffice to leave them general. We now
have the ingredients to define the proposition representing Newton’s third
law: (F1 + F2 = 0) := {(s1 , s2 ) ∈ S1 × S2 |F1 (s1 , s2 ) + F2 (s1 , s2 ) = 0}.
Being a proposition, this is represented as a subobject of S, therefore it
is represented by a map p : {∗} → P (S). Now, let us try and evalu-
ate such a proposition for any given element (s1 , s2 ) ∈ S1 × S2 . Since
this proposition is an axiom it will actually be true for all states, there-
fore the subobject p({∗}) is S1 × S2 . We then obtain the characteristic
function χS1 ×S2 : S × S∈ → {0, 1} which assigns the value true for all
(s1 , s2 ) ∈ S1 × S2 . Putting all these results together, we obtain the desired
map ture = χS1 ×S2 ◦ p : {∗} → {0, 1} which represents the axiom.
So far, we have elucidated a possible way to construct a theory of
physics, namely, (i) first of all, construct an abstract language associated
to the system under investigation, so that one can actually talk about the
system, (ii) choose the appropriate mathematical universe in which to rep-
resent such a system. The choice will depend on whether the system is clas-
sical, quantum or quantum gravity; (iii) represent the abstract language of
the system in terms of the internal language of the topos chosen.
We have seen a simple example using classical theory. In the next sec-
tion, we will explain how quantum theory can be fit into the above proce-
dure. This is called the Isham–Doering schema [8–14].
4. Isham–Doering Schema
We are now interested in applying the procedure described in the previ-
ous section to define a physical theory for a quantum system S. The first
step is to choose which mathematical structure to use for representing the
language L(S). The choice was motivated by the necessity to overcome
234 C. Flori
certain conceptual problems inherent in quantum theory which we briefly

outlined in the introduction. In particular, one possible way to overcome
the problems related to the relative frequency interpretation of quantum
theory is to render quantum theory more realist.d By this, we mean choos-
ing an appropriate mathematical framework whose ensuing interpretation
has a more realist flavor. So, the question then is: what makes a theory a
realist theory? To answer this question, we need to analyze classical theory
and understand which are the mathematical building blocks which render
classical theory a realist theory and how exactly they are defined. In fact,
the way in which the state space, the quantity value object, physical quan-
tities, states, and propositions are mathematically represented in classical
physics implies a realist interpretation of the theory. For example, consider
how physical quantities are defined, they are identified with maps from the
state space to the reals, such that each physical quantity is identified with
the values it can have on a state. So, clearly this presupposes that physical
quantities exist and have real-values. Moreover, propositions are identified
with subsets of the state space, hence by definition, each proposition is
either true or false on every state and the collection of such propositions
satisfies a distributive Boolean logic. Similarly, states are represented as
maps from the Boolean algebra of propositions to the set {0, 1} such that
each state either belongs to a proposition or it doesn’t, i.e. given a state
s every proposition about the values of physical quantities in that state is
either true or false, hence the state is real.
This short discussion shows that the particular choice of mathematical
objects used to represent the types in the abstract language will influence
the interpretation of the ensuing theory. Therefore, the trick to render quan-
tum theory more realist is to choose a mathematical universe which allows
for a representation of the linguistic types by mathematical objects which
resemble those chosen for classical theory. Clearly, from what has been said
in the introduction, such a mathematical universe cannot be Sets with the
associated Hilbert space formalism since the ensuing theory is instrumen-
talist.
It is useful to recall at this point that a general topos looks like Sets,
hence most of the mathematical constructs for classical theory done within
d By a realist theory, we mean a theory which has the following characteristics: (i) prop-
erties belong to the system and can be represented in the theory; (ii) propositions about
the system form a Boolean logic; (iii) there is a space of states such that specifying
the state allows for unequivocally assigning truth values to all propositions. These truth
values are only true and false.
www.ebook3000.com
Sets have an analogue in a general topos. Moreover, the internal logic

of a topos is distributive. Therefore, using a topos seems very promising.
However, there are many toposes, and in order to motivate our choice, we
need to introduce the notion of a context. The notion of contextuality that
emerges, however, is not the canonical notion of contextuality in quan-
tum theory derived by Gleason’s theorem or the Kochen–Specker theorem,
even though it is related to the latter theorem. In particular, although the
Kochen–Specker theorem prohibits us to define values for all quantities at
the same time in a consistent way, it nonetheless allows for the possibility
of assigning values to commuting subsets of quantities. These commuting
subsets can be considered as classical snapshots, since all the peculiarities of
quantum theory arise from non-commuting operators. Thus, with respect to
these classical snapshots (contexts), quantum theory behaves like classical
theory.
The idea is then to define quantum theory locally with respect to these
classical snapshots. In this way, quantum theory can be seen as a collection
of local classical approximations.
Although it seems as if one were cheating by doing this, it turns out that
this is not the case. The reason being that the collection of all the above
mentioned classical snapshots actually forms a category, which means that
it is always possible to relate (compare) any two contexts. This categorical
structure carries the quantum information.
What is happening is the following:
(i) We first consider different contexts which represent classical snapshots.

(ii) We define our quantum theory locally in terms of such classical snap-
shots, therefore in a way performing a classical approximation.
(iii) We then retrieve the quantum information via the categorical structure
of the collection of all the classical contexts.
This prescription allows us to retain all the quantum information which is

lost at the local level.
The category of classical snapshots we will be utilizing is the category
V(H) of Abelian von Neuman subalgebras of the algebra B(H) of bounded
operators on the Hilbert space. This category is actually a partially ordered
set (poset) whose ordering is given by subset inclusion.
It is interesting to understand what this poset structure actually means

from a physics perspective. In particular, if we consider an algebra V such

that V ⊂ V , then the set of self-adjoint operators present in V , which

we denote Vsa , will be smaller than the set of self-adjoint operators in V ,
236 C. Flori

i.e. Vsa ⊂ Vsa . Since self-adjoint operators represent physical quantities, the

context V contains less physical information, so that by viewing the system

from the context V , we know less about it then when viewing it from the
context V . This idea represents a type of coarse graining which takes place
when going from a context with more information V to a context with less

information V .
The issue is now to find a topos which allows for all possible local “classi-
cal” approximations of the quantum system, while still retaining the quan-
tum information given by the categorical structure of V(H). A possible choice
would be the topos of presheaves over V(H). For a detailed definition of such
a topos the reader is referred to Ref. [5]. For now, it suffices to say that a
presheaf consists of a contravariant assignment of a set PV to each context
V ∈ V(H), such that if there exists a map V ⊆ V , then there is a corre-
sponding map at the presheaf level PV → PV . The collection of all such
op
assignments forms a topos which we denote SetsV(H) , and it represents
the mathematical universe within which we will describe quantum theory.
5. Concept of Space–Time in Topos Quantum Theory

Usually, space–time is modeled in terms of the continuum, i.e. locally space–
time is simply defined in terms of R4 . However, as discussed in the intro-
duction, quantum gravity seems to question the use of the continuum to
describe space–time. In particular, it would seem that space–time, at the
Planck length, should be discrete. If one thinks about it, the notion of a
space–time point is hardly justifiable. In fact, when talking about objects
as being in space–time, these tend to occupy regions of space–time rather
than points. This would suggest that perhaps a description of space–time in
terms of extended regions might be more appropriate [15,16]. Such extended
regions would have to satisfy a certain algebra whose operations would be
presumably given in terms of union, intersection, etc. Moreover, given the
uncertainty principle of quantum theory, it would be advisable to consider
open regions rather than closed ones in order to allow for a certain indeter-
minacy.
With these reflections in mind, it seems reasonable to try to utilize
some kind of locale (see next section for a definition) to describe space–
time. A locale is the same thing as a complete Heyting algebra (a Heyting
algebra which is also a complete lattice), and we have seen that the latter
is the internal algebra of a topos. Hence, when considering topos theory as
the mathematical framework of quantum theory, one facilitates a discrete
notion of space–time.
www.ebook3000.com
In this section, we will show how the topos definition of the quantity
value object can be seen as in internal locale.
5.1. Topos definition of the quantity value object

We will now introduce the representation of the quantity value object R in
op
SetsV(H) . In classical theory, the quantity value object is simply the real
numbers since each quantity takes on as its value, an element of the reals.
Similarly, in canonical quantum theory, we have the reals as the quantity
value object. However, in topos quantum theory, the quantity value object
is an object which has the same role as the reals have in standard quantum
theory, but its elements will not be numbers. Clearly, each element of the
quantity value object will be related to the real numbers in some way, but
it will not be a real number itself.
It should be noted that in any topos, there is an object which represents
the real numbers, in fact there are several of them [5, 7]. However, the
quantity value object we use for topos quantum theory is not one of them.
As we will see, the motivations for defining the quantity value object in
topos quantum theory come from physics requirements.
op
In the topos SetsV(H) , the representation of the quantity value object
R is given by the following presheaf [8, 9]:
Definition 5.1. The presheaf R↔ acts as follows:
(i) Objectse :
R↔ V := {(µ, ν)|µ, ν :↓ V → R,
µ is order-preserving, ν is order-reversing ; µ ≤ ν}.

(ii) Arrows: given two contexts V ⊆ V , the corresponding morphism is
R↔ V,V : R↔ V → R↔ V , (µ, ν) → (µ|V , ν|V ).
This presheaf is where physical quantities take their values, thus it has
the same role as the reals in classical physics.
The reason why the quantity value object is defined in terms of order-
reversing and order-preserving functions is because, in general, in quantum
theory, one can only give approximate values to the quantities.

eAmap µ :↓ V → R is said to be order-preserving if V ⊆ V implies that µ(V ) ≤ µ(V ).

A map ν :↓ V → R is order-reversing if V ⊆ V implies that ν(V ) ⊇ ν(V ).
238 C. Flori
Let us analyze the presheaf R↔ in more depth. To this end, we assume

that we want to define the value of a physical quantity A, given a state ψ.
If ψ is an eigenstate of A, then we would get a sharp value of the quantity
A, say a. If ψ is not an eigenstate, then we would get a certain range ∆ of
values for A.
Let us assume that ∆ = [a, b]. Then, what the presheaf R↔ does is to
single out the extreme points a and b, so as to give a range (unsharp) of
values for the physical quantity A. Obviously, since we are in the topos of
presheaves, we have to define each object contextually, i.e. for each context,
V ∈ V(H). It is precisely to accommodate this fact that the pair of order-
reversing and order-preserving functions was chosen to define the extreme
values of our intervals.
To understand this, we consider a context V , such that the self-adjoint
operator Â, which represents the physical quantity A, does belong to V
and such that the range of values of A at V is [a, b]. If we then consider

the context V ⊆ V , such that Â ∈ / V , we will have to approximate Â so as

to fit V . The precise way in which self-adjoint operators are approximated
is defined in Refs. [8, 9]. However, such an approximation will inevitably
coarse-grain Â, i.e. it will deform it.
It follows that the range of possible values of such an approximated oper-
ator, which we denote by δ Â, will be bigger. Therefore, the range of values

of δ Â at V will be [c, d] ⊇ [a, b], where c ≤ a and d ≥ b. These relations
between the extremal points can be achieved by the presheaf R↔ through
the order-reversing and order-preserving functions. Specifically, given that

a := µ(V ), b := ν(V ), V ⊆ V implies that c := µ(V ) ≤ µ(V ) (µ being

order-preserving) and d := ν(V ) ≥ ν(V ) (ν being order-reversing). More-
over, the fact that µ(V ) ≤ ν(V ) by definition, implies that as one goes to
smaller and smaller contexts V , the intervals (µ(V ), ν(V )) keep getting
bigger or stay the same.
An example of the quantity value object can be found in Ref. [14].
This object R↔ can be given the structure of a locale and hence, that
of a complete Heyting algebra. Before showing how this is done, it is worth
giving a rigorous definition of what a locale actually is and how locales can
de constructed.
Locales and their construction. As a first step, we introduce the notion

of a frame [17] which is a lattice L with all finite meets and all joins which
www.ebook3000.com
satisfies the following distributive law:

U∧ Vi = (U ∧ Vi ) ∀ U, Vi ∈ L. (3)
i i
A morphism between frames L1 → L2 is a map of partially ordered sets

which preserves both finite meets and infinite joins, hence we also call it
a frame homomorphism. An example of a frame is given by the collection
of all open subsets O(X) of a topological space X. In this setting, given a
continuous map between two topological spaces f : X1 → X2 , the inverse
image map f −1 : O(X2 ) → O(X1 ) is a morphism of frames.
Clearly, frames have a topological flavour, but sometimes, it is the alge-
braic aspect of a frame which one is more interested in. This algebraic aspect
is encoded in the notion of a locale. In particular, a locale is the same thing
as a frame, but such that to each morphism between frames there cor-
responds to a morphism between locales going the other way. What this
means in mathematical terms is that the category of locales is the opposite
of the category of frames: Loc = (Frames)op . Therefore, given two locales
X, Y with corresponding frames denoted as O(X) and O(Y ), respectively,
a continuous map of locales l : X → Y is defined to be a frame homomor-
phism l−1 : O(Y ) → O(X).
From the above, it is clear that there exists a covariant functor which
maps topological space to locales [5, 7]:
Loc : Spaces → Loc, (4)

X → Loc(X).
The corresponding frame of Loc(X) is O(Loc(X)) := O(X). Therefore,

given any topological space X, we can construct the locale Loc(X) whose
frame consists of all the open subsets of X and to each map f : X → Y
between spaces there corresponds to the locale map Loc(f ) : Loc(X) →
Loc(Y ) with associated map of frames f −1 : O(Y ) → O(X).
From the above description, one can infer that the elements of a locale
are “extended regions” rather than points. It is however possible to define
the notion of a point in a locale [5, 7].
Definition 5.2. A point in a locale X is defined as a morphism p : 1 → X

from the locale corresponding to a one-point space to the locale in question.
240 C. Flori
In terms of frames, a point is defined as a frame homomorphism p−1 :

O(X) → O(1) where O(1) is the frame consisting only of the bottom ele-
ment 0 and the top element 1, therefore O(1) {0, 1} = Ω.f
An open in the locale X is identified with a map 1 → O(X). This
can be used to define a topology on the set of points of X denoted as
P t(X) ⊆ ΩO(X) . In particular, a set in P t(X) is open if it is of the form,
P t(U ) = {p ∈ P t(X)|p−1 (U ) = 1}, (5)
where U ∈ O(X). It turns out that the operation of defining points in a
locale is a covariant functor
P t : Loc → Spaces,
X → P t(X). (6)
Given a map f : X → Y between locales, the associated map between the
corresponding spaces is
P t(X) → P t(Y ),
(p : 1 → X) → (f ◦ p : 1 → X → W ).
As shown in Ref. [5], the functor P t is right adjoint to the functor Loc.
Internal Locales. It is possible to apply the above definitions of frames
and locales to objects inside a topos. One thereby obtains the notions of
internal frame and internal locale (see Definition B.6). Given a topos τ ,
an internal frame is an internal complete Heyting algebra (B.7). A point
of an internal locale X is given by the frame map O(X) → Ω, where now
Ω is the subobject classifier of the topos τ . Opens are defined by maps
1 → X, where 1 is the terminal object in the topos τ . Similarly as for sets,
the collection of all points P t(X) is given the topology expressed by the
analogue of Eq. (5),
P t(U ) = {p ∈ P t(X)|p−1 ◦ (U ) = }, (7)
where U ∈ O(X) and is the ‘maximal’ element of Ω.
f Alternatively, a point P in a frame O(X) is a proper prime element of O(X), i.e. 1 = P
and U ∧ V ≤ P iff U ≤ P or V ≤ P . To understand how these definitions are equivalent,

let us analyze the kernel of the homomorphism p−1 : O(X) → O(1). This is given by the
subset KW:= {U |p−1 (U ) = 0} such that (i) 1 ∈ / K, (ii) U ∧ V ∈ K iff U ∈ K or V ∈ K,
and (iii) Ui`W ∈ K iff Uí ∈ K for all i. Given any such subset K ∈ O(X), this defines an
element P = U ∈K U ∈ O(X). Translating conditions (i)–(iii) above to conditions on
P , we obtain (a) 1 = P and (b) U ∧ V ∈ P iff U ≤ P or V ≤ P (c) U ≤ P iff U ∈ K,
therefore K =↓ P .
www.ebook3000.com
Now that we have the notion of an internal locale, we would like to

show that R↔ is an internal locale in a specific topos τ . The topos we will
be interested is the topos of sheaves over a topological space X which is
denoted by Sh(X).
A sheaf can be thought of as a fiber bundle in which the fibers may vary
from point to point.
Formally, a sheaf is a presheaf F with values in the category of sets that
satisfies the following two axioms [5]:
(i) Given an open set U with open covering Ui , if s, t ∈ F (U ) are such
that s|Ui = t|Ui for all i, then s = t.
(ii) Given an open set U with open covering Ui , and si ∈ F (Ui ) for all i,
such that for each pair Ui and Uj , si |Ui ∩Uj = sj |Ui ∩Uj , then there exists
s ∈ F (U ) such that s|Ui = si for each i. s is called the gluing while the
si are called compatible.
Axioms (i) and (ii) state that compatible sections can be uniquely glued
together.
An important result for us is that presheaves defined over a poset P are
equivalent to sheaves over P equipped with the Alexandroff topology. This
is a topology with a basis given by the collection of all lower sets in the
poset P , i.e. by sets of the form ↓ p := {p ∈ P |p ≤ p}, p ∈ P .g
The dual of such a topology is the topology of upper sets, i.e. the topol-
ogy generated by the sets ↑ p := {p ∈ P |p ≥ p}. Given such a topology, it
is a standard result that, for any poset P ,
SetsP Sh(P + ), (8)
where P + denotes the poset P equipped with the upper Alexandroff topol-
ogy, which are the duals of lower sets. It follows that
op
SetsP Sh((P op )+ ) Sh(P − ), (9)
where P − denotes the set of all lower sets in P . In particular, for the poset
V(H), we have
op
SetsV(H) Sh(V(H)− ). (10)
Thus, every presheaf in our theory is in fact a sheaf with respect to the
topology V(H)− . We will denote both the sheaves over V(H)− and the
g Note that a function α : P1 → P2 between posets P1 and P2 is continuous with respect

to the Alexandroff topologies if and only if it is order-preserving.
242 C. Flori
presheaves over V(H) by A. Moreover, in order to simplify the notation, we

will write Sh(V(H)− ) as just Sh(V(H)).
Having said this, we would now like to interpret the quantity value
object R↔ as a locale in Sh(V(H)).h In order to do this, we will utilize the
fact that, for any topological space X, the category Loc(Sh(X)) of internal
locales in Sh(X) is equivalent to the slice category Loc/X [7]. The slice
category Loc/X has as objects locale maps f : Y → X where Y is a locale
in Sets and given two objects W → X and Y → X, a morphism between
them consists of a commuting triangle of the following form.
Y @ /W
@@ }}
@@ }}
@@ }}
}~ }
X
Here, the map Y → W is a morphisms of locales. The way of defining

the equivalence functor Loc/X → Loc(Sh(X)) is as follows. Consider
an object f : Y → X ∈ Loc/X. This induces a geometric morphismi
f : Sh(Y ) → Sh(X) such that the direct image functor f∗ preserves Heyting
algebras and complete internal posets. Consider now the subobject classi-
fierj ΩY ∈ Sh(Y ). This is defined for each open set U ∈ O(Y ) as ΩY (U ) :=
{V ∈ O(Y )|V ≤ U }. If we apply the functor f∗ to ΩY , then because of
the above-mentioned properties of f∗ , the sheaf f∗ (ΩY ) is a complete Heyt-
ing algebra (internal frame) in Sh(X). Given any open set W ∈ O(X),
then f∗ (ΩY )(W ) = ΩY (f −1 (W )) = {V ∈ O(Y )|V ≤ f −1 (W )}. Therefore,
starting from the object f : Y → X, we have defined the internal locale
f∗ (ΩY ) ∈ Sh(X) whose associated frame we denote by O(Y ).
h In what follows, we will introduce sheaves over locales. These are defined in exactly the
same way as sheaves over a topological space. In fact, from the above discussion, it tran-
spires that the notion of sheaf on a topological space only refers to the open subspaces,
rather than the points and locales are entirely defined in terms of open subspaces with
points, if any, being derived notions.
i A geometric morphism [5], [17] φ : τ → τ between topoi τ and τ is defined to be
1 2 1 2
a pair of functors φ∗ : τ1 → τ2 and φ∗ : τ2 → τ1 , called respectively the inverse image
and the direct image part of the geometric morphism, such that
1. φ∗ φ∗ i.e. φ∗ is the left adjoint of φ∗ ,
2. φ∗ is left exact, i.e. it preserves all finite limits.
j It is a standard result that in every topos, the subobject classifier is a Heyting algebra.
www.ebook3000.com
5.2. Quantity value object as a locale

We would like to apply the above-mentioned procedure to the sheaf R↔ so
as to construct the associated locale in Sh(V(H)). As a first step, we will
need to define a continuous map relating R↔ to V(H) equipped with the
Alexandroff topology. This can be done in various ways depending on what
topology R↔ is considered to have.
Locale associated to R↔ part I. As shown in Ref. [20], it is possible to

view R↔ as a subobject of a locale in Sh(V(H)). In order to understand how
this is indeed possible, we first of all have to consider the product V(H)×IR,
where the set IR is the interval domain consisting of all compact intervals
[a, b] with a, b ∈ R and a ≤ b (see Appendix 6). Normally, the interval
domain comes equipped with the Scott topology (22), however, we will not
consider such a topology when looking at the product V(H) × IR since,
instead of the product topology, we will consider a different topology. In
particular, any algebra V ∈ V(H) and (µ, ν) ∈ R↔ (V ) define a basic open
U [V, (µ, ν)] with (V , [a, b]) ∈ U [V, (µ, ν)] if V ⊆ V and µ(V ) < a ≤
b < ν(V ). It is straightforward to check that such opens form a basis for a
topology on V(H)×IR. In terms of such a topology, the projection map π1 :
V(H) × IR → V(H) is continuous and hence a map between locales. Since
π1 is an element of Loc/V(H), we can now apply the previously defined
procedure to obtain an internal locale Loc(V(H) × IR) whose associated
frame is O(V(H) × IR). The sheaf O(V(H) × IR) is given by, for all V ∈
V(H),
O(V(H) × IR)(↓ V ) = O(V(H) × IR)|↓V ×IR .
The associated presheaf R↔ is a subpresheaf of O(V(H) × IR) given by
φV : R↔ (V ) → O(V(H) × IR)(V )
(µ, ν) → U [V, (µ, ν)].
We would now like to utilize this map to show that R↔ is a sublocale

of Loc(V(H) × IR). However, as a first step, we need to define the locale
structure on R↔ . The first step is to put an ordering on R↔ (V ) for each
V ∈ V(H). We define (µ, ν) ≤ (µ , ν ) iff for all V ⊆ V , µ(V ) ≥ µ (V )
and ν (V ) ≥ ν(V ). The operations of join and meet are then defined as
244 C. Flori
follows: for all V ⊆ V , we have
((µ, ν) ∧ (µ , ν ))(V ) = ((µ ∧ µ )(V ), (ν ∧ ν )(V ))

:= (max{µ(V ), µ (V )}, min{ν(V ), ν (V )})
and
((µ, ν) ∨ (µ , ν ))(V ) = ((µ ∨ µ )(V ), (ν ∨ ν )(V ))

:= (min{µ(V ), µ (V )}, max{ν(V ), ν (V )}),
where the right-hand side of both equations is defined with respect to the
total ordering on R. It is straightforward to then show that (3) holds.
Theorem 5.1. R↔ is a sublocale of Loc(V(H) × IR).
Proof. In this proof, we will utilize the fact that a sublocale is given by
the nucleus of the underlying frame (see Definitions B.4 and B.3). In this
case, therefore, we will show that for each V ∈ V(H), φV defined above
satisfies conditions (B.1). We will first show that φV preserves meets,
i.e. φV ((µ, ν) ∧ (µ , ν )) = φV (µ, ν) ∧ φV (µ , ν ). Considering φV (µ, ν) ∧
φV (µ , ν ) = U [V, (µ, ν)] ∧ U [V, (µ , ν )] from the definition of the opens
U [V, (µ, ν)], it follows that U [V, (µ, ν)] ∧ U [V, (µ, ν)] := U [V, (µ ∧ µ , ν ∧ ν )]
where µ ∧ µ is defined for each context V ⊆ V as (µ ∧ µ )(V ) :=
max{µ(V ), µ (V )} and similarly (ν ∧ν )(V ) := min{ν(V ), ν (V )}. Apply-
ing the definitions, we then get
φV ((µ, ν) ∧ (µ , ν )) = φV ((µ ∧ µ ), (ν ∧ ν )) = U [V, (µ ∧ µ , ν ∧ ν )].
It now remains to show that φV satisfies (B.2) and (B.3). This, however,
is equivalent to the requirement that the image of R↔ under φ is closed
under taking the pseudo-complement, i.e. for all (µ, ν) ∈ R↔ (V ) and for
all W ∈ O(V(H) × IR)(V ), (W → U [V, (µ, ν)]) ∈ φV (R↔ (V )). Here, W →

U [V, (µ, ν)] = {U |U ∧ W ≤ U [V, (µ, ν)]}. Since the opens U [V, (µ, ν)] are
a basis of the topology on V(H) × IR, the result follows.
The reason that the above topology was chosen for V(H) × IR was
because the author in Ref. [20] wanted to analyze the connection between
the topos approach to quantum theory put forward by Isham and Doering
and described above with an alternative formulation by Heunen, Landsman,
and Spitters in Ref. [21]. This latter formulation has as a starting point a
C ∗ -algebra A and an ambient topos SetsC(A) where C(A) is the category of
abelian subalgebras of A. They then promote A to an internal C ∗ -algebra
www.ebook3000.com
A in SetsC(A) and define the spectrum Σ as an internal locale. In this

context, the quantity value object is identified with the internal locale IR.
Through the above construction, it transpires that the quantity value
object R↔ is intimately related to IR.
Locale associated to R↔ part II. We now would like to analyze an
alternative topology for the quantity value object R↔ .

Given such a presheaf, we define the set R := V ∈V(H) R↔ V with
associated map pR : R → V(H) such that pR (µ, ν) = V for (µ, ν) ∈ R↔ V .
We would like to define a topology on R such that the map pR be continuous.
A possibility would be to define the discrete topology on each fiber
↔
p−1
R (V ) = R V which would accommodate for the fact that pR : R → V(H)
is an étale bundle. We could then define the disjoint union topology, but
this would not account for the ‘horizontal’ topology on the base category
V(H) given by the Alexandroff topolofy.
Another possibility would be to consider as a basis for the topology on R
the collection of all open subobjects. Thus, a basis set would be of the form

S = V ∈V(H) S V such that S V is open in R↔ V which is equipped with
the discrete topology. In such a setting, the ‘horizontal’ topology would be
accounted for by the presheaf maps.
Since each R↔ V is equipped with the discrete topology, the topology
on the entire set R would essentially be the discrete topology in which all
subobjects of R are open.
Obviously, with respect to such a topology, the bundle map pR would

be continuous since for each ↓ V , p−1 (↓ V ) = V ∈↓V R↔ V will represent

the open subobject whose value is R↔ V for all V ∈↓ V and ∅ everywhere
else.
We now have at our disposal the continuous map pR : R → V(H) which
is a locale map. This gives rise to the geometric morphism pR : Sh(R) →
Sh(V(H)). We can then apply the procedure defined in the previous section
to construct the internal locale O(R) = pR∗ (ΩR ) such that, for any V ∈
V(H), we obtain
 

R
(pR∗ (ΩR ))V = ΩR (p−1
R (V )) = Ω R↔ V ,
V ∈↓V
where
   
 
ΩR  R↔ V  = S ∈ O(R)|S ⊆ R↔ V .



V ∈↓V V ∈↓V
246 C. Flori
We recall that

S := SV , (11)

V ∈↓V
where S V is an open subset of R↔ V . Since the latter has a discrete topol-

ogy, the basis will be formed of singletons. These are defined, for each
V ⊆ V as a pair (µ, ν) :↓ V → R of order preserving and order revers-
ing functions taking values in R which describe varying intervals of real
numbers. What this means conceptually is the following: for each context,
V ∈ V(H) which represents a classical snapshot of the quantum system,
we obtain a “locale” space–time seen as a collection of unions of varying
intervals of real numbers. Such a space–time has the property that when
considering two contexts V and V such that V ⊆ V , then the intervals
of real numbers describing the space–time associated to the context V are
“bigger”, i.e. are less precise and thus have less information, than the inter-
vals of the space–time associated to the bigger context. Since these intervals
are to be interpreted as the regions of space–time which physical objects
occupy, what the above result signifies is that when going to a smaller
context V which contains less information due to coarse-graining, then
the precision with which one is able to determine the position of physical
objects decreases.
Locale associated to R̆. An alternative way of associating to the quantity

value object of a locale is by first considering the definition of the quantity
value object put forward in Ref. [22]. This new definition was a result of
introducing the notion of a group and a group action in the topos quantum
theory framework. To achieve this, the topos utilized to define quantum the-
ory had to be slightly changed so that now the base category, although still
remaining V(H), was considered to be invariant under any group trans-
formation, i.e. the group acts trivially. This slightly different category is
denoted by Vf (H) where f stands for fixed. All the group actions where
relegated to an intermediate category, resulting in a construction of sheaves
over Vf (H) as sheaves over this intermediate category and then “pushed
down” to sheaves on Vf (H) in the appropriate way. We will not go into
the details of how this is done, the interested reader is referred to Ref. [22].
All that we will do in the present chapter is to state how the resulting new
quantity value object is defined.
www.ebook3000.com
↔
Definition 5.3. The quantity value object R̆ is a presheaf of order-
preserving and order-reversing functions on Vf (H) defined as follows:
• On objects V ∈ Vf (H), we have
↔
R̆V := R↔ φg (V ) , (12)
φg ∈Hom(↓V,V(H))
where eachk
R↔ φg (V ) := {(µ, ν)|µ ∈ OP (↓ φg (V ), R) , µ ∈ OR(↓ φg (V ), R), µ ≤ ν}.

(13)
Here, the maps φg : Vf (H) → V(H) defined by φg (V ) = Ûg V Ûg−1 for g ∈
G represent faithful representations of a given group G. The downward

set ↓ φg (V ) comprises all the subalgebras V ⊆ φg (V ). The condition

µ ≤ ν implies that for all V ∈↓ φg (V ), µ(V ) ≤ ν(V ).

• On morphisms iV V : V → V (V ⊆ V ), we get:
↔ ↔ ↔
R̆ (iV V ) : R̆V → R̆V (14)

R↔ φg (V ) → R ↔
φg (V ) , (15)
φg ∈Hom(↓V,V(H)) φg ∈Hom(↓V ,V(H))
where for each element (µ, ν) ∈ R↔ φg (V ) , we obtain

↔
R̆ (iV V )(µ, ν) := R↔ (iφg (V ),φg (V ) )(µ, ν), (16)
= (µ|φg (V ) , ν|φg (V ) ), (17)

where µ|φg (V ) denotes the restriction of µ to ↓ φg (V ) ⊆↓ φg (V ), and
analogously for ν|φg (V ) .
↔
We are now interested in defining a topology for R̆ . This was done in
Ref. [22]. As a first step, we define the set
↔ ↔
R= R̆V = {V } × R̆V , (18)
V ∈Vf (H) V ∈Vf (H)
↔ ↔
where each R̆V := φg ∈Hom(↓V,V(H)) R φg (V ) .
The above represents a bundle over Vf (H) with bundle map pR : R →

Vf (H) defined by pR (µ, ν) = V , where V is the context such that (µ, ν) ∈
↔
R↔ φg (V ) . In this setting, p−1
R (V ) = R̆V are the fibers of the map pR .
k Here, OP stands for order-preserving while OR stands for order-reversing.

248 C. Flori
We would like to define a topology on R with the minimal requirement

that the map pR is continuous. We know that the category Vf (H) has the
Alexandroff topology whose basis open sets are of the form ↓ V for some
V ∈ Vf (H). Thus, we are looking for a topology such that the pullback

p−1
R (↓ V ) := V ∈↓V R̆V is open in R.

Following the discussion at the end of Ref. [22, Section 2.1], we know
that each R↔ is equipped with the discrete topology in which all subobjects
are open (in particular, each R↔ V has the discrete topology).
↔
Therefore, we define a subsheaf Q̆ of R̆ to be open if for each V ∈
Vf (H), the set Q̆V ⊆ R̆V is open, i.e. each Qφ (V ) ⊆ R↔ φg (V ) is open in
g
↔
the discrete topology on R↔ φg (V ) . It follows that the sheaf R̆ gets induced,
the discrete topology in which all subobjects are open. In this setting, the
‘horizontal’ topology on the base category Vf (H) would be accounted for
by the sheaf maps.
For each ↓ V , we then obtain the open set p−1 R (↓ V ) which has value

R̆V at contexts V ∈↓ V and ∅ everywhere else.
Given the continuous map pR , this can be seen as an element in
Loc/Vf (H), allowing us to construct the corresponding internal locale in
Sh(Vf (H)). In particular, we consider the induced geometric morphism
pR : Sh(R) → Sh(Vf (H)). The internal locale we are looking for is then
given by O(R) = pR∗ (ΩR ). Therefore, for any open ↓ V ∈ Vf (H), we obtain
 

R
O(R)(↓ V ) = ΩR (p−1 R (↓ V )) = Ω R̆V ,
V ∈↓V
where
   
 
ΩR  R̆V  = U ∈ O(R)|U ⊆ R̆V = O(R)|↓V ,

 
V ∈↓V V ∈↓V
where

U := U φg (V ) . (19)
V ∈↓V φg ∈Hom(↓V ,V(H))
In this situation, for each context V , the “locale” space–time is a col-

lection of unions of equivalence classes of varying intervals of real numbers
where such equivalence is defined with respect to a group G. A tentative
interpretation is that these space–time regions represent diffeomorphic
regions of space–time. Since such a locale is a sheaf, once could interpret a
global section as a particular choice of space–time. Each such global section
www.ebook3000.com
(choice of space–time) would then be related to each other by space–time

diffeomorphisms.
Summary. In this section, we have defined three different locales represent-
ing the quantity value object. The first two of these locales are very similar
and only differ in the type of topology used to construct such a locale. The
third locale, on the other hand, carries also information regarding group
transformations and as such it can be seen as a covariant description of R.
5.3. Modeling space–time as a locale

We have seen in the previous section that the quantity value object of topos
quantum theory can be interpreted as an internal locale R and hence as a
complete Heyting algebra.
If we then modeled space–time in terms of the locale R, the concept of
a point would be secondary while the concept of a “region” of space would
be primary. Moreover, since a locale is equivalent to a complete Heyting
algebra, the collection of space–time regions would undergo an intuitionistic
rather than a classical (Boolean) logic. If one equates extended objects with
the space–time regions they occupy, then, at the Planck scale, modeling
space–time in terms of a locale would imply that statements of the form
“an object ψ occupies region A or it does not” would be neither true nor
false. This is another way of stating the fact that in standard quantum
theory, given a vector ψ ∈ H and a subspace W of H, ψ can have non-zero
components in both W and W ⊥ . So, it would seem that modeling space–
time, at least at the quantum level, as a locale, is in agreement with known
facts about quantum theory. Motivated by this, let us try and make the
locale definition of space–time more rigorous.
Since we would like to somehow retrieve the classical concept of space–
time in the appropriate limit, we take quantum space–time to be con-
structed in a similar way as in classical physics, i.e. as the fourth power
of the quantity value object. In order to achieve this, we need to introduce
the notion of tensor product of locales. We will do this in terms of frames.
Definition 5.4 [17]. Given two frames A and B, the tensor product A⊗ B
is defined to be the frame represented by the following presentation:
T a ⊗ b, a ∈ A and b ∈ B|

(ai ⊗ bi ) = ai ⊗ bi , (20)
i i i
250 C. Flori

(ai ⊗ b) = ai ⊗ b, (21)
i i

(a ⊗ bi ) = a ⊗ bi . (22)
i i
In other words, we form the formal products, a ⊗ b, of elements a ∈ A,

b ∈ B and subject them to the relations in Eqs. (20)–(22). We note that
there are injective maps,
i : A → A ⊗ B,
a → a ⊗ true (23)
and
j : B → A ⊗ B,
b → true ⊗ b. (24)
Alternatively, following Ref. [17, Proposition 6.4.2], it is possible to
define the tensor product of frames in a more categorical way as coproductsl
of frames.
Lemma 5.4.1. Given three frames A, B, C and frame homomorphisms f :

A → C and g : B → C, there exists a unique frame homomorphisms
h : A ⊗ B → C such that f = h ◦ i and g = h ◦ j.
An immediate corollary of the above is
Corollary 5.5. P t(A ⊗ B) P t(A) × P t(B)
Proof. The points P t(A ⊗ B) correspond to frame maps p−1 : A ⊗ B → Ω.

By the property of coproducts, such maps are uniquely defined as the prod-
uct maps [p−1 −1 −1 −1
A , pB ] for the frame maps pA : A → Ω and pB : B → Ω.

Equipped with this definition, we can now define space–time to be
the locale R ⊗ R ⊗ R ⊗ R =: R4 constructed by iterations of Definition
5.4. Because of Lemma 5.4.1, the object R4 is defined as a coproduct in
Sh(V(H)).
In this setting, we then interpret space–time as the internal local R4
in Sh(V(H)). Clearly according to which of the above defined locales, we
l Coproducts are the categorical generalization of disjoint unions in Sets.
www.ebook3000.com
consider R to be, we will obtain a slightly different interpretation of space–

time, however all agree on the fact that the basic notions are now given
by extended regions rather than points. It is not clear at this point which
local would be more suitable to represent space–time. We leave the answer
to this question as a topic for a subsequent paper.
Summarizing, so far we have shown how the quantity value object
defined in topos quantum theory can be seen as an internal locale in the
appropriate topos. Such an internal local could be utilized to describe
space–time at the level of quantum theory, the advantage being that its
elements represent extended regions of space–time rather than space–time
points. In our opinion, such a description of space–time might be more suit-
able in the light of quantum gravity since all theories of quantum gravity so
far developed seem to presuppose a discreteness of space–time as opposed
to the continuum notion of space–time put forward by classical physics.
The implications of adopting a localic description of space–time and the
detailed analysis of the physics it might ensue is beyond the scope of the
present chapter. In the present instance, we are only interested in describing
possible alternative mathematical descriptions of space–time which do not
rely on the notion of the continuum.
A natural question to ask is if the notion of a point can be defined
within these locales. In particular, can the notion of a sheaf of points of the
locales be defined? Making use of Corollary 5.5, we know that the presheaf
P t(R4 ) ×4 P t(R), therefore it suffices to analyze the single presheaf
P t(R). We will now analyze such an object for each of the locales defined
in the previous section.
Can we retrieve the notion of space–time points?

It is natural to ask the question of how the points of the locales used
to model space–time are represented assuming they exist. It turns out
that it is possible to retrieve the notion of a point for all three of the
locales defined in the previous section. The “collection” of such points
will itself be a sheaf in Sh(X). In order to define the points of internal
locales, we will make use of the functor IP t : Loc(Sh(X)) → T op(Sh(X))
where Loc(Sh(X)) is the category of internal locales in Sh(X) while
T op(Sh(X)) is the category of internal topological spaces in Sh(X). This
functor is the internal analogue of the functor defined in (5.5). IP t is the
right adjoint of ILoc : T op(Sh(X)) → Loc(Sh(X)) which is the internal
analogue of (4).
252 C. Flori
Internal topological spaces are defined as follows:
Definition 5.6 [18]. A topological space object in a topos τ consists of a

pair (A, TA ) where TA ⊆ P A such that the following conditions are satisfied:
(1) ∅ ∈ TA and A ∈ TA .
(2) For all B, B ∈ P A, if B ∈ TA and B ∈ TA then B ∩ B ∈ TA

(3) For all S ∈ P (P A), if S ⊆ TA then S ∈ TA ,

where : P (P A) → P A is the exponential adjoint of the characteristic
morphism of the subobject of P (P A)×A consisting of those elements (S, a)
such that there exists a B ∈ P A for which a ∈ B and B ∈ S. Similarly

: P (P A) → P A is the exponential adjoint of the characteristic morphism
of the subobject of P (P A) × A consisting of those elements (S, a) such that
for all B ∈ P A if B ∈ S, then a ∈ S.
In the particular case in which the topos τ is Sh(X) for some topological
space X, the above definition reduced to the following:
Definition 5.7 [19]. Given a sheaf A in Sh(X) for some topological space
p
X with topology TX and its associated etalé bundle A −
→ X, then a topolog-
ical structure on A consists of a second topology on A, TA which is courser
than the etalé topology but which still makes the map p continuous.
In the present situation, the topological space we are considering is

V(H) equipped with the Alexandrov topology. In this case, Sh(V(H))
op
SetsV(H) , therefore also an internal topological space will be a presheaf
Y . This implies that for each V ∈ V(H), Y V will be a topological space in
Sets with the appropriate topology. The corresponding etalé space would

then be Y = V ∈V(H Y V and the finer topology TY will be identified
with the appropriate disjoint union topology. Having said that the defi-
nition of ILoc : T op(Sh(X)) → Loc(Sh(X)) is straightforward: given an
internal topological space Y , then Loc(Y ) is a presheaf such that for each
V ∈ V(H), ILoc(Y )V := {U |U is open in Y V } and it is the locale associ-
ated to the topological space Y V in Sets. So, effectively for each V ∈ V(H),
we have ILoc(Y )V = Loc(Y V ).
In the same way, IP t : Loc(Sh(X)) → T op(Sh(X)) is such that, given
a local L then IP t(L)V := P t(LV ). The adjointness relation Loc P t
induces the adjunction ILoc IP t.
We have seen in the previous sections that, for any topological space X,
we have the following equivalence of categories Loc(Sh(X)) Loc/X,
www.ebook3000.com
hence another way of defining a point of an internal local L in Sh(X)

is as a continuous cross-section π : L → X.
In the following, we will consider each local defined in the previous
section and for each of them construct its points. For notational simplicity,
we will denote IP t simply as P t while keeping in mind that we are now,
considering internal locales and topological spaces.
Point of the locale Loc(V(H) × IR). As briefly discussed in Ref. [21],
the points of the locale Loc(V(H) × IR) are given by the presheaf
P t(Loc(V(H) × IR)) such that for each V ∈ V(H), we obtain the set given
by a collection of pairs of subobjects of Q as follows:
P t(Loc(V(H) × IR))(V ) = {(L, U )|↓V | ∀ V ⊆ V, (25)

(L(V ), U (V )) is an element of the locale IR},
where both L and U are subobjects of the constant presheaf Q which assigns
to each V ∈ V(H) the rationals Q. Therefore, for each V ⊆ V , the pair
(L(V ), U (V )) satisfies equation A.1.
However, since we are interested in the sublocale R↔ , we will only
be considering a subpresheaf of P t(Loc(V(H) × IR)). In order to under-
stand how the points in R↔ are defined, let us go back to the locale
Loc(V(H) × IR) with associated frame O(V(H) × IR), where
O(V(H) × IR)(V ) = O(↓ V × IR). Each of these sets is isomorphic to the
set of order-preserving functions OP (↓ V, O(IR)). In particular, given a
topological space X, there is a standard bijection O(X) C(X, S) where
S is the Sierpinski space {0, 1} whose only non-trivial open set is {1}.
This bijection is defined as f : U → χU and f −1 : g → g −1 ({1}). Apply-
ing this result to the case at hand, we obtain that O(↓ V × IR) C(↓
V × IR, S). However, as shown in Ref. [21], because of lambda-abstraction
C(↓ V × IR, S) C(↓ V, S IR ). Moreover, C(IR, S) O(IR), therefore
O(↓ V × IR) C(↓ V, O(IR)). Continuity in this setting is given by mono-
tonicity, therefore we replace C(↓ V, O(IR)) by OP (↓ V, O(IR)) obtaining
the following isomorphisms:
O(↓ V × IR) → C(↓ V × IR, S) → C(↓ V, S IR ) → OP (↓ V, O(IR))

,
U → χU → hχU → ghχU
where

1 iff (V , [a, b]) ∈ U
χU (V , [a, b]) = hχU : V → χU (V , −)
0 otherwise,
254 C. Flori
and

ghχU : V → (χU (V , −))−1 ({1}) = {[a, b]|χU (V , [a, b]) = 1} (26)

= {[a, b]|(V , [a, b]) ∈ U } .
We will now utilize the isomorphism O(↓ V × IR) OP (↓, O(IR)) to

allows us to relate points of Loc(V(H) × IR) defined in 25 to order-reversing
maps ↓ V → O(IR). In particular, to each element of IR there corre-
sponds an open U ∈ O(IR) via the equivalence between p : 1 → IR and
p−1 : O(IR) → Ω and, to each such open U , there corresponds a map
ghχU . Now, since we are in Sets, the element (L(V ), U (V )) defines a com-
pact interval [sup(L(V )), inf(U (V ))] so that P t(IR) can be identified with
the classical Scott interval domain (see Appendix B). Therefore, to each
pair (L(V ), U (V )), there corresponds an open W ∈ O(IR) such that the
interval [sup(L(V )), inf (U (V ))] ∈ W . However, since each such pair is
defined for all ↓ V , then each (L, U )|↓V defines a map ghχU :↓ V → O(IR)
which in turn identifies an open U ∈ O(V(H) × IR). In this way, the set
P t(Loc(V(H) × IR))(V ) can be written as
P t(Loc(V(H) × IR))(V ) = {(L, U)|↓V | ∀ V ⊆ V,

(V , [sup(L(V )), inf(U (V )]) ∈ O(↓ V × IR)}
{ghχU :↓ V → O(IR)}.
However, we are only interested in elements of the sublocale R↔ . Recall-

ing that the sublocale map takes elements (µ, ν) and maps them to base
opens U [V, (µ, ν)], then
P t(R↔ )(V ) = {(L, U)|↓V |∃ (µ, ν) ∈ R↔ V s.t. ∀ V ⊆ V,

µ(V ) < sup(L(V )) ≤ inf(U (V )) < ν(V )}.
Recall that the compact intervals [a, b] seen as elements of IR are ordered
by reverse inclusion. Therefore, the condition µ(V ) < sup(L(V )) ≤
inf(U (V )) < ν(V ) implies that the interval [µ(V ), ν(V )] is the biggest,
i.e. the one with least amount of information.
In this setting, P t(R↔ ) can be seen as a presehaf such that for each
V ∈ V(H), the set P t(R↔ )(V ) is a collection of assignments to each V ⊆
V of an interval domain [sup(L(V )), inf(U (V ))] associated to a global
element (µ, ν) ∈ Γ(R↔ ). This interval domain represents a refinement of
the information contained in the interval [µ(V ), ν(V )].
www.ebook3000.com
Points of the locale R. We are now interested in analyzing the points of

the locale R with associated frame O(R). In particular, the correspondence
between Loc(Sh(X)) and Loc/X for some topological space X implies
that the points of the locale R correspond to continuous cross-sections of
π : R → V(H). These are the locale maps,
φ : V(H) → R,
V → (V, φ(V )),
where φ(V ) ∈ R↔ (V ). At the level of frames, we then have
φ−1 : O(R) → O(V(H)),

UV,(µ,ν) → {V ∈ V(H)|φ(V ) = (µ, ν)|V },
where UV,(µ,ν) = {(V , (µ, ν)|V |V ⊆ V } is a basic open set for the topology
on R. Since φ is continuous, the set φ−1 (UV,(µ,ν) ) is open in the Alexandroff
topology of V(H). This implies that if φ(V ) = (µ, ν) and V ⊆ V , then
φ(V ) = (µ, ν)|V . Hence, a point of the locale R corresponds to a global
section of R↔ , i.e. P t(R) ΓR↔ .
In Ref. [23, Proposition 4.2] it was shown that ΓR↔ OP (V(H), IR).
As a first step, the authors notes that each order-preserving map f :↓ V →
IR can be decomposed into two maps f+ , f− :↓ V → R which pick out the
end points of the interval, i.e. f (V ) := [f− (V ), f+ (V )]. This implies that
f− ≤ f+ and f− is order-preserving while f+ is order-reversing.
On the other hand, each pair (µ, ν) ∈ R↔ gives rise to an order-
preserving map f :↓ V → IR such that f (V ) = [µ(V ), ν(V )] for all
V ⊆ V . This correspondence implies that we can now characterize the
presheaf R↔ in terms of the interval domain as follows:
Theorem 2. The quantity value object R↔ acts on
• Objects: for all V ∈ V(H),
R↔ (V ) = {f :↓ V → IR|f is order-preserving}.
• Morphisms: for all iV V : V ⊆ V ,
R↔ (iV V ) : R↔ (V ) → R↔ (V ); f → f |↓V .
Given this new characterization, one can show that ΓR↔ OP (V(H), IR)
as follows. Consider a global element γ : 1 → R↔ . This, for each V ∈ V(H),
assigns out an element pV := γV ({∗}) :↓ V → IR. Since the global element
256 C. Flori
is a natural transformation, naturality implies that pV = R↔ (iV V )(pV ) =

pV |V . Therefore, each global element γ gives rise to a map
ργ : V(H) → IR,
V → pV |↓V ,

where V ⊆ V . This is well defined since for any other V such that V ⊆ V ,
we have pV |V = pV = pV |V . The fact that it is also order-preserving
follows from naturality.
On the other hand, given an order-preserving map ρ : V(H) → IR, we
can define a global element by setting pV := ρ|↓V for all V ∈ V(H).
This result implies that
P t(R) OP (V(H), IR).
Therefore, the points in the locale utilized to eventually model space–time

are order-preserving functions from the context category to the interval
domain. These describe varying intervals of real numbers.
Points of the locale R. We will now define the points of the last locale
which represents a possible candidate for modeling space–time. Clearly,
such a locale is intimately connected to the locale R and as a consequence
also, the characterization of its points will be very similar. In fact, the locale

R can be written out as R = V ∈V(H) φ:↓V →V(H) R↔ (φ(V )). Therefore,
for each V ∈ V(H), we obtain

P t(R)(V ) P t(R)(φ(V )) OR(↓ φ(V ), IR).
φ:↓V →V(H) φ:↓V →V(H)
What this implies is that the presheaf P t(R) assigns to each V ∈ V(H) a
collection of intervals [µ(V ), ν(V )] all related by a group transformation.
Conceptually, this might be interpreted as stating that the points of the
locale R represent equivalence classes of intervals under some group trans-
formation. Therefore, if we consider space–time to be modeled by such a
locale, then space–time points become equivalence classes of regions under
a symmetry transformation. So, for a point p ∈ P t(R(V )), we will write

p = φ:↓V →V(H) pφ :↓ φ(V ) → IR.
6. Conclusions
Most theories of quantum gravity seem to suggest that, at the fundamental
level, space–time has a discrete structure. As discussed in the introduction,
www.ebook3000.com
this view of space–time seems to contradict the way in which it is described

both in quantum theory and in general relativity. In fact, in both these
two theories, space–time is seen as a continuum modeled on the real num-
bers. This discrepancy of the description of space–time between quantum
theory and general relativity on the one hand and quantum gravity on the
other seems rather surprising since quantum gravity is supposed to combine
general relativity with quantum theory in a coherent framework. This odd
feature motivated some researchers to question the fact that space–time
should be modeled by a continuum and suggested that also at the level
of quantum theory and general relativity, a new mathematical model for
space–time should be created. In this chapter, we proposed an alternative
definition of space–time put forward by topos quantum theory. Such a def-
inition consists of modelling space–time in terms of a locale where now the
fundamental space–time building blocks are regions, not points. This idea
reflects the fact that space–time points are not physically meaningful since
real objects occupy space–time regions.
A locale is equivalent to a complete Heyting algebra. Therefore, model-
ing space–time in terms of a locale can be interpreted, roughly, as modeling
space–time via an algebra of open regions of space–time where the algebraic
operations are interpreted as defining unions and intersections of space–time
regions. The reason open regions are preferred is to account for quantum
indeterminacies given by the generalized uncertainty principle. Therefore,
modeling space–time via a locale seems to resonate more accurately with
our common-sense interpretation of space–time.
In this chapter, we propose various candidate locales which are all ulti-
mately related to one another. Each of these locales could be adopted for
modeling space–time, however, a discussion on which one would be the best
suited is left as a topic for a future paper. To derive the above-mentioned
locales, we started with the quantity value object of topos quantum the-
ory which plays the same role as the Reals in classical physics. This object
being a sheaf, we applied the standard technique to ‘transform’ sheaves
into a locale internal to the appropriate topos. When considering the locale
associated to the quantity value object R↔ , we discovered that, for each
context, the elements of such a locale were identified with collection of
unions of paris of order-preserving and order-reversing functions from the
context category to the interval domain. These describe unions of varying
intervals of real numbers. What this means conceptually is the following:
for each context V ∈ V(H) which represents a classical snapshot of the
quantum system, we obtain a “locale” space–time seen as a collection of
258 C. Flori
unions of varying intervals of real numbers. Such a space–time has the prop-
erty that when considering two contexts V and V such that V ⊆ V , then
the intervals of real numbers describing the space–time associated to the
context V are “bigger”, i.e. are less precise and thus have less information,
than the intervals of the space–time associated to the bigger context. Since
these intervals are to be interpreted as the regions of space–time which
physical objects occupy, what the above result signifies is that when going
to a smaller contexts V which contains less information due to coarse-
graining, then the precision with which one is able to determine the posi-
tion of physical objects decreases. Moreover, when considering the locale
associated to the ‘covariant’ quantity value object R̆, we discovered that
for each context V the “locale” space–time was considered to be a collec-
tion of unions of equivalence classes of varying intervals of real numbers
where such equivalence was defined with respect to a group G. A tenta-
tive interpretation is that these space–time regions represent diffeomorphic
regions of space–time. Since such a locale is a sheaf, once could interpret a
global section as a particular choice of space–time. Each such global section
(choice of space–time) would then be related to each other by space–time
diffeomorphisms. Clearly, this is a very speculative idea and a much more
thorough analysis is needed to determine whether such an interpretation is
physically reasonable. As previously stated, in this chapter, we only wanted
to elucidate possible candidates for an interpretation of space–time which
makes no fundamental use of the continuum. This is a first step towards
defining a mathematical model of space–time. In fact, although the dis-
creteness of space–time in quantum theory is almost universally accepted,
no mathematical model of discrete space–time has been constructed so far.
We hope that this article will provide ideas for such possible models by
offering various candidates. Which, if any, of these candidates would be the
most appropriate choice for representing space–time is a question that still
remains to be answered and will be the topic of a subsequent paper.
Acknowledgments
This work was supported by EPSRC Grant No. EP/J008060/1.
Appendix A. Elements of domain theory

We will now introduce the concept of the intervals domain IR. As a set,
IR consists of all compact intervals [a, b] where a, b ∈ R and a ≤ b. These
www.ebook3000.com
intervals also contain the singletons [a, a] = {a}. The ordering on IR is

given by reverse inclusion. The topology on IR is given by Scott opens.
Definition A.1. Given a complete lattice L, a subset U ⊆ L is Scott-open

if (i) it is an upper set and (ii) give any subset S ⊆ L such that S ∈ U ,

then there is some finite set D ⊆ S such that D ∈ U .
In the case at hand, given any open interval (p, q) in R then
(p, q)S := {[a, b]|p < a and b < q}
is an element of the basis of the Scott-topology of IR.

Interval domains can also be defined as special Dedekind cuts. These are
used to define real numbers in terms of rational numbers. In particular,
given a real number x, this uniquely determines and is uniquely determined
by a pair of sets (Lx , Ux ) where
Lx = {y ∈ Q|y ≤ x} is a lower set of x, (A.1)

Ux = {y ∈ Q|y ≥ x} is an upper set of x.
Such a pair is called a Dedekind cut. The rigorous definition is as follows:
Definition A.2. A pair (L, U ) of sets of rational numbers is a Dedekind

cut if it satisfies the following conditions:
1. Both L and U are inhabited.

2. L is downward-closed while U is upward-closed.
3. L is upward-open: if a ∈ L, a ≤ b for some b ∈ L. Similarly, U is
downward-open: if b ∈ U , then a ≤ b for some a ∈ U .
4. If a ∈ L and b ∈ U , then a ≤ b.
5. If a ≤ b are rational numbers, then a ∈ L or b ∈ U .
An interval domain is a pair (L, U ) satisfying only conditions 1–4.
Appendix B. Frame and locales

Definition B.3. Given a frame L, a nucleus is a function f : L → L such
that the following conditions are satisfied:
f (a ∧ b) = f (a) ∧ f (b), (B.1)

a ≤ f (a), (B.2)
f (f (a)) ≤ f (a). (B.3)
260 C. Flori
Alternatively, it is possible to define a nucleus in terms of a subset S of L

which satisfied the following conditions:

(1) A ∈ S whenever A ⊆ S.
(2) a ⇒ b ∈ S whenever b ∈ S.

The map f is then defined by f (a) := {b ∈ L | b ∈ S, a ≤ b} and
S = {a ∈ L | f (a) = a}.
Definition B.4. A sublocale is given by a nucleus on the underlying frame.
Alternatively, one can define a sublocale S of a locale L as a subset

satisfying the following conditions:
1. S is closed under all meets.
2. For all s ∈ S and all l ∈ L, then pseudo compliment l → s belongs to S.
Definition B.5. Given a topos τ , an internal lattice L is an object in τ

together with two arrows : L × L → L and : L × L → L such that the
following diagrams commute:
• Associativity
V
id×
L×L×L / L×L
V V
×id

L×L V /L

and a similar diagram for .
• Commutativity
V
L×L /L
{{=
{{
{{{
{{
{{ V
δ
{{
{{
{{
L×L

and a similar diagram for . Here, δ : L × L → L × L is defined by
δ(a, b) = (b, a).
www.ebook3000.com
• Idempotent
∆
/ L×L
LC
CC
CC
CC
CC V
C
id CC
CC
CC
C!
L

and a similar diagram for . Here, ∆ : L → L × L is defined by ∆(x) =
(x, x).
• Absorption
∧
LO o L ×O L
pr1 id×∨
∆×id id
L×L / L×L×L / L×L×L
pr1 ∧×id

Lo ∨
L×L
Given the terminal object 1 ∈ τ , the top and bottom elements are defined
by the maps : 1 → L and ⊥ : 1 → L such that the compositions
V
id×
L L × 1 −−−→ L × L −→ L
W
id×⊥
L L × 1 −−−→ L × L −→ L
both give the identity.
Definition B.6. An internal frame F is an internal lattice with all finite

meets and all joins and for which the diagrammatic equivalent of the
following equation is satisfied:

U∧ Vi = (U ∧ Vi ).
i i
Given the definition of an internal lattice, we can now define an internal

Heyting algebra as follows:
Definition B.7. Given a topos τ , an internal Heyting algebra is an internal

lattice L ∈ τ together with a binary operation ⇒: L × L → L such that the
following conditions are satisfied:
262 C. Flori
• Identifying an element of the lattice x with the arrow x : 1 → L, then

: 1 → L is equivalent to the composition
x ∆ ⇒
1−
→ L −→ L × L −→ L.
• The following diagrams commute:
LO o V L ×O L
V
id×⇒
∆×id
L×L / L×L×L
id×δ

pr1 L×L×L
id×⇒

Lo ∆
L×L
V
id× ⇒
L×L×L / L×L /L
O
∆×id×id
V
L×L×L×L
id×δ×id
⇒×⇒
L×L×L×L / L×L
An internal complete Heyting algebra is an internal Heyting algebra which

is a complete internal lattice, as such it is equivalent to an internal frame.
References
1. J. Butterfield and C.J. Isham, Spacetime and the philosophical challenge of
quantum gravity, (1999), [arXiv:gr-qc/9903072].
2. C.J. Isham, Some reflections on the status of conventional quantum theory
when applied to quantum gravity, (2002), [arXiv:quant-ph/0206090].
3. Andreas Doering, Some remarks on the logic of quantum gravity, (2013),
[arXiv:1306.3076 [gr-qc]].
4. R. Goldblatt, Topoi The Categorial Analysis of Logic, North-Holland, Lon-
don, 1984.
5. S. MacLane and I. Moerdijk, Sheaves in Geometry and Logic: A First Intro-
duction to Topos Theory, Springer-Verlag, London 1968.
www.ebook3000.com
6. S. MacLane, Categories for the Working Mathematician, Springer-Verlag,

London, 1997.
7. P. T. Johnstone, Sketches of an Elephant A Topos Theory Compendium I,
II, Oxford Science Publications, 2002.
8. A. Doering and C. Isham. ‘What is a Thing?’: Topos Theory in the Foun-
dations of Physics, in New Structures for Physics, ed. B. Coecke, Springer,
pp. 753–940, Heidelberg, Lect. Notes Phys. 813, 2011.
9. C. Flori, First Course in Topos Quantum Theory, Lect. Notes Phys. 868,
Springer, 2013.
10. A. Doering and C.J. Isham, A topos foundation for theories of physics:
I, formal languages for physics, (2007), [arXiv:quant-ph/0703060].
II, daseinisation and the liberation of quantum theory, (2007), [arXiv:quant-
ph/0703062].
III, the representation of physical quantities with arrows, (2007),
[arXiv:quant-ph/0703064].
IV, categories of systems, (2007), [arXiv:quant-ph/0703066v1].
14. A. Doering, Topos theory and ‘Neo-Realist’ quantum theory’, (2007),
[arXiv:0712. 4003v1 [quant-ph]].
15. C.J. Isham and J. Butterfield, Some possible roles for topos theory in quan-
tum theory and quantum gravity, (1999), [arXiv:quant-ph/9910005].
16. L. Crane, What is the mathematical structure of quantum spacetime? (2007),
[arXiv:0706.4452 [gr-qc]].
17. S. Vickers, Topology Via Logic, Cambridge University Press, 1989.
18. L. Neff Stout, Topological space objects in a topos II: -completeness and
-cocompleteness (manuscripta mathematica 1975, Volume 17, Issue 1,
pp. 1–14).
19. L. Neff Stout, A Topological Structure on the Structure Sheaf, of a Topolog-
ical Ring. Commun. Algebra 5(7): 695–706, 1977.
20. S. Wolters, A Comparison of Two Topos-Theoretic Approaches to Quantum
Theory, (2010), [arXiv:1010.2031 [math-ph]].
21. C. Heunen, N.P. Landsman, B. Spitters, A topos for algebraic quantum the-
ory. Comm. Math. Phys. 291(1):63–110, 2009.
22. C. Flori, Group Action in Topos Quantum Physics, (2011), [arXiv:1110.1650
[quant-ph]].
23. A. Doering, R.S. Barbosa, Unsharp Values, Domains and Topoi, in Quantum
Field Theory and Gravity: Conceptual and Mathematical Advances in the
Search for a Unified Framework, F. Finster et al. (eds.), Birkhaeuser, Basel,
2011, pp. 65–96.
www.ebook3000.com
Chapter 9
From Born Reciprocity to Reciprocal

Relativity: A Paradigm for
Space–Time Physics
Peter Jarvis
School of Physical Sciences, University of Tasmania,
Private Bag 37, Hobart Tas 7001, Australia
Peter.Jarvis@utas.edu.au
Born’s principle of reciprocity — the exchangeability of relativistic

energy–momentum and time-position — can be seen as a discrete ele-
ment of a continuous group of symmetry transformations which tran-
scend relativity. Invariance under the semi-direct product of the Weyl–
Heisenberg group H(4) of canonical commutation relations with the
non-compact unitary group U (3, 1) — the so-called quaplectic group
U (3, 1) H(4) — has been considered by Low as an extension of Born
reciprocity to a fundamental symmetry principle of ‘reciprocal relativity’
for the physics of non-inertial frames and high energy processes.
We review the construction of worldline models for elementary sys-
tems in reciprocal relativity, whereby unitary representations of the
quaplectic group arise as a result of constraint quantization. It is pointed
out that in arbitrary even space–time dimension D, this can always be
achieved by the addition of the canonical one-form to the standard scalar
particle action. This modifies the symmetry group carried by physical
states, from an inhomogeneous orthogonal group (with conformal signa-
ture) in dimension D, to the quaplectic group in dimension 12 D, and
elevates the Heisenberg commutation relations (including the central
extension, given by Planck’s constant) to full symmetry generators of
the system.
Secondly, the Schrödinger–Robertson inequality for the covariance
matrix of position and momentum operator products is interpreted in
terms of relativistic phase space geometry. It is shown that the inequality
is quaplectic invariant, and thus that states which saturate the inequality
are stratified under the action of the reciprocal-relativistic quaplectic
group. The example of the ‘scalar’ quaplectic system, namely the 4D
265
266 P. Jarvis
relativistic oscillator, and the associated minimal uncertainty multimode

squeezed states, are treated in detail. In this case, it is shown that the
relevant parameters for the discussion of semi-classical systems (in formal
contraction limits of the quaplectic group) are the expectation values of
certain rank two tensors.
1. Introduction and Main Results

Almost 80 years ago, Max Born published an article in Nature [1] presenting
what he saw as a fundamental insight into quantum theory and relativity
(see also Ref. [2]). Born took as a guiding principle, the dual roles played by
momentum energy and space–time coordinates in many physical contexts —
in particular, the possibility of point canonical transformations in momen-
tum space, which would entail curvature and equations of motion analogous
to Einstein’s equations of general relativity. The duality idea was evidently
reinforced for Born by the dual roles played by position and momentum
in his own formulation of Heisenberg’s famous exchange relation between
position and momentum operators. Born’s postulate of reciprocity (named
for the terminology of condensed matter physics), the invariance under the
discrete symmetry transformation x → p, p → −x, of course permeates not
only Heisenberg’s relation but also Hamiltonian mechanics, and is at the
fore in the presentation of ordinary wave mechanics in phase space.
Born’s conviction was expressed forcefully to Einstein in their corre-
spondence [3] (Einstein politely but firmly rebuffed the ideas). Born and
collaborators worked intensively for several yearsa on realizations of this
scheme, which was surveyed in a later review article with Green [4], but
they were ultimately unsuccessful in their attempt to use it to predict the
mass spectra of new ‘mesons’ being discovered in cosmic ray emulsion data,
and at the time increasingly emerging at newly established experimental
accelerator facilities.
One of the constructs emerging in this body of work was a “metric equa-
tion” — intended to generate a hierarchy of relativistic wave equations for
new particles — based on the sum x · x + p · p of the quadratic invariants
corresponding to the standard interval in x-space, and its equivalent in p-
space. In the ‘clocks and rods’ language of Einstein’s original presentation
of relativity, a plausible interpretation of the significance of such a combina-
tion might be as follows. Measurements of fundamental space–time events
a A more complete list of references to Born’s work, including some citations of recent
related literature on reciprocity, is given in the concluding remarks.
www.ebook3000.com
From Born Reciprocity to Reciprocal Relativity 267
must always be gained through physical interactions — for example, via

registration of signals (photons or other particles) emitted from a source,
by scattering in a suitable detector. An implication which could be drawn
from this is that the specification of events should be enriched, by tagging
them not only with space–time coordinates, or coordinate differences, ∆xµ ,
but also with energy–momentum transfer, ∆pµ , reflecting the way in which
the space–time coordinates are measured through interaction processes, and
that the relativity principle should accommodate this by an extended invari-
ant interval between events, incorporating the energy–momentum shifts, in
addition to the standard space–time coordinate differences.
With modern particle theory at an impasse, it is opportune to
re-examine reciprocity as a paradigm for space–time physics.
In particular, we consider ‘reciprocal relativity’, an elaboration of Born
reciprocity, developed in recent years by Stephen Low [5–8]. In this for-
mulation, the extended invariant interval is written formally in terms of
differentials of coordinate time, spatial displacement, together with energy
and linear momentum transfer, in physical units, as

c2 1 1 de2
d2 = dτ 2 + 2 dm2 = dt2 − 2 dx2 + 2 − dp2
,
b c b c2
where b is a new universal constant of force which plays the role of maximum
rate of change of momentumb between interacting systems, analogously to
the role of c, the constant of velocity (the speed of light). The discrete
reciprocity transformation symmetry is thus enlarged to a continuous group
of transformations on eight-dimensional phase space, subject however to the
preservation of the Heisenberg algebra H(4)
[X µ , Pν ] = iδ µ ν , µ, ν = 0, 1, 2, 3.
As a result, the fundamental symmetry group is postulated to include the

intersection of the automorphism group of linear transformations on rel-
ativistic phase space which preserve the Heisenberg algebra, namely the
eight-dimensional real symplectic group, with the eight-dimensional orthog-
onal group preserving d2 above. Together with the Heisenberg algebra
itself, this is the semi-direct product of the latterc with the non-compact
unitary group U (3, 1): the so-calledd quaplectic group U (3, 1) H(4).
b Born’s constant b referred to ‘maximum momentum’; c.f. λp .

c The Heisenberg Lie algebra of course exponentiates to the Weyl–Heisenberg group. As
our considerations concern local operators rather than global transformation properties,
the distinction does not require elaboration.
268 P. Jarvis
One can set the notion of a new fundamental constant of nature in the
context of physics at unification or higher energy scales via naı̈ve dimen-
sional analysis. For example, units for time, position, momentum, energy
and acceleration are defined by
√
λt = /bc, λx = c/b, λp = b/c, λe = bc, λa = c bc/.
In the spirit of Born’s original idea, Planck’s constant becomes factorized,

here = λx λp . Further, if it be assumed that reciprocity has its
origins at
the Planck scale, then (for example) the Planck energy mP c2 = c5 /GN
should be equated with the corresponding quaplectic unit λe up to pro-
portionality, so that Newton’s gravitational constant becomes a derived
quantity given by
GN = αG c4 /b.
Provided the proportionality constant αG is of order unity, there is on

dimensional grounds at least, the hint that the physics of reciprocity, may
indeed be entrained with the description of non-inertial frames, and ulti-
mately general relativity and the equivalence principle.
In Sec. 2, we take up the analysis of reciprocal relativity as an extended
symmetry principle for fundamental systems in space–time. Just as there
has been much attention paid over the years to groups containing the basic
Poincaré group, such as the space–time conformal group, in providing orga-
nizing and classifying principles for physics at high energy (in the confor-
mal case, for massless systems), so too it should be possible to analyze
the unitary, irrreducible representations of the full quaplectic symmetry
group in order to discern their particle content. This is set up in Sec. 2 for
representations arising in the context of first-quantized models, following
the well-trodden path from (super)particle and (super)string action prin-
ciples, to constraint algebras and physical spectra. It turns out that in
arbitrary even space–time dimensions, the scalar particle worldline action
admits a simple extension, whose constraint quantization irrevocably leads
to reciprocity. This new reciprocal-relativistic variant of the standard par-
ticle action indeed reveals an unexpectedly close relationship between these
model systems.
Notwithstanding these interesting constructs, the hope of generalizing
standard physics to reveal reciprocal-relativistic ‘corrections’ at high energy
d See S.G. Low, math-ph/0502018 for an explanation of this name.
www.ebook3000.com
within this program seems seriously confounded by the fact that in a par-
ticle basis, there are many undesired constituents such as continuous spin
or mass representations within unitary ireducible representations of the
quaplectic group. This analysis is not carried out in detail here, but is of
course a matter of the structure of the relevant unirreps, and has been
presented elsewhere in previous papers on quaplectic-invariant worldline
models [9, 10].
In Sec. 3, we present an alternative attempt [11] to characterize ‘ele-
mentary systems’ in reciprocal relativity. We turn to the examination of
typical state spaces arising in the first quantized methods described above.
As indicators of the essential structure of a ‘semi-classical’ formulation,
we examine parametric descriptions of certain special states within these
spaces, which have the property of saturating various indeterminacy rela-
tions. Namely, we point out that the Schrödinger–Robertson relation for the
product of uncertainties in relativistic energy–momentum and position is in
fact quaplectic invariant. In relativistic oscillator representations, it can be
saturated by so-called multimode squeezed states. Therefore, the ‘elemen-
tary’ semi-classical systems are characterized by the stratification of such
squeezed states under the action of the quaplectic group. The conclusion is
that in contraction limits of the quaplectic algebra, which could be expected
to be an experimentally accessible regime, it is the expectation values of
various second rank tensor operators, that is, (symmetrical) tensors qµν and
tµν , which encapsulate the appropriate parameters to describe fundamental
reciprocal-invariant systems. We identify these quantities with quadrupole-
like and energy–momentum like attributes, respectively (a third tensor rµν
is also involved). Indeed, given covariance with respect to Lorentz transfor-
mations, classes of reciprocal-relativistic invariant systems will therefore be
distinguished by the Lorentz normal forms of these numerical tensors.
In Sec. 4, we close with prospects for future work, and for complete-
ness, we provide a brief discussion of some related recent works on Born
reciprocity.
2. Reciprocal Relativity and Elementary Systems

As pointed out above, the idea of enriching the study of the represen-
tation theory of the symmetry groups of physical systems by construct-
ing space–time models has a noble parentage which includes of course
standard particles, spinning particles, superparticles as well as strings and
superstrings in various presentations. The aim is to construct dynamical
270 P. Jarvis
models, whose gauge fixing constraint conditions are realized on appro-

priate physical states carrying unitary representations of such symmetry
groups. In the case of reciprocal relativity, models have been presented
[9, 10] for ‘scalar’ particles, with world-line action a functional of a nine-
dimensional extended space. This incorporates an additional scalar quantity
on the worldline, θ(τ ), consistent with the parametrization of a homoge-
neous factor space of the quaplectic group (dimension 25) over the unitary
group U (3, 1) (dimension 16). In addition to Noether charges for the gener-
ators of the unitary group, the conserved canonical momentum πθ appears
as a central term in the algebra of translation generators, and its eigenvalue
plays the role of in the resultant Heisenberg algebra.
Here, we provide a new presentation of a modified scalar particle action
in even space–time dimensions, starting from the following Lagrangian
in 2D coordinates xµ (τ ), y ν (τ ), µ, ν = 0, 1, . . . , D and their derivatives
ẋν (τ ) = dxν /dτ , ẏ ν (τ ) = dy ν /dτ which are smooth functions of proper
time along the worldline,
1 µ
L(x, y, ẋ, ẏ) = ẋ ηµν ẋν + ẏ µ ηµν ẏ ν + c ẋµ ηµν y ν − ẏ µ ηµν xν . (1)
2
Here, ηµν = diag(+1, −1, −1, . . . , −1) is the standard Minkowski metric in
D dimensions, and c is a non-zero positive real constant. Compared with the
standard scalar particle action, corresponding to the first term (quadratic
in velocities on the worldline), there is a new contributione proportional
to c which treats the components of the 2D-dimensional vector (xµ , y µ )
asymmetrically, and which will prove to be the key to the modification of the
symmetry of the system, from inhomogeneous conformal to quaplectic. In
contrast to the extended coordinate system of the above-mentioned models,
there is no additional scalar coordinate (in physical dimensions, there are
thus eight, not nine, coordinates on the worldline).
We turn to an analysis of this model, firstly its classical symmetries,
and then to its quantization and physical states. Clearly (1) is invari-
ant under global homogeneous linear transformations of the coordinates
(xµ , y µ ), regarded as a vector of total dimension 2D, which preserve both
the symmetric bilinear form corresponding to the first term, as well as
the antisymmetric bilinear form appearing in the second term (which nec-
essarily entails two distinct 2D-dimensional vectors, namely (xµ , y µ ) and
e Anticipating the identification of y with ‘momentum’ p, this term is simply the canonical
one-form x dp up to a total derivative.
www.ebook3000.com
(ẋν , ẏ ν )). It is well known that the intersection of the corresponding trans-
formation groups, O(2, 2D − 2) and Sp(2D, R), respectively, is the non-
compact unitary group U (1, D −1) in half of the total dimension, namely
1
2 ·2D ≡ D. At the level of infinitesimal transformations, the group elements
are easily described. Lorentz invariance obviously entails
xµ → xµ + ω µ ν xν , pµ → pµ + pν ω ν µ , ωµν = −ωνµ
and in addition, there exists a set of ‘reciprocal boost’ transformations,
xµ → xµ + cαµ ν pν , pµ → pµ + c−1 xν αν µ , αµν = +ανµ .
Turning now to inhomogeneous transformations, clearly the leading

quadratic part of (1) is invariant under global translations xµ → xµ + aµ ,
y ν → y ν + bν , whereas the second term is not. Instead, the total variation
is by construction a total derivative:
d µ
δL = c(ẋµ bµ − ẏ µ aµ ) ≡ c −y aµ + xµ bµ . (2)
dτ
The procedure for constructing conserved quantities generating the
above symmetry transformations is well known, the only modification
required in this case being the adjustment of the standard Noether charges
generating global translations by the addition of appropriate terms from
(2) to restore the overall conservation laws. We complete the above analy-
sis by giving the expressions for the various conserved charges in terms of
the appropriate phase space variables: the coordinates (xµ , y µ ), and their
canonical momenta πµx = ∂L/∂ ẋµ = ẋµ + cyµ , πνy = ∂L/∂ ẏ ν = ẏν − cxν .
The Lorentz generators are
Lµν = (xµ πνx − xν πµx ) + (yµ πνy − yν πµy ) ;
the reciprocal boost generators are
Mµν = (yµ πνx + yν πµx ) − (xµ πνy + xν πµy )
and the adjusted translation generators are (for a particular choice of labels
and convenient normalization)
X µ = πyµ − cxµ ; P µ = πxµ + cy µ .
Remarkably, at the classical level of Poisson brackets, the algebra of these

charges is the full quaplectic Lie algebra (see below): in particular, the
272 P. Jarvis
adjusted Noether charges for translation non-invariance fulfil the Heisenberg

algebra:
{Xµ , Pν }P B = −2c ηµν .
The above results are consistent with elevating the status of the quaplec-
tic group of reciprocal relativity to a fundamental symmetry, specifically as
a semi-direct product involving symmetry generators which fulfil a Heisen-
berg algebra. Quantizing the system in the usual way by taking Poisson
brackets over into Lie brackets, it is clear that Planck’s constant appears
together with the central term cI, allowing the generators of the Heisen-
berg algebra to be identified with relativistic energy–momentum and time-
position operators.
The meaning of the relativistic time component X 0 within this world-
line formulation is resolved in the usual way through gauge fixing within
the time-reparametrization invariant extension of the model, which we now
briefly discuss. Consider the extended Lagrangian derived from (1) by intro-
ducing an additional scalar Einbein field e(τ ),

y, ẋ, ẏ) = 1 ẋµ ηµν ẋν + ẏ µ ηµν ẏ ν + c ẋµ ηµν y ν − ẏ µ ηµν xν + e Λ. (3)
L(x,
2e
Here, Λ is an additional fixed real parameter (‘cosmological constant’).
Clearly, the extended action S = Ldτ now admits invariance under arbi-
trary time reparametrizations τ → τ (τ ) provided e(τ ) transforms covari-
antly, e (τ ) = e(τ ) dτ/dτ . Note that the c term is by itself invariant under
such local diffeomorphisms without additional modification.
The quantization of the extended system now proceeds in a standard
way [12] consistent with the fact that there is now a primary constraint
πe = 0 as well as additional secondary constraints. As usual, the physical
Hilbert space is the kernel of the Dirac first class constraint algebra. We cite
a result formulated by Govaerts [13]:
Lemma. Given a dynamical system with Lagrangian L(q, q̇), introduce a
modified action and Lagrangian function

≡ dτ e(τ ) L q, 1 dq + Λ .
S = Ldτ
e dτ
Then, the equations of motion are expressible in the form
∂H ∂H
ṗ = −e , q̇ = e , H − Λ = 0,
∂q ∂p
where q, p H are the dynamical quantities of the original system, with
H = pq̇ − L(q, q̇).
www.ebook3000.com
The implication of this result for the quantization of such systems is

that the symmetry algebra must be realized on physical state spaces for
which the ‘cosmological constant’ Λ is a selected element of the spectrum
of the Hamiltonian H. As we show presently, in the present case of the mod-
ified world line action, as in the previous worldline models [9, 10], the lat-
ter is (up to a constant) simply the quadratic Casimir operator of the
Lie algebra of the quaplectic symmetry group [14]. In this way, we indeed
recover the standard scenario, that the value of Λ must lie in the spectrum
of the Casimir operator, in such a way that physical states (elementary
reciprocal-relativistic systems) should be classified by the associated uni-
tary (irreducible) representations of the quaplectic symmetry group.
For completeness, we present the Lie algebra of the quaplectic group in
unitary representations, after constraint quantization of worldline models
such as (1) or variants [9, 10]. Identifying one of the D-dimensional vector
operators with relativistic position, and the other with energy–momentum,
we use the re-scaled Heisenberg algebra in dimensionless form

X µ Pν
, = i δµν , [X µ , X ν ] = 0 = [P µ , P ν ] (4)
λx λp
with = λx λp = 0. The remaining generators comprised the Hermitean

generators of the Lorentz group, an antisymmetric tensor Lµν = −Lµν ,
together with additional non-compact generators Mµν = Mνµ fulfilling the
commutation relations,
[Lκλ , Lµν ] = i (ηλµ Lκν − ηκµ Lλµ − ηλν Lκµ + ηκν Lλµ ) ,
[Lκλ , Mµν ] = i (ηλµ Mκν − ηκµ Mλµ + ηλν Mκµ − ηκν Mλµ ) ,
[Mκλ , Mµν ] = i (ηλµ Lκν + ηκµ Lλµ + ηλν Lκµ + ηκν Lλµ ) ,
[Lκλ , Xµ ] = i (ηλµ Xκ − ηκµ Xλ ) ,
[Lκλ , Pµ ] = i (ηλµ Pκ − ηκµ Pλ ) ,
[Mκλ , Xµ ] = − i (ηλµ Pκ − ηκµ Pλ ) ,
[Mκλ , Pµ ] = i (ηλµ Xκ − ηκµ Xλ ) . (5)
A more succinct notation is achieved in a complex basis with generators

Z µ , defined by

1 X0 P0 1 Xi Pi
Z0 = √ −i , Zi = √ −i , (6)
2 λx λp 2 λx λp
274 P. Jarvis
such that Z µ = η µρ Z ρ ≡ (Z µ )† which fulfil the Heisenberg algebra (with

central generator I)
[Z µ , Z ν ] = − η µν I,
[Z µ , Z ν ] = 0 = [Z µ , Z ν ] (7)
and similarly with Lµν and Mµν combined into a matrix E µ ν , µ, ν =

0, 1, 2, 3, via Lµν = i(Eµν −Eνµ ), Mµν = Eµν +Eνµ , satisfying the hermitic-
ity conditions (E µ ν )† = η µρ ηνσ (E σ ρ ) which generate the real Lie algebra
of U (3, 1),
[E µ ν , E ρ σ ] = (δν ρ E µ σ − δ µ σ E ρ ν ), (8)
together with
[E µ ν , Z ρ ] = δν ρ Z µ , [E µ ν , Z ρ ] = −η µρ Zν . (9)
Returning to the model, we note finally the form of the Hamiltonian

and confirm its relationship to the Casimir operator of the quaplectic Lie

algebra, as claimed above. The canonical Hamiltonian q̇p − L following
from (1) is
1
H = ẋµ πµx + ẏ ν πνy − L(x, ẋ, y, ẏ) ≡ ηµν ẋµ ẋν + ẏ µ ẏ ν .
2
In terms of conserved quantities, this has the form,f
1 µ
H= X Xµ + P µ Pµ − cM µ µ , (10)
2
but it can more succinctly be written in terms of an auxiliary set of (non-
conserved) generators Xµ = πxµ − cy µ , Pν = πyµ + cxµ , which also fulfil the
Heisenberg algebra (with the same central extension as the conserved X µ ,
Pν ), namely,
1 µ
H= X Xµ + P µ P µ . (11)
2
At the operator level, taking into account the form of the quaplectic group
generators (5), (7), (8), (9) it can be seen that the auxiliary generators
eµ ν := E µ ν − {Z µ , Zν } generate U (3, 1) and commute with Z µ , Z ν , and
µ µ λ
that trace combinations such as C1 := e µ , C2 := e λ e µ are there-
fore quaplectic Casimir operators (see for example, Refs. [10, 11]). Indeed,
f Each term herein is separately a generator of the discrete Born reciprocity transforma-
tion on X µ , Pν ; the difference of their actions of course cancels.
www.ebook3000.com
from (10), the linear invariantg C1 is precisely the Hamiltonian H up to a

constant multiple.
With this observation, we make contact with the original vision of
Born that an extended interval of the type “x · x + p · p” might be used
to generate field equations (the intention having been to build a meson
theory). In the present construction, the operator in question is derived
from a Casimir invariant of a generalized symmetry group (the quaplec-
tic group). Albeit in a first-quantized context, this is consistent with the
modern understanding of the role of field equations as projecting out from
local fields, the components which carry the appropriate physical degrees of
freedom.
3. Reciprocal Relativity and the Granularity

of Space–Time
The hope of reciprocal relativity — of generalizing standard physics to
reveal reciprocal-relativistic corrections for high energy processes — is seri-
ously confounded by the fact that, in a particle basis, within unirreps of the
quaplectic group there are invariably many undesired constituents such as
continuous spin or mass representations, or worse still, space-like represen-
tations (see Refs. [9, 10]). This is an inevitable consequence of the presence
of the Heisenberg algebra and its influence on induced representations. In
this respect, the situation is similar to that for symmetries such as the
conformal group, or even for relativistic wave equations based on certain
infinite-dimensional representations of the Lorentz group.
In this section, we take a different approach (the following discussion is
based on Ref. [11]). We consider here the ‘elementary’ reciprocal-relativistic
systems for which up to a constant, E µ ν = {Z µ , Z ν } (these correspond to
“scalar” type systems arising from the Hamiltonian quantization approach
of the previous section; see also Ref. [14]). We attempt to characterize
such systems by claiming that states which saturate appropriate indeter-
minacy relations will provide a semi-classical picture of the structure and
parametrization of such systems.
It is well known that the Heisenberg relation is but one of a hierar-
chy of inequalities relating means and variances of observables. Stronger
than the Heisenberg relation for two observables A, B, but equivalent for
g Closely
related to the number operator in oscillator representations; see Sec. 3, Refs. [14]
and [9–11].
276 P. Jarvis
uncorrelated states, is the Schrödinger inequality [15]

1
(∆A)2 · (∆B)2 − (cov(A, B))2 ≥ |− i[A, B]|2 ,
2
where
1
cov(A, B) = (AB + BA) − AB
2
and
cov(A, A) = (∆A)2 .
Regarding both the left- and the right-hand sides as determinants, this
is in turn but the 2 × 2 case of a more general relation, the Schrödinger–
Robertson inequality [16] for any even number of observables, which asserts
the dominance of the determinant of the covariance matrix over that of
the commutator expectation values (see also Ref. [17]). Exploiting the fact
that reciprocal relativity singles out the relativistic energy–momentum and
time-position operators as key physical observables, we therefore consider
the Schrödinger–Robertson inequality in the form,
8
cov(X µ , X ν ) cov(X µ , P ν ) 1
Det ≥ , (12)
cov(P µ , X ν ) cov(P µ , P ν ) 2
where the right-hand determinant of commutators has been evaluated using

the Heisenberg algebra −12 i[X µ , Pν ] = 12 δ µ ν .
The inequality thus rides on the evaluation of the left-hand determinant,
of the above operator product expectation values covariance matrix, which
we can write in an extended index notation with eight-vector Zm , m, n =
0, 1, 2, 3, 0, 1 , 2 , 3 arranged as

Zµ Zµ
Zm = = , (13)
Zµ Zµ
namely
1 1
Cmn = Zm Zn + Zn Zm − Zm Zn .
2 2
In the present, scalar case, the quaplectic group Hilbert space admits
an action under a larger group, the extension of the non-compact unitary
group to the full symplectic group of dimension 36, generated by the bilinear
combinations Zmn := {Zm , Zn }. Transformations on state space under this
www.ebook3000.com
group action, |ψ → |ψ = U |ψ, will necessarily generate homogeneous

linear transformations

Zm = U −1 Zm U = Mmm (U )Zm ,
m
which (being elements of the group of inner automorphisms of the Lie alge-
bra) must themselves be linear canonical transformations, that is, 8 × 8
symplectic matrices [18] (compare the discussion of the invariances of the
model worldline Lagrangian, (1)). Thus, in view of the bilinear structure of
the covariance matrix, we have immediately (using the fact that symplectic
matrices are unimodular)
Det(C ) = Det(M CM T ) = Det(C)(Det(M ))2 = Det(C).
On the other hand, under the action of the Weyl–Heisenberg group, the
Zm undergo inhomogeneous transformations — phase rotations and global
translations — which of course also leave the covariance matrix unaffected.
The conclusion is that the Schrödinger–Robertson inequality (12) is
invariant under the full inhomogeneous symplectic transformation group,
and in particular, under the quaplectic group. Specifically, we wish to exam-
ine the states |ψ which saturate the inequality (S–R minimal uncertainty
states), in claiming that these will represent ‘semi-classical’ elementary
quaplectic systems. However, in view of the quaplectic symmetry, our inter-
est here is therefore to study the stratification of the state space — enu-
merating possible orbit classes, and attributing physical significance to the
parametrizations and coordinates for the distinct orbits within such strata.
This strategy will be elaborated on more concretely below.
Now it is well known in quantum optics that for systems of several
oscillators, the S–R minimal uncertainty states are the so-called multimode
squeezed states [19]. The same construction applies for the present ‘rela-
tivistic oscillator’ case with appropriate identification of the ‘creation’ and
‘annihilation’ parts (compare (6)h ). The multimode squeezed states are of
the form,
P P
1
ϕmn Zmn ζ m Zm
|ϕ, ζ = e 2 ·e · |0 (14)
for some parameters ϕmn and ζ m , where the vacuum |0 is the standard
ground state for each oscillator mode.
h Leading to standard expressions in the coordinate representation; note however that

xi = −xi .
278 P. Jarvis
In the present context, it is assumed that these S–R minimal uncertainty

states do describe the quantum-classical limit of reciprocal relativity. Dif-
ferent S–R minimal uncertainty states |ϕ, ζ, |ϕ , ζ , only correspond to
physically distinct semi-classical states if the unitary transformation relat-
ing them does not belong to the quaplectic group U (3, 1) H(4). This result
follows irrevocably from the reciprocal relativity principle, and quaplectic
symmetry. Thus, we conjecture that in the case of the scalar quaplectic
system, there are various physically different semi-classical states, associ-
ated with classes of S–R minimal uncertainty states, parametrized by the
geometry of the homogeneous space Sp(8, R) H(4)/U (3, 1) H(4)
Sp(8, R)/U (3, 1).
Insight into the nature of these classes of S–R minimal uncertainty
states can be gained by reference to ordinary (non-relativistic) quantum
mechanics. Recall that the multimode squeezed states (14) can equiva-
lently be regarded as generalized coherent states, characterized by their
property of diagonalizing a linear combination of the annihilation and cre-
ation operators,
aa + ba† |z = z|z (15)
and for such a system, we can of course (to the extent that these are sharp
subject to the Heisenberg principle) give average quantities ‘position’ and
‘momentum’ via the expectation values X, P , where the operators are
defined in the usual way by appropriate complex combinations of a and a†
(compare (6)). For standard (non-squeezed) coherent states b = 0, and the
position and momentum variances are equal. In the general case, these vari-
ances are of course scaled by the squeezing parameter, while maintaining
the uncertainty product at its minimum. Finally, the position and momen-
tum eigenstates themselves can be regarded as singular (non-normalizable)
limits of such squeezed states, where the squeezing goes to 0 or ∞, for
inadmissible values of the ratio a/b [19].
In the relativistic context, given the reciprocal boost transformations
which can rotate unitarily between X µ and Pν , the attribution of corre-
sponding operator expectation values for these quantities in semi-classical
limit states, whether or not sharp, is purely frame-dependent from the
reciprocal-relativistic perspective. Instead, we must carefully re-examine
the relevant semi-classical limit states (14), with a view to understand-
ing what attributes can be regarded as intrinsic, rather than merely frame
dependent.
www.ebook3000.com
Recall that the symplectic generators Zmn include all momentum-

position bilinears: the 36 operators Xµ Xν , Pµ Pν , and Xµ Pν , wherein the
16 unitary generators Eµν (complex combinations of Lµν and Mµν ) are
derived from the number conserving combinations Zµ Z ν , leaving 20 addi-
tional objects (the real and imaginary parts of Zµν and Zµ ν ≡ Z µν in the
above notation). Thus, an obvious generic set of orbits for Sp(8, R)/U (3, 1)
is parametrized by the states derived by acting with the latter combinations
(the generators of the factor space),
1 µν 1 µν
|Φ = e 2 ϕ Zµν
· e2ϕ Z µν
· |Ψ (16)
for some fixed vector |Ψ.

So far we have been referring to the semi-classical limit → 0, as is
appropriate for discussing coherent states and minimal uncertainty, and
the quantum-classical transition. At the same time, reciprocal relativity
is expected to govern interactions at high energy (connections with non-
inertial reference frames and the equivalence principle have been noted
above). Seen another way, there should be evidence of reciprocal-relativistic
corrections to standard physics, in the limit where the new dimensionful
constants, in particular the universal force constant, is large, b → ∞.
It is therefore legitimate to adopt a restricted formulation of the struc-
ture of the S–R minimal uncertainty states, using an appropriate contrac-
tion limit of the full symmetry algebra. Noting from (1) that = λp λx ,
and b = cλp /λx , we define in the standard way, generators re-scaled by
(inverse) powers of a corresponding formal parameter β → ∞ which will
reflect the physical b → ∞ limit, and we investigate the resulting commuta-
tor brackets, assuming smooth limits of the new generators.i The following
is a brief description of two such admissible contraction limits, accompa-
nied by a plausible interpretation of the implications for the structure of
the S–R minimal uncertainty states.
On physical grounds, we assume that the correct contraction of the
quaplectic group in this limit must leave unchanged the subalgebra con-
sisting of the semi-direct product of the Lorentz with the Heisenberg
Lie algebra, generated by unscaled operators L◦µν ≡ Lµν and Pµ◦ ≡ Pµ ,
Xµ◦ ≡ Xµ . Because observations give primacy to the Poincaré group however
(with momentum generators Pµ ), it must be the case that the reciprocal-
relativistic, position-momentum equivalence will be restored at high energy.
i Fora Lie algebra L, this corresponds to taking finite dimensional invariant subalgebras
of the loop algebra L[[β]].
280 P. Jarvis
We therefore discount this apparent asymmetry, and we do not further

discuss standard particle states or position operators (note that the non-
compact U (3, 1) unitary group remains as an outer automorphism group of
◦
the Heisenberg algebra). The remaining generators Mµν , the limit of scaled
◦
Mµν := Mµν /β, as a result of the commutation relations (5) now become a
set of ten abelian operators (a slightly more refined scaling is given below,
which does not however affect the quaplectic group contraction). We infer
that in the β → ∞ regime, states of a quaplectically covariant system will
be characterized by some eigenvalue of this tensor.
A further question is then to identify the relevant β → ∞ contraction
limit of the associated full symplectic algebra, given the behavior of its
U (3, 1) subalgebra. From the commutation relations,
[Z µν , Zρσ ] = ηµρ Mνσ + ηνρ Mµσ + ηµσ Mνρ + ηνσ Mµρ , (17)
there are two possibilities, involving the decomposition of Mµν (see (5)) into
its traceless part, defined by Nµν := Mµν − 12 ηµν N , N := 12 M ν ν = E ν ν ,
so that
1
Mµν = Nµν + ηµν N ,
2
with Z ◦ µν = Z µν /β, Z ◦ µν = Zµν /β; (18)
and either (I) M ◦ µν = Mµν /β ,
or (II) N ◦ µν = Nµν /β, N ◦ = N /β 2 .
In case (I), the Z ◦ µν , Z ◦ µν become abelian, and appropriate physical states
could be expected to be eigenstates with some complex eigenvalues zµν , z µν .
In case (II), these generators fulfil a 10-dimensional Lorentz-tensor Heisen-
berg algebra, with central generator N ◦ (from (17), (18)) ,
[Z ◦ µν , Z ◦ ρσ ] = (ηµρ ηνσ + ηµσ ηνρ )N ◦ . (19)
Then, the states (16) (with |Ψ = |0) have precisely the structure of mul-
timode squeezed states of this large Heisenberg algebra (compare (14) and
the equivalent presentation in (15) in the standard case), this time at the
level of eigenvalues of linear combinations of the tensor generators Z ◦ µν
and Z ◦ µν , with the real and imaginary parts of their expectation values
Z ◦ µν and Z ◦ µν playing the role of mean ‘position’ and ‘momentum’.
Either casej thus leads (for scalar quaplectic systems) to tensor parame-
ters — whether eigenvalues or expectation values — associated with these
j Without the β → ∞ limit, the states (16) can be considered in the context of group
coherent states (see Ref. [19]).
www.ebook3000.com
contracted operators, together with eigenvalues of M ◦ µν in this regime of

‘reciprocal-relativistic corrections to relativistic processes’.
Resolving the above objects with respect to relativistic time-position
and energy–momentum bilinears, and taking linear combinations, the con-
clusion is that to the state of the elementary scalar quaplectic system
can be associated attributes qµν = Xµ Xν , rµν = Xµ Pν + Xν Pµ , and
tµν = Pµ Pν . It is reasonable to suggest therefore that the significance
of the orbit stratification in characterizing ‘semi-classical’ parameters of a
quaplectic system (as b → ∞) is the emergence of structure, in the sense
of a quadrupole-like tensor qµν , an energy–momentum-like tensor tµν and
an additional tensor rµν . Moreover, distinct types of such systems will be
distinguished at low energy by classifying these symmetric tensor quan-
tities by their Lorentz-normal form. In contrast, the distinction between
the position and momentum space quantities Xµ , Pν , in the reciprocal-
relativistic regime, is frame-dependent and not physically significant.
4. Conclusions
In this chapter, we have re-examined the credentials of Born reciprocity
through the lens of modern developments relating to symmetry principles
in relativistic quantum theory, and with the benefit of some hindsight. We
have seen (Sec. 2) that, in a first quantized formulation, models exhibit-
ing invariance under the extension to reciprocal relativity are surprisingly
closely related to the standard wordline particle system. Nonetheless, anal-
ysis of the occurring representations in a particle basis turns out to give
unphysical spectra. However, analysis of relativistic oscillator spaces via
the Schrödinger–Robertson inequality (Sec. 3) suggests the emergence of
physical attributes associated with certain tensor invariants (whose physi-
cal interpretation is not clear at this stage).
From a modern perspective, the fundamental work of Wigner [20] on uni-
tary representations of the Poincaré group, and its central importance for
relativistic wave equations, was not incorporated into Born’s early work (the
first publication on reciprocity appears to be Ref. [21]). Moreover, renormal-
izable, relativistic local quantum field theory as the established foundation
of particle theory was of course not yet available to Born’s group in post-
war Edinburgh. In addition to Refs. [1, 2] and the later review article in
collaboration with Green [4], the group’s work can be traced through their
papers [22–29]. Despite one of Born’s original motivations — the removal
of infinities from quantum field theory — having been circumvented by the
282 P. Jarvis
advent of renormalization theory, the ideas continued to be of interest to

him and co-workers in subsequent years [30]. Indeed, Green returned to
the topic with a presentation at the 1965 international Kyoto meeting cel-
ebrating meson theory [31] (see also Ref. [32]), and different aspects were
intermittently taken up and promoted by other authors [33, 34].
In more recent years, Born reciprocity has repeatedly surfaced as a
key idea for physics at the Planck scale. The following list is a very cur-
sory selection from the literature. We refer to Refs. [35, 36] which pur-
sue Born’s idea of momentum space curvature in the context of ‘relative
locality’; Ref. [37, 38] for an approach of ‘holomorphic gravity’, with posi-
tion/momentum polarization via a local almost complex structure in phase
space (see also Ref. [39]); Refs. [40,41] for gauge theories of gravitation based
on a local U (3, 1) structure group; and Ref. [42] and references therein,
for models with the distinction between position and momentum in phase
space resolved via a local gauged Sp(2, R) symmetry. See also Ref. [43] for
an exploration of the implications of reciprocity for worldline models.
Whether reciprocal relativity, as discussed here, or a further elaboration
along the lines of some of the above-mentioned works, proves the most
fruitful avenue for future work, it is clear that Born’s idea of reciprocity
will continue to provide an enduring impetus to the development of a deeper
understanding of physics at fundamental levels.
Acknowledgments
PDJ thanks A. Bracken, J. Govaerts, S. Low and S. Morgan [44] for con-
structive suggestions and comments. The Physics Department, University
of Austin at Texas, in particular hosts A. Bohm and S. Low, are thanked for
hospitality during a visit where part of this work was carried out. Financial
support from the Australian–American Fulbright Commission during this
stay is gratefully acknowledged.
References
1. M. Born, Relativity and quantum theory, Nature 141 (3564), 327–328 (1938),
doi: http://dx.doi.org/10.1038/141327a0.
2. M. Born, A suggestion for unifying quantum theory and relativity, Proc. Roy.
Soc. Lond. Ser. A, Math. Phys. Sci. 165, 291–303 (1938).
3. M. Born, A. Einstein, and I. Born, The Born Einstein Letters: correspondence
between Albert Einstein and Max and Hedwig Born from 1916 to 1955 with
Commentaries by Max Born. Translated by Irene Born. Basingstoke, Macmil-
lan Press, (1971), http://www.archive.org/details/TheBornEinsteinLetters.
www.ebook3000.com
4. M. Born, Reciprocity theory of elementary particles, Rev. Mod. Phys. 21(3),

463–473 (1949), doi: http://dx.doi.org/10.1103/RevModPhys.21.463.
5. S.G. Low, Canonically relativistic quantum mechanics: Representations of
the unitary semidirect Heisenberg group, U (1, 3) ⊗s H(1, 3), J. Math. Phys.
38, 2197–2209 (1997).
6. S.G. Low, Representations of the canonical group, (the semidirect product
of the unitary and Weyl–Heisenberg groups), acting as a dynamical group
on noncommutative extended phase space, J. Phys. A Math. Gen. 35(27),
5711–5729 (2002).
7. S.G. Low, Reciprocal relativity of noninertial frames and the quaplectic
group, Found. Phys. 36, 1036–1069 (2006).
8. S.G. Low, Reciprocal relativity of noninertial frames: Quantum mechanics,
J. Phys. A Math. Theor. 40 (14), 3999–4016 (2007).
9. J. Govaerts, P.D. Jarvis, S.O. Morgan, and S.G. Low, World-line quantisation
of a reciprocally invariant system, J. Phys. A Math. Theor. 40, 12095–12111
(2007).
10. P.D. Jarvis and S.O. Morgan, Constraint quantization of a worldline system
invariant under reciprocal relativity: II, J. Phys. A Math. Theor. 41(46)
(2008).
11. P.D. Jarvis and S.O. Morgan, Born reciprocity and the granularity of
spacetime, Found. Phys. Lett. 19(6), 501–517 (2006).
12. J. Govaerts, Hamiltonian Quantisation and Constrained Dynamics, Leuven,
University Press, 1991.
13. J. Govaerts, The cosmological constant of one-dimensional matter coupled
quantum gravity is quantised, Proc. 3rd Int. Workshop on Contemporary
Problems in Mathematical Physics (Cotonour, Republic of Benin), eds. J.
Govaerts, M.H. Hounkonnou and A.Z. Msezane, World Scientific, Singapore,
2003, pp. 244–272.
14. A. Shamaly, Invariants of Born reciprocity theory, J. Math. Phys. 15(8),
1178–1180 (1974).
15. E. Schrödinger, Zum Heisenbergschen Unschärfeprinzip, Sitzungsberichte,
Preuss. Akad. Wiss. (Physikalisch-Mathematische Klasse) 296–303 (1930).
16. H.R. Robertson, An indeterminacy relation for several observables and its
classical interpretation, Phys. Rev. 46, 794 (1934).
17. J.L. Synge, Geometrical approach to the Heisenberg uncertainty relation and
its generalisation, Proc. Roy. Soc. Lond. A 325, 151–156 (1971).
18. S.G. Low, Maximal quantum mechanical symmetry: Projective representa-
tions of the inhomogeneous symplectic group, J. Math. Phys. 55(2), 022105
(2014).
19. D.A. Trifonov, Robertson intelligent states, J. Phys. A 30, 5941 (1997).
20. E. Wigner, On unitary representations of the inhomogeneous Lorentz group,
Ann. Math. 40(1), 149–204 (1939).
21. M. Born, Quantised field theory and the mass of the proton, Nature
136(3450), 952–953 (1935), doi: http://dx.doi.org/10.1038/136952a0.
22. M. Born, Application of reciprocity to nuclei, Proc. Roy. Soc. Lond. Ser.
A Math. Phys. Sci. 166, 552–557 (1938).
284 P. Jarvis
23. M. Born, Some remarks on reciprocity, Proc. Indian Acad. Sci. Sec.
A 8, 309–314 (1938).
24. M. Born, Reciprocity and the number 137. I, Proc. Roy. Soc. Edinburgh. 59,
219–223 (1939).
25. M. Born and K. Fuchs, Reciprocity II. Scalar wave functions, Proc. Roy. Soc.
Edinburgh. 60, 100–116 (1940).
26. M. Born and K. Fuchs, Reciprocity III. Reciprocal wave functions, Proc. Roy.
Soc. Edinburgh. 60, 141–146 (1940).
27. M. Born and A.E. Rodriguez, Meson masses and the principle of reciprocity,
Nature 163(4139), 320–321 (1949), doi: http://dx.doi.org/10.1038/163320b0.
28. M. Born, K.C. Cheng, and H.S. Green, Reciprocity theory of electrody-
namics, Nature. 164(4163), 281–282 (1949), doi: http://dx.doi.org/10.1038/
164281b0.
29. M. Born and H.S. Green, Quantum theory of rest masses, Proc. Roy. Soc.
Edinburgh. 62, 470–488 (1949). With appendices by K.C. Cheng and A.E.
Rodriguez.
30. J. Bernstein, Max Born and the quantum theory, Am. J. Phys. 73(11),
999–1008 (November, 2005).
31. H.S. Green. Theory of Reciprocity, Broken SU (3) symmetry, and strong
Interactions. In Proc. Int. Conf. Elementary Particles, Kyoto, 1965, p. 159
(1966).
32. A. J. Bracken. Group-theoretical applications in a tri-local model for baryons,
PhD thesis, University of Adelaide, 1970.
33. E.R. Caianello, Is there a maximal acceleration?, Lettere al Nuovo Cimento.
32, 65–70 (1981).
34. A. Jadczyk, Born’s reciprocity in the conformal domain. In Spinors, Twistors,
Clifford Algebras and Quantum Deformations, pp. 129–140. Springer, New
York, 1993.
35. G. Amelino-Camelia, Relativity in spacetimes with short-distance structure
governed by an observer-independent (Planckian) length scale, Int. J. Mod.
Phys. D 11(01), 35–59 (2002).
36. G. Amelino-Camelia, L. Freidel, J. Kowalski-Glikman, and L. Smolin, Rel-
ative locality: A deepening of the relativity principle, Int. J. Mod. Phys. D
20(14), 2867–2873 (2011).
37. C.L. Mantz and T. Prokopec, Resolving curvature singularities in holomor-
phic gravity, Found. Phys. 41(10), 1597–1633 (2011).
38. C.L. Mantz, Holomorphic gravity, Utrecht University Master’s Thesis (2007).
39. J.G. Burgers, C.L.M. Mantz, and T. Prokopec, The Newtonian limit of her-
mitian gravity, General Relativity and Gravitation 45(1), 155–187 (2013).
40. C. Castro, On Born’s deformed reciprocal complex gravitational theory and
noncommutative gravity, Phys. Lett. B 668(5), 442–446 (2008).
41. C. Castro, The extended relativity theory in Clifford phase spaces and mod-
ifications of gravity at the Planck/Hubble scales, Adv. Appl. Cliff. Algeb.
24(1), 29–53 (2014).
www.ebook3000.com
42. I. Bars, Gauge symmetry in phase space, consequences for physics and
spacetime, Int. J. Mod. Phys. A25, 5235–5252 (2010), doi: 10.1142/
9789814335614 0026,10.1142/S0217751X10051128.
43. M. Pavšič, Beyond the relativistic point particle: A reciprocally invariant
system and its generalisation, Phys. Lett. B 680(5), 526–532 (2009).
44. S. Morgan. A Modern Approach to Born Reciprocity, PhD thesis, University
of Tasmania, 2011.
www.ebook3000.com
Chapter 10
On Non-Equilibrium
Thermodynamics of Space–Time
and Quantum Gravity
Joakim Munkhammar
Department of Engineering Sciences, Uppsala University,
SE-751 21 Uppsala, Sweden.
joakim.munkhammar@angstrom.uu.se
Based on recent results from general relativistic statistical mechanics

and black hole information transfer limits, a space–time entropy-action
equivalence is proposed as a generalization of the holographic principle.
With this conjecture, the action principle can be replaced by the second
law of thermodynamics, and for the Einstein–Hilbert action the Ein-
stein field equations are conceptually the result of thermodynamic equi-
librium. For non-equilibrium situations, Jaynes’ information-theoretic
approach to maximum entropy production is adopted instead of the sec-
ond law of thermodynamics. As it turns out for appropriate choices of
constants, quantum gravity is obtained. For the special case of a free par-
ticle the Bekenstein–Verlinde entropy-to-displacement relation of holo-
graphic gravity and thus the traditional holographic principle emerges.
Although Jacobson’s original thermodynamic equilibrium approach pro-
posed that gravity might not necessarily be quantized, this particular
non-equilibrium treatment might require it.
1. Introduction
Gravity, thermodynamics and quantum mechanics are deeply connected.
The partition function for a grand canonical ensemble and the quantum
partition function derived from Feynman’s path integral formulation are
analogous with the substitution of time and inverse temperature it/ ↔
1/kb T [1]. While such a substitution — formally corresponding to a Wick
287
288 J. Munkhammar
rotation — might be useful for computations, it is also a key to connect

both theories via Shannon information theory [1–4].
As regards thermodynamics and gravity, Jacobson showed that the
Einstein field equations could be derived based on space–time thermody-
namic assumptions of Unruh temperature and Rindler coordinates [5]. This
connection was deepened with the emergence of the holographic princi-
ple from string theory, loop quantum gravity and entropic gravity theories
[6–14]. This suggests that entropy — and thus Shannon information theory
— might potentially be an important aspect in the connection between
thermodynamics and gravity as well.
However, as regards entropy, a complete covariant theory of thermody-
namics and statistical mechanics in a full general relativistic context is yet
to be established [15]. Initial investigations have shown that for a constant
“thermal time” there seems to be a direct relation between temperature and
inverse proper time [15–17]. With the assumption that entropy is negative
information I, there is also a direct connection between entropy, energy E
and time t in upper limits of the bound on the information I transfer in
black hole thermodynamics [18]:
I < γEt (1)
for the constant γ = π/ ln 2. Although this is in line with the holographic
principle, it is perhaps worthwhile to consider an alternative holographic
principle based on these results. A principle which connects entropy dynam-
ically to energy and time such that an equilibrium configuration is a special
case, like in Ref. [19]. Non-equilibrium is after all believed to be the most
common state of a thermodynamics system [19, 20].
2. Entropy-Action Equivalence Conjecture

If there is a potential proportionality between inverse temperature and time
the thermodynamic relation dH = δQ/T proposes a direct proportionality
between entropy H, energy E and time t:
H ∝E·t (2)
which is up to a constant the upper information transfer limit for black

holes (1) derived by Bekenstein [18]. This general entropy is dependent
on coordinate variables of space and time. For most purposes, the second
law of thermodynamics is enforced to constrain the dynamics. In effect, this
means a maximization of the entropy. Keep in mind that the form of energy
www.ebook3000.com
On Non-Equilibrium Thermodynamics of Space–Time and Quantum Gravity 289
E is not constrained here and that basically it is set up to match units of

the dynamical components of (1) and that of thermal time.
Thus consider, for the time being, the possibility that E is the
Lagrangian L of a system (in classical terms) and that the Lagrangian
is integrated over time t to bring the entropy H:

H ∝ L(q, q̇, t) dt (3)
with a generalized coordinate q. This is dramatically different from

Bekenstein’s information transfer bound, but the units of the dynamical
components remain. And for the special case of L = E and that E is inde-
pendent of time t brings H ∝ E · t which is the upper limit of Bekenstein’s
information transfer bound. Also, if one applies the second law of thermo-
dynamics to (3), with respect to the generalized coordinates q and q̇, it
would be equivalent to the variational principle. This is worthwhile investi-
gating deeper, since if applicable it could potentially dispense with the use
of certain action principles in physics.
The most general classical description of physics is given by general rel-
ativity, whose dynamics is traditionally derived from the Einstein–Hilbert
action. As the main conjecture of this chapter, let us assume, for the time
being, that the space–time thermodynamic entropy H for a space–time
region D can be identified with the Einstein–Hilbert action SEH with a
proportionality constant κ:
4
c √
H(gµν , LM , D) ≡ κSEH (gµν , LM , D) = κ R + LM −gd4 x,
16πG
D
(4)
where R = R(gµν ) is the Ricci-scalar based on the metric tensor gµν and
LM is the matter-Lagrangian. The entropy is then a function of the metric
gµν (as a set of generalized coordinates), the classical matter-Lagrangian
LM and space–time region D. This means that the entropy is dependent
on both space–time configuration and matter content within a domain of
space–time. Entropy, as defined here, is still in practice in proportionality
with the dynamical variables of energy and time.
If entropy H is varied with respect to the generalized coordinates, that
is the metric gµν , then a “thermodynamic equilibrium state” is conceptu-
ally obtained. Formally, the Einstein field equations are given since that
variation corresponds to the traditional variation of the Einstein–Hilbert
action. These derivations are formally equivalent, but there is a conceptual
difference for the variation: extremization of entropy instead of traditional
290 J. Munkhammar
variation of the action, which will be of importance as the chapter pro-

gresses.
There is no formal novelty in varying the Einstein–Hilbert action with
respect to the metric, but there is a conceptual gain via the connection
to entropy and thermodynamics. In thermodynamics, equilibrium is not
always achieved. In fact, a state of maximum entropy is rarely achieved,
and most systems are likely in a non-equilibrium state [19, 20]. Theoreti-
cally, for thermodynamic applications, in non-equilibrium situations a max-
imum entropy production approach is adopted instead of the second law of
thermodynamics which postulates a maximum entropy approach [19, 20].
In Jacobson’s original thermodynamic approach to gravity the Einstein
field equations were derived on the assumption of thermodynamic equilib-
rium [5]. This ansatz was extended to non-equilibrium space–time thermo-
dynamics by Eling et al. in Ref. [19]. This study assumed that the entropy
was a function of the Ricci scalar:
H ∝ σ + f (R). (5)
As a means to estimate the rate of entropy change semi-classical compo-

nents such as the Unruh temperature T and “internally developed entropy”
dHi were used in the entropy balance equation dH = δQ/T + di H. Gener-
ally, see [20] for an overview of entropy production methodologies. Albeit a
semi-classical approach, the study in Ref. [19] revealed interesting connec-
tions between shear viscosity of the horizon in the Einstein field equations
and entropy. While the entropy-action equivalence (4) is a function of space
and time, it is also — in similarity with (5) — a function of the Ricci scalar.
Eling et al.’s approach regarded a general setup, while the approach here is
more specified and by onset not based on semi-classical concepts. In their
study, gravity was based on semi-classical concepts, but remained, however,
classic.
An alternative approach for describing non-equilibrium thermodynam-
ics is to utilize Janyes’ information theoretic setup [21–24]. As a macroscopic
approach — similar to that of thermodynamics itself — that method devises
a partition function based on all possible states of entropy and uses a max-
imum entropy production principle to constrain probabilities for different
states. If applied to Eling et al.’s study it would imply setting up a parti-
tion function for the unknown function of the Ricci scalar. Formally, in this
approach, we have the following.
Assume the proposed space–time entropy of the classical Einstein–
Hilbert action (4), but resist the hesitation to apply variation to obtain
www.ebook3000.com
thermodynamic equilibrium. Instead, setup all possible states of configura-

tion for H in terms of gµν . Based on these states, a new form of entropy
H2 can be identified. In traditional non-equilibrium thermodynamics, this
is labeled “the second entropy” [20, 25], which is the pure information-
theoretic entropy of all states of entropy H[gµν ]:

H2 = p[gµν ] log p[gµν ] = − Dgµν p[gµν ] log p[gµν ], (6)
All gµν
where p[gµν ] is an unknown probability function defined for each configu-

ration of gµν . This produces expected values, for example, for the (“first”)
entropy H:

H = p[gµν ]H[gµν ] = Dgµν p[gµν ]H[gµν ] (7)
All gµν
which is proportional to the expected action via (4). The probability

function p[gµν ] is found by using Lagrange multipliers, as was done for
non-equilibrium thermodynamics in Refs. [21–24] and analogously for
information-theoretic interpretations of quantum mechanics in Refs. [1, 3,
4, 26]. By employing complex Lagrange multipliers, λ and α, the second
entropy H2 is maximized in terms of “first” entropy H by:

H2 = − Dgµν p[gµν ] log p[gµν ] + λ 1 − Dgµν p[gµν ]

+ α H[gµν ] − Dgµν p[gµν ]H[gµν ] . (8)
This is simplified as:

H2 = λ + αH − Dgµν (p[gµν ] log p[gµν ] + λp[gµν ] + αp[gµν ]H[gµν ]). (9)
If we perform variation on the probability distribution, we get:

δH2 = − Dgµν (δp[gµν ])(log p[gµν ] + 1 + λ + αH[gµν ]) (10)
which is extremized when δH2 = 0, which corresponds to the probability

distribution:
1
p[gµν ] = e−1−λ e−αH[gµν ] = e−αH[gµν ] . (11)
Z
By varying the Lagrange multipliers, one enforces the two constraints, giv-
ing λ and its connection to α. Especially, one gets: e−1−λ = Z1 where Z is
292 J. Munkhammar
a form of partition function:

−αH[gµν ]
Z= e = Dgµν e−αH[gµν ] . (12)
gµν
This yields the expected entropy:
1
H = p[gµν ]H[gµν ] = e−αH[gµν ] H[gµν ]
gµν g
Z
µν

1 −αH[gµν ]
= Dgµν e H[gµν ] (13)
Z
which by (4) is proportional to the expected action with proportionality

constant κ:
1
S = κ p[gµν ]SEH [gµν ] = κ e−ακSEH [gµν ] SEH [gµν ]
gµν g
Z
µν

1 −ακSEH [gµν ]
=κ Dgµν e SEH [gµν ]. (14)
Z
The second entropy — also called the “entropy of entropy” — arises in

Jaynes’ non-equilibrium formulation here and is defined as:

H2 = − p[gµν ] log(p[gµν ]) = − Dgµν p[gµν ] log([gµν ]) (15)
gµν

which by the criterion All gµν p[gµν ] = 1 and algebraic manipulations
become (see derivation in Refs. [3, 4]):
H2 = −αH + log Z. (16)
This characterizes the complete description of possible states of entropy

and second entropy in this setup with undefined constants κ connecting
action and entropy, and α related to entropy production maximization.
Sections 3 and 4 will outline the pursuit of these constants.
In short, this approach represents the complete non-equilibrium dynam-
ics of general relativity based on the entropy-action equivalence conjecture
and Jayne’s second entropy approach for maximum entropy production.
www.ebook3000.com
3. ακ and Quantum Gravity

In this setup, two constants are undefined so far: κ which connects entropy
H to the action S, and α which connects entropy H to second entropy H2 .
By replacing H with κS in the partition function, according to (4), and
assuming that ακ ≡ i/, one gets:
i
i
Z= e S[gµν ] = Dgµν e S[gµν ] (17)
gµν
which is the quantum partition function for quantum gravity [27]. This also
renders the second entropy equivalent to an information-theoretic entropy
in quantum mechanics, also called “quantropy”, as the term was coined by
Baez and Pollard [3, 4].
Since this connects to information theory, and previous studies on infor-
mation theoretic approaches to quantum mechanics, an explicit probability
for each configuration of gµν is provided (see Refs. [1, 3, 4] for general con-
figurations):
1 − i S[gµν ]
p[gµν ] = e . (18)
Z
Because ακ = i/ is complex the probability for each state is complex,
which is problematic from the probability theory perspective. However,
that can be mended with the following setup, which was devised by Lisi
for general quantum systems [1]. The probability for the system to be on a
specific path in a set of possible paths in configuration space is:

set
p(set) = δpath p[path] = Dgµν δ(set − gµν )p[gµν ]. (19)
paths
Typically, the system reverses sign under inversion of parameter integration

limits [1]:

t

St = dt L(gµν ) = − dt L(gµν ) = −St . (20)
t
This implies that the probability for the system to pass through configura-

tion gµν at parameter value t is:

p(gµν , t ) = Dgµν δ(gµν (t ) − gµν )p[gµν ]
294 J. Munkhammar
  
gµν (t )=gµν

  
= Dgµν pt [gµν ]  Dgµν pt [gµν ]
gµν (t )=gµν

= ψ(gµν , t )ψ † (gµν

, t ) (21)
in which we can identify the quantum wave function:
gµν (t )=gµν

gµν (t )=gµν

1 t
ψ= Dgµν pt [gµν ] = √ Dgµν e−αS
Z
gµν (t )=gµν

1 St
= √ Dgµν ei . (22)
Z
This gives the probability amplitude for the probability of a system to pass

through metric gµν at time t . See the similarity with quantum mechan-
ics in Ref. [1].
It should be noted that that the quantum wave function in quantum
mechanics is subordinate to the partition function since it only works when
t is a physical parameter of the system and that the system is t symmetric,
which provides a real partition function Z.
Thus, with appropriate constants it is perhaps unexpectedly possible to
obtain quantum gravity from this non-equilibrium ansatz. However, only
the product of constants ακ has been determined so far. In order to obtain
the equivalence proportionality constant κ a connection to the holographic
principle is made in the following section.
4. α and Holographic Gravity

The entropy-action equivalence conjecture with the Einstein–Hilbert action
here proposes that the action principle is equivalent to the second law of
thermodynamics. That is, classical dynamics of a system is obtained for
thermodynamic equilibrium situations and quantum dynamics arise in non-
equilibrium situations. However, all of this requires that the entropy-action
equivalence conjecture in fact is valid and indeed supersedes the holographic
principle.
A first step on that path is to show equivalence under certain conditions.
According to the entropy-action equivalence conjecture (4), the entropy of
www.ebook3000.com
a stationary particle with mass m is:

H = κ mc2 dt = κmc2 t. (23)
If we assume that time can be expressed as ∆x/c for a “time-like” distance

∆x then we get:
∆S = κmc∆x (24)
and if we set κ = −2πkB /, where kB is Boltzmann’s constant, then we

get the Bekenstein–Verlinde expression from holographic gravity [14]:
mc
∆S = −2πkB ∆x. (25)

This expression was originally derived by Bekenstein (up to a constant) [18]
for particles falling into black holes, and it was more recently slightly altered
by Verlinde to develop his approach to entropic gravity [14]. If this special
case of the entropy can be assumed equivalent to the Bekenstein–Verlinde
expression, it is possible to identify α as well since ακ = i :
i i
α= =− . (26)
κ 2πkB
This defines both quantum dynamics in (17) and the entropy-action pro-
portionality for the Einstein–Hilbert action SEH (4):
2πkb
H ≡− SEH (27)

which completes the investigation on the generalization of the holographic
principle by means of an entropy-action equivalence conjecture in this chap-
ter.
5. Discussion and Conclusions

Based on recent developments in general relativistic statistical mechanics
and black hole thermodynamics, an entropy-action equivalence is conjec-
tured. With it, Einsteinian gravity is obtained for equilibrium situations —
analogous to Jacobson’s results — and a non-equilibrium theory of grav-
ity is also developed based on Jaynes’ information-theoretic approach to
maximum entropy production.
This approach to maximum entropy production, in thermodynamics
and other fields, is one among several different approaches to characterize
non-equilibrium dynamics, and the non-equilibrium concept is not without
296 J. Munkhammar
controversy [20]. But the fact that for a specific choice of the coupling
constant and a Lagrange multiplier the result is quantum gravity is perhaps
interesting enough to consider this possibility. And if true, it also presents
a form of information-theoretic quantization principle: maximum entropy
production. Another interesting feature is also that the correspondence to
classical physics is obtained by enforcing the second law of thermodynamics,
which dispenses with the use of a classical action principle.
One advantage of this approach is the derivation of an explicit probabil-
ity for each potential metric configuration, which could perhaps be useful
for certain calculations [1]. The entropy-action equivalence is conjectured as
a generalization of the holographic principle and shown to correspond to the
Bekenstein–Verlinde entropy-to-displacement relation of holographic grav-
ity for the special case of a stationary mass. That being said, this conjecture
likely violates the holographic principle, even under classical considerations,
for non-stationary situations.
The entropy of any object, even a black hole, is universal for all observers
in this approach (since the classical Lagrangian is universal) and based on
the energy and matter content of space–time. This information-theoretic
quantization of gravity suggests that space–time is encoded with the fun-
damental stochastic nature of quantum mechanics. Even though entropy
is invariant for all observers, the second entropy — quantropy — is not.
Since observers may or may not gain information regarding the state of
any object, quantropy is by necessity observer dependent [1, 4].
In a foundational sense, since the theory is based on accessible informa-
tion and observer dependence, it seems to demand some form of principle of
information covariance: the laws of physics can only be defined on the basis
of the information accessible to each observer [4]. This suggests that per-
haps Rovelli’s relational interpretation for quantum mechanics is favorable
for this theory [28].
Acknowledgments
The author wishes to thank Dr. Garrett Lisi, Dr. Ingemar Bengtsson,
Dr. Lars Mattsson and Dr. Jacob Bekenstein for valuable comments on
the theory and the chapter.
References
1. G. Lisi, Quantum mechanics from a universal action reservoir, (2006), arXiv:
physics/0605068v1.
www.ebook3000.com
2. D. Acosta, P.F. de Cordóba, J.M. Isidro and J.L.G. Santander, A holographic

map of action onto entropy, J. Phys. 361, 1–9 (2012).
3. J.C. Baez and B.S. Pollard, Quantropy, Entropy 17, 772–789, (2013), arXiv:
1311.0813.
4. J.D. Munkhammar, Canonical relational quantum mechanics from informa-
tion theory, EJTP 8, 93–108, arXiv:1101.1417 (2011).
5. T. Jacobson, Thermodynamics of spacetime, the Einstein equation of state,
Phys. Rev. Lett. 75, 1260–1263, (1995), arXiv:gr-qc/9504004v2.
6. J.D. Bekenstein, Black holes and entropy, Phys. Rev. D 7, 2333–2346 (1973).
7. G. ’t Hooft, Dimensional reduction in quantum gravity, THU-93/26, (1993),
arXiv:gr-qc/9310026v2.
8. J. Maldacena, The large N limit of superconformal field theories and super-
gravity, Adv. Theor. Math. Phys. 2, 231–252, (1998), arXiv:hep-th/9711200.
9. R. Bousso, The holographic principle, Rev. Mod. Phys. 74, 825–874, (2002),
arXiv:hep-th/0203101v2.
10. T. Padmanabhan, Equipartition of energy in the horizon degrees of freedom
and the emergence of gravity, Mod. Phys. Lett. A 25, 1129–1136, (2010),
arXiv:0912.3165.
11. T. Padmanabhan, Gravitational entropy of static space-times and micro-
scopic density of states, Class. Quant. Grav. 21, 4485–4494, (2004), arXiv:gr-
qc/0308070.
12. T. Padmanabhan, Thermodynamical aspects of gravity, New insights, Rep.
Prog. Phys. 73, (2010), arXiv:0911.5004v2[gr-qc].
13. L. Susskind, The world as a hologram. J. Math. Phys. 36, 6377–6396, (1995),
arXiv:hep-th/9409089v2.
14. E. Verlinde, On the origin of gravity and the laws of Newton, JHEP
1104(029), (2011), arXiv:1001.0785v1 [hep-th].
15. C. Rovelli, General relativistic statistical mechanics, Phys. Rev. D 87, 084055,
(2013), arXiv:1209.0065v2 [gr-qc].
16. H.M. Haggard and C. Rovell, Death and resurrection of the zeroth principle
of thermodynamics, Phys. Rev. D 87, (2013), arXiv:1302.0724 [gr-qc].
17. P.F. De Córdoba, J.M. Isidro and M.H. Perea, Emergent quantum mechanics
as a thermal ensemble, Int. J. Geomet. Meth. Modern Phys. 11, 1–16 (2014).
18. J.D. Bekenstein, Energy cost of information transfer, Phys. Rev. Lett. 46,
623–626 (1981).
19. C. Eling, R. Guedens and T. Jacobson, Non-equilibrium thermodynamics of
spacetime, Phys. Rev. Lett. 96, 121301, (2006), arXiv:gr-qc/0602001v1.
20. L.M. Martyushev and V.D. Seleznev, Maximum entropy production principle
in physics, chemistry and biology, Phys. Rep. 426, 1–45 (2006).
21. E.T. Jaynes, Information theory and statistical mechanics, Phys. Rev. 106,
620–630 (1957).
22. E.T. Jaynes, Information theory and statistical mechanics II, Phys. Rev. 108,
171–190 (1957).
298 J. Munkhammar
23. E.T. Jaynes, Macroscopic prediction, in Complex Systems — Operational

Approaches in Neurobiology, ed. H. Haken, Springer-Verlag, Berlin, (1985),
pp. 254–269, ISBN 3-540-15923-1.
24. E.T. Jaynes, Gibbs vs. Boltzmann entropies, Am. J. Phys. 33, 391–398
(1965).
25. P. Attard, The second entropy, a variational principle for time-dependent
systems, Entropy 10, 380–391 (2008).
26. J.W. Lee, On the origin of entropic gravity and inertia, Found. Phys. 42,
1153–1164 (2012).
27. G.W. Gibbons and S.W. Hawking, Action integrals and partition functions
in quantum gravity, Phys. Rev. D 15, 2752–2756 (1977).
28. C. Rovelli, Relational quantum mechanics, Int. J. Theor. Phys. 35, 1637–
1678, (1996), arXiv:quant-ph/9609002.
www.ebook3000.com
Chapter 11
World Crystal Model of Gravity

Hagen Kleinert
Institut für Theoretische Physik, Freie Universität Berlin 14195 Berlin,
Germany
ICRANeT Piazzale della Repubblica, 10-65122 Pescara, Italy
h.k@fu-berlin.de
We show that a crystalline model with defects can reproduce Einstein’s

theory of gravity down to the Planck scale.
1. Introduction
The natural length scale of gravitational physics is the Planck length
lP ≈ 1.616 × 10−33 cm formed from combinations of Newton’s gravita-
tional constant GN , the light velocity c, and Planck’s constant . It is the
Compton wavelength
lP ≡ /mP c (1)
associated with the Planck mass
mP ≈ 2.177 × 10−5 g = 1.22 × 1022 MeV/c2 . (2)
The Planck length is an extremely small quantity which presently lies

beyond any experimental resolution and will probably be so in the not too
distant future. Particle accelerators are presently able to probe distances
which are still 10 orders of magnitude larger than lP . Considering the fast
growing costs of accelerators with energy, it is unimaginable that they will
get close to the Planck length for many generations to come. This length
may therefore be considered as the shortest length accessible to experimen-
tal physics.
It has been known for a long time that space–time with curvature and
torsion may be imagined as a world crystal of lattice spacing lP with defects
without contradicting any experiment. In fact, it makes no physical sense to
299
300 H. Kleinert
produce theories which predict properties of the universe at smaller length

scales. Since the times of Galileo Galilei, such theories fall into the realm
of philosophy, or even religion. The history of science shows us that nature
has always surprised us with new discoveries as observations invaded into
shorter and shorter distances. So far, all theories in the past which claimed
for a while to be theories of everything have been falsified by completely
unexpected discoveries.
The presently most popular examples for the claim to be the theory
of everything are string theories. Mathematically, they describe the physics
ranging from cosmological length scales down to the trans-Planckian regime
of zero length. In the experimentally accessible energy range, these theories
require a large number of extra space–time dimensions which have so far
unobserved effects. Unfortunately, they also predict many particles which
are not found in nature. In particular, the assumption of a string represent-
ing fundamental particles makes only sense if there are overtones, which any
string must have. In string theory, these overtones lie all in the inaccessible
Planck regime. Being unable to explain correctly the observed low-energy
particle spectra, it is unclear how anybody can believe the predictions for
this ultra-high energy regime.
One of the most important features of string theories is that they predict
the validity of Lorentz invariance at all energies in the trans-Planckian
regime. In this chapter, we would like to point out that at this level of
speculations, an entirely different scenario is possible.
2. Gravity in the World Crystal

Let us suppose, just for fun, that we live in a world crystal with a lattice con-
stant of the order of the Planck length [1]. Up to now, we would have been
unable to notice this. And this would remain so for a long time to come.
None of the present-day relativistic physical laws would be observably vio-
lated. The gravitational forces and their geometric description would arise
from variants of the plastic forces in this world crystal . The observed cur-
vature of space–time would be just a signal of the presence of disclinations
in the crystal. Matter would be sources of disclinations [2].
For simplicity, we shall present such a construction only for a system
without torsion. If the world crystal is distorted by an infinitesimal dis-
placement field
xµ → xµ = xµ + uµ (x), (3)
www.ebook3000.com
it has a strain energy

µ
A= d4 x (∂µ uν + ∂ν uµ )2 , (4)
4
where µ is some elastic constant. We assume the second possible elastic con-
stant, the Poisson ratio, to be zero, to shorten calculations. If the distortions
are partly plastic, the world crystal contains defects defined by Volterra sur-
faces, where crystalline sections have been cut out. The displacement field
is multivalued [3]. The Euclidean action of the world crystal becomes the
space–time generalization of the crystal energy (without the λ-term (4)):

A = µ d4 x (uµν − upµν )2 , (5)
where uµν ≡ (∂µ uν +∂ν uµ )/2 is the elastic strain tensor, and upµν the plastic
strain tensor. As explained in the textbook [3], the plastic strain tensor is
a gauge field of plastic deformations. The energy density is invariant under
the single-valued defect gauge transformations
uµν p → uµν p + (∂µ λν + ∂ν λµ )/2, 3uµ → uµ + λµ . (6)
Physically, they express the fact that defects are not affected by elastic
distortions of the crystal. Only multivalued gauge functions λµ change the
defect content in the plastic gauge field upµν .
We now rewrite the action (5) in a canonical form by introducing an
auxiliary symmetric stress tensor field σµν as

1
A = d3 x σµν σ µν + iσ µν (uµν − upµν ) . (7)
4µ
After a partial integration and extremization in uµ , the second term in the
action yields the equation
∂ν σ µν = 0. (8)
This may be guaranteed identically, as a Bianchi identity, by an ansatz

σµν = µ κλσ ν κλ σ ∂λ ∂λ χσσ . (9)
Inserting (9) into (7), we obtain

4 1 µκλσ νκλ σ 2 νκλσ µκλ σ p
A= d x [ ∂λ ∂λ χσσ ] − i ∂λ ∂λ χσσ uµν .
4µ
(10)
302 H. Kleinert
A further partial integration brings this to the form

4 1 µκλσ νκλ σ 2 σκλν σ κλ µ p
A= d x [ ∂λ ∂λ χσσ ] − iχσσ ∂λ ∂λ uµν .
4µ
(11)
This action is now double-gauge theory invariant under the defect gauge
transformation (6), and under stress gauge transformations
χστ → χστ + ∂σ Λσ + ∂τ Λσ . (12)
The action can further be rewritten as

4 1 µν µν
A= d x σµν σ − iχµν η , (13)
4µ
where ηµν is the four-dimensional (4D) extension of the standard defect
density ηij of plasticity theory [3]

ηµν = µ κλσ νκ λ τ ∂λ ∂λ upστ . (14)
This is invariant under defect gauge transformations (6), and satisfies the
conservation law
∂ν η µν = 0. (15)
We may now replace upσσ by half the metric field gµν , and we recognize
the tensor ηµν as the Einstein tensor associated with the metric tensor gµν .
Let us eliminate the stress gauge field from the action (13). For this
we use a standard identity for the product of two Levi–Civita tensors, and
rewrite the stress field (9) as

σµν = µ κλσ ν κλ τ ∂λ ∂λ χστ
= −(∂ 2 χµν + ∂µ ∂ν χλ λ − ∂µ ∂λ χµ λ − ∂ν ∂λ χµ λ )
+ ηµν (∂ 2 χλ λ − ∂λ ∂κ χλκ ). (16)
Introducing the field φµ ν ≡ χµ ν − 12 δµ ν χλ λ , and going to the Hilbert gauge

∂ µ φµ ν = 0, the stress tensor reduces to
σµν = −∂ 2 φµν (17)
and the action of an arbitrary distribution of defects becomes

4 1 2 µν 2 ν µ 1 µ λ
A= d x ∂ φ ∂ φµν + iφµ η ν − δ ν η λ . (18)
4µ 2
www.ebook3000.com
Extremization with respect to the field φµν yields the interaction of an

arbitrary distribution of defects

1
A = µ d4 x (η µ ν − 12 δ µ ν η λ λ ) 2 2 (ηµ ν − 12 δµ ν η λ λ ). (19)
(∂ )
This is not the Einstein action for a Riemann space–time. It would be

so if the derivatives (∂ 2 )2 in the denominator would be replaced by
σµν (−∂ 2 )σ µν , which would change (∂ 2 )2 in (19) into −∂ 2 . An index rear-
rangement would then lead to the interaction

1
A = µ d4 x (η µ ν − 12 δ µ ν η λ λ ) ηµ ν . (20)
−∂ 2
The defect tensor ηµν is composed of the plastic gauge fields upµν in the same
way as the stress tensor is in terms of the stress gauge field in Eq. (16), i.e.

ηµν = µ κλσ ν κλ τ ∂λ ∂λ upστ
= −(∂ 2 upµν + ∂µ ∂ν upλ λ − ∂µ ∂λ upµ λ − ∂ν ∂λ upµ λ )
+ ηµν (∂ 2 upλ λ − ∂λ ∂κ upλκ ). (21)
If we introduce the auxiliary field wµp ν ≡ upµ ν − 12 δµ ν upλ λ , and chose the
Hilbert gauge ∂ µ wµν
p
= 0, the defect density reduces to
ηµν = −∂ 2 wµν
p
, ηµ ν − 12 δµ ν η λ λ = −∂ 2 upµν (22)
and the interaction (20) of an arbitrary distribution of defects would become

A=µ d4 x up µν (x)η µν (x). (23)
This coincides with the linearized Einstein–Hilbert action

1 √
A=− d4 x −gR̄, (24)
2κ
if we identify the elastic constant µ with 1/4κ, where κ is the gravita-

tional constant. Indeed, in the linear approximation gµ ν = δµ ν + g µ ν with
|g µ ν | 1, where the Christoffel symbols can be approximated by
1

Γ̄µν λ ≈ ∂µ g νλ + ∂ν g µλ − ∂λ g µν (25)
2
304 H. Kleinert
and the Riemann curvature tensor becomes

1
R̄µνλκ ≈ ∂µ ∂λ g νκ − ∂ν ∂κ g µλ − (µ ↔ ν) . (26)
2
This gives the Ricci tensor
1
R̄µκ ≈ (∂µ ∂λ hλκ + ∂κ ∂λ g λµ − ∂µ ∂κ g − ∂ 2 g µκ ), (27)
2
where g ≡ g λ λ is the trace of the tensor g µν . The associated scalar cur-
vature is
µν
R̄ ≈ −(∂ 2 g − ∂µ ∂ν g ). (28)
In combination with (27), we obtain the Einstein tensor
1
Ḡµκ = R̄µκ − gµκ R̄
2

1 2
λ λ
≈− ∂ g µκ + ∂µ ∂κ g − ∂µ ∂λ g κ − ∂κ ∂λ g µ
2
1 νλ
+ ηµκ (∂ 2 g − ∂ν ∂λ g ). (29)
2
This can be written as a 4D version of a double curl
1
Ḡµκ = µδ νλ κ δστ ∂ν ∂σ g λτ . (30)
2
Thus, the Einstein–Hilbert action has the linear approximation

1
A≈ d4 x g µν Gµν . (31)
4κ
Recalling the previously established identifications of plastic field and
defect density with metric and Einstein tensor, respectively, the interaction
between defects (23) is indeed the linearized version of the Einstein–Hilbert
action (24) if µ = 1/4κ.
The world crystal with the elastic energy (5) does not lead to the inter-
action (20). But it is easy to modify it to do so. We may simply introduce
two more derivatives and assign to the world crystal the higher-gradient
elastic energy

A = µ d4 x [∂(uµν − upµν )]2 .

(32)
In the canonical form (7) of the energy it changes the term σµν σ µν to
σµν (−∂ 2 )σ µν . This has the desired effect of removing one power of −∂ 2
from the denominator in the interaction (19).
www.ebook3000.com
A world crystal whose elastic energy is governed by the higher-gradient

action (32) will be called floppy world crystal . It is really the analogue of
a liquid crystal [2]. Thus, we have shown that defects in the floppy world
crystal are capable of creating a Riemannian space–time with an action of
the Einstein–Hilbert type.
The floppyness of the world crystal has an important physical conse-
quence. If we calculate the correlation function of the displacement field
uµ (x),

d4 k 1 ik(x−x )
uµ (x)uν (x ) = gµν e
(2π)4 k 4
1
≈ − 2 log(|x − x |a), (33)
8π
we see that its long-distance behavior is logarithmic in |x − x |. This implies
that the field uµ (x) is fluctuating violently, so violently that over long dis-
tances the crystal orientation is completly lost. This phenomenon is familiar
from ordinary elastic crystals in two dimensions. They have 1/k 2 -correlation
function in Fourier space, so that the spatial correlation function

d2 k 1 ik(x−x )
ui (x)uj (x ) = δij e
(2π)2 k 2
1
≈− log(|x − x |a), (34)
2π
has a logarithmic behavior. This implies that two-dimensional (2D) crystals
can be in an ordered state only if they are allowed to perform vertical
fluctuations into the third dimension. Otherwise, the crystalline order exists
only at short distance, and they do not exhibit the typical δ-function-like
Bragg peaks, but only power-like peaks.
Another famous example is provided by the phase fluctuations of the
order parameter in thin layers of superfluid helium. These destroy a properly
ordered uniform phase at low temperatures, and lead to what is called
a quasi-long-range-ordered state [4, 5], in a so-called Kosterlitz–Thouless
transition.
For more work on the world-crystal, see Refs. [6–13] and the Wikpedia
remarks on the term World-Crystal [14].
References
1. H. Kleinert, Gravity as theory of defects in a crystal with only second-
gradient elasticity, Ann. der. Physik 44, 117 (1987) (kl/172).
306 H. Kleinert
2. For an attempt to explain why no torsion is found in the world crystal see
H. Kleinert and J. Zaanen, World nematic crystal model of gravity explaining
the absence of torsion, Phys. Lett. A 324, 361 (2004), gr-qc/0307033.
3. H. Kleinert, Multivalued Fields in Condensed Matter, Electromagnetism, and
Gravitation, World Scientific, Singapore, 2008, pp. 1–497, http://klnrt.
de/b11.
4. V.L. Berezinskii, Sov. Phys. JETP 32, 493 (1971), Sov. Phys. JETP 34, 610
(1972); J.M. Kosterlitz and D.J. Thouless, J. Phys. C 6, 1181 (1973).
5. H. Kleinert, Gauge fields in Condensed Matter , Vol. I: Superflow and Vortex
Lines, Disorder Fields, Phase Transitions, World Scientific, Singapore, 1989,
kl/b1.
6. M. Danielewski, The Planck–Kleinert crystal, Z. Naturfoschung 62a, 546
(2007).
7. H. Kleinert, New Gauge Symmetry and the Evanescent Role of Torsion in
Gravity, Berlin preprint 2010.
8. H. Kleinert, Phys. Lett. A 130, 443 (1988), http://klnrt.de/174.
9. F.J. Belinfante, Physica 6, 887 (1939). For more details see Sec. 17.7 in the
textbook [3].
10. See einstein.stanford.edu.
11. H. Kleinert, Gen. Rel. Grav. 32, 1271 (2000), http://klnrt.de/271/271j.
pdf.
12. A.D. Sakharov, DAN SSSR 177, 70 (1967). Reprinted in Gen. Rel. Grav. 32,
365 (2000).
13. A cosmological model based on Sakharov’s idea is discussed in H. Kleinert and
H.-J. Schmidt, Cosmology with curvature-saturated gravitational Lagrangian
R/(1 + l4 )1/2 , Gen. Rel. Grav. 34, 1295 (2002), gr-qc/0006074.
14. For cosmological data see the internet page
http://super.colorado.edu/~michaele/Lambda/links.html.
www.ebook3000.com
Chapter 12
Quantum Features of Natural

Cellular Automata
Hans-Thomas Elze
Dipartimento di Fisica “Enrico Fermi”
Largo Pontecorvo 3, 56127 Pisa, Italia
elze@df.unipi.it
We review the properties of discrete and integer-valued, hence “nat-

ural”, cellular automata (CA), a particular class of which comprises
“Hamiltonian CA” with equations of motion that bear strong similari-
ties to Hamilton’s equations, despite presenting discrete updating rules.
The resulting dynamics is linear in the same sense as unitary evolution
described by the Schrödinger equation. Employing Shannon’s Sampling
Theorem, we construct an invertible map between such CA and con-
tinuous quantum mechanical models which incorporate a fundamental
discreteness scale. This leads to one-to-one correspondence of quantum
mechanical and CA conservation laws. In order to illuminate the all-
important issue of linearity, we presently introduce an extension of the
class of CA incorporating nonlinearities. We argue that these imply non-
local effects in the continuous quantum mechanical description of intrin-
sically local discrete CA, enforcing locality entails linearity. We recall
the construction of admissible CA observables and the existence of solu-
tions of the modified dispersion relation for stationary states, besides dis-
cussing next steps of the deconstruction of quantum mechanical models
in terms of deterministic CA.
1. Introduction
A novel interpretation of quantum mechanics (QM) — which amounts to
the attempt to redesign the foundations of quantum theory in accordance
with “classical” concepts, foremost with determinism — has recently been
laid out by G. ’t Hooft [1]. The hope for a comprehensive theory expressed
307
308 H.-T. Elze
in this far-reaching article is founded on the observation that quantum

mechanical features arise in a large variety of deterministic “mechanical”
models. While practically all of these models have been singular cases, i.e.
which cannot easily be generalized to cover a realistic range of phenom-
ena incorporating interactions, cellular automata (CA) promise to provide
the necessary versatility, as we shall discuss in the following [2, 3]. For an
incomplete list of various earlier attempts in this field, see, for example,
Refs. [4–12] and further references therein.
The linearity of QM is a fundamental feature most notably embodied
in the Schrödinger equation. This linearity does not depend on the partic-
ular object under study, provided it is sufficiently isolated from anything
else. It is naturally reflected in the superposition principle and entails such
“quantum essentials” as interference and entanglement.
The linearity of QM has been questioned repeatedly and nonlinear mod-
ifications have been proposed not only as suitable approximations for com-
plicated many-body dynamics, but especially in order to test experimentally
the robustness of QM against such nonlinear deformations. This has been
thoroughly discussed by T.F. Jordan who presented a proof ‘from within’
quantum theory that the theory has to be linear, given the essential separa-
bility assumption “. . . that the system we are considering can be described
as part of a larger system without interaction with the rest of the larger
system [13].”
Recently, we have considered a seemingly unrelated discrete dynamical
theory, i.e. which deviates drastically from quantum theory, at first sight.
However, we have shown with the help of sampling theory that the deter-
ministic mechanics of the class of Hamiltonian CA can be related to QM
in the presence of a fundamental time scale. This relation demonstrates
that consistency of the action principle of the underlying discrete dynamics
implies, in particular, the linearity of both theories. We will briefly review
the results obtained and discuss in more detail the undesirable consequences
that follow immediately, if one generalizes the action principle and, thus,
enlarges the class of CA to be admitted.
Our CA approach may offer additional insight into interference and
entanglement in the limit when the discreteness scale is negligible. Further-
more, future developments are conceivable which can address the dynam-
ics of QM measurement processes about which standard quantum theory
remains notouriously silent.
www.ebook3000.com
Quantum Features of Natural CA 309
2. The Cellular Automaton Action Principle

The state of a classical CA with countably many degrees of freedom (DOF)
is described by discrete integer-valued “coordinates” xα
n , τn and “conjugated
α
momenta” pn , πn , where α ∈ N0 denote different DOF and n ∈ Z differ-
ent states. xn and pn might be higher-dimensional vectors, while τn and
Pn are assumed one-dimensional. The “coordinate” τn has been separated
from the xα α
n ’s (correspondingly πn from the pn ’s), since this DOF repre-
sents a dynamical time variable, discussed in Refs. [2,3,14,15], with further
references given there.
For any one of the dynamical variables, say fn , the finite difference
operator ∆ is defined by:
∆fn := fn − fn−1 . (1)
Furthermore, we introduce the quantities (assuming the summation

convention for Greek indices, rα sα ≡ α rα sα ):
an := cn πn , (2)
1
Hn := Sαβ (pα β α β α β
n pn + xn xn ) + Aαβ pn xn + Rn , (3)
2
An := ∆τn (Hn + Hn−1 ) + an , (4)
where the constants, cn , the symmetric, Ŝ ≡ {Sαβ }, and the antisymmetric,

Â ≡ {Aαβ }, matrices are all integer-valued; Rn stands for higher than
second powers in xα α
n or pn . The definition (2) determines the behavior of
the variable τn and only the most simple choice (involving a single constant)
will be relevant for our present purposes, cf. below.
In terms of these definitions, the integer-valued CA action is given by:

S := [(pα α α
n + pn−1 )∆xn + (πn + πn−1 )∆τn − An ]. (5)
n
Furthermore, we introduce integer-valued variations δfn which are applied

to a polynomial g as follows:
δfn g(fn ) := [g(fn + δfn ) − g(fn − δfn )]/2δfn (6)
and δfn g ≡ 0, if δfn = 0. Then, the CA dynamics is determined by postu-

lating the following variational principle.
The CA Action Principle: The discrete evolution of a CA is determined by
the stationarity of its action under arbitrary integer-valued variations of all
dynamical variables, δS = 0.
310 H.-T. Elze
We observe some characteristic aspects of this CA Action Principle:

(i) Variations of terms that are constant, linear, or quadratic in integer-
valued variables yield analogous results as standard infinitesimal vari-
ations of corresponding terms in the continuum.
(ii) While infinitesimal variations do not conform with integer-valuedness,
there is a priori no restriction of integer variations, hence arbitrary
integer-valued variations must be admitted.
(iii) However, for arbitrary variations δfn , the remainder of higher powers
Rn in Hn , which ultimately enters the action, has to vanish for con-
sistency. Otherwise, the number of equations of motion generated by
variation of the action, according to Eq. (6), would exceed the number
of variables. (Yet, a suitably chosen R0 or a sufficient small number
of such remainder terms can serve to encode the initial conditions for
the CA evolution.)
We may anticipate here that the features (i)–(iii) turn out to be essential
in constructing a map between the considered CA and equivalent quantum
mechanical models. Therefore, we will further discuss these points in Sec. 4.
by studying a generalization of the variational principle incorporating non-
linearities and the, generally, undesirable consequences.
2.1. The discrete updating rules (CA equations

of motion)
Employing the notation Ȯn := On+1 − On−1 , discrete analogues of Hamil-
ton’s equations follow from the variations of the action S (keeping Rn ≡ 0):
ẋα β β
n = τ̇n (Sαβ pn + Aαβ xn ), ṗα β β
n = −τ̇n (Sαβ xn − Aαβ pn ), (7)
τ̇n = cn , π̇n = Ḣn , (8)
where all terms are integer-valued. Discreteness of the automaton time n is

reflected by finite difference equations here. Their appearance has motivated
the name Hamiltonian CA.
Equation (7) is time reversal invariant; the state n + 1 can be calcu-
lated from knowledge of the earlier states n and n − 1 and the state n − 1
from the later ones n + 1 and n. Note that τ̇n presents parameters for the
evolving x, p-variables, as a consequence of Eq. (8). More generally, τ̇ can
be a dynamically coupled lapse function in Eq. (7).
Surprisingly, there are conservation laws respected by the discrete equa-
tions of motion (7), which are in one-to-one correspondence with those of the
www.ebook3000.com
related Schrödinger equation in the continuum, as we shall see. Introducing

the self-adjoint matrix Ĥ := Ŝ + iÂ, these equations can be combined into:
ẋα α β β
n + iṗn = −iτ̇n Hαβ (xn + ipn ) (9)
and its adjoint. Thus, we recover a discrete analogue of Schrödinger’s equa-

tion, with ψnα := xα α
n + ipn as the amplitude of the “α-component” of “state
vector” |ψ at “time” n and with Ĥ as the pertinent Hamilton operator.
(We continue to use the QM terminology freely in the following and con-
centrate on new effects arising here.) Then, Eq. (7) implies the following
theorem.
Theorem A. For any matrix Ĝ that commutes with Ĥ, [Ĝ, Ĥ] = 0, there
is a discrete conservation law:
ψn∗α Gαβ ψ̇nβ + ψ̇n∗α Gαβ ψnβ = 0. (10)
For self-adjoint Ĝ, with complex integer elements, this relation concerns
real integer quantities.
Corollary A. For Ĝ := 1̂, Eq. (10) implies a conserved constraint on the
state variables:
ψn∗α ψ̇nα + ψ̇n∗α ψnα = 0. (11)
For Ĝ := Ĥ, an energy conservation law follows.

We remark that Eqs. (10) and (11) cannot be “integrated” in the usual
way, since the Leibniz rule is modified in the present context. Recalling

Ȯn := On+1 − On−1 , we have, for example, On+1 On+1 − On−1 On−1 =
1
2 (Ȯn [On+1 + O n−1 ] + [On+1 + On−1 ]Ȯn ), instead of the product rule of
differentiation.
Furthermore, note that a continuum limit of the equations of motion
and related conservation laws do not simply follow from letting the dis-
creteness scale l → 0. This is obstructed by the integer-valuedness here,
which conflicts with continuous time and related derivatives. This is one of
the reasons why a more elaborate mapping is needed in order to identify
CA with continuum models, in particular, quantum mechanical ones (cf.
Sec. 3).
2.2. The CA action and shadows of unitary symmetry

The CA action is invariant under suitable unitary transformations. This is
easily recognized by considering an equivalent form of the action, i.e. which
312 H.-T. Elze
generates the same discrete equations of motion as before. The definition

of Eq. (5) can be replaced by:

S := [Im(ψnα ∗ ψn−1
α
) + (πn + πn−1 )∆τn − An ] (12)
n
with ψnα
:= xα α
and ImX := (X − X ∗ )/2i, together with a correspond-
n + ipn
ingly modified expression for Hn , Eq. (3), given by (recall Rn ≡ 0):
1
Hn := Hαβ ψnα ∗ ψnβ (13)
2
which enters the action through An , Eq. (4). Then, n-independent unitary
transformations Û , with ψn = Û ψn and [Û , Ĥ] = 0, leave the action S
invariant.
The self-adjoint matrices Ĝ featuring in Theorem A generate unitary
transformations which leave S invariant. However, the CA variables ψnα :=
xα α
n + ipn are complex integer-valued, by definition. Therefore, only those
transformations are admissible which preserve this property. Thus, we find
here CA conservation laws, according to Theorem A or Corollary A, which
are related to discrete sets of admissible symmetry transformations. They
may form discrete subgroups of the familiar continuous symmetry groups
in the case of QM.
2.3. Discrete Poisson brackets and CA observables

Finally, another surprising parallel exists between the discrete and contin-
uum models, CA and QM, respectively, in the form of a Poisson bracket.
It may be defined with respect to the dynamical variables of the CA, which
are denoted collectively here by X, P and which represent the xα n , τn and
pα
n , πn , respectively:
{A, B} := δX A δP B − δX B δP A, (14)
where A and B are polynomials depending on the dynamical variables and

it is tacitly assumed that the expressions are to be evaluated at a partic-
ular CA state n. We employ the variational derivative defined in Eq. (6),
since discreteness prevents the use of ordinary derivatives incorporating
infinitesimals, as before.
Now, inspection shows that such polynomials A and B cannot be arbi-
trarily chosen — generally, the bracket would not be consistent, namely,
fail on one or the other of the defining properties of a Lie bracket opera-
tion: bilinearity, antisymmetry, derivation-like product formula, and Jacobi
www.ebook3000.com
identity. The problem arises from the fact that the result of the bracket
operation might depend on the integer-valued variations δf , which enter
through the definition of the variational derivative, Eq. (6). This would
prohibit a closed algebra of polynomials. However, recalling observation
(i) following the CA Action Principle above, we may restrict the polyomi-
als to be constant, linear, or quadratic (in dynamical variables). Thus, they
form a closed algebra with respect to the bracket operation, which becomes
fully consistent in this way.
These observations are remarkable for two reasons. First, the Hamilton
operator Ĥ, introduced with Eq. (9), defines a quadratic form Hn in terms
of the xα α
n and pn , which is compactly written in Eq. (13). It is analogous
to the expectation ψ|Ĥ|ψ in QM, which can always be reexpressed in the
X, P phase space representation developed by Heslot [16] that underlies
our approach. This expectation belongs to the observables of a quantum
mechanical object and may correspondingly be assumed to belong to the
CA observables as well. This seems justified, in particular, since Eq. (9)
amounts to ψ̇ α = τ̇ {ψ α , H}, evaluated at a particular CA state n. Second,
defining and, thus, restricting CA observables as quadratic forms in the
dynamical X, P variables (eliminating constant and linear forms that would
lead to inhomogeneous evolution equations), we arrive at a closed algebra
of observables with respect to the Poisson bracket operation (14).
We recall that all quantum mechanical observables are generated by
Hermitean operators in this way as quadratic forms [16]. Thus, insisting
on the Hamiltonian structure of CA dynamics, including a suitably defined
Poisson bracket, we are able to extend the close correspondence between CA
and quantum mechanical systems to cover also the structure of observables.
In Sec. 3, it will be shown that this correspondence is not accidental
and can be understood with the help of an invertible map that relates
the descriptions of Hamiltonian CA and of quantum mechanical objects
incorporating a fundamental discreteness scale l.
3. Sampling Theory and the CA ↔ QM Map

Despite the similarities of the Hamiltonian CA with QM systems that we
have observed, we may wonder whether the discreteness of a CA can be
reconciled with a continuum description at all and, in particular, with QM?
We have argued earlier that physical fields, including wave functions,
could be simultaneously discrete and continuous, represented by sufficiently
smooth functions containing a finite density of DOF [2, 3]. Related ideas
314 H.-T. Elze
have been discussed by Lee and collaborators and more recently by Kempf
in attempts to introduce a covariant ultraviolet cut-off for quantum field
theories and, last not least, for gravity, see, for example, Refs. [14, 17] with
further references there and in Refs. [2, 3]. However, neither integer-valued
CA nor the structure of QM had been addressed in this context.
That information can have simultaneously continuous and discrete char-
acter has been pointed out by Shannon in his pioneering work [18] and is
routinely applied in signal processing, converting analog to digital encoding
and vice versa. Sampling theory demonstrates that a bandlimited signal can
be perfectly reconstructed, provided discrete samples of it are taken at the
rate of at least twice the band limit (Nyquist rate). For an extensive review
of sampling theory, see Ref. [19].
We recall the basic version [19] of the Sampling Theorem: Consider
square integrable bandlimited functions f , i.e. which can be represented
ω
as f (t) = (2π)−1 −ωmax max
dω e−iωt f˜(ω), with bandwidth ωmax . Given the
set of amplitudes {f (tn )} for the set {tn } of equidistantly spaced times
(spacing π/ωmax ), the function f is obtained for all t by:
sin[ωmax (t − tn )]
f (t) = f (tn ) . (15)
n
ωmax (t − tn )
Since the CA state is labeled by the integer n, the automaton time, the
corresponding discrete physical time is obtained by multiplying with the
fundamental scale l, tn ≡ nl, and the bandwidth by ωmax = π/l.
When attempting to map invertibly Eq. (7) on reconstructed continuum
equations, according to Eq. (15), the nonlinearity on the right-hand side is
problematic, since the product of two functions with bandwidth ωmax each
is not a function with the same bandwidth. It is sufficient to assume here
that τ̇n is a constant. While we postpone a discussion of nonlinearities to
Sec. 4.
Let us recall the discrete time Eq. (9). Inserting ψnα := xα α
n + ipn and
applying the Sampling Theorem, we obtain the equivalent continuous time
equation:
1
(D̂l − D̂−l )ψ α (t) = 2 sinh(l∂t )ψ α (t) = Hαβ ψ β (t), (16)
i
where we employ the translation operator defined by D̂T f (t) := f (t + T )
and have implemented the natural choice τ̇n ≡ 1, for all n.
In this way, we recover the Schrödinger equation. However, it is modified
in important ways as given in the following paragraphs.
www.ebook3000.com
First of all, by construction, the continuous time wave function ψ α is

bandlimited (by ωmax ). Therefore, knowing the wave function at the discrete
times of a set {t0 + nl|n ∈ Z}, with t0 arbitrary, it can be reconstructed for
all times by a slight generalization of Eq. (15). Furthermore, since Eq. (16)
is of the form f (t + l) = f (t − l) − iĤf (t), it is sufficient to know f at
two times, say t0 and t0 − l, in order to obtain it for all times of the set
{t0 + nl|n ∈ Z}. Thus, we learn that two initial conditions (separated by
a time step l) have to be specified to define the solution of Eq. (16). This
agrees precisely with the requirements of the discrete description of the CA,
cf. Sec. 2.1. In order that sampling reproduces the integer-valuedness of the
CA, both initial values have to be integer-valued.
If instead the modified Schrödinger equation is written in terms of the
infinite series of odd powers of time derivatives, it might give the false
impression that an infinity of initial conditions is required. In any case,
the higher-order derivatives are negligible for low-energy wave functions,
which vary little with respect to the cutoff scale, i.e. |∂ k ψ/∂tk | l−k =
(ωmax /π)k .
Secondly, the bandlimit |ω| ≤ ωmax leads to an ultraviolet cut-off of the
energy E of stationary states of the generic form ψE (t) := exp(−iEt)ψ̃.
Diagonalizing the self-adjoint Hamiltonian, Ĥ → diag(0 , 1 , . . . ), Eq. (16)
yields the eigenvalue equation:
2 sin(Eα l) = α (17)
and a modified dispersion relation, Eα = l−1 arcsin(α /2) = (2l)−1 α [1 +

α2 /24 + O(α4 )]; e.g. if α labels the momentum modes of a given spatial
lattice. Thus, the spectrum {Eα } is cut off by the condition |α | ≤ 2,
entailing |Eα | ≤ π/2l = ωmax /2, i.e. half the bandlimit.
Tracing our argument backwards, the precise form of the spectral cut-
off or of the left-hand side of Eq. (16) depends, of course, on the left-hand
sides of the finite difference equations of motion obtained in Sec. 2.1. Thus,
it pertains to the class of Hamiltonian CA selected by our Action Principle.
In any case, the cut-off introduced through the eigenvalue Eq. (17) is not
as restrictive as it may seem at first sight. Most importantly, discrete Hamil-
tonians do indeed exist which have their spectrum bounded between −2 and
+2, such that Eq. (17) has real solutions. A complete classification of such
integer-valued symmetric matrices has recently been given with an inter-
esting reformulation as the problem of classifying certain “charged signed
graphs”, and comprises infinite families of arbitrarily large matrices as well
as a number of sporadic small ones [20]. Partial results are also available for
316 H.-T. Elze
self-adjoint complex integer matrices. This is certainly a subject for future

extension and physical interpretation, in particular, considering the subset
of matrices that are adjacency matrices of certain graphs.
Interestingly, the quantum harmonic oscillator derived from a cellu-
lar automaton — a ‘particle’ performing uniform jumps on a periodic
one-dimensional lattice of N sites [21], see also Refs. [8, 22] — provides
an example of a quantum model emerging from an automaton which is not
covered by the above eigenvalue equation, since the spectrum of the dimen-
sionless Hamiltonian (before considering any form of continuum limit) is
given by:

2π 1
m = m+s+ , (18)
2s + 1 2
where 2s + 1 ≡ 2N and −s ≤ m ≤ s, for half-integer m. Clearly, there

are always values m such that |m | > 2, even if one shifts the spectrum
by a constant term. This may remind us that more general than Hamilto-
nian CA could be quite relevant for emerging QM features and need to be
studied.
Finally, we remark that Planck’s constant does not interfere with our
map between discrete and continuum dynamics and remains independent
of the discreteness scale l. To illustrate this, we rewrite the Schrödinger
equation as i∂t ψ = phys ĥψ, where by phys we factor out a physical energy
scale, such that the dimensionless Hamiltonian ĥ is given by numbers that
are (loosely speaking) “of O(1)”. Rescaling the time variable t /M =: t,
with some M 1, we obtain: i∂t ψ = phys M ĥψ = l−1 M ĥψ, where we
introduced the bandwidth limit, l−1 ≡ ωmax /π := phys M /M phys ,
for M M 1. At this point, units can be chosen such that = 1, as
usual. Thus, we arrive at il∂t ψ = M ĥψ, corresponding to the leading terms
of Eq. (16). The right-hand side here may be approximated by a complex
integer-valued Hamiltonian, Ĥ, producing errors in the matrix elements
“of O(1/M )” (loosely speaking), as compared to M ĥ. Similarly, one can
set up a possibly useful approximation scheme for quantum mechanical
calculations based on sampling theory, which maps dynamics described by
the Schrödinger equation on a cellular automaton [23].
3.1. Continuous time CA conservation laws

The correct continuous time conservation laws can be easily inferred by
observing that the Schrödinger equation (16) is obtained from the analogous
www.ebook3000.com
CA Eq. (9) through the replacement

1
ψ̇n := ψn+1 − ψn−1 −→ sin(il∂t )ψ(t). (19)
i
Applying this to Eqs. (10) and (11), respectively, we obtain:
Theorem B. For any matrix Ĝ with [Ĝ, Ĥ] = 0, there is a continuous time
conservation law:
ψ ∗α Gαβ sin(il∂t )ψ β + [sin(il∂t )ψ ∗α ]Gαβ ψ β = 0 (20)
and, in particular, for Ĝ := 1̂,
ψ ∗α sin(il∂t )ψ α + [sin(il∂t )ψ ∗α ]ψ α = 0, (21)
which expresses a modification of the conservation of the QM wave function
normalization, referring to a basis labeled by α.
This theorem is indeed easily verified with the help of Eq. (16). It is
now possible to remove the ultraviolet cut-off, with l → 0, thus recovering
the familiar QM results from the leading order terms in Eqs. (20) and (21).
If l is a fundamental constant, this limit leads to the regime where QM is
known to work perfectly fine.
To further illustrate these results, we consider the real symmetric two-
time function,
2CĜ (t2 , t1 ) := ψ ∗α (t2 )Gαβ ψ β (t1 ) + c.c., (22)
where X + c.c. := X + X ∗ and Ĝ is a self-adjoint matrix, with [Ĝ, Ĥ] = 0.
Applying Theorem B, we obtain:
Corollary B. The two-time function CĜ is invariant under discrete trans-
lations, t → t + l:
CĜ (t, t − l) = CĜ (t + l, t). (23)
Therefore, given CĜ (t, t − l) for all t ∈ [t0 , t0 − l[, with arbitrary t0 , it is
fixed everywhere.
In particular, if the two-time function is constant for t ∈ [t0 , t0 − l[,
for some t0 , it is constant everywhere. Furthermore, since the equations of
motion are linear, the wave function can be rescaled such that this constant
equals one, without changing the dynamics. The familiar wave function
normalization arises in this way from the coincidence limit of a two-time
function with the property C1̂ (t + l, t) ≡ 1, for all t:
1 = lim C1̂ (t + l, t) = ψ ∗α (t)ψ α (t) (24)
l→0
318 H.-T. Elze
which is consistent with Eq. (21) and becomes essential for the probabil-
ity interpretation in QM. An analogous equal-time constraint, instead of
Eq. (11), generally does not exist for the discrete CA description. The con-
straint ψn∗α ψnα = xα α α α
n xn + pn pn = 1, for example, is only compatible with
rather trivial evolution, since all variables are integer-valued.
It is remarkable how properties of Hamiltonian CA produce familiar
QM results, even if modified by the finite scale l. The operators or matrices
that generate the QM conservation laws do so for the bandwidth limited
continuum theory as well, as stated by Theorem B. Since the same vanish-
ing commutators are responsible for the CA conservation laws, Eqs. (10)
and (11), they correspond to each other one-to-one.
Yet, the QM symmetry transformations generally comprise a larger set
than the admissible discrete ones for CA, which have to respect complex
integer-valuedness of the dynamical variables, as we discussed in Sec. 2.2.
We are left here with the intriguing question: What is the physical inter-
pretation of conservation laws for Hamiltonian CA that are related to dis-
crete subgroups of the continuous symmetry groups of QM models?
4. On Nonlinear Hamiltonian CA
The general properties and the quantum features, in particular, of the
Hamiltonian CA that we discussed in the previous sections derive from
the CA Action Principle introduced in Sec. 2. One of the most important
aspects has been the linearity of the resulting equations of motion, Eq. (7)
or Eq. (9). We have argued that additional higher-order terms in the action,
which would lead to nonlinear (in xα α α
n , pn or ψn ) terms in the equations of
motion, would simultaneously enlarge the set of equations. In this way, the
CA dynamics tends to become overdetermined. In this section, we will take
another look at this problem.
4.1. A generalized variational derivative

The potentially troublesome additonal equations are related to two aspects
of the dynamics. One is the additional higher than quadratic powers of
dynamical variables in the action, the terms summarized by Rn in def-
inition (3). The other is the arbitrary integer-valued variations δfn of all
dynamical variables appearing in the action, which are invoked by the varia-
tional principle. They are applied as defined in Eq. (6) and this can produce
additional terms which involve powers of δfn . Since the coefficients of such
terms need to vanish independently, we are faced with additonal equations
of motion, possibly turning the whole set to become inconsistent.
www.ebook3000.com
The described problem can be avoided by a suitably generalized defini-

tion of the variations, cf. Eq. (6). We choose:

δf g (N ) (f ) := γk [g (N ) (f + mk δf ) − g (N ) (f − mk δf )]/2δf, (25)
k≥1
where f stands for a dynamical variable entering the N th degree polynomial

g (N ) and γk and mk (mk = mk , for k = k ) are constant real and positive
integer-valued coefficients, respectively, to be determined.
Our aim is to arrange these coefficients in such a way that δf g (N ) (f ) =
(N −1)
ḡ (f ), i.e. the variation results in a polynomial ḡ (N −1) of degree N − 1,
while all terms proportional to powers of δf cancel by construction. This
will be sufficient to eliminate the possibility of an overdetermined set of
(eventually nonlinear) equations of motion.
Writing the polynomial g (N ) explicitly,
g (N ) (f ) := g0 + g1 f 1 + · · · + gN f N , (26)
we expand the difference appearing in Eq. (25):

g (N ) (f + mk δf ) − g (N ) (f − mk δf ) /2
= g1 · mk δf + g2 · 2mk f δf

+g3 · 3mk f 2 δf + (mk δf )3

+g4 · 4mk f 3 δf + 4f (mk δf )3

+g5 · 5mk f 4 δf + 10f 2 (mk δf )3 + (mk δf )5
+··· . (27)
We observe that the terms ∝ δf correspond to the ones obtained by ordi-

nary differentiation of the polynomial. However, since any dynamical vari-
able f of our CA is integer-valued, such derivatives in the following are
to be interpreted as an appropriate shorthand notation. Thus, we have

δf g (N ) (f ) = k γk mk · (d/df )g (N ) (f ) + · · · , where the additional terms
involving powers of δf are not explicitly written. The latter can be made
to vanish always by suitably adjusting the coefficients γk and mk , if we
restrict the maximal order of polynomials to be dealt with.
We illustrate this by considering all polynomials of order ≤ 4, i.e. g (4) .
Here, the terms ∝ δf 3 are eliminated always, cf. Eq. (27), if the following
320 H.-T. Elze
condition is fullfilled:
!
γk (mk )3 = 0. (28)
k≥1
A solution is provided by m1 = 1, m2 = m ≥ 2, γ1 = 1/(1 − m−2 ),

γ2 = −m−3 /(1 − m−2), and all other coefficients vanishing. This results in:
δg (4) (f ) = (d/df )g (4) (f ), cf. Eq. (25). Other solutions are possible.
This elementary method to find a satisfactory variational derivative,
which does not lead to an overdetermined set of equations of motion, can
be generalized to polynomials of arbitrary order.
Consequently, by generalizing the variational derivative employed, the
CA Action Principle can be generalized such that consistent finite differ-
ence equations of motion incorporating nonlinear potentials result, which
maintain the similarity with Hamilton’s equations as in Sec. 2.1.
4.2. Some undesirable consequences of nonlinearity

The generalized variational derivative of Eq. (25) can be employed to define
a Poisson bracket, as before in Sec. 2.3. Since this variational derivative
effectively acts like an ordinary derivative, there is no need anymore to
restrict the algebra of observables to constant, linear or quadratic forms in
the dynamical variables. However, it remains consistent to do so, recover-
ing the previous results and analogous symmetry properties as in QM, in
particular.
Once higher order polynomials (in xα α α
n , pn or ψn ) are admitted in the
action and equations of motion or as observables, however, it does not
remain consistent to limit the set of relevant polynomials at any finite
order. For example, the Poisson bracket of two polynomials of order N and
N , respectively, may result in a polynomial of order N + N − 2 > N, N .
Thus, we observe here a qualitatively profound ‘bifurcation’ in the prop-
erties of Hamiltonian CA depending on whether nonlinearities are present in
their equations of motion or not. To emphasize this point, we note that the
discrete or continuous conservation laws and traces of QM unitary sym-
metry, discussed in Secs. 2.1 and 2.2, or 3.1, will generally be absent in
nonlinear CA.
Next, we turn to an illustration of the behaviour of nonlinear terms
under the map relating the discrete description of CA and its continuum
counterpart employing Shannon’s Sampling Theorem as in Sec. 3.
www.ebook3000.com
4.2.1. Useful properties of sinus cardinalis

We will make use of several results for the sinus cardinalis function,
sinc(x) := sin(x)/x, which enters the reconstruction formula, Eq. (15).
Introducing sn (t) := sinc[π(t/l − n)], its Fourier transform is:

∞
dt e−iωt sn (t) = lθ(π/l − ω)θ(π/l + ω)e−iωln , (29)
−∞
where θ denotes the Heaviside step function. Thus, the function sn is ban-
dlimited. Furthermore, it presents a “nascent” Dirac delta function, which
is characterized by:

∞
−1
l dt sn (t) = 1, (30)
−∞

∞
−1
lim l dt sn (t)f (t) = f (0). (31)
l→0
−∞
These results are easily obtained with the help of the Fourier transform
of sn , Eq. (29), also assuming that f has a well-behaved Fourier transform.
Employing the inverse Fourier transformation of Eq. (29), one demonstrates
the orthogonality relation:

∞
−1
l dt sm (t)sn (t) = δmn . (32)
−∞
Finally, we evaluate the following sum:

sn (ml − t )sm (t) = sn (t − t ), (33)
m∈Z
where we applied the Sampling Theorem, Eq. (15), noting that the factor
of sn under the sum can be interpreted as the function on the right-hand
side sampled at the times ml; all functions here have the same bandwidth
ωmax = π/l.
4.2.2. Nonlinearity leads to non-locality

Now, suppose that the discrete analogue of the Schrödinger equation,
Eq. (9), is modified by a genuine nonlinearity, such that we have instead:
ψ̇nα = −iHαβ ψnβ + Mαβγ (ψn∗β + ψnβ )(ψn∗γ + ψnγ ), (34)
322 H.-T. Elze
keeping τ̇n ≡ 1 and where the coefficients Mαβγ are real and totally sym-
metric in the indices. A corresponding potential can be incorporated into
the action, cf. Eqs. (12) and (13), and the modification of Eq. (34) follows
by applying a suitably generalized variational derivative, cf. Sec. 4.1.
Here we are not interested in the physical (ir) relevance of such a
cubic potential, which serves only as an example for the following obser-
vations. We would like to see what happens with the nonlinear terms, e.g.
Mαβγ ψnβ ψnγ , when the Sampling Theorem is applied to Eq. (34), similarly
as before in Sec. 3.
Omitting irrelevant greek indices, we consider ψn =: ψ(tn ) and ψn ψn =:
ψ(2) (tn ). Through the reconstruction formula (15) the discrete time val-
ues ψ(tn ) and ψ(2) (tn ) are replaced by continuous time functions ψ(t) and
ψ(2) (t), respectively. What is the relation between the latter?
To answer this, we employ the orthogonality relation (32), in order to
invert the reconstruction formula:

∞
−1
ψ(tn ) = l dt sn (t)ψ(t). (35)
−∞
This means simply:

∞
∞
ψ(2) (tn ) = l−2 dt sn (t )ψ(t ) dt sn (t )ψ(t ). (36)
−∞ −∞
Applying the reconstruction formula to ψ(2) (tn ) produces a relation between

ψ(2) (t) and ψ(t):

∞
∞
−2
ψ(2) (t) = l dt dt sn (t)sn (t )sn (t )ψ(t )ψ(t ), (37)
−∞ −∞ n∈Z
where we interchanged summation and integrations. Invoking the same

trick as in Eq. (33) allows us to perform the summation. The function
s(2)n (t , t ) := sn (t )sn (t ) = sinc[π(nl − t )/l]sinc[π(nl − t )/l] is of the
bandlimited kind and sampled here at the times nl; it is reconstructed by
the summation including the factor sn (t), in accordance with the Sam-
pling Theorem. However, the relevant bandwidths need some attention.
By Fourier transformation, one verifies that s(2)n has a doubled bandwidth,
www.ebook3000.com
(2)
ωmax = 2π/l = 2ωmax , in comparison with sn , which is not surprising. This
is implemented by evaluating the summation as follows:

nl/2 − t/2
sn (t)s(2)n (t , t ) = sinc π
l/2
n∈Z n∈Z

nl/2 − t /2 nl/2 − t /2
· sinc π sinc π
l/2 l/2
= sinc[π(t − t )/l]sinc[π(t − t )/l], (38)
i.e. rewriting all appearances of l in terms of l/2 and applying Eq. (15).
Thus, we obtain:
 2

∞
ψ(2) (t) = l−2  dt sinc[π(t − t )/l]ψ(t ) (39)
−∞
which expresses ψ(2) in terms of ψ. Using Eq. (31), we obtain a simple

quadratic term in the limit of vanishing discreteness scale:
lim ψ(2) (t) = (ψ(t))2 (40)

l→0
which presents the expected result that is local in time.

However, we observe that the quadratic term on the right-hand side
of Eq. (39) involves factors which are non-local in time, in particular, the
function ψ is integrated over all times, weighted by the oscillating but
slowly decaying sinc kernel. Inserting this result (and corresponding addi-
tional terms) into the continuous time version of the discrete analogue of
a nonlinear Schrödinger equation, Eq. (34), would change the character of
this equation profoundly: it does not present a discrete CA updating rule
anymore! Even if the left-hand side of Eq. (16) is non-local as well, this
non-locality is rather mild and refers only to two instants in such a way
that the linear equation can be solved forwards (or backwards) in time step
by step, recalling the discussion of initial conditions following Eq. (16) in
Sec. 3. With the non-locality here, due to an anharmonic potential, inserted
on the right-hand side of the continuous time equation, this would fail.
Furthermore, applying the Sampling Theorem to a nonlinear equation,
such as Eq. (34), yields an inconsistent continuous time equation, since
resulting linear and nonlinear terms have different bandwidth, unless an
additional cutoff on nonlinear terms is introduced by hand.
324 H.-T. Elze
Clearly, the present observations can be generalized for any kind of

polynomial nonlinearity. Generally, this would lead to non-local (in time)
effects in the same way.
Such non-locality may very well be a persistent qualitative feature
appearing in any continuous description of an underlying discrete CA
dynamics, unless the CA updating rules are linear in the dynamical vari-
ables (as in Sec. 2). We anticipate this to carry over to a discretization in
space, besides in time, which can be implemented explicitly along similar
lines [23].
This leaves us with a second intriguing and speculative question: Could
it be that unitary linear evolution in continuous time — which appears to
hold universally in QM (leaving aside measurement processes) — is dictated
by a local perspective on more general, possibly nonlinear underlying CA
dynamics? Or, in short: Does locality filter for linearity?
5. Conclusions
We have seen in this review how the description of a class of deterministic
discrete CA can be mapped via Shannon’s sampling theory on a continuous
time picture, which resembles in many respects the description of non-
relativistic quantum mechanical objects [2, 3].
Foremost stands the resulting relation between the discrete CA updat-
ing rules, which are closely analogous to Hamilton’s equations of motion in
mechanics, and a modified Schrödinger equation, which incorporates correc-
tions due to the finite discreteness scale characterizing the CA. This extends
to a one-to-one correspondence between the associated conservation laws,
between continuous unitary symmetries and their discrete counterparts.
On the other hand, familiar concepts in mechanics, such as Poisson brack-
ets and observables, can be extended to apply for our class of CA.
We have reported a restriction of the infinite sets (besides sporadic ones)
of admissible CA Hamiltonian functions such that the modified Schrödinger
equation allows stationary states. This calls for interpretation.
Our derivation of the dynamics has been founded on a variational prin-
ciple for a suitably defined action. Presently, we paid particular attention
to a generalization which incorporates polynomial nonlinearities into the
action and equations of motion in a consistent way, which otherwise tends
to be spoiled by overdetermining the CA updating rules.
However, we have also seen that allowing a nonlinear generalization of
the discrete CA dynamics leads to non-locality (in time) of the resulting
continuous time picture, requiring new consistency checks.
www.ebook3000.com
Which leads us to the follow-up question: What could be the effects

of such nonlinear CA processes affecting only some of the dynamical vari-
ables? Which would be the different ones than those commonly described
as quantum mechanical, i.e. by a local and linear evolution law?
Another interesting result has been that the normalization of a wave
function or state vector, which is deeply connected to the concept (and
evaluation) of probabilities in QM and may be seen as one of the primitives
of the theory, corresponds to a CA two-time correlation function having a
fixed value. This may lead to further physical interpretation.
In order to turn our observations of such surprising connections between
the properties of CA and apparently quantum mechanical features of more
familiar physical objects into a theory, as advocated by G. ’t Hooft [1],
for example, several immediate problems need to be addressed. We should
understand how composite systems fare in this context, which is a prereq-
uisite to analyze the CA analogue of QM measurement processes. Above
all, it has to be studied how aspects of relativity and of the physics of
space–time come into play here, which we tacitly bypassed so far.
Acknowledgments
It is a pleasure to thank Tom Kibble and Jonathan Halliwell for discussions,
as well as for the kind invitation to present this work in the theory group
at Imperial College (London).
References
1. G. ’t Hooft, The cellular automaton interpretation of quantum mechanics.
A View on the Quantum Nature of our Universe, compulsory or impossible?
(2014), preprint arXiv:1405.1548.
2. H.-T. Elze, Action principle for cellular automata and the linearity of quan-
tum mechanics, Phys. Rev. A 89, 012111 (2014).
3. H.-T. Elze, The linearity of quantum mechanics from the perspective of
Hamiltonian cellular automata, J. Phys. Conf. Ser. 504, 012004 (2014).
4. G. ’t Hooft, Quantization of discrete deterministic theories by Hilbert space
extension, Nucl. Phys. B 342, 471 (1990).
5. G. ’t Hooft, K. Isler and S. Kalitzin, Quantum field theoretic behavior of a
deterministic cellular automaton, Nucl. Phys. B 386, 495 (1992).
6. G. ’t Hooft, Quantummechanical behaviour in a deterministic model, Found.
Phys. Lett. 10, 105 (1997).
7. Z. Haba and H. Kleinert, Towards a simulation of quantum computers by
classical systems, Phys. Lett. A 294, 139 (2002).
8. H.-T. Elze and O. Schipper, Time without time: a stochastic clock model,
Phys. Rev. D 66, 044020 (2002).
326 H.-T. Elze
9. G. Grössing, From classical Hamiltonian flow to quantum theory: derivation

of the Schrödinger equation, Found. Phys. Lett. 17, 343 (2004).
10. M. Blasone, P. Jizba, F. Scardigli and G. Vitiello, Dissipation and quantiza-
tion for composite systems, Phys. Lett. A 373, 4106 (2009).
11. M. Sakellariadou, A. Stabile and G. Vitiello, Noncommutative spectral geom-
etry, algebra doubling and the seeds of quantization, Phys. Rev. D 84, 045026
(2011).
12. D. Acosta, P. Fernandez de Cordoba, J.M. Isidro and J.L.G. Santander, An
entropic picture of emergent quantum mechanics, Int. J. Geom. Meth. Mod.
Phys. 9, 1250048 (2012).
13. T.F. Jordan, Assumptions that imply quantum dynamics is linear, Phys. Rev.
A 73, 022101 (2006); do., Why quantum dynamics is linear, J. Phys. Conf.
Ser. 196, 012010 (2009).
14. T.D. Lee, Can time be a discrete dynamical variable? Phys. Lett. 122B, 217
(1983).
15. H.-T. Elze, Discrete mechanics, time machines and hybrid systems, EPJ Web
Conf. 58, 01013 (2013).
16. A. Heslot, Quantum mechanics as a classical theory, Phys. Rev. D 31, 1341
(1985).
17. A. Kempf, Spacetime could be simultaneously continuous and discrete in the
same way that information can, New J. Phys. 12, 115001 (2010).
18. C.E. Shannon, Communications in the presence of noise, Proc. IRE 37, 10
(1949).
19. A.J. Jerri, The Shannon Sampling theorem — Its various extensions and
applications: a tutorial review, Proc. IEEE 65, 1565 (1977).
20. J.F. McKee and C.J. Smyth, Integer symmetric matrices having all their
eigenvalues in the interval [−2, 2], J. Algebra 317(1), 260 (2007).
21. G. ’t Hooft, Quantum mechanics and determinism, presented at PASCOS
2001, arXiv:hep-th/0105105.
22. H.-T. Elze, Emergent discrete time and quantization: relativistic particle with
extradimensions, Phys. Lett. A 310, 110 (2003).
23. D. Gigli, Application of Shannon’s Sampling Theorem in Quantum Mechan-
ics, Master Thesis, University of Pisa, December 2014, unpublished.
www.ebook3000.com
Chapter 13
Structurally Dynamic
Cellular Networks as Models
for Planck Scale Physics
and the Quantum Vacuum
Manfred Requardt
Institut fuer Theoretische Physik
Universitaet Goettingen
Friedrich-Hund-Platz 1
37077 Goettingen, Germany
requardt@theorie.physik.uni-goettingen.de
Starting from the working hypothesis that both physics and the corre-
sponding mathematics have to be described by means of discrete con-
cepts on the Planck scale, one of the many problems one has to face in
this enterprise is to find the discrete protoforms of the building blocks
of our ordinary continuum physics and mathematics. We regard these
continuum concepts and continuum space–time (S-T) in particular as
being emergent, coarse-grained and derived relative to an underlying
erratic and disordered microscopic substratum which is expected to play
by quite different rules. A central role in our analysis is played by a geo-
metric renormalization group which creates (among other things) a kind
of sparse translocal network of correlations in classical continuous space–
time and underlies in our view such mysterious phenomena as holography
and the black hole entropy-area law. The same point of view holds for
quantum theory which we also regard as a low-energy, coarse-grained
continuum theory, being emergent from something more fundamental.
1. Introduction
In the beautiful book [1], the title of Chapter 12 reads: “Is Nature, under-
neath it All, a CA?”. Such ideas have in fact been around for quite some
time (cf. for example, Refs. [2–4] or Ref. [5], to mention a few references
327
328 M. Requardt
or, as an example for a discrete non-CA approach the work of Lee, [6]). A
little bit later, ’t Hooft analyzed the possibility of deterministic CA under-
lying models of quantum field theory or quantum gravity (Refs. [7] and [8]
are two examples from a long list of papers; see also the recent [9]). For
more detailed historical information, see Ref. [1] or [10]. A nice collection
of references can also be found in Ref. [11]. However, we would like to issue
a warning against an overly optimistic attitude. While we share the gen-
eral philosophy uttered in these works, there are some subtle points as ’t
Hooft remarks correctly [12]. It is no easy task to incorporate something
as complex as the typical entanglement structure of quantum theory into
the, at first glance, quite simple and local CA-models. We would like to
emphasize that it is not sufficient to somehow simulate or reproduce these
quantum phenomena numerically on a computer or CA. What is actually
called for is a structural isomorphism between those phenomena and cor-
responding emergent phenomena in CA. This problem has been one of the
reasons underlying our interest in CA having a fluctuating time-dependent
geometry (see below). Let us note that surprisingly similar ideas about the
discrete fine structure of space–time (S-T) similar to our own working phi-
losophy have been uttered in Ref. [10] Chapter 9, in particular, concerning
the existence of what we like to call shortcuts or whormhole structure (there
are absolutely no references given in the voluminous book; therefore we are
unable to make up our mind concerning priorities).
Still another interesting point is discussed by Svozil [13], i.e. the
well-known problem of species doubling of fermionic degrees on regular lat-
tices, which, as he argues, carries over to CA. Among the various possibil-
ities to resolve this problem, he suggests a kind of dimensional reduction
(“dimensional shadowing”), which leads in the CA one is actually interested
in, to non-local behavior (see also Ref. [1, p. 649ff]). It is perhaps remarkable
that, motivated by completely different ideas, we came to a similar conclu-
sion concerning the importance of non-local behavior (cf. Refs. [14, 15], see
also Ref. [16]). In Ref. [17], we clarified how these findings are related to
modern topics in quantum gravity like e.g. holography and the black-hole
entropy-area law. The connection is established by what we call wormhole
spaces.
While presently the discussion in the physics community, when it comes
to the high-energy end of fundamental physics, is dominated by string the-
ory and/or loop quantum gravity, frameworks which are in a conceptual
sense certainly more conservative, we nevertheless regard an approach to
these primordial questions via networks and/or CA as quite promising.
www.ebook3000.com
SDCN as Models for Planck Scale Physics 329
In contrast to the above-mentioned (more conservative) approaches which

start from continuum physics and hope to detect discrete S-T behavior at
the end of the analysis (for example, by imposing quantum theory as a
quasi-God-given absolute framework on the underlying structures over the
full range of scales), we prefer a more bottom-up-approach (which is, how-
ever, presumably more difficult as almost everything has to be developed
nearly from scratch). One of our reasons for this preference is that we do
not believe that quantum theory holds sway unaltered over the many scales
addressed by modern physics down to the pristine Planckian regime. Like
’t Hooft, we regard quantum theory rather as a kind of effective interme-
diate framework, which emerges from some more primordial structure of
potentially very different nature. We start from some underlying dynamic,
discrete and highly erratic network substratum consisting of (on a given
scale) irreducible agents interacting (or interchanging pieces of information)
via elementary channels. On a more macroscopic (or, rather, mesoscopic)
scale, we then try to reconstruct the known continuum structures as emer-
gent phenomena via a sequence of coarse graining and/or renormalisation
steps (see Refs. [18] and [19]).
While CA have been widely used in modeling complex behavior of
molecular agents and the like (a catchword being artificial life or Conway’s
game of life; for a random selection see e.g. Refs. [1,20–23] or [24], a general
discussion of some of the concepts being of relevance in this context can also
be found in Ref. [25] and further references given there), papers on the more
pristine and remote regions of Planck scale physics are understandably less
numerous.
When we embarked on such a programme in the early nineties of the
last century, we soon realized that the ordinary framework of CA, typically
living on fixed and quite regular geometric arrays, appears to be far to
rigid and regular in this particular context (whereas the hope is frequently
uttered that complexity is able to emerge from quite simple microscopic
laws, which is certainly correct in principle). In order to implement the
lessons of general relativity we think we have to make the wiring structure of
CA dynamical, i.e. not only the local states at the vertices of the lattice but
also the local states attached to the links have to become dynamical degrees
of freedom (DOF). A fortiori, we would like the whole wiring diagram of
links to be “clock-time dependent”. To put it briefly: matter shall act on
geometry and vice versa, where we, tentatively, associate the pattern of local
vertex states with the matter distribution and the geometric structure of
the network with geometry.
330 M. Requardt
Remark 1.1. In order to be able to implement the lessons from general

relativity, it is our central idea that the wiring diagram of the network and
the node states (matter degrees) coevolve.
Definition 1.2. If this central concept is realized, we call the network a

structurally dynamic cellular network (SDCN).
We want to briefly mention a superficially related approach by Konopka

et al. (see Refs. [26–28]) which uses several of the ideas and concepts which
have already been introduced or further developed somewhat earlier by us
in this particular context (e.g. dynamic graphs, the random graph approach,
the existence of translocal links, etc.) but this approach is more reminis-
cent of the spin networks models, deriving from the loop quantum gravity
framework. For example, the local states at the links are the well-known
angular momentum states occurring also in the spin network business while
the nodes do not carry elementary DOF. That is, there is no coevolution
of geometric and matter DOF. Furthermore, the introduced Hamiltonian is
mainly used to define statistical averages over graph configurations. That
is, most of the typical CA or SDCN framework is missing. One may perhaps
rather mention in this context the paper by Antonsen [29].
Before we go into the technical details of our framework, we want to
repeat what we consider to be the central points.
1. We want to derive the continuum concepts of ordinary mathematics and

physics and/or their discrete protoforms from our SDCN. To mention
just a few examples, discrete differential and functional analysis, oper-
ator theory (e.g. graph Laplacians graph Dirac operators together with
their eigen values), generalizations of dimensional concepts, etc.
2. We want, in particular, to reconstruct continuous S-T via a geomet-
ric renormalization process, where it may happen that the various
coarse-graining levels of the network have scale dependent (possibly non-
integer) dimensions.
3. An important consequence of this renormalization process is that our
geometry may develop a so-called near- and far-order. While the near-
order leads to ordinary continuous space, the far-order (i.e. correlations
or entanglement between regions which are some distance apart with
respect to the ordinary distance metric) may be responsible for such
important effects like quantum entanglement, the holographic principle
and/or the entropy-area law black hole physics.
www.ebook3000.com
We begin our investigation by introducing our underlying network model

and the necessary concepts and notions. We then proceed with the deriva-
tion of various concepts and tools of discrete (functional) analysis and
operator theory. In a next step, we introduce the concept of generalized
dimension on such discrete and irregular spaces. Then follows the ambi-
tious enterprise to define a geometric renormalization process with classical
S-T emerging as some coarse-grained limit. In this context, we provide a
brief introduction to the random graph framework. We will conclude the
presentation with providing arguments how these steps will lead to a better
understanding of various crucial concepts of modern physics (in particular
concerning quantum gravity). Examples being the mysterious phenomenon
of holography, quantum entanglement, the black hole entropy-area law, etc.,
the novel idea being the concept of wormhole space.
Let us make a final remark concerning the various mathematical fields
which are involved in our enterprise as this variety of connections is
frequently also invoked as a special merit on the side of string theory.
To mention just a few fields: advanced graph theory (e.g. clique graphs,
random graphs), Connes’ non-commutative analysis and geometry, oper-
ator theory on discrete spaces (graph Laplacians, Dirac operators, their
eigenvalues), generalizations of dimensional concepts, leading even to con-
nections with geometric group theory (via the so-called Cayley graphs),
Gromov–Haussdorff Limit of irregular spaces, etc.
2. The Microscopic SDCN-Substratum

Our networks are defined on general graphs, G, with V (G) the set of its
vertices (sites or nodes) and E(G) its set of edges (links or bonds) of the
graph G.
Definition 2.1. Here are some graph-theoretical notions and concepts (for
more details, see e.g. Ref. [30]).
1. We write the simple, or directed labeled graph as G := (V, E), where V

is the countable set of vertices {ni } and E the set of edges. The graph is
called simple if there do not exist elementary loops and multiple edges,
in other words: each existing edge connects two different vertices and
there exists at most one edge between two vertices. (We could of course
also discuss more general graphs). Furthermore, for reasons of simplicity,
we assume the graph to be connected, i.e. two arbitrary vertices can be
connected by a sequence of consecutive edges called an edge sequence
332 M. Requardt
or walk. A minimal edge sequence, that is, one with each intermediate
vertex occurring only once is called a path.
2. For convenience, we assume the graph to be locally finite, that is, each
vertex is incident with only a finite number of edges. Sometimes, it is use-
ful to make the stronger assumption that this vertex degree, vi , (number
of edges being incident with ni ), is globally bounded away from ∞.
3. One can give the edges both an orientation and a direction (these two
slightly different geometric concepts are frequently intermixed in the lit-
erature). In an undirected graph, the edges eik correspond to unordered
pairs of vertices {ni , nj } while in a directed graph the edges have a direc-
tion represented by an ordered pair of vertices (ni , nj ), i.e. the edge dik
points from ni to nj . In our context, we adopt the following convention:
If two vertices ni , nk are connected by an edge in an unordered graph,
we interpret this as follows: There exists a directed edge, dik , pointing
from ni to nk and a directed edge, dki , pointing in the opposite direc-
tion. In an algebraic sense, which will become clear below, we call their
superposition
bik := dik − dki = −bki (1)
the corresponding oriented edge (for obvious reasons; the directions are
fixed while the orientation can change its sign). In a sense, the above
reflects the equivalence of an undirected graph with a directed multi-
graph having two directed edges pointing in opposite directions for each
undirected edge.
We now associate states si and Jik with the vertices and edges ni , eik .
The local vertex states can assume values in a certain discrete set. In the
examples we have studied, we follow the philosophy that the network should
be allowed to find its typical range of states via the imposed dynamics.
That is, we allow the si to vary in principle over the set q · Z, with q
a certain discrete quantum of information, energy or whatever. The edge
states can assume the values Jik ∈ {−1, 0, +1} (we regard the edge states
as representing a kind of elementary coupling).
Viewed geometrically, we can associate the states Jik = +1, −1, 0 with
directed edges pointing from vertex ni to nk , or from nk to ni , or, in the
Jik = 0 case, with an empty edge. Our network will be updated (as in the
case of CA) after each discrete step of the evolution parameter t (which
may be called somewhat sloppily, as in computer science, clock time. That
is, at each clock time step, t · τ (τ an elementary quantum of clock time and
www.ebook3000.com
t ∈ Z), we have as underlying substratum a clock time dependent directed

graph, Gt . Our physical idea is that at each clock time step an elementary
quantum q is transported along each existing directed edge in the indicated
direction.
To implement our general working philosophy of mutual interaction of
overall vertex states and network geometry, we now describe some particular
network laws, which we investigated in greater detail in Ref. [31] together
with a lot of numerical simulation and analysis. We mainly consider two
different classes of evolution laws for vertex and edge states (for reasons of
simplicity we choose units so that q, τ are set equal to one):
• Network Type I

si (t + 1) = si (t) + Jki (t), (2)
k

Jik (t + 1) = sign(∆sik ) for [|∆sik | ≥ λ2 ∨ |∆sik | ≥ λ1 ∧ Jik (t) = 0 ],
(3)
Jik (t + 1) = 0 otherwise. (4)
• Network Type II

si (t + 1) = si (t) + Jki (t), (5)
k
Jik (t + 1) = sign(∆sik ) for

[0 < |∆sik | < λ1 ∨ 0 < |∆sik | < λ2 ∧ Jik (t) = 0 ], (6)
Jik (t + 1) = Jik (t) for ∆sik = 0, (7)
Jik (t + 1) = 0 otherwise, (8)
where ∆sik = si (t) − sk (t) and λ2 ≥ λ1 ≥ 0 (∨, ∧ meaning or, and, respec-
tively). We see that in the first case, vertices are connected that have very
different internal states, leading to large local fluctuations, while for the
second class, vertices with similar internal states are connected.
Remark 2.2. The role of the λ parameters is the following. They define
kind of a hysteresis interval, regulating the switching off and on of edges.
We hope that they can be tuned so that the network can perform phase
transitions. We studied the λ dependence of various network properties in
the computer simulations being discussed in Ref. [31].
We proceed by making some remarks in order to put our approach into

the appropriate context.
334 M. Requardt
Remark 2.3.
1. It is important that, generically, laws, as introduced above, do not lead
to a reversible clock time evolution, i.e. there will typically be attractors
or state-cycles in total phase space (the overall configuration space of
the node and edge states). On the other hand, there exist strategies (in
the context of cellular automata!) to design particular reversible network
laws (cf. e.g. Ref. [32]) which are, however, typically of second order in
clock time. Usually, the existence of attractors is considered to be impor-
tant for pattern formation. On the other hand, it may suffice that the
phase space, occupied by the system, shrinks in the course of evolution,
that is, that one has a flow into smaller subvolumes of phase space.
2. In the above class of laws, a direct edge–edge interaction is not yet
implemented. Note that such a term would be a direct nonlinear action
of geometry on geometry or, as in gauge theory, a pure interaction of the
gauge fields. We are prepared to incorporate such a (possibly important)
contribution in a next step if it turns out to be necessary. In any case
there are not so many ways to do this in a sensible way. Stated differently,
the class of possible, physically sensible interactions, is perhaps not so
large.
3. We would like to emphasize that the (undynamical) clock-time, t, should
not be confused with the notion of physical time, i.e. the time opera-
tionally employed on much coarser scales. The latter is rather supposed
to be a collective variable and is expected (or hoped!) to emerge via
a cooperative effect. Clock-time may be an ideal element, i.e. a notion
which comes from outside, so to speak, but — at least for the time
being — we have to introduce some mechanism, which allows us to label
consecutive states or describe change or evolution.
We make the following observation because it is relevant if one follows
the general spirit of modern high energy physics.
Observation 2.4 (Gauge Invariance). The above dynamical law
depends nowhere on the absolute values of the vertex charges si but only
on their relative differences. By the same token, charge is nowhere created
or destroyed. We have

∆ si = 0, (9)
i∈I
where for reasons of simplicity, we denote the set of vertices by their set of
indices, I and ∆ denote the difference between consecutive clock-time steps.
www.ebook3000.com
Put differently, we have conservation of the global vertex charge. To avoid

artificial ambiguities we can, e.g. choose a fixed reference level and take as
initial condition the constraint

si = 0. (10)
i∈I
We summarize the main steps of our working philosophy:
Remark 2.5. Irrespective of the technical details of the dynamical evolu-

tion law under discussion, the following in our view that are crucial princi-
ples that should be emulated in order to match fundamental requirements
concerning the capability of emergent and complex behavior.
1. As is the case with, say, gauge theory or general relativity, our evolu-
tion law on the surmised primordial level should implement the mutual
interaction of two fundamental substructures, put sloppily: “geometry”
acting on “matter” and vice versa, where in our context “geometry” is
assumed to correspond in a loose sense with the local and/or global bond
states and “matter” with the structure of the node states.
2. By the same token, the alluded self-referential dynamical circuitry of
positive feedback structure being present in the network is expected to
favor a kind of undulating behavior or self-excitation in contrast to a
return to some uninteresting equilibrium state as is frequently the case
in systems consisting of a single component which directly feeds back
on itself. This propensity for the autonomous generation of undulation
patterns is in our view an essential prerequisite for some form of “proto-
quantum behavior” we hope to recover on some coarse grained and less
primordial level of the network dynamics.
3. In the same sense, we expect the overall pattern of switched-on and -off
bonds to generate a kind of “protogravity”.
We want to comment on a particular intriguing result from our numer-

ical simulation performed in Ref. [31], that is, the phenomenon of limit
cycles. Because of the finite phase space of the CA (technically it is infinite,
but the vertex states only fill a finite interval of Z due to the nature of the
network laws), network states will eventually repeat, which leads to a limit
cycle because of the memoryless dynamics. We tested for the appearance
of such limit cycles for different network size n (number of vertices) and
to our surprise, networks of Type I (as defined above) had with very few
exceptions extremely short limit cycles of period 6. The exceptions we were
able to find, had periods of a multiple of 6, the longest found (in a network
336 M. Requardt
with n = 810 vertices) was 36. The prevalence of such short limit cycles
is still an open question and beyond this work. We note in this context
that already Kauffman observed such short cycles in his investigation of
switching nets [20, 21] and found it very amazing.
Remark 2.6. Our computer simulations employed the following initial con-
ditions. We started with a maxmal complete graph, i.e. each pair of vertices
is connected by an undirected edge. The vertex states were chosen from a
uniform distribution scattered over some interval of integers (we tried also
other distributions but did not find any significantly different results). The
initial values of the edge states were chosen from the set {+1, −1} with
equal weight. By this choice, we wanted to simulate the initial condition
prevalent in the big bang era.
This phenomenon of a short limit cycle is remarkable in the face of the

huge accessible phase spaces of typical models and points to some hidden
ordering tendencies in these model classes. What is even more startling is
that this phenomenon prevails also in our case for model Class 1 when we
introduce a further element of possible disorder by allowing edges to be
dynamically created and deleted. We formulate the following hypothesis.
Conjecture 2.7. We conjecture that this important phenomenon has its
roots in the self-referential structure (feedback mechanisms) of many of the
used model systems.
It is instructive to observe the emergence of such short cycles in very small
models on paper, setting for example, λ1 = λ2 = 0, i.e. no switching-off
of edges and taking n = 2, 3, or 4. Taking, e.g. n = 2 and starting from
s2 (0) = s1 (0) mod 2, the network will eventually reach a state s1 (t0 ) =
s2 (t0 ). Without loss of generality, we can assume s1 (t0 ) = s2 (t0 ) = 0 and
J12 (t0 ) = 1. This state develops into a cycle of length 6 (cf. Table 2a(1) in
[31]). For s1 (0) = s2 (0) + 1 mod 2 the state eventually becomes s1 (t1 ) =
s2 (t1 ) + 1, without loss of generality s1 (t1 ) = 1, s2 (t1 ) = 0, J12 (t1 ) = 1
resulting in the dynamics of Table 2a(2) of [31]. Again, the length of the
cycle is six. Hence, six is a good candidate for a short cycle length, which —
of course — does not explain why such a short length should appear at all.
The transients (i.e. the clock-time interval after which the network has
forgotten the random initial conditions or after which it arrives on an attrac-
tor) in networks of Type I are also rather short and grow slowly with the
network size. On the other hand, networks of Type II have much longer
limit cycles and transients. Because of numerical limitations we were only
www.ebook3000.com
able to determine cycle lengths for small networks. We observed that the
typical transient and cycle lengths both grow approximately exponentially
(cf. Table 2b in Ref. [31]).
3. Differential and Operator Calculus on Graphs

In the following section, we introduce differential and operator calculus on
graphs. To some extent, this topic carries the flavor of our own ideas (i.e.
we surmise that not everything we introduce below can be found in the
standard mathematical literature, we note for example that, as a minor
point, the use of matrices instead of operators is widespread in the lit-
erature). A classical text is, for example [33], a nice more recent source
is Ref. [34]. Our own framework can be found in the papers Refs. [35–37],
where more references are given. In a first step, we introduce the vertex and
edge Hilbert spaces for directed and undirected graphs (for reasons of math-
ematical simplicity we restrict ourselves to locally finite graphs; the more
general situation can of course also be dealt with. For a directed graph, we
then have ingoing and outgoing edges at each vertex.
Definition 3.1. We denote the in-vertex degree at vertex xi by viin , the
out-vertex degree by viout and the local vertex degree by vi = viin + viout .
For such a graph, we can introduce two Hilbert spaces, a vertex Hilbert
space, H0 , and an edge Hilbert space, H1 , with orthonormal bases the set
of vertices, xi , and the set of directed edges, dij . This means, we introduce
a formal scalar product on H0 , H1 respectively with
xi , xj = δij , dij , dlm = δil δjm (11)
and with vectors being the formal sums
∞ ∞

f= fi xi , g = gij dij with fi , gij ∈ C (12)
1 i,j=1

with |fi |2 < ∞ and |gik |2 < ∞.
Remark 3.2. We treat the vertices and edges as abstract basis elements (in
a way similar to the group algebra of a group). One can of course consider
the abstract vectors equally well as discrete functions over the vertex- or
edge-set, respectively and the basis vectors as elementary indicator func-
tions. Therefore, we replace from now on the nodes ni by the corresponding
indicator functions xi , having the value one at the respective node under
discussion and zero else.
338 M. Requardt
If we deal with an undirected but orientable graph we find it convenient

to introduce the superposition
bij := dij − dji = −bji (13)
and relate it to an undirected but orientable edge. We now introduce two

operators, interpolating between H0 and H1 . We define them on the basis
vectors:
d :Ho → H1 , (14)

d(xi ) := dki − dik (15)
k k
with the first sum running over the ingoing edges relative to xi , the second
sum running over the outgoing edges. In the case of a symmetric (or undi-
rected graph), we have

d(xi ) := (dki − dik ) = bki . (16)
k k
This operator is closely related to a sort of non-commutative discrete dif-

ferential calculus on graphs as we have

df = (fk − fi )dik . (17)
i,k
A simple calculation shows that the adjoint, d∗ : H1 → H0 , acts on the

basis vectors of H1 as follows:
d∗ (dik ) = xk − xi . (18)
Remark 3.3. Note that these operators are closely related to the boundary
and coboundary operator in algebraic topology.
In algebraic graph theory (finite graphs), the so-called incidence matrix,

B, is introduced, having the entry 1 if vertex xi is the positive end of a cer-
tain (ingoing) edge, and having a −1 if it is the negative end (outgoing edge)
(see, for example, Ref. [34]). This matrix corresponds to our operator d∗.
Another important operator is the adjacency matrix, A, being a map
from H0 to H0 and having (in ordinary graph theory of (un)oriented graphs)
a + 1 at entry (i, j) if xi and xj are connected by an edge. This matrix is a
symmetric operator, aij = aji . In our more general context (which includes
however the ordinary situation as a special case) of directed graphs, one
can introduce the in-adjacency matrix, Ain , and the out-adjacency matrix,
www.ebook3000.com
Aout , with A = Ain + Aout , Ain having a +1 at entry (k, i) if xk and xi are
connected by a directed edge dki and correspondingly for Aout .
Remark 3.4. Note that our A differs slightly from the classical A. The
classical A for an undirected graph equals our Ain or Aout in that case. Our
operators apply to more general situations with our A being even symmetric
for arbitrary directed graphs.
In our (operator)-notation, they are given by

A xi = ki xk , Ain xi = xk , Aout xi = xk (19)
k∼i k→i i→k
with ∼ designating the unordered pair {xi , xk }, k → i the ordered pair

(k, i) and k,i is either one or two depending on the two possible cases
of one directed edge between node xi and node xk or two directed edges,
pointing in opposite directions.
These operators can be built up from more elementary operators
(cf. Refs. [36] or [37]).

d1 xi = dki , d2 xi = dik , (20)
k k
d∗1 dik = xk , d∗2 dik = xi , (21)
so that
d = d1 − d2 , d∗ = d∗1 − d∗2 , (22)

d∗1 d1 xi = viin
· xi , d∗2 d2 xi = viout
· xi , (23)

d∗1 d2 xi = xk , d∗2 d1 xi = xk , (24)
i→k k→i
where viin , viout is the in-, out degree of vertex xi respectively. We hence
have
Lemma 3.5. The in-, out-vertex degree matrices read
V in = d∗1 d1 , V out = d∗2 d2 . (25)
The in-, out-adjacency matrices read
Ain = d∗2 d1 , Aout = d∗1 d2 , (26)
A = Ain + Aout is symmetric.

340 M. Requardt
Proposition 3.6. The so-called graph Laplacian is the following positive

operator

−∆ := d∗ d = V in + V out − Ain + Aout = V − A. (27)
Note that for an undirected graph (i.e. both dik and dki being present) the
above Laplacian is simply twice the classical (matrix) Laplacian.
The reason to call this operator a Laplacian stems from the observation
that it acts like a second order partial difference operator on functions
of H0 .

−∆ f = fi viin xi + viout xi − xk − xk (28)
i k→i i→k
and after a simple relabeling of indices

in out
−∆ f = − fk + fk − vi fi − vi fi xi
i k→i i→k

=− (fk − fi ) + (fk − fi ) xi
i k→i i→k

=− ki (fk − fi ) xi (29)
i k∼i
which reduces to the ordinary expression in the undirected case.

Forming now the direct some H := H0 ⊕ H1 , we can introduce
yet another important graph operator which closely entangles geometric
and functional analytic properties of graphs (and similar structures); see
Refs. [36] and [37].
Definition 3.7. We define the graph Dirac operator as follows

0 d∗ H0
D : H → H with D := , H= . (30)
d 0 H1
Observation 3.8. We have

∗
2 d d 0
D = DD = (31)
0 dd∗
with d∗ d = −∆.
www.ebook3000.com
The action of dd∗ on a basis vector dik reads

dd∗ dik = d (xk − xi ) = dk k − dkk − dli + dil (32)
k k l l
which, after some relabeling and introduction of the Kronecker delta func-
tion, can be written as

dd∗ dik = (dmj δjk − djm δjk − dmj δij + djm δij ). (33)
m,j

For a function g = gik dik , we hence get

∗
dd g = gim − gmi − gil + gli dlm . (34)
l,m i
There is a pendant in the calculus of differential forms on general Rieman-

nian manifolds where, with the help of the Hodge-star operation, we can
construct a dual, δ, to the ordinary exterior derivative. The generalized
Laplacian then reads
−∆ = δd + dδ (35)
with δ (modulo certain combinatorial prefactors) corresponding to our d∗

(see for example [38] or [39]).
With the help of the machinery we have introduced above, further prop-
erties of the graph Laplacian and Dirac operator can be derived, some of
them being directly related to geometric and/or wiring propertis of the
graph under discussion. Of particular relevance are spectral properties of
−∆ and D. In most of the graph literature the graphs are assumed to be
finite, hence the corresponding operators are automatically bounded and
−∆ and D are self-adjoint. A fortiori the spectrum is discrete as the cor-
responding Hilbert spaces are finite dimensional. This is the reason why
most of the graph literature used the matrix calculus (which is in our view
a little bit clumsy). Investigation of the unbounded case are less numerous.
We discussed this more general case in Refs. [36] and [37], where also more
references are mentioned. A particular result is for example:
Theorem 3.9. For a globally bounded vertex degree −∆ and D are bounded
self-adjoint operators with their bounds being explicitly computable. (For a
proof, see Ref. [36] and/or Ref. [37].)
342 M. Requardt
It is noteworthy that with the help of the tools we have introduced and
developed above we can successfully deal with various interesting modern
topics of mathematical physics in this particular context of irregular dis-
crete spaces. To mention a few cases, in Refs. [35–37], we treated graphs
as models of non-commutative geometry and supersymmetry. Among other
things, we introduced an example of Connes’ spectral triple, and calculated
the so-called Connes distance metric on graphs. That our networks/graphs
carry automatically and naturally a supersymmetric structure may perhaps
be a further hint that this approach and its continuum limit has something
to do with the real physical world of high-energy physics.
4. Dimensional Concepts on Graphs or Networks

There exist a variety of concepts in modern mathematics which extend
the ordinary or naive notion of dimension one is accustomed to in e.g.
differential topology or linear algebra. In fact, topological dimension and
related concepts are notions which are perhaps even closer to the underlying
intuition (cf. e.g. Ref. [40]).
Apart from the purely geometric concept there is also an important
physical role being played by something like dimension, having pronounced
effects on the behavior of, say, many-body-systems near their phase transi-
tion points or in the critical region.
But even in the case of e.g. lattice systems they are usually treated
as being embedded in an ambient continuous background space (typically
euclidean) which supplies the concept of ordinary dimension so that the
intrinsic dimension of the discrete array itself does usually not openly enter
the considerations.
Anyway, it is worthwhile even in this relatively transparent situation
to have a closer look on the situations where attributes of something like
dimension really enter the physical stage. Properties of models of, say, sta-
tistical mechanics are typically derived from the structure of the microscopic
interactions of their constituents. This then is more or less the only place
where dimensional aspects enter the calculations.
Naive reasoning might suggest that it is something like the number of
nearest neighbors (in e.g. lattice systems) which encodes the dimension
of the underlying space and influences via that way the dynamics of the
system. However, this surmise, as we will show in the following, does not
reflect the crucial point which is more subtle.
This holds more so for systems which cannot be considered as being
embedded in a smooth regular background and hence do not inherit their
www.ebook3000.com
dimension from the embedding space. A case in point is our primordial

network in which Planck scale physics is assumed to take place. In our
approach, it is in fact exactly the other way round: Smooth S-T is assumed
to emerge via a (geometric) phase transition or a certain cooperative behav-
ior and after some ‘coarse graining’ from this more fundamental structure.
That is, our task is to formulate an intrinsic notion of dimension for model
theories living on quite irregular spaces without making recourse to the
dimension of some continuous embedding space.
In the first step, we will show that graphs or networks as introduced in
the preceding sections carry a natural metric structure. We have already
introduced a certain neighborhood structure in a graph with the help of the
minimal number of consecutive edges connecting two given vertices. In a
connected graph, any two vertices can be connected by a sequence of edges.
Without loss of generality, one can restrict oneself to paths (all vertices
being distinct). One can then define the length of a path (or sequence of
bonds) by the number l of consecutive edges making up the path.
Observation 4.1. Among the paths connecting two arbitrary vertices
xi , xk there exists at least one (a geodesic path) with minimal length which
we denote by d(xi , xk ). This d has the properties of a metric, i.e.
d(xi , xi ) = 0, (36)
d(xi , xk ) = d(xk , xi ), (37)
d(xi , xl ) ≤ d(xi , xk ) + d(xk , xl ). (38)
(The proof is more or less evident).
Corollary 4.2. With the help of the metric one gets a natural neighborhood
structure around any given vertex, where Bm (xi ) comprises all the vertices,
xk , with d(xi , xk ) ≤ m, ∂Bm (xi ) the vertices with d(xi , xk ) = m.
This natural neighborhood structure enables us to develop the concept
of an intrinsic dimension on graphs and networks. To this end one has at
first to realize what property really matters physically (e.g. dynamically),
independently of the kind of model or embedding space.
Observation 4.3. The crucial and characteristic property of, say, a graph
or network which may be associated with something like dimension is the
number of ‘new vertices’ in Bm+1 compared to Bm for m sufficiently large
or m → ∞. The deeper meaning of this quantity is that it measures the kind
of ‘wiring’ or ‘connectivity’ in the network and is therefore a ‘topological
invariant’.
344 M. Requardt
Regarding the graph as an example of a metric space we shall replace

the discrete value m by the continuous parameter r, hence writing the ball-
neighborhoods as B(x, r). We then define the growth function and spherical
growth function on G relative to some arbitrary but fixed vertex x. (We use
here the notion more common in e.g. geometric group theory. In other fields,
it is also called the distance degree sequence, cf. Ref. [41].)
Definition 4.4. The growth function β(G, x, r) is defined by
β(G, x, r) = |B(x, r)| (39)
with |B(x, r)| denoting the number of nodes in B(x, r).

Correspondingly, we define
∂β(G, x, k) := β(G, x, k) − β(G, x, k − 1). (40)
With the help of the limiting behavior of β , ∂β, we introduce two-

dimensional (2D) concepts.
Definition 4.5. The (upper, lower) internal scaling dimension with respect
to the vertex x is given by
Ds (x) := lim sup(ln β(x, r)/ ln r), Ds (x) := lim inf (ln β(x, r)/ ln r). (41)
r→∞ r→∞
The (upper, lower) connectivity dimension is defined correspondingly as
Dc (x) := lim sup(ln ∂β(x, k)/ ln k)+1, Dk (x) := lim inf (ln β(x, k)/ ln k)+1.
k→∞ k→∞
(42)
If upper and lower limit coincide, we call it the internal scaling dimension,
the connectivity dimension, respectively.
Remark 4.6.
1. The two notions are not entirely the same in general whereas they coin-
cide for many models (this is quite similar to the many different fractal
dimensions).
2. For regular lattices, both yield the expected result, i.e. the embedding
dimension. In general, however, upper and lower limit are different and
non-integer. Similarities to fractal dimensions are not accidental. For
more thorough discussion of all these points, see Ref. [42].
www.ebook3000.com
Some historical remarks are perhaps in order. We developed and inves-

tigated these concepts in Ref. [42] almost from scratch. We later observed
that there existed some scattered remarks in the literature using similar
concepts but, as to our knowledge, nowhere were the extremely interesting
properties of this concept studied in detail in the physics literature (cf. the
remarks and references in Ref. [42]). On the other hand, we later (roughly
at the time of writing [19]) came upon similar concepts employed in a com-
pletely different context, i.e. a field of pure mathematics called geometric
group theory (see e.g. Ref. [43]). We shall come back to this point below
when we shall introduce the renormalization and coarse graining process
on graphs or networks.
It is important that these notions display a marked rigidity against all
sorts of deformations of the underlying graph and are independent of the
reference vertex for locally finite graphs. We mention only two properties
in this direction.
Observation 4.7. If the vertex-degree of the graph is locally finite, the

numerical values of the above quantities are independent of the reference
vertex. (The simple proof can be found in Ref. [42].)
In the following theorem, we prove stability of graph dimension under

local perturbations of the wiring of graphs. In the first step, we add edges
in the k-neighborhoods of each vertex. In the second part of the theorem,
the local deformations are slightly more complicated.
Definition 4.8. We pass over from a graph G to a new graph G , living on

the same vertex set, by means of a number of edge deletions. These edge
deletions are called local of order k if only edges between nodes x, y are
deleted which have a distance in G globally bounded by k (note that the
distance metrics in G and G will differ in general!).
Theorem 4.9. With G locally finite the following holds:

1. Insertions of arbitrarily many edges within a k-neighborhood of any ver-
tex do not alter the dimension.
(The proof and a slight generalization can be found in Lemma 4.10 of
Ref. [42]).
2. Edge deletions fulfilling the above described property also do not alter the
graph dimension.
(This can be proved by reversing the process, i.e. we pass from the graph
G to G by k-local edge insertions; cf. Theorem 6.8 in Ref. [15]).
346 M. Requardt
Such deformation results are very useful because it turns out to be surpris-
ingly difficult to construct sufficiently irregular large graphs with presrcibed
properties, for example, having a prescribed (possibly non-integer) dimen-
sion. The above theorem guarantees that irregular graphs which can be
constructed via appropriate deformations starting from e.g. regular graphs
will have the same dimension. For more results in this direction, see the
following sections (or Refs. [42] and [19]).
We want to conclude this section with a remark concerning the nature of
the above defined graph dimension. At first glance, it may remind the reader
of the various fractal dimensions (see e.g. Ref. [44]) but this impression is
not entirely correct. In a sense, it is just the opposite of a fractal dimension.
While fractal dimension is related to the infinitesimal structure of (irreg-
ular) sets, it is in our case the large distances which matter, therefore the
notion growth degree is a better description. The reason for this duality
stems from our working philosophy to construct the continuum from some
discrete irregular underlying structure by performing a continuum limit
via coarse graining and scaling (cf. the following section). This is just the
opposite from going into the infinitesimal small as in fractal geometry.
It is a characteristic of our construction that we go to large distances
on the underlying graphs. After all, to arrive at a rigorous definition the
graphs have to be infinite. For large but finite graphs, we can of course
use the concept in an approximate way. Going to large and at the end
infinite distances is also crucial when we take the continuum limit in order
to reconstruct a corresponding continuum theory. In this process, we resale
the original graph distance metric, that is, we go over from the original
distance d(x, y) to λ · d(x, y) with λ → 0. Consequently, points which lie
very far apart in the underlying graph or network G become infinitesimal
neighbors in the continuum limit. That is, the growth degree characterizes
in the end the infinitesimal neighborhood of points in the continuum which
is a property of the notion of dimension in the continuum.
As a last remark, we want to mention another concept of dimension
which is frequently employed in the physics of statistical and critical sys-
tems on irregular geometric structures. It is called the spectral dimension.
As far as we are aware, early attempts can be found in Ref. [45]. A careful
mathematical analysis is made in Ref. [46]. A nice chapter, relating the
spectral dimension to our scaling dimension, is Ref. [47]. Another inves-
tigation of these dimensions is made in Ref. [48] (we emphasize that our
list is quite incomplete). The spectral dimension is closely related to dif-
fusion processes on the underlying networks and the return probability of
www.ebook3000.com
random walks. Both types of dimension encode certain geometric proper-

ties of the network which can be associated with something like dimension.
There are some connections between these notions but generically they are
numerically different.
5. Dynamical Networks as Random Graphs

5.1. The statistical hypothesis
As we are dealing with very large graphs, which are, a fortiori, constantly
changing their shape, i.e. their distribution of (active) bonds, we expect the
dynamics to be sufficiently stochastic so that a point of view may be appro-
priate, which reminds of the working philosophy of statistical mechanics.
This does, however, not imply that our evolving network is nothing but a
simple random graph as introduced below (cf. the remarks at the end of this
section). It rather means that some of its geometric characteristics can, or
should, be studied within this well-developed context.
Visualizing the characteristics and patterns being prevalent in large and
“typical” graphs was already a notorious problem in combinatorial graph
theory and led to the invention of the random graph framework (see the
more complete discussion in Ref. [49]). The guiding idea is to deal with
graphs of a certain type in a probabilistic sense. This turns out to be partic-
ularly fruitful as many graph characteristics (or their absence) tend to occur
with almost certainty in a probabilistic sense (as has been first observed by
Erdös and Rényi). The standard source is Ref. [50] (for further references,
see Ref. [49]).
Another strand of ideas stems from the theory of dynamical systems and
cellular automata, where corresponding statistical and ensemble concepts
are regularly employed. Typically, we are looking for attractors in phase
space, which are assumed to correspond to large scale, i.e. after coarse grain-
ing and rescaling, quasi-continuous or macroscopic patterns of the system.
Experience shows that such a structure or the approach toward attractors
is in many cases relatively robust to the choice of initial configurations or
microscopic details and, hence, suggests an ensemble picture.
Furthermore, since the early days of statistical mechanics, the ensemble
point of view is, at least partly, corroborated by the philosophy that time
averages transform (under favorable conditions) into ensemble averages.
In our context, this means the following. Denoting the typical length/time
scale of ordinary quantum theory by [lqm ], [tqm ], we have
[lqm ] [lpl ], [tqm ] [tpl ] (43)

348 M. Requardt
the latter symbols denoting the Planck scale. Under renormalization, the
mesoscopic scales comprise a huge number of microscopic clock time inter-
vals and degrees of fredom of the network under discussion.
A fortiori, the networks, we are interested in, correspond to graphs,
having a large vertex degree, i.e. channels, entering a given typical node
of the graph. That is, we expect large local fluctuations in microscopic
grains of space or time. Put differently, the network locally traverses a large
number of different microscopic states in a typical mesoscopic time interval,
[tqm ]. This observation suggests that, on a mesoscopic or macroscopic scale,
microscopic patterns will be washed out or averaged over.
5.2. The random graph framework

One kind of probability space is constructed as follows. Take all possible
labeled graphs over n nodes as probability space G (i.e. each graph rep-
resents an
elementary event). The maximal possible number of bonds is
N := n2 , which corresponds to the unique simplex graph (denoted usu-
ally by Kn ). Give each bond the independent probability 0 ≤ p ≤ 1, (more
precisely, p is the probability that there is a bond between the two nodes
under discussion). Let Gm be a graph over the above vertex set, V , having
m bonds. Its probability is then
pr(Gm ) = pm · q N −m , (44)
N
where q := 1 − p. There exist m different labeled graphs Gm , having m
bonds, and the above probability is correctly normalized, i.e.
N
N m N −m
pr(G) = p q = (p + q)N = 1. (45)
m=0
m
This probability space is sometimes called the space of binomially random

graphs and denoted by G(n, p). Note that the number of edges is binomially
distributed, i.e.

N m N −m
pr(m) = p q (46)
m
and

m = m · pr(m) = N · p. (47)
The really fundamental observation made already by Erdös and Rényi (a

rigorous proof of this deep result can e.g. be found in Ref. [50]) is that there
www.ebook3000.com
are what physicists would call phase transitions in these random graphs.
To go a little bit more into the details, we have to introduce some more
graph concepts.
Definition 5.1 (Graph Properties). Graph properties are certain par-
ticular random variables (indicator functions of so-called events) on the
above probability space G, i.e. a graph property, Q, is represented by the
subset of graphs of the sample space having the property under discussion.
To give some examples: (i) connectedness of the graph (ii) existence

and number of certain particular subgraphs (such as subsimplices, etc.)
(iii) other geometric or topological graph properties, etc.
In this context, Erdös and Rényi made the following important
observation.
Observation 5.2 (Threshold Function). A large class of graph proper-
ties (e.g. the monotone increasing ones, cf. Refs. [50] or [51]) have a so-called
threshold function, m∗ (n), with m∗ (n) := N · p∗ (n), so that for n → ∞ the
graphs under discussion have property Q almost shurely for m(n) > m∗ (n)
and almost shurely not for m(n) < m∗ (n) or vice versa (more precisely: for
m(n)/m∗ (n) → ∞ or 0; for the details see the above cited literature). That
is, by turning on the probability p, one can drive the graph one is interested
in beyond the phase transition threshold belonging to the graph property
under study. Note that, by definition, threshold functions are only unique
up to “factorization”, i.e. m∗2 (n) = O(m∗1 (n)) is also a threshold function.
Calculating these graph properties is both a fascinating and quite intri-

cate enterprise. In our context, we are mainly interested in properties of
cliques, their distribution (with respect to their order, r, i.e. number of ver-
tices), frequency of occurence of cliques of order r, degree of mutual overlap,
etc. (cf. also Refs. [18] and [49]). These cliques shall be the building blocks
of our geometric renormalization process being described in the following
section. We shall relate these properties to the various assumed stages and
phases of our S-T manifold.
Definition 5.3. A subsimplex is a subgraph with all its vertices being

connected with each other. A clique is a maximal subsimplex, i.e. adding
another vertex to the subsimplex will destroy the property of being a sub-
simplex.
We can introduce various random function on the above probability

space. For each subset Vi ⊂ V of order r, we define the following random
350 M. Requardt
variable:

1, if Gi is an r-simplex
Xi (G) := , (48)
0, else
where Gi is the corresponding induced subgraph over Vi in G ∈ G (the prob-

ability space). Another random variable is then the number of r-simplices
in G, denoted by Yr (G) and we have:
(nr)

Yr = Xi (49)
i=1

with nr the number of r-subsets Vi ⊂ V . With respect to the probability
measure introduced above we have for the expectation values:

Yr = Xi (50)
i
and

Xi = Xi (G) · pr(Gi = r-simplex in G). (51)
G∈G
These expectation values were calculated in Ref. [49]. We have, for example,
r
Xi = p(2) . (52)
The probability that such a subsimplex is maximal, i.e. is a clique is

then (cf. Ref. [49])
r
pr(Gr is a clique) = (1 − pr )n−r · p(2) . (53)
n
As there exist exactly r possible different r-sets in the node set V , we
arrive at the following conclusion:
Conclusion 5.4 (Distribution of Subsimplices and Cliques). The
expectation value of the random variable ‘number of r-subsimplices’ is

n r
Yr = · p(2) . (54)
r
For Zr , the number of r-cliques (i.e. maximal! r-simplices) in the random
graph, we have then the following relation

n r
Zr = · (1 − pr )n−r · p(2) . (55)
r
www.ebook3000.com
These quantities, as functions of r (the order of the subsimplices) have

quite a peculiar numerical behavior. We are interested in the typical order
of cliques occurring in a generic random graph (where typical is understood
in a probabilistic sense.
Definition 5.5 (Clique Number). The maximal order of occurring
cliques contained in G is called its clique number, cl(G). It is another ran-
dom variable on the probability space G(n, p).
It is remarkable that this value is very sharply defined in a typical ran-

dom graph. Using the above formula for Zr , we can give an approximative
value, r0 , for this expectation value and get
r0 ≈ 2 log(n)/ log(p−1 ) + O(log log(n)), (56)
(cf. Chap. XI.1 of Ref. [50]). It holds that practically all the occurring
cliques fall in the interval (r0 /2, r0 ). We illustrate this with the following
tables. Our choice for n, the number of vertices, is 10100 . The reason for
this seemingly very large number is that we want to deal with systems
ultimately simulating our whole universe or continuous S-T manifolds (see
the more detailed discussion in Ref. [49]). We first calculate r0 .
p 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1
, (57)
r0 4370 2063 1291 901 664 502 382 286 200
(for reasons we do not understand we made some numerical errors in the
original Table 1 in [49, p. 2043], the correct numerical calculations can be
found in Ref. [18]).
It is more complicated to give numerical estimates of the distribution
of cliques, that is Zr . After some manipulations and approximations, we
arrived in ([49, p. 2051f]) at the following approximative formula and numer-
ical table (the numerical values are given for p = 0.7; note that for this
parameters the maximal order of occurring cliques, r0 , was approximately
1291)
log(Zr ) ≈ r · log(n) + n · log(1 − pr ) + r2 /2 · log(p), (58)
(with r2 /2 an approximation of r(r − 1)/2) for r sufficiently large).
r 600 650 800 1000 1200 1300 1400
log(Zr ) −5.7 · 106 3.2 · 104 3.2 · 104 2.5 · 104 8.4 · 103 −0.75 · 102 −1.1 · 104
(59)
(in the original Table 2 of Ref. [49], the numerical values for small and
large r’s, lying outside the interval (r0 /2, r0 ), were wrong as we neglected
352 M. Requardt
numerical contributions which are only vanishingly small in the above inter-
val, the correct calculations are taken from Ref. [18]). The above table nicely
illustrates how fast the frequency of cliques of order r drops to zero outside
the above interval.
As to the interpretation of these findings, one should remind the reader
that the above results apply to the generic situation, that is, do hold for
typical graphs (in very much the same sense as in corresponding discussions
in the foundations of statistical mechanics). An evaluation of the combina-
torial expressions in this and the following sections show that frequently
the same kind of extreme probabilistic concentration around, for example,
most probable values occurs as in ordinary statistical mechanics.
What is not entirely clear is how far the random graph approach can
be applied to our complex dynamical networks. Our working philosophy is
that these results serve to show, what we hope, is the qualitative behavior of
such systems. As our systems follow deterministic dynamical laws, starting
from certain initial conditions, the behavior cannot be entirely random in
the strict sense. This holds more so since we expect the systems to evolve
toward attracting sets in phase space and/or generate some large scale
patterns. On the other hand, due to the constant reorientation of the edges,
being incident with an arbitrary but fixed vertex and the generically large
vertex degrees of the vertices, one may assume that the system is sufficiently
random on small scales, so that the random graph picture reproduces at
least the qualitative behavior of such extremely complex systems.
To make this picture more quantitative, the general strategy is as fol-
lows. We count the typical number of active edges in our evolving network
at a given clock time t, calculate from this the corresponding edge prob-
ability, p(t), and relate this snapshot of our network to a random graph
with the same! edge probability. This should yield at least some qualitative
clues. That is, we expect that qualitative characteristics of our evolving
network can, at each given clock time, be related to the characteristics of
a corresponding random graph. In this specific sense, one may regard the
edge probability, p(t), as the crucial dynamical parameter of our network,
regarded as a statistical system.
6. A Geometric Renormalization Group

and the Continuum Limit of Discrete Geometries
In the preceding section, we introduced the notion of cliques as maximal
complete subgraphs of a given (random) graph. We are interested in them
www.ebook3000.com
because physically they represent lumps of vertices which are maximally

entangled among each other. That means, viewing the graph as the geo-
metric substrate on which our dynamical network lives, the vertices or the
respective DOF of a clique are directly interacting with each other. Invok-
ing the Wilson/Kadanoff picture of the renormalization group in statistical
mechanics these cliques are assumed to act as dynamical entities of their
own in the larger network.
In this context, we want to mention an older approach developed by
Menger and coworkers some time ago and which we discussed in greater
detail in Ref. [52]. Ideas in this direction (a geometry of lumps in which
points are not primary entities) were also briefly mentioned in Mengers con-
tribution to the anthology: Albert Einstein:Philosopher Scientist, Volume
II ([53]); note also the comments of Einstein in the same volume.
Remark 6.1. The cliques in our dynamical networks may change their
shape under the imposed dynamics which may create and/or delete edges.
This was one of the reasons why we developed the concept of fuzzy lumps
or fuzzy cliques in Ref. [52].
Similar to the block spin approach in the theory of critical phenomena

we promote the cliques of the initial (random) graph, G (level zero), to
the vertices of the next level (level one). We draw an edge between these
vertices of level one if the corresponding cliques have an overlap of a certain
order of vertices of level zero (see below). In this sense, we get a new graph,
the so-called clique graph, Cl(G) or Gcl . We can repeat this process, i.e. we
can form the graphs Cl(G), Cl2 (G) = Cl(Cl(G)) . . . and study the change
of geometric characteristics and/or emergence of new collective patterns on
the various levels of this process.
So what is the physical picture underlying this process? The idea of
the ordinary (infrared ) renormalization process in e.g. condensed matter
physics is to integrate out the microscopic details of the model and its
dynamics and, via coarse graining and rescaling, make visible the collective
and large scale properties of the model. This is accomplished by integrating
out in each step the DOF in the blocks or lumps of the preceding level and
make the new averaged DOF into the constituents of the next level, thus
establishing a new coarse grained and rescaled model together with a new
coarse grained Hamiltonian.
Slightly deviating from this philosophy, we want to concentrate in our
geometric renormalization process, for the time being, primarily on the
underlying geometric substratum. While we presume that one can treat
354 M. Requardt
e.g. gravitation (including a metric field gij ) as being emergent from some
dynamical network model as we have introduced in Sec. 2, we want at
the moment to deal only with the geometric large scale properties of our
dynamic graphs.
Remark 6.2. The following point is important. In the ordinary renormal-

ization process in, for example, some spin model the spins in a certain block
are regarded as behaving in a coherent way, i.e. provided the correlation
length is larger than the diameter of the blocks, the spins in some block are
expected to be more or less aligned. Therefore, it makes sense to build an
average spin as a new block variable. In our network we want, on the con-
trary, to implement both the wild vacuum fluctuations which are expected
to be very large on small scales and create, on the other hand, some smooth
classical S-T continuum on large scales. As a consequence of the dynamical
laws we introduced above, the individual vertex states in a certain clique
fluctuate considerably but, due to the close entanglement induced by the
edges, we expect them to cooperate strongly so that some global coherent
pattern may emerge. The picture we have in mind is that of an array of
coupled phase oscillators as will be described in the following section about
wormhole spaces and small world networks.
Observation 6.3. As a result of the process described in the following,

we arive at a picture where lumps or cliques are contained in larger cliques
(of the next level) which are again contained in larger cliques of the following
level etc. That is, we get a hierarchical picture of the concept of physical
points which have a rich internal structure.
Definition 6.4. We want to denote this double structure of an underlying

erratic network coexisting with a conjectured coherently behaving smooth
classical macroscopic surface structure by QX/S-T (quantum space plus
classical S-T).
We talked in Sec. 5.2 about the existence of threshold functions which

resemble phase transitions and phase transition lines in statistical mechan-
ics. Furthermore, the different phases are frequently characterized by non-
vanishing order parameters. In this spirit, we make the following definition:
Definition 6.5. If such a superstructure S-T emerges in our dynami-

cal network in a certain parameter regime we call it an order parameter
manifold.
www.ebook3000.com
Remark 6.6. This implies that the existence of such a S-T is a particular
property prevailing only in a certain region of the phase space of the model,
i.e. we may also have such a microscopic substratum without an overlying
classical S-T.
It is interesting to consider what takes the place of the so-called critical

systems of the ordinary renormalization group. Critical systems converge
to fixed points which, on their part, describe the large scale properties of
the critical system under discussion.
Conjecture 6.7. In our geometric case, where cliques and their entangle-
ment are expected to be the crucial building blocks, geometric critical systems
are conjectured to display some geometric long range order and a certain
self-similarity. This will be worked out in more detail in the following.
It turns out that it is advisable to split the investigation into two sub-
sections. In the first, which deals with the clique structure, we develop
the geometric coarse graining process. In the second section, we discuss
the rescaling process which leads to a continuum limit. Both parts of the
renormalization process have problems of their own and lead to quite deep
mathematics. The material of the following two subsections consists mainly
of a review of the content of the papers [18,49], parts of Refs. [14] and [19].
That is, due to length limitations, we have to refer the reader as to the
more complicated and partly quite intricate combinatorial and numerical
calculations to these papers.
6.1. The geometric coarse-graining

As indicated above, we assume that presently we live in a network scenario
where in the underlying microscopic network, called QX, there exists a
superstructure or order parameter manifold, S-T (classical S-T). The emer-
gence of S-T signals the transition from a chaoic initial phase, QX0 , to
a phase developing a near/far-order, i.e. a causal structure and relatively
stable physical points or (fuzzy) lumps (Menger).
Our physical picture concerning the initial scenario (big bang era) is the
following. The network QX started from a presumably densely entangled
initial phase QX0 , in which on average every pair of vertices, xi , xk , is
connected by an active edge with high probability p ≈ 1 or a Jik = 0
(in our examples of dynamical laws).
We then envisage two main epochs of our evolving network, a so-called
embryonic epoch with a still large edge probability near p = 1 and an
356 M. Requardt
unfolded epoch with a much smaller value of p and a large diameter at least
on some higher clique level.
Observation 6.8. One should note that pure random graphs have a very
small diameter for a large range of p-values (cf. e.g. Ref. [50]) so that p
must be quite small if we want to have an unfolded epoch.
Definition 6.9. The diameter of a graph is defined by maxxi ,xk d(xi , xk ).
In Sec. 5.2, we calculated the typical order and number of cliques and
provided two tables for the order of the largest expected cliques (the clique
number of the random graph), r0 , and number of r-cliques. The numerical
values were p = 0.7, n = 10100 . We showed that almost all cliques have an
order between r0 and r0 /2.
In the first step, we want to clarify the mathematical and physical pro-
cess of constructing the clique graph Cl(G) of a graph G.
Definition 6.10. We employ two methods of constructing the clique graph:
1. The mathematical clique graph operator Clm is defined as above.

2. The physical clique graph operator Clp is constructed in the following
way. We delete too small (marginal) cliques which do not lie in the above
described interval (r0 , r0 /2). An edge between cliques Ci , Ck is drawn if
the overlap is non-marginal (we discuss the physical implications of this
notion below). That is, we say, the overlap of the respective cliques is
non-marginal if it is larger than a certain value l0 (r) which depends on
r and the underlying physics.
Observation 6.11. In this way, the original graph and the mathematical
clique graph are purified, put differently, the iterated mathematical clique
graph is coarse-grained, i.e. on each level some marginal structure is deleted.
Remark 6.12. To give examples for l0 ; after referring to Sec. V.B in

Ref. [18] we took n = 10100 , p = 0.7 which yielded r0 = 1291. Our choice
for l0 was l0 = 50. We, however, convinced ourselves that the physical pic-
ture does not critically depend on this choice. For example, l0 = 30, we got
similar results.
The physical motivation behind this procedure is as follows. Our aim

is to construct a S-T substratum in which we have, on the one hand, a
near order structure concerning interaction and flow of information among
neighboring physical points of S-T (i.e. the classical S-T concept under a
www.ebook3000.com
certain magnification, so that the internal structure of the points of the

manifold becomes visible). On the other hand, we have weak, translocal ties
between lumps, which are with respect to the ordinary S-T metric some
distance apart. These weak ties result from edges which have been deleted
in the original graph G or on some lower clique level, say Clm (G), relative
to some clique level Cln (G) with n > m or a level which is already near our
presumed classical S-T. This dual picture will become clearer in the next
section on wormhole spaces and small world networks.
Conjecture 6.13. We conjecture that these weak translocal ties are respon-
sible for some aspects of quantum theory. This will be discussed in more
detail in the following section about wormhole spaces and small world
networks.
In this context, we regard the cliques of some given level as approximately

autonomous subunits which are coupled both with their neighbors via
strong ties and with more distant cliques via some weak ties. We conjecture
that this dual structure will lead to a sort of global coherence which results
in the emergence of the order parameter manifold S-T.
We now want to briefly discuss some analytic and numerical results
of our construction of the (purified) clique graph. For the details of the
mostly quite intricate calculations we refer to the above mentioned papers.
We begin with the notion of the embryonic epoch (cf. Sec. 4.1 of Ref. [49]).
In this epoch most of the edges are still active, i.e. p ≈ 1. In Sec. 4.1 of
Ref. [49] we dealt with the question under what conditions all the cliques
of the graph do have a common non-empty overlap. On the other hand,
for sufficiently small p we could show that this overlap is empty with high
probability. This regime describes presumably the so-called unfolded epoch.
The main work consists of providing combinatorial/probabilistic expressions
for these graph properties.
In the next step, we want to calculate the order of the socalled local group
of a fixed clique. This local group comprises the cliques which have non-
marginal overlap with an arbitrary but fixed clique, C0 . This is an interest-
ing graph property in the unfolded epoch because it defines the infinitesimal
neighborhood of a vertex in the clique graph. The necessary combinatori-
cal/probabilistic analysis is made in Sec. 4.1 of Ref. [49] and Sec. V.B of
Ref. [18]. The number of cliques overlapping with some given clique C0 is,
as all these properties, a random variable in our random graph picture. We
denote it by N (C0 ; r , l) and its expectation value by N (C0 ; r , l). In this
expression, C0 is a clique of fixed order r. Its overlap is considered with
358 M. Requardt
respect to cliques of some order r , both lying in the interval (r0 , r0 /2).
The overlap is denoted by l ≥ l0 with l0 some minimal value (denoting the
non-marginal overlap). In the end, this expression is summed over all r
lying in then interval (r0 , r0 /2).
In the same way, we analyze the other defining parameters of the clique
graph (cf. Sec. V.B of Ref. [18]).
Remark 6.14. One should note that one can infer from our second table
in Sec. 5.2 that, generically, there do exist much more cliques than vertices
in a typical random graph, i.e. Cl(G) has typically much more vertices than
G.
Observation 6.15. Our numerical analyis shows after some intricate cal-
culations and estimates that the clique graph of a random graph G with
n = 10100 , p = 0.7, clique overlap l ≥ l0 = 50, (60)

4 3
has typically 1010 nodes, the edge probability is approximately 10−7·10
4
and the vertex degree is of the order of 100.3·10 .
Remark 6.16. Note that the small edge probability is compensated by the
huge vertex degree.
Proceeding in the same way, while however readjusting the non-marginal

overlap l0 (r) on each level, we get the iterated purified clique graph. We con-
vinced ourselves that the gross parameters of the clique graphs on the con-
secutive levels seem to reach stable values after only a few renormalization
steps.
Observation 6.17. As the whole scenario is much more irregular in our

case compared to the situation in e.g. statistical mechanics, the existence
of a fixed limit network under the geometric renormalization group (more
properly, modulo graph isomorphisms) should presumably be refined a lit-
tle bit. We introduce the important concept of quasi-isometry or coarse
isometry in the following subsection. We hence expect that the limit of an
iterated clique graph is only invariant under the physical clique operator
Clp modulo quasi-isometries.
Definition 6.18. An ordinary graph isomorphism I is a bijective map

between graphs so that the graphs are forminvariant, i.e. an edge exists
between two image vertices iff there exists an edge between the preimages.
www.ebook3000.com
Remark 6.19.
1. In Sec. VI of Ref. [18], we gave simple examples which show that our
expectations are aparently not completly far fetched. These examples
should, however, not be regarded as examples of possible S-Ts. We see
that both limit points and limit cycles are possible.
2. In pure mathematics, the study of the iterated clique graph is indeed an
interesting field in graph theory (cf. for example, Refs. [54–56]).
We want to close this subsection by giving some important results for the
unpurified, mathematical clique graph. We begin with the graph property
connectedness.
Lemma 6.20. If G is connected (i.e. each two vertices can be connected by

a path), the same holds for Clm (G).
Proof. Let C, C be two cliques and x, x two vertices with x ∈ C , x ∈ C .

There exists a path connecting x and x . We denote the corresponding
vertices by x = x0 , x1 , . . . , xk = x . There exist cliques, Ci , containing the
pairs (xi , xi+1 ) with i = 0, . . . , k − 1. Note that the cliques Ci , Ci+1 have
non-zero overlap. This proves the statement.
Remark 6.21. The property of connectedness need of course no longer

hold for the coarse-grained clique graph, Clp (G). In the random graph
framework, this property becomes a random variable and its probability
can be calculated (cf. e.g. Conclusion 5.10 in Ref. [18]).
Another interesting property is the possible change of distance under the

clique operator. The following lemma shows that distances do not change
significantly.
Lemma 6.22. Let G be a connected graph and C, C two cliques in Clm (G),
having the distance dCl(G) (C, C ). Then we have
dCl(G) (C, C ) = min d(x, x ) + 1, x ∈ C, x ∈ C . (61)
x,x
Proof. The proof is essentially already contained in the proof of the following
theorem (cf. Theorem 7.4 in Ref. [18] or see Ref. [56]). Choose a minimal-
length path in Clm (G) connecting C and C and consisting of the cliques
C = C0 , C1 , . . . , Ck = C i.e. dCl(G) (C, C ) = k (62)
implying that Ci , Ci+1 have non-empty overlap. Choose a path in G starting
at some x1 ∈ C0 ∩ C1 , with x2 ∈ C1 ∩ C2 , . . . . This is a path with initial
360 M. Requardt
point in C, endpoint in C , having the length k − 1. It is easy to show that

such a path is minimal under this condition, i.e. it holds k = (k − 1) + 1,
which proves the statement.
The following observation is non-trivial and important.
Theorem 6.23. We assume that G, having a globally bounded vertex
degree, has scaling dimension D (cf. Sec. 4). It follows that Clm (G) has
also a dimension with
DClm (G) = D. (63)
The proof is more intricate and can be found in Sec. VII of Ref. [18].
Remark 6.24. While there exists the restriction of a globally vertex
degree, the result shows that when building the iterated unpurified clique
graph the microscopic dimension remains fixed in the transitions to higher
levels. That is, if we surmise that our physical S-T is smoother than the
underlying microscopic substratum, implying among other things that also
the graph dimension on the different levels of the iterated clique graph is
expected to vary, we have to employ the concept of the physical clique graph
operator Clp with its purification and coarse-graining properties. That it
is not easy to change the graph dimension at all is shown in Sec. VIII of
Ref. [18]. We will briefly come back to this point in the following section.
6.2. The rescaling process and the continuum limit

In the preceding subsection, we remained within the class of discrete net-
works or graphs. i.e. at all levels of our constructions the models under
discussion were discrete. We will now develop the framework which allows
to construct continuum limit models of our discrete networks. While such a
process is perhaps transparent in the context of models living on a periodic
Bravais lattice, it becomes very intricate in the case of general irregular net-
works with a relatively deep amount of mathematics being involved. The
general context is the theory of general metric spaces. We shall make heavy
use of material being developed for example in Refs. [57, 58] and [43]. This
subsection referees the content of Ref. [19] where more references can be
found.
An important conceptual tool is the notion of quasi-isometry. This is
the appropriate generalization of the notion of isometry to dissordered and
irregular spaces where marginal details and variations are partly ignored.
Definition 6.25. Let F be a map from a metric space, X, to a metric
space, Y with metrics dX , dY . It is called a quasi-isometric embedding if
www.ebook3000.com
the following holds: There exist constants, λ ≥ 1, ≥ 0, such that
λ−1 · dX (x, y) − ≤ dY (F (x), F (y)) ≤ λ · dX (x, y) + . (64)
If, furthermore, there exists a constant such that for all y ∈ Y we have
dY (y, F (X)) ≤ , i.e. Y ⊂ U (F (X)) (for the definition of U (A) see below)
it is called a quasi-isometry; the spaces are then called quasi-isometric.
There is an equivalent definition which shows that the preceding definition
is in fact symmetric between X and Y (see for example [43]). That is, there
exists a quasi-isometric map G from Y to X with corresponding constants
and dX (G ◦ F (x), x) ≤ ρ and dY (F ◦ G(y), y) ≤ ρ for some ρ. If λ = 1, it is
called a rough isometry.
It is an important observation that in our framework of networks and

graphs many properties are stable under quasi-isometries. We have, for
example, the following results:
Observation 6.26. The “growth type” of graphs with globally bounded

vertex degree is stable under quasi-isometry (cf. Sec. 4 and Ref. [43]). We
have in particular that quasi-isometric graphs have the same dimension, i.e.
(with G1 , G2 quasi-isometric graphs)
D̄1 = D̄2 , D1 = D2 , (65)
(see Ref. [19] for a proof).
A further interesting observation is the following. We introduced in Sec. 4

the concept of local edge insertions/deletions and showed that these proce-
dures do not alter the dimension in the case of graphs with globally bounded
vertex degree. We have the following observation.
Observation 6.27. Local edge insertions/deletions lead to quasi-

isometries. By the same token, we see that via this method we get a rich
class of examples of quasi-isometric graphs.
It is of great interest to derive criteria under what conditions a network

has a finite growth degree (in contrast to e.g. an exponential growth) and
a fortiori, an integer dimension. This is discussed in greater detail in Sec. 3
of Ref. [19]. We want furthermore to emphasize that perhaps rather sur-
prisingly our physically motivated interests turn out to be closely related
to a field of pure mathematics, i.e. geometric group theory via the concept
of Cayley graph (cf. Ref. [43]).
362 M. Requardt
We now come to the construction of the continuum limit of an infinite

graph (for finite graphs the construction is not particularly interesting as it
leads to a single point). To this end, we have in a first step to construct a
metric on a set of graphs or sets as elements (i.e. a metric on a set of metric
spaces). We begin with the definition of the so-called Hausdorff-metric on
a space of subsets of a metric space.
Definition 6.28. Let X be a metric space, U (A) the -neighborhood of a

subset A ⊂ X (U (A) := {y : d(x, y) < for some x ∈ A}. The Hausdorff-
distance between A, B ⊂ X is then given by
dH (A, B) := inf{; A ⊂ U (B), B ⊂ U (A)}. (66)
We have the following lemma:
Lemma 6.29. With X a compact metric space, the closed subsets of X

form a compact (i.e. complete) metric space with respect to dH (see e.g.
[58] or [59]).
In the following, it is sometimes useful to make a slight generalization to

pseudo metric spaces as we will encounter situations where spaces or sets
have zero Gromov–Hausdorff-distance (for example, the one being a dense
subset of the other) while they are not strictly the same. Everything we
will state for metric spaces in the following will also hold for pseudo metric
spaces.
Definition 6.30. A pseudo metric fulfills the same axioms as a metric with
the exception that d(a, b) = 0 → a = b does not necessarily hold.
The above distance concept is too narrow to be useful in a more general

context. It was considerably generalized by Gromov in an important way
(see Ref. [61]) and later slightly modified by himself and other authors
([57, 58, 60]). What is really beautiful in our view is that while it seems
to be more abstract, it encodes the really important and crucial aspects
of similarity or “nearness” of spaces in a more satisfying way. That is, it
measures their structural similarity and not simply the nearness of two
structureless sets of points in a space. In general, it is a pseudo metric
which may even take the value infinity. For compact spaces, it is always
finite. If one forms equivalence classes of compact spaces under isometries,
it becomes a true metric.
The Gromov–Hausdorff distance, dGH , can be formulated in two equiv-
alent ways.
www.ebook3000.com
Definition 6.31. dGH (X, Y ) between two metric spaces, X, Y is defined

as the infimum of dZ H (f (X), g(Y )) over all metric spaces Z and isometric
embeddings, f, g, of X, Y into Z.
Equivalently, one can define dGH by the infimum over dH (X, Y ) in X Y
(disjount union) equipped with the metrics dXY which extend the respec-
tive metrics dX , dY in X, Y .
The crucial part of the distance concept is always the triangle-inequality.

Furthermore, we have to show that the above infimum is again a metric.
This is proved in Sec. 4 of Ref. [19] where quite a few more interesting results
are discussed. Due to lack of space, we directly embark on the deep results
of Gromov concerning the formulation of convergence of spaces toward each
other.
We now present the fundamental Gromov-compactness theorem, first
for compact spaces, then for more general cases.
Definition 6.32. We call a family of compact spaces, Xλ , uniformly com-

pact if their diameters are uniformly bounded and if for each > 0 Xλ is
coverable by N < ∞ balls of radius independent of the index λ.
Theorem 6.33 (Gromov). A sequence {Xi } contains a convergent sub-

sequence in dGH iff {Xi } is uniformly compact.
Proof (see Refs. [60, 61] or [58]). Typically, an Arzela–Ascoli–Cantor-

diagonal-sequence-like argument is used in the proof.
In our framework, we are mainly interested in infinite graphs, i.e. non-
compact metric spaces being however frequently proper.
Definition 6.34. A metric space, X, is called proper if all its closed balls,
B(x, r), are compact.
We can then extend the above result in the following way. Ordinary GH-
convergence works well in the category of compact metric spaces. If the
spaces are non-compact, a slightly modified approach is more satisfactory.
One problem which may arise is that things in unbounded spaces can “wan-
der away” to infinity. So it is reasonable to pin down the members of the
sequence of spaces at certain points, so that they can be better compared.
More precisely, we work in the category of pointed metric spaces, (X, x),
which is, a fortiori pretty normal from the physical point of view as it is
like introducing a reference point or a coordinate system.
364 M. Requardt
Definition 6.35. The sequence of pointed metric spaces, (Xi , xi ), is said

to converge to (X, x) in pointed GH-sense if for every r > 0 the sequence
of closed balls, B(xi , r), converges to B(x, r) in dGH .
The Gromov-uniform-compactness theorem now reads:
Theorem 6.36. If for all r and > 0 the balls B(xi , r) of a given sequence
(Xi , xi ) are uniformly compact, then a subsequence of spaces converges in
pointed GH-sense.
Remark 6.37. There exist various slightly different notions of pointed con-
vergence in the literature. One can, for example, define pointed GH-distance
by admitting only isometries which map the base points onto each other
[57]. Another possibility is to include the distance of the images of the base
points in the definition [62]. The above definition is used in Ref. [58].
We now apply these techniques to the following sequence of graphs.

We start with a graph, G, of globally bounded vertex degree, v, and, taking
G with the original graph metric, d, as initial metric space, generate a
sequence, or, more generally, a directed system of metric spaces, λG, by
taking the same graph, G, but now with the scaled metric, λd, defined as
λd(x, y) := λ · d(x, y) (67)
and (usually) taking λ → 0. One may, in particular, take subsequences of

the kind
Gn , dn := n−1 · d, n → ∞ (68)
or replace n by 2−k .
In the next step, we have to show that all the above criteria are fulfilled
in this case which is a non-trivial task (see Sec. 5 of Ref. [19]). Among other
things, several new notions and concepts have to be introduced like e.g.
doubling measures, etc. Finally, we can show that our sequence of rescaled
graphs has a continuum limit! It is now very important to learn something
about the structure of this limit space. Some steps are done in Sec. 5 of
Ref. [19]. It is of particular importance to understand under what conditions
this limit space is a smooth manifold or, on the other hand, a chaotic space
of rather fractal type. We are very interested in the possibility of a limit
space having a superficially smooth structure together with an internal
infinitesimal more erratic structure “around” the “classical” points of the
base manifold, being kind of a generalized fiber space.
www.ebook3000.com
7. Wormhole Spaces, Holography and the Translocal

Structure of Quantum Theory
In this last short section, which is also kind of a conclusion, we want to dis-
cuss in a very sketchy way various fundamental (open) questions in modern
physics and show how our above framework can be applied to them. As we
shall treat these important topics in a quite cursory way, we refer the inter-
ested reader to Refs. [14, 17, 63] for a more thorough discussion.
In Sec. 6, we came already to the conclusion that presumably the cellular
network substratum which has the propensity to lead to a continuum limit,
resembling our physical (quantum) S-T, has to be in a peculiar critical
state which resembles a scale free small world network as we described it in
Ref. [14]. This means that apart from a certain nearorder leading to a lumpy
local structure in the network on the various scales of coarse graining and
rescaling there remains on all scales a certain sparse network of so-called
translocal connections (i.e. edges in the graph language relating distant
regions with respect to the ordinary local metric). Rather surprisingly such
a structure was analyzed roughly at the same time in a quite different
area of science, dubbed small world networks (for a detailed treatment see
e.g. [64]. A brief discussion can also be found in Ref. [16]. In Ref. [16], we
described such models in the following way.
We start with a regular graph having a nearorder structure like e.g. Zn
with edges to nearest neighbors in the vertical and horizontal direction. In a
scaling limit this would lead to a space like Rn with the Euclidean distance
metric. In the next step, we superimpose Zn or a similar regular space with
a sparse random graph on the same vertex set and a small edge probability
p. These random edges now lead to an additional translocal structure on
Zn . We conjecture that this serves as a toy model for the kind of substratum
underlying our physical (quantum) S-T which we dubbed QX/S-T (i.e. an
underlying discrete substratum having both a near and a farorder together
with a continuous surface structure on a low resolution of S-T).
Observation 7.1. A characteristic of small world networks is their surpris-

ingly small diameter or mean distance (cf. e.g. Refs. [14, 16] or Ref. [64]),
given the sparseness of additional translocal edges.
In Ref. [17], we developed and studied this phenomenon in quite some

detail and coined the notion wormhole spaces for such structures like our
QX/S-T. We showed that the BH entropy area law and the holographic
principle follow quite naturally from our framework.
366 M. Requardt
Finally, in Ref. [63], we undertook to develop a framework which

describes the quantum phenomena as emergent properties on a mesoscopic
scale. More precisely, at various places in our chapter we indicated that our
dynamic network of local lumps or cliques can be associated to a network
of coupled phase oscillators (cf. Remark 2.5 or Remark 6.2).
Remark 7.2. Many nonlinear systems approach limit cycles in their evo-
lution on which they then evolve according to a law like θ̇ = ω, ω a certain
specific natural frequency, θ the phase of the system. For more details, see
Refs. [64, 65] or [63].
The famous Kuramoto model desribes a large population of coupled

limit cycle or phase oscillators whose natural frequencies are drawn from
some prescribed distribution (see for example, [66]). The hallmark of such
models is that there may occur a particular kind of phase transition in
which all the initially different natural frequencies, ωi , become dynamically
synchronized.
Conjecture 7.3. We expect that our dynamical network models show a

similar behavior with the cliques or lumps representing the limit cycle or
phase oscillators.
It is an important observation [67] that this emergent property of syn-

chronization is strongly enhanced and stabilized by a certain sparse non-
local network of random couplings superimposed on the prevailing network
of local couplings in the array of oscillators. This is exactly what we found
in our network models.
Conjecture 7.4. This possibility of phase locking may be a hint how a

global time function emerges from the array of initially different local times
with the phase oscillators viewed as local clocks.
It is remarkable that Bohm, starting from a different direction, also

speculates about the existence of a hierarchy of coupled oscillators on con-
secutive scales of (quantum) S-T (see Ref. [68]). He came to conclusions
similar to the ones uttered in Ref. [63] concerning the consequences for
quantum theory as being emergent from such a deep structure. We dis-
cussed this in greater detail in Ref. [63].
www.ebook3000.com
References
1. A. Ilachinski, Cellular Automata, a Discrete Universe, World Scientific Publ.,
Singapore, 2001.
2. K. Zuse, The computing universe, Int. J. Theor. Phys. 21, 589 (1982).
3. R.P. Feynman, Simulating physics with computers, Int. J. Theor. Phys. 21,
467 (1982).
4. E. Fredkin, Digital mechanics, Physica D 45, 254 (1990) or the (program-
matic) material on his homepage http://www.digitalphilosophy.org.
5. D. Finkelstein, The space-time code, Phys. Rev. 184, 1261 (1969).
6. T.D. Lee in Selected Papers, From Random Lattices to Gravity, Vol. 3, ed.
G. Feinberg, Birkhaeuser Boston, 1986.
7. G. ’t Hooft, Quantization of discrete deterministic theories, Nucl. Phys. B
342, 471 (1990).
8. G. ’t Hooft, Quantum gravity as a dissipative deterministic system, Class.
Quant. Grav. 16, 3263 (1999), arXiv: gr-qc/9903084.
9. G. ’t Hooft, The cellular automaton approach of quantum mechanics,
arXiv:1405.1548.
10. S. Wolfram, A New Kind of Science, Wolfram Media, 2002.
11. K. Svozil, Computational universe, Chaos, Solitons, Fractals 25, 845 (2005),
arXiv: physics/0305048.
12. G. ’t Hooft, Can quantum mechanics be reconciled with CA? Int. J. Theor.
Phys. 42, 349 (2003).
13. K. Svozil, Are quantum fields cellular automata? Phys. Lett. A 119, 153
(1987).
14. M. Requardt, Scale free small-world networks and the structure of quantum
space-time, arXiv:gr-qc/0308089.
15. M. Requardt, Wormhole spaces, connes’ points, speaking to each other, and
the translocal structure of quantum theory, arXiv:hep-th/0205168.
16. A. Lochmann and M. Requardt, An analysis of the transition zone between
the various scaling regimes in the small-world model, J. Stat. Phys. 122, 255
(2006), arXiv:cond-mat/0409710.
17. M. Requardt, Wormhole spaces: The common cause for the black hole
entropy-area law, the holographic principle and quantum entanglement,
arXiv:0910.4017.
18. M. Requardt, A geometric renormalisation group and fixed-point behavior
in discrete quantum space-time, J. Math. Phys. 44, 5588 (2003), arXiv:gr-
qc/0110077.
19. M. Requardt, The continuum limit of discrete geometries, Int. J. Geom.
Meth. Mod. Phys. 3, 285 (2006), arXiv:math-ph/0507017.
20. S. Kauffman, Origins of Order: Self-Organisation and Selection in Evolution,
Oxford Univ. Pr., Oxford, 1993.
21. S. Kauffman, At Home in the Universe, The Search for Laws of Self-
Organisation and Complexity, Oxford Univ. Pr., Oxford, 1995.
22. S. Kauffman, Antichaos and adaption, Sci. Am. 265, 76 (1991).
368 M. Requardt
23. C.G. Langton: ed. Artificial Life: Proc. of an Interdisciplinary Workshop

on the Synthesis and Simulation of Living Systems, September 1987, Los
Alamos, Addison-Wesley Publ., N.Y., 1989.
24. M. Gardner, On CA, Self-Reproduction. The garden of eden and the game
of life, Sci. Am. 224, 112 (1971).
25. I. Licata, Almost-anywhere Theories, Reductionism and Universality of
Emergence, arXiv:o912.5012.
26. T. Konopka, F. Markopoulou and L. Smolin, Quantum graphity,
arXiv:0611197.
27. T. Konopka, F. Markopoulou and S. Severini, Quantum graphity: A model
of emergent locality, PR D77, 104029 (2008), arXiv:0801.0861.
28. T. Konopka, Statistical mechanics of graphity models, PR D78, 044032
(2008), arXiv:0805.2283.
29. F. Antonsen, Random graphs as a model for pregeometry, Int. J. Theor.
Phys. 33, 1189 (1994).
30. B. Bollobas, Modern Graph Theory, Springer, N.Y., 1998.
31. T. Nowotny and M. Requardt, Emergent properties in structurally dynamic
disordered cellular networks, J. Cell. Automata 2, 273 (2007), arXiv:cond-
mat/0611427.
32. T. Toffoli and N. Margolus, Cellular Automaton Machines, MIT Pr., Cam-
bridge Mass., 1987.
33. N. Biggs, Algebraic Graph Theory, 2nd Editon, Cambridge Univ. Pr., Cam-
bridge, 1993.
34. C. Godsil and G. Royle, Algebraic Graph Theory, Springer, Berlin, 2001.
35. M. Requardt, Cellular networks as models for planck scale physics, J. Phys.
A: Math. Gen. 31, 7997 (1998), arXiv:hep-th/9806135.
36. M. Requardt, Dirac operators and the calculation of the connes metric
on arbitrary (infinite) graphs, J. Phys. A. Math. Gen. 35, 759 (2002),
arXiv:math-ph/0108007.
37. M. Requardt, Supersymmetry on graphs and networks, Int. J. Geom. Meth.
Mod. Phys. 2, 585 (2005), arXiv:math-ph/0410059.
38. B. Felsager, Geometry, Particles, and Fields, Springer, N.Y., 1998.
39. W. Hodge, The Theory and Applications of Harmonic Integrals, 2nd Edition,
Cambridge Univ. Pr., Cambridge, 1952.
40. G.A. Edgar, Measure, Topology, and Fractal Geometry, Springer, N.Y., 1990,
or K. Kuratowski, Topology, Vol. 1, Acad. Press., N.Y., 1966.
41. F. Buckley and F. Harary, Distance in Graphs, Addison-Wesley, N.Y., 1990.
42. T. Nowotny and M. Requardt, Dimension theory on graphs and networks, J.
Phys. A: Math. Gen. 31, 2447 (1998), arXiv:hep-th/9707082.
43. P. de la Harpe, Topics in Geometric Group Theory, Univ. Chicago Pr.,
Chicago, 2000.
44. K.J. Falconer, Fractal Geometry, Wiley, Chichester, 1990.
45. D. Dhar, Lattices of effectively nonintegral dimensionality, J. Math. Phys.
18, 577 (1977).
www.ebook3000.com
46. K. Hattori, T. Hattori and H. Watanabe, Gaussian field theories on general

networks and the spectral dimension, Progr. Theor. Phys. Suppl. 92, 108
(1987).
47. T. Filk, Equivalence of massive propagator distance and mathematical dis-
tance on graphs, Mod. Phys. Lett. A 7, 2637 (1992).
48. B. Durhuus, Hausdorff and spectral dimension of infinite random graphs,
Act. Phys. Pol. B 40, 3509 (2009).
49. M. Requardt, (Quantum) Space-Time as a statistical geometry of lumps in
random networks, Class. Quant. Grav. 17, 2029 (2000), gr-qc/9912059.
50. B. Bollobas, Random Graphs, Acad. Pr., N.Y., 1985.
51. B. Bollobas, Combinatorics, Cambridge Univ. Pr., London, 1986.
52. M. Requardt and S. Roy, (Quantum) Spacetime as a statistical geometry of
fuzzy lumps and the connection with random metric spaces, CQG 18, 3039
(2001), arXiv:gr-qc/9912059.
53. K. Menger in Albert Einstein:Philosopher Scientist, ed. P.A. Schilpp, 3rd
Edition, Cambridge University Pr., London, 1970.
54. F. Larrion, V. Neumann-Lara, M.A. Pizana and T.D. Porter, Recognizing
self-clique graphs, Mathematica Contemporanea 25, 125 (2003).
55. M.A. Pizana, The icosaheron is clique divergent, Discr. Math. 262, 229
(2003).
56. M.A. Pizana, Distances and diameters on iterated clique graphs, Discr. Appl.
Math. 141, 255 (2004).
57. M. Gromov, Metric Structures for Riemannian and Non-Riemannian Spaces,
Birkhaeuser, N.Y., 1998.
58. M.R. Bridson and A. Haeflinger, Metric Spaces of Non-Positive Curvature,
Springer, N.Y., 1999.
59. G.A. Edgar, Measure,Topology, and Fractal Geometry, Springer, Berlin, 1990.
60. P. Petersen, Gromov-Hausdorff convergence of metric spaces, AMS Proc.
Pure Math. 54(3), 489 (1993).
61. M. Gromov, Groups of polynomial growth and expanding maps, Publ. Math.
IHES 53, 53 (1981).
62. P. Petersen, Riemannian Geometry, Chapter 10, Springer, Berlin, 1991.
63. M. Requardt, Quantum theory as emergent from an undulatory translocal
sub-quantum level, arXiv:1205.1619.
64. D. Watts, Small Worlds, Princeton Univ. Pr., Princeton, 1999.
65. S.H. Strogatz, Nonlinear Dynamics and Chaos, Perseus Books, Cambridge
(USA) 1994.
66. S.H. Strogatz, From Kuramoto to Crawford: Exploring the onset of synchro-
nization in populations of coupled oscillators, Physica D 143, 1 (2000).
67. E. Niebur, H.G. Schuster, D.M. Kammen and C. Koch, Oscillator-phase cou-
pling for different two-dimensional network connectivities, PR A 44, 6895
(1991).
68. D. Bohm, Wholeness and the Implicate Order, Routhledge and Kegan,
London 1980.
www.ebook3000.com
Chapter 14
On a Time–Space Operator
(and other Non-Self-Adjoint
Operators) for Observables
in QM and QFTa
Erasmo Recami
INFN-Sezione di Milano, Milan, Italy
Facoltà di Ingegneria, Università statale di Bergamo,
Bergamo, Italy
recami@mi.infn.it
Michel Zamboni-Rached
DECOM, FEEC, UNICAMP, Campinas, SP, Brazil
mzamboni@decom.fee.unicamp.br
Ignazio Licata
ISEM, Institute for Scientific Methodology, Palermo, Italy
ignazio.licata@ejtp.info
The aim of this paper is to show the possible significance, and usefulness,
of various non-self-adjoint operators for suitable Observables in non-
relativistic and relativistic quantum mechanics (QM), and in quantum
electrodynamics. More specifically, this work deals with: (i) the Hermi-
tian (but not self-adjoint) Time operator in non-relativistic QM and in
quantum electrodynamics; (ii) idem, the introduction of Time and Space
operators; and (iii) the problem of the four-position and four-momentum
operators, each one with its Hermitian and anti-Hermitian parts, for
relativistic spin-zero particles. Afterwards, other physical applications
a Work supported in part by INFN, Italy. One of us (ER) acknowledges a recent
research fellowship (No. 2013/12025-8) by FAPESP; while another author (MZR) wishes
to acknowledge partial support from the brazilian Institutions FAPESP (under grant
11/51200-4), and CNPq (under Grant 307962/2010-5).
371
372 E. Recami, M. Zamboni-Rached and I. Licata
of non-self-adjoint (and even non-Hermitian) operators are briefly dis-

cussed. We mention how non-Hermitian operators can indeed be used
in physics [as it was done, elsewhere, for describing Unstable States];
and some considerations are added on the cases of the nuclear optical
potential, of quantum dissipation, and in particular of an approach to
the measurement problem in QM in terms of a chronon. This paper is
largely based on work developed, along the years, in collaboration with
V.S. Olkhovsky, and, in smaller parts, with P. Smrz, with R.H.A. Farias,
and with S.P. Maydanyuk.
1. Introduction
This chapter is largely based on work developed in a large part, along the
years, with V.S. Olkhovsky, and, in smaller part, with P. Smrz, with R.H.A.
Farias, and with S.P. Maydanyuk.
Time, as well as three-position, sometimes is a parameter, but sometimes
is an observable that in quantum theory would be expected to be associated
with an operator. However, almost from the birth of quantum mechanics
(QM) (cf., e.g. Refs. [1, 2]), it is known that time cannot be represented by
a self-adjoint operator, except in the case of special systems (such as an
electrically charged particle in an infinite uniform electric field).b The list
of papers devoted to the problem of time in QM is extremely large (see, for
instance, Refs. [3–38], and references therein). The same situation had to be
faced also in quantum electrodynamics and, more in general, in relativistic
quantum field theory (see, for instance, Refs. [3, 4, 26, 27, 38]).
As to QM, the very first relevant articles are probably Refs. [3–15], and
references therein. A second set of papers on time in quantum physics [16–
37] appeared in the 90s, stimulated partially by the need of a consistent
definition for the tunneling time. It is noticeable, and let us stress it right
now, that this second set of papers seems however to have ignored Naimark’s
theorem [39], which had previously constituted (directly or indirectly) an
important basis for the results in Refs. [3–15]. Moreover, all the papers
in Refs. [16–23] attempted at solving the problem of time as a quantum
observable by means of formal mathematical operations performed outside
b This is a consequence of the semi-boundedness of the continuous energy spectra from

below (usually from zero). Only for an electrically charged particle in an infinite uniform
electric field, and other very rare special systems, the continuous energy spectrum is not
bounded and extends over the whole axis from −∞ to +∞. It is curious that for systems
with continuous energy spectra bounded from above and from below, the time operator
is however, self-adjoint and yields a discrete time spectrum.
www.ebook3000.com
On a Time–Space Operator for Observables in QM and QFT 373
the usual Hilbert space of conventional QM. Let us recall that Naimark’s
theorem states [39] that the non-orthogonal spectral decomposition of a
Hermitian operator can be approximated by an orthogonal spectral function
(which corresponds to a self-adjoint operator), in a weak convergence, with
any desired accuracy.
The main goal of the first part of the present chapter is to justify the
use of time as a quantum observable, basing ourselves on the properties
of the Hermitian (or, rather, maximal Hermitian) operators for the case of
continuous energy spectra: cf., e.g. the Refs. [24–27, 38].
The question of time as a quantum-theoretical observable is conceptu-
ally connected with the much more general problem of the four-position
operator and of the canonically conjugate four-momentum operator, both
endowed with an Hermitian and an anti-Hermitian part, for relativistic
spin-zero particles: This problem is analyzed in the second part of this
chapter.
In the third part of this work, it is briefly mentioned that non-Hermitian
operators can be meaningfully and extensively used in physics [as it was
done, elsewhere, for describing unstable states (decaying resonances)]. And
some considerations are added on the cases of the nuclear optical potential,
of quantum dissipation, and in particular of an approach to the measure-
ment problem in QM in terms of a chronon.
2. Time Operator in Non-Relativistic QM

and in Quantum Electrodynamics
2.1. On Time as an observable in non-relativistic QM
for systems with continuous energy spectra
The last part of the above-mentioned list [17–37] of papers, in particu-
lar Refs. [18–37], appeared in the 90s, devoted to the problem of Time in
non-relativistic QM, essentially because of the need to define the tunnel-
ing time. As already remarked, those papers did not refer to the Naimark
theoremc [39] which had mathematically supported, on the contrary, the
results in [3–15] and afterwards in [24–28, 38]. Indeed, already in the 70s
(in Refs. [3–9] while more detailed presentations and reviews can be found
c The Naimark theorem states in particular the following [39]: The non-orthogonal
spectral decomposition of a maximal Hermitian operator can be approximated by an

orthogonal spectral function (which corresponds to a self-adjoint operator), in a weak
convergence, with any desired accuracy.
in [10–13] and independently in [14, 15]), it was proved that, for systems
with continuous energy spectra, time is a quantum-mechanical observable,
canonically conjugate to energy. Namely, it had been shown that the time
operator

t, in the time (t-) representation, (a),
t̂ = ∂ (1)
−i , in the energy (E-) representation (b),
∂E
was not to be self-adjoint, but Hermitian, and to act on square-integrable

space–time wave packets in the representation (1a), and on their Fourier
transforms in (1b), once point E = 0 is eliminated (i.e. once one deals only
with moving packets, excluding any non-moving rear tails and the cases
with zero fluxes).d In Refs. [10–13] and [24–28, 38], the operator t̂ (in the
t-representation) had the property that any averages over time, in the one-
dimensional (1D) scalar case, were to be obtained by use of the following
measure (or weight):
j (x, t) dt
W (t, x) dt = +∞ , (2)
j (x, t) dt
−∞
where the flux density j (x, t) corresponds to the (temporal) probability for
a particle to pass through point x during the unit time centered at t, when
traveling in the positive x-direction. Such a measure is not postulated, but is
a direct consequence of the well-known probabilistic spatial interpretation of
ρ (x, t) and of the continuity relation ∂ρ (x, t)/∂ t + divj (x, t) = 0. Quantity
ρ(x, t) is, as usual, the probability of finding the considered moving particle
inside a unit space interval, centered at point x, at time t.
Quantities ρ(x, t) and j (x, t) are related to the wave-function
Ψ (x, t) by the ordinary definitions ρ (x, t) = |Ψ (x, t)|2 and j (x, t) =
[Ψ∗ (x, t) (/iµ) Ψ (x, t))]). When the flux density j (x, t) changes its sign,
quantity W (x, t) dt is no longer positive-definite and, as in Refs. [10, 24–
28], it acquires the physical meaning of a probability density only during
those partial time-intervals in which the flux density j (x, t) does keep its
d Such a condition is enough for operator (1a, b) to be a Hermitian, or more precisely a

maximal Hermitian [2–8] operator (see also [24–28, 38]; but it can be dispensed with by
recourse to bilinear forms (see, e.g. Refs. [8, 9, 38, 40] and references therein), as we shall
see below.
www.ebook3000.com
sign. Therefore, let us introduce the two measures [24–27, 38] by separating
the positive and the negative flux-direction values (that is, the flux signs)
j± (x, t) dt
W± (t, x) dt = +∞ (3)
j± (x, t) dt
−∞
with j± (x, t) = j (x, t) θ(±j).

Then, the mean value t± (x) of the time t at which the particle passes
through position x, when traveling in the positive or negative direction, is,
respectively,
+∞
t j± (x, t) dt
t± (x) = −∞

+∞
j± (x, t) dt
−∞
+∞
1 ∗
G (x, E) t̂ v G (x, E) + v G∗ (x, E) t̂ G (x, E) dE
0 2
= +∞ ,(4)
2

v G (x, E) dE
0
where G (x, E) is the Fourier transform of the moving 1D wave-packet

+∞
Ψ (x, t) = G (x, E) exp(−iEt/) dE

0

+∞
= g(E) ϕ(x, E) exp(−iEt/) dE,

0
when going on from the time to the energy representation. For free motion,
one has G(x, E) = g(E) exp(ikx), and ϕ(x, E) = exp(ikx), while E =
µ 2 k 2 / 2 = µ v 2 / 2. In Refs. [24–27, 38], the mean time durations were
defined for the particle 1D transmission from xi to xf > xi , and reflection
from the region (xi , +∞) back to the interval xf ≤ xi . Namely,
τT (xi , xf ) = t+ (xf ) − t+ (xi ) (5)
and
τR (xi , xf ) = t− (xf ) − t+ (xi ), (6)

respectively. The 3D generalization for the mean durations of quantum colli-

sions and nuclear reactions appeared in [10–13]. Finally, suitable definitions
of the averages tn on time of tn , with n = 1, 2, . . . , and of f (t), quantity
f (t) being any analytical function of time, can be found in [27,38,41], where
single-valued expressions have been explicitly written down.
The two canonically conjugate operators, the time operator (1) and the
energy operator

E, in the energy (E-) representation, (a)
Ê = ∂ (7)
i , in the time (t-) representation (b)
∂t
do clearly satisfy the commutation relation [8, 9, 27, 38, 41]
[Ê, t̂] = i. (8)
The Stone and von Neumann theorem [42] has been always interpreted
as establishing a commutation relation like (8) for the pair of the canonically
conjugate operators (1) and (7), in both representations, for self-adjoint
operators only. However, it can be generalized for (maximal) Hermitian
operators, once one introduces t̂ by means of the single-valued Fourier trans-
formation from the t-axis (−∞ < t < ∞) to the E-semi-axis (0 < E < ∞),
and utilizes the properties [43,44] of the “(maximal) Hermitian” operators:
This has been shown, e.g. in Ref. [4], as well as in Refs. [27, 38, 41].
Indeed, from Eq. (8), the uncertainty relation
∆E ∆t ≥ /2, (9)
√
(where the standard deviations are ∆a = Da, quantity Da being the
variance Da = a2 −a2 , and a = E, t, while . . . denotes the average over
t with the measures W (x, t) dt or W± (x, t) dt in the t-representation) can be
derived also for operators which are simply Hermitian, by a straightforward
generalization of the procedures which are common in the case of self-
adjoint (canonically conjugate) quantities, like coordinate x̂ and momentum
p̂x . Moreover, relation (8) satisfies [27, 38, 41] the Dirac “correspondence”
principle, since the classical Poisson brackets {q0 , p0 }, with q0 = t and
p0 = −E, are equal to 1. In Refs. [6–10], and [27, 38, 41], it was also shown
that the differences, between the mean times at which a wave-packet passes
through a pair of points, obey the Ehrenfest correspondence principle.
As a consequence, one can state that, for systems with continuous energy
spectra, the mathematical properties of (maximal) Hermitian operators,
like t̂ in Eq. (1), are sufficient for considering them as quantum observ-
ables. Namely, the uniqueness [43] of the spectral decomposition (although
www.ebook3000.com
not orthogonal) for operators t̂, and t̂n (n > 1), guarantees the “equiva-
lence” of the mean values of any analytical function of time when evaluated
in the t and in the E-representations. In other words, such an expansion is
equivalent to a completeness relation, for the (approximate) eigenfunctions
of t̂n (n > 1), which with any accuracy can be regarded as orthogonal, and
corresponds to the actual eigenvalues for the continuous spectrum. These
approximate eigenfunctions belong to the space of the square-integrable
functions of the energy E (cf., for instance, Refs. [8–13, 27, 38] and refer-
ences therein).
From this point of view, there is no practical difference between self-
adjoint and maximal Hermitian operators for systems with continuous
energy spectra. Let us repeat that the mathematical properties of t̂n (n > 1)
are enough for considering time as a quantum mechanical observable (like
energy, momentum, space coordinates, etc.) without having to introduce any
new physical postulates.
It is remarkable that von Neumann himself [45], before confining him-
self for simplicity to self-adjoint operators, stressed that operators like our
time t̂ may represent physical observables, even if they are not self-adjoint.
Namely, he explicitly considered the example of the operator − i ∂/∂x
associated with a particle living in the right semi-space bounded by a rigid
wall located at x = 0; that operator is not self-adjoint (acting on wave
packets defined on the positive x-axis only), nevertheless it obviously cor-
responds to the x-component of the observable momentum for that particle:
See Fig. 1.
Fig. 1. For a particle Q free to move in a semi-space, bounded by a rigid wall

located at x = 0, the operator −i∂/∂x has the clear physical meaning of the
particle momentum x-component even if it is not self-adjoint (cf. von Neumann
[45] and Refs. [8, 9]): See the text.
At this point, let us emphasize that our previously assumed boundary

condition E = 0 can be dispensed with by having recourse [3, 4, 8, 9] to the
bilinear Hermitian operator
↔
−i ∂
t̂ = (10)
2 ∂E
where the meaning of the symbol ↔ is clear from the accompanying defi-
nition

ih ∂ ih ∂
(f, t̂ g) = f, − g + − f, g .
2 ∂E 2 ∂E
By adopting this expression for the time operator, the algebraic sum of the
two terms in the RHS of the last relation results to be automatically zero at
point E = 0. This question will be exploited below in Sec. 3 (when dealing
with the more general case of the four-position operator). Incidentally, such
an “elimination” [3,4,8,9] of point E = 0 is not only simpler, but also more
physical, than other kinds of elimination obtained much later in papers
like [33, 34].
In connection with the last quotation, let us briefly comment on the
so-called positive-operator-value-measure (POVM) approach, often used or
discussed in the second set of papers on time in quantum physics mentioned
in our introduction. Actually, an analogous procedure had been proposed,
since the 60s [46], in some approaches to the quantum theory of measure-
ments. Afterwards, and much later, the POVM approach has been applied,
in a simplified and shortened form, to the time-operator problem in the
case of 1D free motion: for instance, in Refs. [16, 18, 21, 29–37] and espe-
cially in [33, 34]. These papers stated that a generalized decomposition of
unity (or “POV measure”) could be obtained from self-adjoint extensions
of the time operator inside an extended Hilbert space (for instance, adding
the negative values of the energy, too), by exploiting the Naimark dilation-
theorem [47]: But such a program has been realized till now only in the
simple cases of 1D particle free motion.
By contrast, our approach is based on a different Naimark’s theorem
[39], which, as already mentioned above, allows a much more direct, simple
and general — and at the same time non-less rigorous — introduction of a
quantum operator for Time. More precisely, our approach is based on the
so-called Carleman theorem [48], utilized in Ref. [39], about approximat-
ing a Hermitian operator by suitable successions of “bounded” self-adjoint
operators: That is, of self-adjoint operators whose spectral functions do
weakly converge to the non-orthogonal spectral function of the considered
www.ebook3000.com
hemitian operator. And our approach is applicable to a large family of 3D

particle collisions, with all possible Hamiltonians. Actually, our approach
was proposed in the early Refs. [3–10] and in Ref. [24], and applied therein
for the time analysis of quantum collisions, nuclear reactions and tunneling
processes.
2.2. On the momentum representation

of the time operator
In the continuous spectrum case, instead of the E-representation, with
0 < E < +∞, in Eqs. (1)–(4) one can also use the k-representation [14,15],
with the advantage that −∞ < k < +∞:

+∞
Ψ (x, t) = g(k) ϕ(x, k) exp(−iEt/) dk (11)

−∞
2 2
with E = k / 2µ, and k = 0.
For the extension of the momentum representation to the case of tn ,
with n > 1, we confine ourselves here to refer the reader to the papers
[27, 38, 41].
2.3. An alternative weight for time averages

(in the cases of particle dwelling inside
a certain spatial region)
We recall that the weight (2) [as well as its modifications (3)] has the
meaning of a probability for the considered particle to pass through point
x during the time interval (t, t + dt). Let us follow the procedure presented
in Refs. [24–28] and references therein, and analyze the consequences of the
equality

+∞
+∞

j (x, t) dt = Ψ(x, t)2 dx (12)
−∞ −∞
obtained from the 1D continuity equation. One can easily realize that a
second, alternative weight can be adopted:

Ψ(x, t)2 dx
d P (x, t) ≡ Z (x, t) dx = +∞ , (13)

Ψ(x, t)2 dx
−∞
which possesses the meaning of probability for the particle to be located

(or to sojourn, i.e. to dwell) inside the infinitesimal space region (x, x + dx)
at the instant t, independently of its motion properties. Then, the quantity
x2
Ψ(x, t)2 dx
x1
P (x1 , x2 , t) = +∞ (14)

Ψ(x, t)2 dx
−∞
will have the meaning of probability for the particle to dwell inside the
spatial interval (x1 , x2 ) at the instant t.
As it is known (see, for instance, Refs. [24–27,38] and refs. therein), the
mean dwell time can be written in the two equivalent forms:
+∞ xf
dt |Ψ(x, t)|2 dx
−∞ xi
τ (xi , xf ) = +∞ (15)
jin (xi , t) dt
−∞
and
+∞ +∞
t j(xf , t) dt − t j(xi , t) dt
−∞ −∞
τ (xi , xf ) = +∞ , (16)
jin (xi , t) dt
−∞
where it has been taken account, in particular, of relation (12), which fol-
lows — as already said — from the continuity equation.
Thus, in correspondence with the two measures (2) and (13), when inte-
grating over time one gets two different kinds of time distributions (mean
values, variances, etc.), which refer to the particle traversal time in the case
of measure (2), and to the particle dwelling in the case of measure (13).
Some examples for 1D tunneling are contained in Refs. [24–27].
2.4. Time as a quantum-theoretical observable

in the case of photons
As is known (see, for instance, Refs. [26, 49, 50]), in first quantization the
single-photon wave-function can be probabilistically described in the 1D
www.ebook3000.com
case by the wave-packete

d3 k
A(r, t) = χ(k) ϕ(k, r) exp(−ik0 t), (17)
k0
k0
where as usual, A(r, t) is the electromagnetic vector potential, while

r = {x, y, z}, k = {kx , ky , kz }, k0 ≡ w/c = ε/ c, and k ≡ |k| = k0 .
Axis x has been chosen as the propagation direction. Let us note that

χ(k) = i=y,z χi (k) ei (k), with ei ej = δij , and xi , xj = y, z, while χi (k)

is the probability amplitude for the photon to have momentum k and polar-
ization ej along xj . Moreover, it is ϕ(k, r) = exp(ikx x) in the case of plane
waves, while ϕ(k, r) is a linear combination of evanescent (decreasing) and
anti-evanescent (increasing) waves in the case of “photon barriers” (i.e.
band-gap filters, or even undersized segments of waveguides for microwaves,
or frustrated total-internal-reflection regions for light, and so on). Although
it is not easy to localize a photon in the direction of its polarization [49, 50],
nevertheless for 1D propagations it is possible to use the space–time prob-
abilistic interpretation of Eq. (17), and define the quantity

S0 dx
ρem (x, t) dx = , S0 = s0 dy dz (18)
S0 dx
(s0 = [E ∗ · E + H ∗ · H]/ 4π being the energy density, with the electro-

magnetic field H = rot A, and E = −1/c ∂A/∂t), which represents the
probability density of a photon to be found (localized) in the spatial interval
(x, x + dx) along the x-axis at the instant t; and the quantity

Sx dt
jem (x, t) dt = , Sx (x, t) = sx dy dz (19)
Sx (x, t) dt
(sx = c [E ∗ ∧ H]x / 8π being the energy flux density), which represents
the flux probability density of a photon to pass through point x in the time
interval (t, t + dt): in full analogy with the probabilistic quantities for non-
relativistic particles. The justification and convenience of such definitions is
self-evident, when the wave-packet group velocity coincides with the veloc-
ity of the energy transport; in particular: (i) the wave-packet (17) is quite
similar to wave-packets for non-relativistic particles, and (ii) in analogy with
conventional non-relativistic QM, one can define the “mean time instant”
e The gauge condition divA = 0 is assumed.

for a photon (i.e. an electromagnetic wave-packet) to pass through point x,

as follows
+∞

+∞ t Sx (x, t) dt
t(x) = t Jem, x dt = −∞
+∞ .
−∞ Sx (x, t) dt
−∞
As a consequence [in the same way as in the case of Eqs. (1) and (2)], the
form (1) for the time operator in the energy representation is valid also
for photons, with the same boundary conditions adopted in the case of
particles, that is, with χi (0) = χi (∞) and with E = c k0 .
The energy density s0 and energy flux density sx satisfy the relevant
continuity equation
∂s0 ∂sx
+ =0 (20)
∂t ∂x
which is Lorentz-invariant for 1D spatial propagation [26, 27, 38] processes.
2.5. Introducing the analogue of the “Hamiltonian”

for the case of the Time operator:
a new Hamiltonian approach
In non-relativistic quantum theory, the Energy operator acquires (cf., e.g.
∂
Refs. [11–13, 27, 38]) the two forms: (i) i ∂t in the t-representation and
(ii) Ĥ (p̂x , x̂, . . .) in the hamiltonianian formalism. The “duality” of these
two forms can be easily inferred from the Schröedinger equation itself,
ĤΨ = i ∂Ψ ∂t . One can introduce in QM a similar duality for the case of
Time: Besides the general form (1) for the Time operator in the energy
representation, which is valid for any physical systems in the region of con-
tinuous energy spectra, one can express the time operator also in a “Hamil-
tonian form”, i.e. in terms of the coordinate and momentum operators, by
having recourse to their commutation relations. Thus, by the replacements
Ê → Ĥ (p̂x , x̂, . . .),

(21)
t̂ → T̂ (p̂x , x̂, . . .),
and on using the commutation relation [similar to Eq. (3)]
[Ĥ, T̂ ] = i, (22)
www.ebook3000.com
one can obtain [51], given a specific ordinary Hamiltonian, the correspond-
ing explicit expression for T̂ (p̂x , x̂, . . .).
Indeed, this procedure can be adopted for any physical system with a
known Hamiltonian Ĥ (p̂x , x̂, . . .), and we are going to see a concrete exam-
ple. By going on from the coordinate to the momentum representation, one
realizes that the formal expressions of both the Hamiltonian-type operators
Ĥ (p̂x , x̂, . . .) and T̂ (p̂x , x̂, . . .) do not change, except for an obvious change
of sign in the case of operator T̂ (p̂x , x̂, . . .).
As an explicit example, let us address the simple case of a free particle
whose Hamiltonian is

 2 ∂
p̂x / 2µ, p̂x = −i , in the coordinate representation, (a)
Ĥ = ∂x (23)
 p2 / 2µ, in the momentum representation, (b)
x
Correspondingly, the Hamilton-type time operator, in its symmetrized form,

will write
µ

 p̂−1 −1 −2
x x + xp̂x + i ; p̂x , in the coordinate representation, (a)
2
T̂ =

 − µ p−1 x̂ + x̂p−1 2
x + i/px , in the momentum representation, (b)
2 x
(24)
where

−1 i ∂
p̂x = dx . . . , x̂ = i .
∂px
∂
Incidentally, operator (24b) is equivalent to −i ∂E , since E = p2x / 2µ; and
therefore it is also a (maximal) Hermitian operator. Indeed, by applying the
operator T̂ (p̂x , x̂, . . .), for instance, to a plane-wave of the type exp(ikx),
we obtain the same result in both the coordinate and the momentum rep-
resentations:
x
T̂ exp(ikx) = exp(ikx), (25)
v
quantity x/v being the free-motion time (for a particle with velocity v ) for
traveling the distance x.
On the basis of what precedes, it is possible to show that the wave-
function Ψ(x, t) of a quantum system satisfies the two (dual) equations
∂Ψ
Ĥ Ψ = i and T̂ Ψ = t Ψ. (26)
∂t
In the energy representation, and in the stationary case, we obtain again

two (dual) equations
∂ϕt
Ĥ ϕt = ε ϕt and T̂ ϕt = −i , (27)
∂ε
quantity ϕt being the Fourier transform of Ψ:

+∞
1
ϕt = Ψ(x, t) eiεt/ dt. (28)
2π
−∞
It might be interesting to apply the two pairs of the last dual equa-
tions also for investigating tunneling processes through the quantum grav-
itational barrier, which appears during inflation, or at the beginning of
the big-bang expansion, whenever a quasi-linear Schrödinger-type equation
does approximately show up.
2.6. Time as an observable (and the time-energy

uncertainty relation), for quantum-mechanical
systems with discrete energy spectra
For describing the time evolution of non-relativistic quantum systems
endowed with a purely discrete (or a continuous and discrete) spectrum,
let us now introduce wave-packets of the form [11–13, 27, 38, 41]:

ψ (x, t) = gn ϕn (x) exp[−i(εn − ε0 )t/], (29)
n=0
where ϕn (x) are orthogonal and normalized bound states which satisfy
the equation Ĥ ϕn (x) = εn ϕn (x), quantity Ĥ being the Hamiltonian of

the system; while the coefficients gn are normalized: n=0 |gn |2 = 1. We

omitted the non-significant phase factor exp(−iε0 t/) of the fundamental
state.
Let us first consider the systems whose energy levels are separated by
intervals admitting a maximum common divisor D (for e.g. harmonic oscil-
lator, particle in a rigid box, and spherical spinning top), so that the wave
packet (29) is a periodic function of time possessing as period the Poincaré
cycle time T = 2π/D. For such systems, it is possible [11–13, 27, 41] to
www.ebook3000.com
1.0 1.0
ρ (t) ρ (t)
0.5 0.5
0.0 0.0
0.0 5.0 × 10–8 1.0 × 10–7 1.5 × 10–7 2.0 × 10–7 2.5 × 10–7 0 1 × 10–12 2 × 10–12 3 × 10–12 4 × 10–12
t (s) t (s)
(a) (b)
Fig. 2. Damping of the non-diagonal terms of the density operator for two dif-
ferent values of τ . For both cases we used ∆E = 4 eV. (a) Slower damping for
τ = 6.26 × 10−24 s; (b) faster damping for τ = ×10−19 s. This figure is taken
from Ref. [73].
construct a self-adjoint time operator with the form (in the time represen-
tation) of a saw-function of t, choosing t = 0 as the initial time instant:
∞
∞

t̂ = t − T Θ(t − [2n + 1]T /2) + T Θ(−t − [2n + 1]T /2. (30)
n=0 n=0
This periodic function for the time operator is a linear (increasing) function
of time t within each Poincar cycle: Cf., e.g. Fig. 2 in Ref. [38], where the
periodic saw-tooth function for the time operator, in the present case of
quantum mechanical systems with discrete energy spectra [i.e. of Eq. (30)],
is explicitly shown.
The commutation relations of the Energy and Time operators, now both
self-adjoint, acquires in the case of discrete energies and of a periodic Time
operator the form
∞

[Ê, t̂] = i 1 − T δ(t − [2n + 1]T ) , (31)
n=0
wherefrom the uncertainty relation follows in the new form

 
2
T |ψ(T /2 + γ)|
(∆E)2 (∆t)2 = 2 1 − T /2 , (32)
2
−T /2 |ψ(t)| dt
where a parameter γ has been introduced with −T /2 < γ < T /2, in order
to assure that the RHS integral is single-valued [27, 41].
When ∆E → 0 (that is, when |gn | → δnn ), the RHS of Eq. (32) tends
to zero too, since |ψ(t)|2 tends to a constant value. In such a case, the distri-
bution of the time instants at which the wave-packet passes through point x
becomes flat within T /2each Poincaré cycle. When, by contrast, ∆E D and
|ψ(T + γ)|2 ( −T /2 |ψ(t)|2 dt)/T , the periodicity condition may become
inessential whenever ∆t t. In other words, our uncertainty relation (32)
transforms into the ordinary uncertainty relation for systems with contin-
uous spectra.
In more general cases, for excited states of nuclei, atoms and molecules,
the energy-level intervals, for discrete and quasi-discrete (resonance) spec-
tra, are not multiples of a maximum common divisor, and hence the
Poincaré cycle is not well defined for such systems. Nevertheless, even for
those systems one can introduce an approximate description (sometimes,
with any desired degree of accuracy) in terms of Poincaré quasi-cycles and
a quasi-periodical evolution; so that for sufficiently long time intervals the
behavior of the wave-packets can be associated with a a periodical motion
(oscillation), sometimes — e.g. for very narrow resonances — with any
desired accuracy. For them, when choosing an approximate Poincaré-cycle
time, one can include in one cycle as many quasi-cycles as it is necessary
for the demanded accuracy. Then, with the chosen accuracy, a quasi-self-
adjoint time operator can be introduced.
3. On Four-Position Operators in Quantum Field

Theory, in Terms of Bilinear Operators
In this section, we approach the relativistic case, taking into considera-
tion — therefore — the space–time (4D) “position” operator, starting how-
ever with an analysis of the 3D (spatial) position operator in the simple
relativistic case of the Klein–Gordon equation. Actually, this analysis will
lead us to tackle already with non-Hermitian operators. Moreover, while
performing it, we shall meet the opportunity of introducing bilinear opera-
tors, which will be used even more in the next case of the full four-position
operator.
Let us recall that in Sec. 2.1 we mentioned that the boundary condi-
tion E = 0, therein imposed to guarantee (maximal) hermiticity of the
time operator, can be dispensed with just by having recourse to bilinear
forms. Namely, by considering the bilinear Hermitian operator [8, 9, 40]
↔
t̂ = (−i ∂ /∂E)/2, where the symbol ↔ is defined through the accompa-
nying equality (f, t̂ g) = (f, − ih ∂ ih ∂
2 ∂E g) + (− 2 ∂E f, g).
www.ebook3000.com
3.1. The Klein–Gordon case: Three-position operators

The standard position operators, being Hermitian and moreover self-
adjoint, are known to possess real eigenvalues: i.e. they yield a point-like
localization. J.M. Jauch showed, however, that a point-like localization
would be in contrast with “unimodularity”. In the relativistic case, more-
over, phenomena such as the pair production forbids a localization with pre-
cision better than one Compton wavelength. The eigenvalues of a realistic
position operator ẑ are therefore expected to represent space regions, rather
than points. This can be obtained only by having recourse to non-Hermitian
(and therefore non-self-adjoint) position operators ẑ (a priori, one can have
recourse either to non-normal operators with commuting components, or
to normal operators with non-commuting components). Following, e.g. the
ideas in Refs. [52–56], we are going to show that the mean values of the Her-
mitian (self-adjoint) part of ẑ will yield a mean (point-like) position [57,58],
while the mean values of the anti-Hermitian (anti-self-adjoint) part of ẑ will
yield the sizes of the localization region [3, 4].
Let us consider, e.g. the case of relativistic spin-zero particles, in natural
units and with metric (+ − − −). The position operator, i ∇p , is known
to be actually non-Hermitian, and may be in itself a good candidate for an
extended-type position operator. To show this, we want to split [52–56] it
into its Hermitian and anti-Hermitian (or skew-Hermitian) parts.
Consider, then, a vector space V of complex differentiable functions on
a 3D phase space [40] equipped with an inner product defined by

d3 p ∗
(Ψ, Φ) = Ψ (p) Φ(p), (33)
p0

quantity p0 being p2 + m20 . Let the functions in V satisfy moreover the
condition

dS ∗
lim Ψ (p) Φ(p) = 0, (34)
R→∞ p0
SR
where the integral is taken over the surface of a sphere of radius R. If

U : V → V is a differential operator of degree one, condition (34) allows a
definition of the transpose U T by
(U T Ψ, Φ) = (Ψ, U Φ) for all Ψ, Φ ∈ V, (35)
where U is changed into U T , or vice versa, by means of integration by parts.

This allows, further, to introduce a dual representation [40] (U1 , U2 ) of

a single operator U1T + U2 by
(U1 Ψ, Φ) + (Ψ, U2 Φ) = (Ψ, (U1T + U2 ) Φ). (36)
With such a dual representation, it is easy to split any operator into its
Hermitian and anti-Hermitian parts
1 1
(Ψ, U Φ) = (Ψ, U Φ) + (U ∗ Ψ, Φ) + (Ψ, U Φ) − (U ∗ Ψ, Φ) . (37)
2 2
Here, the pair
1 ∗ ↔
(U , U ) ≡ U h , (38)
2
corresponding to (1/2) (U + U ∗T ) represents the Hermitian part, while
1 ↔
(−U ∗ , U ) ≡ U a (39)
2
represents the anti-Hermitian part.
Let us apply what precedes to the case of the Klein–Gordon position-
operator ẑ = i ∇p . When
∂
U =i , (40)
∂pj
we have [3, 4]
↔
1 ∗ 1 ∂ ∂ i ∂
(U , U ) = −i , i ≡ , (a)
2 2 ∂pj ∂pj 2 ∂pj
↔
(41)
1 1 ∂ ∂ i ∂+
(−U ∗ , U ) = i , i ≡ . (b)
2 2 ∂pj ∂pj 2 ∂pj
And the corresponding single operators turn out to be
1 ∂ i pj
(U + U ∗T ) = i − , (a)
2 ∂pj 2 p2 + m20
(42)
1 i pj
(U − U ∗T ) = . (b)
2 2 p + m20
2
It is noteworthy [3,4] that, as we are going to see, operator (42a) is nothing

but the usual Newton–Wigner operator, while (42b) can be interpreted
[3, 4, 31, 52–56] as yielding the sizes of the localization-region (an ellipsoid)
via its average values over the considered wave-packet.
www.ebook3000.com
Let us underline that the previous formalism justifies from the mathe-
matical point of view the treatment presented in papers like [52–58]. We can
split [3, 4] the operator ẑ into two bilinear parts, as follows:
i ↔ i ↔(+)
ẑ = i ∇p = ∇p + ∇p , (43)
2 2
↔ ↔(+)
where Ψ∗ ∇p Φ ≡ Ψ∗ ∇p Φ − Φ∇p Ψ∗ and Ψ∗ ∇p Φ ≡ Ψ∗ ∇p Φ + Φ∇p Ψ∗ ,
and where we always referred to a suitable [8, 9, 40, 52–58] space of wave
packets. Its Hermitian part [52–58]
i ↔
x̂ = ∇p , (44)
2
which was expected to yield an (ordinary) point-like localization, has been
derived also by writing explicitly
3
d p ∗
(Ψ, x̂ Φ) = i Ψ (p) ∇p Φ(p) (45)
p0
and imposing hermiticity, i.e. imposing the reality of the diagonal elements.
The calculations yield
3
d p ∗ ↔
Φ, x̂ Φ = i Φ (p) ∇p Φ(p), (46)
p0
suggesting to adopt just the Lorentz-invariant quantity (44) as a bilinear

Hermitian position operator. Then, on integrating by parts (and due to the
vanishing of the surface integral), we verify that Eq. (44) is equivalent to
the ordinary Newton–Wigner operator :
i ↔ i p
x̂h ≡ ∇p ≡ i ∇p − ≡ N − W. (47)
2 2 p + m2
2
We are left with the (bilinear) anti-Hermitian part
i ↔(+)
ŷ = ∇ , (48)
2 p
whose average values over the considered state (wave-packet) can be
regarded as yielding [8, 9, 40, 52–58] the sizes of an ellipsoidal localization-
region.
After the digression associated with Eqs. (43)–(48), let us go back to
the present formalism, as expressed by Eqs. (33)–(42).
In general, the extended-type position operator ẑ will yield
Ψ| ẑ |Ψ = (α + ∆α) + i (β + ∆β), (49)
where ∆α and ∆β are the mean-errors encountered when measuring the

point-like position and the sizes of the localization region, respectively. It is
interesting to evaluate the commutators (i, j = 1, 2, 3):
↔ ↔
i ∂ i ∂ (+) i 2 pi pj
, = δij − , (50)
2 ∂pi 2 ∂pj 2 p20 p20
from which the noticeable “uncertainty correlations” follow:

1 1 2 pi pj
.
∆αi ∆βj ≥ 2 δ ij − 2 (51)
4 p0 p0
3.2. Four-position operators

It is tempting to propose as four-position operator the quantity ẑ µ = x̂µ +
i ŷ µ , whose Hermitian (Lorentz-covariant) part can be written
↔
µ i ∂
x̂ = − , (52)
2 ∂pµ
to be associated with its corresponding “operator” in four-momentum space

↔
i ∂
p̂µh =+ . (53)
2 ∂xµ
Let us recall the proportionality between the four-momentum opera-

tor and the four-current density operator in the chronotopical space, and
then underline the canonical correspondence (in the four-position and four-
momentum spaces, respectively) between the “operators” (cf. the previous
subsection):
↔
i ∂
m0 ρ̂ ≡ p̂0 = , (a)
2 ∂t
(54)
↔
i ∂
m0 ĵ ≡ p̂ = − , (b)
2 ∂r
www.ebook3000.com
and the operators

↔
i ∂
t̂ ≡ − , (a)
2 ∂p0
↔
(55)
i ∂
x̂ ≡ , (b)
2 ∂p
where the four-position “operator” (55) can be considered as a four-current

density operator in the energy-impulse space. Analogous considerations can
be carried on for the anti-Hermitian parts (see Ref. [4]).
Finally, by recalling the properties of the time operator as a maximal
Hermitian operator in the non-relativistic case (Sec. 2.1), one can see that
the relativistic time operator (55a) (for the Klein–Gordon case) is also a
self-adjoint bilinear operator for the case of continuous energy spectra and a
(maximal) Hermitian linear operator for free particles [due to the presence
of the lower limit zero for the kinetic energy, or m0 for the total energy].
4. Some Considerations on Non-Hermitian

Hamiltonians
As to the important issue of Unstable States, and of their association with
quasi-Hermitian Hamiltonians, let us confine ourselves to refer the inter-
ested reader to Sec. 4 in Ref. [38], a section based on previous work per-
formed in collaboration with A. Agodi, M. Baldo, and A. Pennisi di Floris-
tella [59].
Here, we shall only mention the case of the nuclear optical model and of
microscopic quantum dissipation, and deal with an approach to the mea-
surement problem in QM in terms of the chronon.
Actually, we shall deal with the chronon formalism [73] — where the
chronon τ0 is a “quantum” of time, in the sense specified below — not
only for its obvious connection with our view of time, and of space–time,
but also because that discrete formalism has a non-Hermitian character, as
stressed e.g. in the Appendices of Ref. [73]. For instance, in its Schrödinger
representation (see the following), proper continuous equations can repro-
duce the outputs obtained with the discretized equations once we replace
the (discrete) conventional Hamiltonian with a suitable (continuous) non-
Hermitian Hamiltonian, that can be called the “equivalent Hamiltonian”.
One important point is that non-Hermitian Hamiltonians imply non-unitary
evolution operators.
4.1. Nuclear optical model

Since the fifties, the so-called optical model has been frequently used for
describing the experimental data on nucleon-nucleus elastic scattering, and,
not less, on more general nuclear collisions: see, e.g. Refs. [60–63]; while for
a generalized optical model — namely, the coupled-channel method with
an optical model in any channel of the nucleon-nucleus (elastic or inelastic)
scattering — one can see Ref. [64] and references therein.
In these cases, the Hamiltonian contains a complex potential, its
imaginary part describing the absorption processes that take place by
compound-nucleus formation and subsequent decay. As to the Hamiltonian
with complex potential, we confine ourselves at referring to work of ours
already published, where it was studied the non-unitarity and analytical
structure of the S-matrix, the completeness of the wave-functions, and so
on: see Ref. [65], and also [66, 67].
4.2. Microscopic quantum dissipation

Before going forward, let us inform the interested reader that in the
appendix to the already quoted Ref. [38] some discussions and details can be
found related to the time-dependent Schrödinger equation with dissipative
terms.
Various different approached are known, aimed at getting dissipation —
and possibly decoherence — within QM. First of all, the simple introduction
of a “chronon” (see, e.g. Refs. [68–73]) allows one to go on from differential
to finite-difference equations, and in particular to write down the quan-
tum theoretical equations (Schrödinger’s, Liouville–von Neumann’s (LvN),
etc.) in three different ways: Symmetrical, retarded, and advanced. The
retarded “Schrödinger” equation describes in a rather simple and natural
way a dissipative system, which exchanges (loses) energy with the envi-
ronment. The corresponding non-unitary time-evolution operator obeys a
semi-group law and refers to irreversible processes. The retarded approach
furnishes, moreover, an interesting way for proceeding in the direction of
solving the “measurement problem” in QM, without any need for a wave-
function collapse: See Refs. [73–77]. The chronon theory can be regarded
as a peculiar “coarse grained” description of the time evolution.
Let us stress that it has been shown that the mentioned discrete
approach can be replaced with a continuous one, at the price of introducing
a non-Hermitian Hamiltonian: See, e.g. Ref. [78].
www.ebook3000.com
Further relevant work can be found, for instance, in papers like [83–93]
and references therein.
Let us add that much work is still needed, however, for the description
of time irreversibility at the microscopic level. Indeed, various approaches
have been proposed, in which new parameters are introduced (regulation
or dissipation) into the microscopic dynamics (building a bridge, in a sense,
between microscopic structure and macroscopic characteristics). Besides the
Caldirola–Kanai [90, 91] Hamiltonian
2 ∂ 2 −γt
ĤCK (t) = − e + V (x) eγt , (56)
2m ∂x2
(which has been used, e.g. in Ref. [92]), other rather simple approaches,
based of course on the Schrödinger equation
∂
i Ψ(x, t) = Ĥ Ψ(x, t), (57)
∂t
and adopting a microscopic dissipation defined via a coefficient of extinction

γ, are known. In Sec. 5 of Ref. [38], we gave some details on: (A) Nonlin-
ear (non-Hermitian) Hamiltonians, with “potential” operators of Kostin’s,
Albrecht’s, and Hasse’s types; and (B) Linear (non-Hermitian) Hamiltoni-
ans, of Gisin’s, and Exner’s types.
One may here recall also the important, so-called “microscopic models”
[93], even if they are not based on the Schröedinger equation.
All such proposals are to be further investigated, and completed, since
till now they do not appear to have been exploited enough. Let us remark,
just as an example, that it would be desirable to take into deeper consider-
ation other related phenomena, like the ones associated with the “Hartman
effect” (and “generalized Hartman effect”) [24–26, 94–97], in the case of
tunneling with dissipation: a topic faced in few papers, like [98, 99].
As already mentioned, in the appendix to Ref. [38], we presented, for
example, a scheme of iterations (successive approximations) as a possible
tool for explicit calculations of wave-functions in the presence of dissipation.
At last, let us incidentally recall that two generalized Schröedinger equa-
tions, introduced by Caldirola [80, 100–102] in order to describe two differ-
ent dissipative processes (behavior of open systems, and the radiation of a
charged particle) have been shown — see, e.g. Ref. [103]) — to possess the
same algebraic structure of the Lie-admissible type [104].
4.3. Approaching the “Measurement Problem”

in QM in terms of a chronon τ0
In the previous subsection, we addressed Caldirola’s theory “of the
Chronon”.
With that theory as inspiration, we now wish to present a simple quan-
tum (finite difference) equation for dissipation and decoherence on the basis
of work performed in collaboration with R.H.A. Farias [73, 74].
Namely, as said above, the mere introduction (not of a “time-lattice”,
but simply) of a ‘chronon’ τ0 allows one to go on from differential to finite-
difference equations; and in particular to write down the Schrödinger equa-
tion (as well as the LvN equation) in three different ways: “retarded”,
“symmetrical”, and “advanced”. One of such three formulations — the
retarded one — describes in an elementary way a system which is exchang-
ing (and losing) energy with the environment. In its density-matrix version,
indeed, it can be easily shown that all non-diagonal terms go to zero very
rapidly.
We already mentioned that we are interested in the chronon formal-
ism [73] not only for its obvious connection with our view of time, and of
space–time, but also because the discrete formalism has a non-Hermitian
character (as clarified e.g. in the appendices of Ref. [73]). For instance, in its
Schrödinger representation (see the following), proper continuous equation
can reproduce the outputs obtained with the discretized equations once we
replace the (discrete) conventional Hamiltonian by a suitable (continuous)
non-Hermitian Hamiltonian that can be called the “equivalent Hamilto-
nian”. Indeed, in some special cases, the finite-difference equations can be
solved by one of the (not easy) existing methods. An interesting alternative
method is, however, finding out a new Hamiltonian H̃ such that the new
continuous Schrödinger equation
∂Ψ (x, t)
i = H̃Ψ (x, t)
∂t
reproduces, at the points t = nτ0 (see below), the same results obtained
from the discretized equations. As it was shown by Casagrande and Mon-
taldi, it is always possible to find a continuous generating function which
makes it possible to obtain a differential equation equivalent to the original
finite-difference one, such that at every point of interest their solutions are
identical [this procedure is useful since it is generally very difficult to work
with the finite-difference equations on a qualitative basis; except for some
very special cases, they can be only numerically solved]. This equivalent
www.ebook3000.com
Hamiltonian H̃ is, however, non-Hermitian and it is often quite difficult to

be obtained: Happily enough, for the special case where the Hamiltonian
is time independent, the equivalent Hamiltonian is quite easy to calculate.
For example, in the symmetric equation case, it would be given by
τ
H̃ = sin−1 Ĥ .
τ
Of course, H̃ → Ĥ when τ0 → 0.
Since the introduction of the chronon has various consequences for Clas-
sical and Quantum Physics (also, as we have argued, for the decoherence
problem), let us open a new section about all that.
5. The Particular Case of the “Chronon” —

Its Consequences for Classical and Quantum
Physics (and for Decoherence)
As we were saying, let us devote a brief section to the consequence of the
introduction of a Chronon for Classical Physics and for QM (and for a
new approach to Decoherence); without forgetting what has been stated
in the previous two subsebsections. There are various consequences and
applications of the “Chronon”, an example being Ref. [105], where it was
suggested that the chronon approach can account also for the origin of the
internal DOF of the particles. In the last subsection of this section, we shall
also mention the possible role of the chronon in Cosmology.
Let us recall first of all that the interesting Caldirola’s “finite difference”
theory forwards — at the classical level — a solution for the motion of a
particle endowed with a non-negligible charge in an external electromag-
netic field, overcoming all the known difficulties met by Abraham–Lorentz’s
and Dirac’s approaches (and even allowing a clear answer to the question
whether a free falling charged particle does or does not emit radiation),
and — at the quantum level — yields a remarkable mass spectrum for
leptons.
In Ref. [73] (where also extensive references can be found), after having
reviewed Caldirola’s approach, we worked out, discussed, and compared
with one another the new representations of QM resulting from it, in the
Schrödinger, Heisenberg and density-operator LvN pictures, respectively.
For each representation, three (retarded, symmetric and advanced) for-
mulations are possible, which refer either to times t and t − τ0 , or to times
t−τ0 /2 and t+τ0 /2, or to times t and t+τ0 , respectively. It is interesting to
note that when the chronon tends to zero, the ordinary QM is obtained as
the limiting case of the “symmetric” formulation only; while the “retarded”
one does naturally appear to describe QM with friction, i.e. to describe dis-
sipative quantum systems (like a particle moving in an absorbing medium).
In this sense, discretized QM is much richer than the ordinary one.
In the mentioned work [73], we also obtained the (retarded) finite-
difference Schrödinger equation within the Feynman path integral
approach, and studied some of its relevant solutions. We have then derived
the time-evolution operators of this discrete theory, and used them to get
the finite-difference Heisenberg equations. [Afterward, we studied some typ-
ical applications and examples: as the free particle, the harmonic oscillator
and the hydrogen atom; and various cases have been pointed out, for which
the predictions of discrete QM differ from those expected from “continu-
ous” QM].
We want to pay attention here to the fact that, when applying the
density matrix formalism towards the solution of the measurement problem
in QM, some interesting results are met, as, for instance, a possible, natural
explication of the “decoherence” [74] due to dissipation: this reveals some
of the power of dicretized (in particular, retarded ) QM.
5.1. Outline of the classical approach

If ρ is the charge density of a particle on which an external electromagnetic
field acts, the Lorentz’s force law

1
f =ρ E+ v∧B
c
is valid only when the particle charge q is negligible with respect to the
external field sources. Otherwise, the classical problem of the motion of a
(non-negligible) charge in an electromagnetic field is still an open question.
For instance, after the known attempts by Abraham and Lorentz, in 1938
Dirac [106] obtained and proposed his known classical equation
duµ
m = Fµ + Γµ , (58)
ds
where Γµ is the Abraham four-vector

2 e2 d2 uµ uµ uν d2 uν
Γµ = + 2 , (59)
3 c ds2 c ds2
www.ebook3000.com
that is, the (Abraham) reaction force acting on the electron itself; and Fµ
is the four-vector that represents the external field acting on the particle
e
Fµ = Fµν uν . (60)
c
At the non-relativistic limit, Dirac’s equation formally goes into the one
previously obtained by Abraham–Lorentz:

dv 2 e2 d2 v 1
m0 − =e E+ v∧B . (61)
dt 3 c3 dt2 c
2 2
The last equation shows that the reaction force equals 23 ec3 ddtv2 .
Dirac’s dynamical Eq. (58) is known to present, however, many troubles,
related to the infinite many non-physical solutions that it possesses. Actu-
ally, it is a third-order differential equation, requiring three initial conditions
for singling out one of its solutions. In the description of a free electron, e.g.
it yields “self-accelerating” solutions (runaway solutions), for which velocity
and acceleration increase spontaneously and indefinitely. Moreover, for an
electron submitted to an electromagnetic pulse, further non-physical solu-
tions appear, related this time to pre-accelerations: If the electron comes
from infinity with a uniform velocity v0 and at a certain instant of time t0 is
submitted to an electromagnetic pulse, then it starts accelerating before t0 .
Drawbacks like these motivated further attempts to find out a coherent
(not point-like) model for the classical electron.
Considering elementary particles as points is probably the sin plaguing
modern physics (a plague that, unsolved in classical physics, was transferred
to quantum physics). One of the simplest ways for associating a discreteness
with elementary particles (let us consider, e.g. the electron) is just via the
introduction (not of a “time-lattice”, but merely) of a “quantum” of time,
the chronon, following Caldirola. [107] Like Dirac’s, Caldirola’s theory is
also Lorentz invariant (continuity, in fact, is not an assumption required
by Lorentz invariance). This theory postulates the existence of a universal
interval τ0 of proper time, even if time flows continuously as in the ordinary
theory. When an external force acts on the electron, however, the reaction
of the particle to the applied force is not continuous: The value of the
electron velocity uµ is supposed to jump from uµ (τ − τ0 ) to uµ (τ ) only
at certain positions sn along its world line; these “discrete positions” being
such that the electron takes a time τ0 for traveling from one position sn−1 to
the next one sn . The electron, in principle, is still considered as point-like,
but the Dirac relativistic equations for the classical radiating electron are
replaced: (i) by a corresponding finite-difference (retarded) equation in the

velocity uµ (τ )

m0 uµ (τ ) uν (τ )
uµ (τ ) − uµ (τ − τ0 ) + [u ν (τ ) − u ν (τ − τ0 )]
τ0 c2
e
= Fµν (τ ) uν (τ ), (62)
c
which reduces to the Dirac equation (58) when τ0 → 0; and (ii) by a second
equation [the transmission law ] connecting this time the discrete positions
xµ (τ ) along the world line of the particle:
τ0
xµ (nτ0 ) − xµ [(n − 1) τ0 ] = {uµ (nτ0 ) − uµ [(n − 1) τ0 ]} (62 )
2
which is valid inside each discrete interval τ0 , and describes the internal
motion of the electron. In these equations, uµ (τ ) is the ordinary four-vector
velocity, satisfying the condition uµ (τ )uµ (τ ) = −c2 for τ = nτ0 , where
n = 0, 1, 2, . . . and µ, ν = 0, 1, 2, 3; while F µν is the external (retarded)
electromagnetic field tensor, and the chronon associated with the electron
(by comparison with Dirac’s equation) resulted in
τ0 2 ke2
≡ θ0 = 6.266 × 10−24 s,
2 3 m 0 c3
depending, therefore, on the particle (internal) properties [namely, on its
charge e and rest mass m0 ].
As a result, the electron happens to appear eventually as an extended-
like [108] particle, with internal structure, rather than as a point-like object.
For instance, one may imagine that the particle does not react instanta-
neously to the action of an external force because of its finite extension
(the numerical value of the chronon is of the same order as the time spent
by light to travel along an electron classical diameter). As already said,
Eq. (62) describes the motion of an object that happens to be point-like
only at discrete positions sn along its trajectory; even if both position and
velocity are still continuous and well-behaved functions of the parameter τ ,
since they are differentiable functions of τ . It is essential to note that a dis-
creteness character is given in this way to the electron without any need of
a “model” for the electron. Actually, it is well known that many difficulties
are met not only by the strictly point-like models, but also by the extended-
type particle models (“spheres”, “tops”, “gyroscopes”, etc.). We deem the
answer stays with a third type of models, the “extended-like” ones, as the
www.ebook3000.com
present approach; or as the (related) theories [108] in which the center of

the point-like charge is spatially distinct from the particle center-of-mass.
Let us repeat, anyway, that the worst troubles met in quantum field the-
ory, like the presence of divergencies, are due to the point-like character
still attributed to (spinning) particles; since — as we already remarked —
the problem of a suitable model for elementary particles was transported,
unsolved, from classical to quantum physics. One might consider that prob-
lem to be the most important even in modern particle physics.
Equation (62) and the following one, together, provide a full description
of the motion of the electron; but they are free from pre-accelerations, self-
accelerating solutions, and problems with the hyperbolic motion.
In the non-relativistic limit, the previous (retarded) equations get sim-
plified into the form

m0 1
[v (t) − v (t − τ0 )] = e E (t) + v (t) ∧ B (t) , (63)
τ0 c
τ0
r (t) − r (t − τ0 ) = [v (t) − v (t − τ0 )]. (63 )
2
The point is that Eq. (62) or Eq. (63) allow to overcome the difficulties met
with the Dirac classical equation. In fact, the electron macroscopic motion
is completely determined once velocity and initial position are given. The
explicit solutions of the above relativistic-equations for the radiating elec-
tron — or of the corresponding non-relativistic equations — verify that the
following questions can be regarded as having been solved within Caldirola’s
theory: (A) exact relativistic solutions: (1) free electron motion; (2) elec-
tron under the action of an electromagnetic pulse; (3) hyperbolic motion;
(B) non-relativistic approximate solutions: (4) electron under the action of
time-dependent forces; (5) electron in a constant, uniform magnetic field;
(6) electron moving along a straight line under the action of an elastic
restoring force.
In Ref. [73], we studied the electron radiation properties as deduced
from the finite-difference relativistic Eq. (62), and their series expansions,
with the aim of showing the advantages of the present formalism w.r.t. the
Abraham–Lorentz–Dirac one.
5.2. The three alternative formulations

Two more (alternative) formulations are possible of Caldirola’s equations,
based on different discretization procedures. In fact, Eqs. (62) and (63)
describe an intrinsically radiating particle. And, by expanding Eq. (62)

in terms of τ0 , a radiation reaction term appears. Caldirola called these
equations the retarded form of the electron equations of motion.
On the contrary, by rewriting the finite-difference equations in the form:

m0 uµ (τ ) uν (τ )
uµ (τ + τ0 ) − uµ (τ ) + [uν (τ + τ0 ) − uν (τ )]
τ0 c2
e
= Fµν (τ ) uν (τ ), (64)
c
xµ [(n + 1) τ0 ] − xµ (nτ0 ) = τ0 uµ (nτ0 ), (64 )
one gets the advanced formulation of the electron theory, since the motion is
now determined by advanced actions. At variance with the retarded formu-
lation, the advanced one describes an electron which absorbs energy from
the external world.
Finally, by adding together retarded and advanced actions, Caldirola
wrote down the symmetric formulation of the electron theory:

m0 uµ (τ ) uν (τ )
uµ (τ + τ0 ) − uµ (τ − τ0 ) + [uν (τ + τ0 ) − uν (τ − τ0 )]
2τ0 c2
e
= Fµν (τ )uν (τ ), (65)
c
xµ [(n + 1) τ0 ] − xµ ((n − 1) τ0 ) = 2τ0 uµ (nτ0 ) (65 )
which does not include any radiation reactions, and describes a non-
radiating electron.
Before closing this introduction to the classical “chronon theory”, let
us recall at least one more result derivable from it. If we consider a free
particle and look for the “internal solutions” of the Eq. (63 ), we get — for
a periodical solution of the type

2πτ 2πτ
ẋ = −β0 c sin ; ẏ = −β0 c cos ; ż = 0
τ0 τ0
(which describes a uniform circular motion) and by imposing the kinetic
energy of the internal rotational motion to equal the intrinsic energy m0 c2
of the particle — that the amplitude of the oscillations is given by β02 = 34 .
Thus, the magnetic moment corresponding to this motion is exactly the
anomalous magnetic moment of the electron, obtained in a purely classical
1 e3
context: µa = 4π m0 c2 . This shows, by the way, that the anomalous mag-
netic moment is an intrinsically classical, and not quantum, result; and the
absence of in the last expression is a confirmation of this fact.
www.ebook3000.com
5.3. Discretized QM
Let us pass to a topic we are more interested in, which is a second step for
our eventual application of the discretization procedures towards a possi-
ble solution of the measurement problem in QM, without having to make
recourse to the reduction (wave-packet instantaneous collapse) postulate.
Namely, let us focus our attention now on the consequences for QM of the
introduction of a chronon. In Ref. [73], we have extensively examined such
consequences. Here, we shall recall only some useful results.
There are physical limits that, even in ordinary QM, seem to prevent
the distinction of arbitrarily close successive states in the time evolution of
a quantum system. Basically, such limitations result from the Heisenberg
relations in such a way that, if a discretization is to be introduced in the
description of a quantum system, it cannot possess a universal value (since
those limitations depend on the characteristics of the particular system
under consideration): In other words, the value of the fundamental interval
of time has to change a priori from system to system. All these points are
in favor of the extension of Caldirola’s procedure to QM. Time will still
be a continuous variable, but the evolution of the system along its world
line will be regarded as discontinuous. In analogy with the electron theory
in the non-relativistic limit, one has to substitute the corresponding finite-
difference expression for the time derivatives; e.g.:
df (t) f (t) − f (t − ∆t)
→ , (66)
dt ∆t
where proper time is now replaced by the local time t. The chronon pro-
cedure can then be applied to obtain the finite-difference form of the
Schrödinger equation. As for the electron case, there are three different
ways to perform the discretization, and three “Schrödinger equations” can
be obtained:

i [Ψ (x, t) − Ψ (x, t − τ )] = ĤΨ(x, t), (67)
τ

i [Ψ (x, t + τ ) − Ψ (x, t − τ )] = ĤΨ(x, t), (67b)
2τ

i [Ψ (x, t + τ ) − Ψ (x, t)] = ĤΨ(x, t) (67c)
τ
which are, respectively, the retarded, symmetric and advanced Schrödinger
equations, all of them transforming into the (same) continuous equation
when the fundamental interval of time (that can now be called just τ ) goes
to zero.
Since the equations are different, the solutions they provide are also
fundamentally different. As we have already seen in the classical theory of
the electron, the symmetric equation represented a non-radiating motion,
providing only an approximate description of the motion (without taking
into account the effects due to the self-fields of the electron). However, in the
quantum theory it plays a fundamental role. In the discrete formalism too,
the symmetrical equation constitutes the only way to describe a bound non-
radiating particle. Let us remark that, for a time independent Hamiltonian,
the outputs obtained in the discrete formalism by using the symmetric
equation resulted to be [73] very similar to those obtained in the continuous
case. For these Hamiltonians, the effect of discretization appears basically
in the frequencies associated with the time dependent term of the wave-
function; and, in general, seem to be negligible.
However, the solutions of the retarded (and advanced ) equations show a
completely different behavior. For a Hamiltonian explicitly independent of
time, the solutions have a general form given by
τ −t/τ
Ψ (x, t) = 1 + i Ĥ f (x)

and, expanding f (x) in terms of the eigenfunctions of Ĥ:
Ĥun (x) = Wn un (x),

that is, writing f (x) = cn un (x), with |cn |2 = 1, one can obtain that
n n
τ −t/τ
Ψ (x, t) = cn 1 + i Wn un (x).
n

The norm of this solution is given by

2
2
|Ψ (x, t)| = |cn | exp (−γn t)
n
with

1 τ2 2 W2
γn = ln 1 + 2 Wn = 2n τ + O(τ 3 ),
τ
where it is apparent that the damping factor depends critically on the value
τ of the chronon. This dissipative behavior originates from the character
of the retarded equation; in the case of the electron, the retarded equation
possesses intrinsically dissipative solutions, representing a radiating system.
The Hamiltonian has the same status as in the ordinary (continuous) case:
www.ebook3000.com
It is an observable, since it is a Hermitian operator and its eigenvectors

form a basis of the state space. However, as we have seen, the norm of
the state vector is not constant any longer, due to the damping factor. An
opposite behavior is observed for the solutions of the advanced equation,
in the sense that they increase exponentially.
One of the achievements due to the introduction of the chronon hypoth-
esis in the realm of QM has been obtained in the description of a bound
electron by using the new formalism. In fact, Caldirola found for the excited
state of the electron the value E 105.55 MeV, which is extremely close
(with an error of 0.1%) to the measured value of the rest mass of the
muon. For this, and similar questions, we just refer the reader to the quoted
literature.
5.4. Discretized (retarded) Liouville equation

and the measurement problem:
Decoherence from dissipation
Suppose we want to measure the dynamical variable R of a (microscopic)
object O, by utilizing a (macroscopic) measuring apparatus A. The eigen-
value equation R|rO = r|rO defines a complete eigenvector-basis for the
observable R; so that any state |ψO of O can be given by the expansion

|ψO = r cr |rO .
As to the apparatus A, we are interested only in its observable A, whose
eigenvalues α represent the value indicated by a pointer ; then, we can write
A|α, N A = α|α, N A , quantity N representing the set of internal quantum
numbers necessary to specify a complete eigenvector-basis for it.
Let the initial state of A be |0, N A ; in other words, the pointer is
assumed to indicate initially the value zero. The interaction between O
and A is expressed by a time–evolution operator U , which is expected to
relate the value of r with the measurement α.
In conventional (“continuous”) QM, the density operator, ρ, obeys the
LvN equation
dρ i
= − [H, ρ] ≡ −i L ρ(t),
dt
where L is the Liouville operator; so that, if the Hamiltonian H is indepen-
dent of time, the time evolution of ρ is

i i
ρ(t − t0 ) = exp − H(t − t0 ) ρ0 exp H(t − t0 ) .

Let us consider the case in which the compound system O plus A is

initially, for instance,f in the mixed state

ρin = CM |ψM in in
ψM |,
M
where quantities CM are (classical) probabilities associated with the states

in
|ψM .
The “continuous” approach is known to forward

ρfin ≡ U ρU † = CM |ψM fin fin
ψM |
M

= c∗r1 cr2 CM {|αr1 ; r1 , M αr2 ; r2 , M |},
r1 ,r2 M
where the off-diagonal terms yield a coherent superposition of the corre-

sponding eigenvectors. In this case, the ordinary reduction postulate does
usually imply that, in the measurement process, the non-diagonal terms
vanish instantaneously due to the wave-function collapse; while smoother
approaches to de-phasing must normally have recourse to statistical consid-
erations, associated, e.g. to thermal baths.
On the contrary, in the discrete case, with the interaction embedded in
the Hamiltonian H, the situation is rather different and simpler; and one
does not have to call any statistical approaches into the play. Indeed, let
us consider the energy representation, where |n are the states with defined
energy: H|n = En |n. Since the time evolution operator is a function of
the Hamiltonian, and commutes with it, the basis of the energy eigenstates
will be a basis also for this operator.
The discretized (retarded ) LvN equation is
ρ(t) − ρ(t − τ )
= −i L ρ(t) (68)
τ
which reduces to the LvN equation when τ → 0. The essential point is that,
following e.g. a procedure similar to Bonifacio’s [75, 126], one gets in this
f By contrast,
N if we consider as initial state for the system O plus A the pure state |ψN in =
|ψO |0, N A ≡ |ψO |0, N A , then, within the ordinary “continuous” approach, the
time evolution leads necessarily to a coherent
P superposition of (macroscopically distinct)
eigenvectors: U (t, t0 ) |ψO |0, N A = r c r |α r ; r, N ≡ |ψN . As a consequence, as
fin
well known, one has to postulate a state collapse from |ψN fin to |α ; r , N , where r is
r0 0 0
the value indicated by the pointer after the measurement.
www.ebook3000.com
case a non-unitary time-evolution operator :

−t/τ
iτ L
V (t, 0) = 1 + (69)

which, as all non-unitary operators, does not preserve the probabilities asso-
ciated with each of the energy eigenstates (that make up the expansion of
the initial state in such a basis of eigenstates). We are interested in the
time instants t = kτ , with k an integer.g Thus, the time-evolution opera-
tor (13) takes the initial density operator ρin to a final state for which the
non-diagonal terms decay exponentially with time; namely, to
−t/τ
ρfin in
rs = r|V (t, 0)|s = ρrs [1 + iωrs τ ] , (70)
where
1 1
ωrs ≡ (Er − Es ) ≡ (∆E)rs . (71)

Expression (70) can be written as
ρrs (t) = ρrs (0)e−γrs t e−iνrs t (72)
with
1 2 2

γrs ≡ ln 1 + ωrs τ ; (73)
2τ
1
νrs ≡ tan−1 (ωrs τ ). (74)
τ
One can observe, indeed, that the non-diagonal terms tend to zero with
time, and that the larger the value of τ , the faster the decay becomes.
Actually, the chronon τ is now an interval of time related no longer to
a single electron, but to the whole system O + A. If one imagines the time
interval τ to be linked to the possibility of distinguishing two successive,
different states of the system, then τ can be significantly larger than 10−23 s,
implying an extremely faster damping of the non-diagonal terms of the
density operator: See Fig. 2.
g Let us emphasize that the appearance of non-unitary time-evolution operators is not

associated with the coarse graining approach only, since they also come out from the
discrete Schrödinger equations.
5.5. Further comments

It should be noticed that the time-evolution operator (69) preserves trace,
obeys the semi-group law, and implies an irreversible evolution towards a
stationary diagonal form. In other words, notwithstanding the simplicity of
the present “discrete” theory, that is, of the chronon approach, an intrinsic
relation is present between measurement process and irreversibility: Indeed,
the operator (13), meeting the properties of a semi-group, does not possess
in general an inverse (and non-invertible operators are, of course, related
to irreversible processes). For instance, in a measurement process in which
the microscopic object is lost after the detection, one is just dealing with an
irreversible process that could be well described by an operator like (69).
In our (discrete and retarded) theory, the “reduction” to the diagonal
form
t→0

ρ(t) → ρnn (0)|nn|
n
is not instantaneous, but depends — as we have already seen — on the

characteristic value τ . More precisely, the non-diagonal terms tend expo-
nentially to zero according to a factor which, to the first order, is given by

−ωnm
2
τ t

exp
2 . (75)
Thus, the reduction to the diagonal form occurs, provided that τ possesses
a finite value, no matter how small, and provided that ωnm τ , for every n,m,
is not much smaller than 1; where
ωnm = (En − Em )/
are the transition frequencies between the different energy eigenstates (the
last condition being always satisfied, e.g. for non-bounded systems).
It is essential to note that decoherence has been obtained above, with-
out having recourse to any statistical approach, and in particular without
assuming any “coarse graining” of time. The reduction to the diagonal
form illustrated by us is a consequence of the discrete (retarded) Liouville–
von Neumann equation only, once the inequality ωnm τ 1 is not verified.
Moreover, the measurement problem is still controversial even with
regard to its mathematical approach: In the simplified formalization
introduced above, however, we have not included any consideration beyond
those common to the quantum formalism, allowing an as clear as possible
recognition of the effects of the introduction of a chronon. Of course, we
www.ebook3000.com
have not fully solved the quantum measurement problem, since we have not
yet clearly found a model for determing which one of the diagonal values
the actual experiment will reveal . . . .
Let us, however, repeat that the introduction of a fundamental interval
of time in approaching the measurement problem made possible a simple
but effective formalization of the diagonal reduction process (through a
mechanism that can be regarded as a decoherence caused by interaction
with the environment [see Ref. [74] and references therein]) only for the
retarded case. This is not obtainable, when taking into account the sym-
metric version of the discretized LvN equation.
It may be worthwhile to stress that the retarded form (68) of the direct
discretization of the LvN equation is the same equation obtained via the
coarse grained description (extensively adopted in [75, 126]). This led us
to consider such an equation as a basic equation for describing complex
systems, which is always the case when a measurement process is involved.
Let us add some brief remarks. First : The “decoherence” does not
occur when we use the time evolution operators obtained directly from the
retarded Schrödinger equation; the dissipative character of that equation,
in fact, causes the norm of the state vector to decay with time, leading
again to a non-unitary evolution operator: However, this operator (after
having defined the density matrix) yields damping terms which act also
on the diagonal terms! We discussed this point, as well as the question of
the compatibility between Schrödinger’s picture and the formalism of the
density matrix, in an appendix of Ref. [73]. Second : The new discrete for-
malism allows not only the description of the stationary states, but also
a (space–time) description of transient states: The retarded formulation
yields a natural quantum theory for dissipative systems; and it is not with-
out meaning that it leads to a simple explication of the diagonal reduction
process. Third : Since the composite system O + A is a complex system, it
is suitably described by the coarse grained description (exploited by Boni-
facio in some important papers of his [75, 126]): it would be quite useful
to increase our understanding of the relationship between the two men-
tioned pictures in order to get a deeper insight on the decoherence processes
involved.
A further comment is the following. We have seen that the chronon for-
malism [73] has obvious connection with our view about time, and space–
time. But let us remind that the discrete formalism bears a further element
of interest, since it possesses a non-Hermitian character (as better clar-
ified e.g. in the appendices of Ref. [73]). We know by now, for instance,
that in the Schrödinger representation of such formalism, proper contin-

uous equations can reproduce the outputs obtained with the discretized
equations, once we replace the (discrete) conventional Hamiltonian by the
suitable (continuous) equivalent, non-Hermitian Hamiltonian. Indeed, one
can find out a new Hamiltonian H̃ such that the new continuous Schrödinger
equation
∂Ψ (x, t)
i = H̃Ψ (x, t)
∂t
reproduces, at the points t = nτ , the same results obtained from the dis-
cretized equations. Let us recall that Casagrande and Montaldi [78] showed
it to be always possible to find out a continuous generating function that
allows obtaining a differential equation equivalent to the original finite-
difference one, such that at every point of interest their solutions are iden-
tical. This procedure, as we know, is useful also because it is often rather
difficult to work with the finite-difference equations on a quantitative (and
qualitative) basis. This equivalent Hamiltonian H̃ is non-Hermitian; even
if, as expected, H̃ → Ĥ when τ → 0.
Let us finally recall that, as previously mentioned, the chronon can have
consequences in several different areas of physics: for instance, in Ref. [105]
spin was derived within a discrete-time approach. As a further example, in
the next subsection, we want to report with some details on the possible
role of the chronon in Cosmology.
5.6. On the chronon in quantum cosmology

As we were saying, the chronon can play a role also in recent theories refer-
ring, e.g. to the “archaic” universe: theories which are group-theoretical
approaches to quantum cosmology based on works by L. Fantappié and
G. Arcidiacono. These classical, interesting (and often forgotten) publica-
tions by Fantappié and by Arcidiacono form such a large theoretical back-
ground, that here, as far as it is concerned, we can only refer the readers to
papers like the ones quoted in this subsection, as well as to Ref. [109] and
references therein.
Let us here recall that, in terms of the Penrose terminology, the structure
of QM can be regarded as represented essentially by a unitary evolution
operator U , acting on the wave-function Ψ, and by the Ψ-collapse that we
indicate by R. Some of the problems of QM are known to come out from
the difficulty in connecting, loosely speaking, U and R; indeed, the collapse
does not seem to be derivable from U . A possible way out for conciliating
www.ebook3000.com
U and R is by the introduction of the “pilot wave”, which leads however to

problems with the meaning of Ψ. A view on QM which can help is the new
Transactional Interpretation of QM [110–112].
Its first version, due to Cramer [113], regarded the non-local connections
as a link between advanced and retarded potentials a la Wheeler–Feynman:
But this aroused of course a lot of mathematical and conceptual problems,
connected also to its too classical context. Intuitively, the idea was rather
simple: each particle “responded” to all its future possibilities. In the new
version of the so-called Transactional Interpretation, one does not meet
any longer complications of this kind; and one just needs some simple rules
about the opening and closing of the “transactions” in order to be able to
fix in a univocal way the evolution operators.
Actually, at a fundamental level only the transactions between the field-
modes take place, and the wave-function manifests itself as a statistical
coverage of a large amount of elementary transitions.
In this context, the adoption of the chronon as a minimum duration
of the transaction opening/ending is a possibly useful hypothesis, justified
for instance by the role — as in Caldirola’s papers — of the classical elec-
tron radius, and the very range of strong interactions in particle physics;
even if future developments in quantum gravity might shift the chronon
value towards the Planck scale [114]. According to these views, physical
processes whose duration is not larger than a chronon are possible only as
virtual processes so that cosmology could result to be connected with the
foundations of QM. Indeed, when the age of the “cosmos” (or rather of its
precursor) did not exceed a chronon, it may be expected that all matter
was associated with quantum virtual processes. By contrast, when the age
of such a “cosmos” exceeded a chronon, the transactional processes became
possible and conversion of matter from the virtual to the real state could
have taken place: this conversion might be nothing but what we call “big
bang”.
Such an idea plays an important role in the theories of the Archaic Uni-
verse, when one refers indeed to a quantum vacuum still populated solely by
virtual processes (without ordinary particles); and gets, among the others,
that the geometry of such a vacuum becomes then a de Sitter Euclidean 5D
(hyper)spherical surface. More specifically, the “archaic universe” theories
go back to the group theoretical approaches proposed in the mentioned,
classical works of Arcidiacono and Fantappié [115–120] wherein the Projec-
tive Relativity was introduced.
Let us recall some basic concepts. Projective Relativity differs from the
usual einsteinian Relativity in the existence of a de Sitter horizon, located
at the same chronological distance from any observer. Because this distance
does not depend on cosmic time, it is now the same as it was at the big bang
time. But the existence of a de Sitter horizon in the past of an observer
who emerged out from the big bang does imply in its turn the pre-existence
of some form of space–time, even before the big bang. In other words,
before big-bang the aforementioned conversion process had to take place.
In the meantime, no real matter existed; as a consequence, the geometry
of this “pre-spacetime” must be that of the de Sitter space (according to
the gravitational equations of Projective Relativity itself in the absence of
matter). The inexistence of real processes could be seen, if you preferred,
as the inexistence of time. It is therefore possible to assume that such an
archaic universe was the 4D surface of a 5D hemisphere (cf. also Ref. [109]
and references therein), that is, the Wick-rotated version of the de Sitter
space. The “precursor” of time was, then, the 5D distance from the plane
of the equator; and the big bang happened when this time became equal to
a chronon. Afterward, matter became real and real physical processes were
started, requiring a radical change of geometry.
The new geometry will be connected to the “archaic” geometry via a
Wick rotation (with the emergence of time); why the gravitational equa-
tions in presence of matter involved a scale reduction. Using the Milne ter-
minology, the public archaic space–time now breaks down into a multitude
of single private space–times (one for each “fundamental observer”), con-
nected at the beginning by the de Sitter group. It may be even shown that
this nucleation from the pre-vacuum can naturally recover, as a consequence
of the geometry one had to adopt, the Hartle–Hawking condition [121].
6. Some Conclusions
1. We have shown that the Time operator (1), Hermitian even if non-
self-adjoint, works for any quantum collisions or motions, in the case of
a continuum energy spectrum, in non-relativistic QM and in 1D quantum
electrodynamics. The uniqueness of the (maximal) Hermitian time operator
(1) directly follows from the uniqueness of the Fourier-transformations from
the time to the energy representation. The time operator (1) has been fruit-
fully used in the case, for instance, of tunneling times (see Refs. [24–28]),
and of nuclear reactions and decays (see Refs. [10–13] and also Ref. [122]).
We have discussed the advantages of such an approach with respect to
www.ebook3000.com
POVM’s, which is not applicable for 3D particle collisions, within a wide

class of Hamiltonians.
The mathematical properties of the present Time operator have actu-
ally demonstrated — without introducing any new physical postulates —
that time can be regarded as a quantum-mechanical observable, at the
same degree of other physical quantities (spatial coordinates, energy,
momentum, . . .).
The commutation relations (Eqs. (8), (22), and (31)) analyzed here, and
the uncertainty relations (9), are found to be analogous to those known
for other pairs of canonically conjugate observables (as for coordinate x̂
and momentum p̂x , in the case of Eq. (9)). Of course, our new relations
do not replace, but merely extend the meaning of the classic time and
energy uncertainties, given e.g. in Ref. [41]. In Sec. 2.6, we have studied
the properties of Time, as an observable, for quantum-mechanical systems
with discrete energy spectra.
2. Let us stress that the Time operator (1), and relations (2)–(4), (15), (16),
have been used for the temporal analysis of nuclear reactions and decays in
Refs. [10–13]; as well as of new phenomena, about time delay-advances in
nuclear physics and about time resonances or explosions of highly excited
compound nuclei, in Refs. [122–125]. Let us also recall that, besides the
time operator, other quantities, to which (maximal) Hermitian operators
correspond, can be analogously regarded as quantum-physical observables:
For example, von Neumann himself [8, 9, 45]) considered the case of the
momentum operator −i∂/∂x in a semi-space with a rigid wall orthogonal
to the x-axis at x = 0, or of the radial momentum −i∂/∂r, even if both act
on packets defined only over the positive x- or r-axis, respectively.
Section 2.5 has been devoted to a new “Hamiltonian approach”: namely,
to the introduction of the analogue of the “Hamiltonian” for the case of the
Time operator.
3. In Sec. 3, we have proposed a suitable generalization for the Time opera-

tor (or, rather, for a Space–Time operator) in relativistic QM. For instance,
for the Klein–Gordon case, we have shown that the Hermitian part of the
three-position operator x̂ is nothing but the Newton–Wigner operator, and
corresponds to a point-like position; while its anti-Hermitian part can be
regarded as yielding the sizes of an extended-type (ellipsoidal) localiza-
tion. When dealing with a four-position operator, one finds that the Time
operator is self-adjoint for unbounded energy spectra, while it is a (max-
imal) Hermitian operator when the kinetic energy, and the total energy,
are bounded from below, as for a free particle. We have extensively made
recourse, in the latter case, to bilinear forms, which dispense with the neces-
sity of eliminating the lower point — corresponding to zero velocity — of
the spectra. It would be interesting to proceed to further generalizations of
the three- and four-position operator for other relativistic cases, and ana-
lyze the localization problems associated with Dirac particles, or in 2D and
3D quantum electrodynamics, etc. Work is in progress on time analyses
in 2D quantum electrodynamic, for application, e.g. to frustrated (almost
total) internal reflections. Further work has still to be done also about the
joint consideration of particles and antiparticles.
4. Non-Hermitian Hamiltonians can play an important role in the descrip-

tion of Unstable States, by associating the decaying “resonances” with the
eigenvectors of quasi-Hermitian Hamiltonians: But on this point we just
referred the interested reader to Refs. [40, 59, 122]. In Secs. 4 and 5 of
this work we paid attention, instead, to the possible role non-Hermitian
Hamiltonians, and non-unitary time-evolution operators, in the cases of
the nuclear optical model, of microscopic quantum dissipation, and partic-
ularly in an approach to the measurement problem in QM in terms of the
chronon. We have particularly devoted Sec. 5 to the chronon formalism —
where the chronon is a “quantum” of time, in the sense specified above —
for its obvious connection with our view of time, and of space–time; and
also because that discrete formalism has a non-Hermitian character [73].
As to quantum dissipation [68–93], we discussed e.g. a particular approach
for getting decoherence through interaction with the environment [73, 74].
In Sec. 4, we had however touched also questions related with collisions
in absorbing media; mentioning the case of the optical model in nuclear
physics, without forgetting that non-Hermitian operators show up even in
the case of tunneling — e.g. in fission phenomena — with quantum dis-
sipation, and of quantum friction. As we were saying, among the many
approaches to quantum irreversibility, in Sec. 4.3, we already anticipated
the possible way for obtaining a “reduction to diagonal form” by the intro-
duction of finite-difference equations (in terms of the “chronon”): And we
have subsequently exploited at length this issue, for showing the conse-
quences of the introduction of a chronon in classical physics and in QM
(Sec. 5).
5. Let us eventually observe that the “dual equations” (26) and (27) seem
to be promising also for the study the initial stage of our cosmos, when
www.ebook3000.com
tunnelings can take place through the barriers which appear in quantum
gravity in the limiting case of quasi-Schrödinger equations [127].
Acknowledgments
This paper is largely based on work developed by one of us (ER), along
the years, in collaboration with V.S. Olkhovsky, and, in smaller parts, with
P. Smrz, with R.H.A. Farias, and with S.P. Maydanyuk; while another of
us (IL) acknowledges the collaboration of L. Chiatti. Thanks are more-
over due, for stimulating discussions or kind collaboration, to Y. Aharonov,
A. Agodi, M. Baldo, R. Bonifacio, E.O. Capelas, H.E. Hernández-Figueroa,
A.S. Holevo, V.L. Lyuboshitz, C. Meroni, R. Mignani, S. Paleari, A. Pen-
nisi, V. Petrillo, U.V.G. Recami, P. Riva, G. Salesi, A. Santambrogio, and
B.N. Zakhariev. One of the authors’ home-page is www.unibg.it/recami.
References
1. W. Pauli, Handbuch der Physik vol. 5/1 Ed. by S. Fluegge, Springer, Berlin,
p. 60, 1926.
2. W. Pauli, General Principles of Quantum Theory, Springer, Berlin, 1980.
3. V.S. Olkhovsky and E. Recami, Nuovo Cimento A 53, 610 (1968).
4. V.S. Olkhovsky and E. Recami, Nuovo Cimento A 63, 814 (1969).
5. V.S. Olkhovsky and E. Recami, Lett. Nuovo Cim. (First Series) 4, 1165
(1970).
6. V.S. Olkhovsky, Ukrainskiy Fiz. Zhurnal [in Ukrainian and Russian] 18,
1910 (1973).
7. V.S. Olkhovsky, E. Recami, and A. Gerasimchuk, Nuovo Cimento A 22,
263 (1974).
8. E. Recami, An operator for the observable time 1976 in Recent Develop-
ments in Relativistic Q.F.T. and its Application, (in Proc. XIII Karpatz
Winter School on Theor. Phys.), Vol. II, ed. by W. Karwowski, Wroclaw
Univ. Press, Wroclaw, 1976, pp. 251–265.
9. E. Recami, A time operator and the time-energy uncertainty relation
1977, in The Uncertainty Principle and Foundation of Quantum Mechanics,
Chap. 4, ed. by C. Price and S. Chissik, J. Wiley, London, 1977, pp. 21–28.
10. V.S. Olkhovsky, Sov. J. Part. Nucl. 15, 130 (1984).
11. V.S. Olkhovsky, Nukleonika 35, 99 (1990).
12. V.S. Olkhovsky, Atti Accademia Peloritana dei Pericolanti, Sci., Mat. e
Nat. 70, 21 and 135 (1992).
13. V.S. Olkhovsky, Mysteries, Puzzles and Paradoxes in Quantum Mechanics,
Ed. by R. Bonifaccio, AIP, Woodbury, N.Y., 1998, pp. 272–276
14. A.S. Holevo, Rep. Math. Phys. 13, 379 (1978).
15. A.S. Holevo, Probabilistic and Statistical Aspects of Quantum Theory, North
Holland, Amsterdam, 1982.
16. M.D. Srinivas and R. Vijayalakshmi, Pramana J. Phys. 16, 173 (1981).
17. P. Busch, M. Grabowski, and P.J. Lahti, Phys. Lett. A 191, 357 (1994).
18. D.H. Kobe and V.C. Aguilera-Navarro, Phys. Rev. A 50, 933 (1994).
19. P. Blanchard and A. Jadczyk, Helv. Phys. Acta 69, 613 (1996).
20. N. Grot, C. Rovelli, and R.S. Tate, Phys. Rev. A 54, 4676 (1996).
21. J. Leo’n, J. Phys. A 30, 4791 (1997).
22. Y. Aharonov, J. Oppenhem, S. Popescu, B. Reznik, and W. Unruh,
Phys. Rev. A 57, 4130 (1998).
23. H. Atmanspacher and A. Amann, Internat. J. Theor. Phys. 37, 629 (1998).
24. V.S. Olkhovsky and E. Recami, Phys. Rep. 214, 339 (1992).
25. V.S. Olkhovsky, E. Recami, F. Raciti, and A.K. Zaichenko, J. de Phys.
(France) I 5, 1351 (1995).
26. V.S. Olkhovsky, E. Recami, and J. Jakiel, Phys. Rep. 398, 133 (2004).
27. V.S. Olkhovsky and E. Recami, Intern. J. Mod. Phys. A 22, 5063 (2007).
28. V.S. Olkhovsky and A. Agresti, Proc. of the Adriatico Research. Conf. on
Tunneling and its Implications World Sci., 1997, pp. 327–355.
29. R. Giannitrapani, Int. J. Theor. Phys. 36, 1575 (1997).
30. J. Kijowski, Phys. Rev. A 59, 897 (1999).
31. M. Toller, Phys. Rev. A 59, 960 (1999).
32. V. Delgado, Phys. Rev. A 59, 1010 (1999).
33. J. Muga, J. Palao, and C. Leavens, Phys. Lett. A 253, 21 (1999).
34. I.L. Egusquiza and J.G. Muga, Phys. Rev. A 61, 012104 (1999).
35. P. Kocha’nski and K. Wo’dkievicz, Phys. Rev. A 60, 2689 (1999).
36. A. Góźdź and M. Dȩbicki, Phys. At. Nuclei 70, 529 (2007).
37. Zhi-Yong Wang and Cai-Dong Xiong, Annals of Physics 322, 2304 (2007).
38. E. Recami, V.S. Olkhovsky and S.P. Maydanyuk, Int. J. Mod. Phys. A 25,
1785–1818 (2010).
39. M.A. Naimark, Izvestiya Akademii Nauk SSSR, seriya matematicheskaya
[in Russian, partially in English] 4, 277 (1940).
40. E. Recami, W.A. Rodrigues, and P. Smrz, Hadronic Journal 6, 1773–1789
(1983).
41. V.S. Olkhovsky and E. Recami, Int. J. Mod. Phys. B 22, 1877 (2008).
42. M.H. Stone, Proc. Nat. Acad. Sci. USA 16(1), (1930).
43. N.I. Akhiezer and I.M. Glazman, The Theory of Linear Operators in Hilbert
Space Pitman, Boston, Mass, 1981.
44. D. ter Haar, Elements of Hamiltonian Mechanics, Oxford, 1971.
45. J. Von Neumann, Mathematical Foundations of Quantum Mechanics,
Princeton Univ. Press, Princeton, N.J., 1955.
46. Y. Aharonov and D. Bohm, Phys. Rev. A 122, 1649 (1961).
47. M.A. Naimark, Izvestiya Akademii Nauk SSSR, seriya Matematika 7, 237–
244 (1943).
48. T. Carleman, Sur les Equations Integrales Anoyau Rel et Symtrique, Upp-
sala, 1923.
49. S. Schweber, An Introduction to Relativistic Quantum Field Theory,
Chapt. 5.3, Row, Peterson and Co., Evanston, IL, 1961.
www.ebook3000.com
50. A.I. Akhiezer and V.B. Berestezky, Quantum Electrodynamics [in Russian],
Fizmatgiz, Moscow, 1959.
51. D.M. Rosenbaum, J. Math. Phys. 10, 1127 (1969).
52. A.J. Ka’lnay, Boletin del IMAF (Co’rdoba) 2, 11 (1966).
53. A.J. Ka’lnay and B.P. Toledo, Nuovo Cimento A 48, 997 (1967).
54. J.A. Gallardo, A.J. Ka’lnay, B.A. Stec and B.P. Toledo, Nuovo Cimento A
48, 1008 (1967).
55. J.A. Gallardo, A.J. Ka’lnay, B.A. Stec and B.P. Toledo, Nuovo Cimento A
49, 393 (1967).
56. J.A. Gallardo, A.J. Ka’lnay and S.H. Risenberg, Phys. Rev. 158, 1484
(1967).
57. M. Baldo and E. Recami, Lett. Nuovo Cim. 2, 613 (1969).
58. E. Recami, Atti Accad. Naz. Lincei (Roma) 49, 77 (1970).
59. A. Agodi, M. Baldo and E. Recami, Annals of Physics 77, 157 (1973).
60. H. Feshbach, C.E. Porter and V.F. Weisskopf, Phys. Rev. 96, 448 (1954).
61. P.I. Hodgson, The Optical Model of Elasstic Scattering, Clarendon Press,
Oxford, UK, 1963.
62. P.A. Moldauer, Nucl. Phys. 47, 65 (1963).
63. A.G. Koning and J.P. Delaroche, Nucl. Phys. A 713, 231 (2003).
64. S. Kunieda, S. Chiba, K. Shilata, A. Ichihara and E. Sh. Suchovitski, J.
Nucl. Sc. and Techn. 44, 838 (2007).
65. M.V. Nikolaiev and V.S. Olkhovsky, Theor. Mathem. Phys. 31, 418 (1977).
66. V.S. Olkhovsky, Theor. Mathem. Phys. 20, 774 (1975).
67. V.S. Olkhovsky and A.K. Zaichenko, Nuovo Cimento A 63, 155 (1981).
68. P. Caldirola, G. Casati, and A. Prosperetti, Nuovo Cimento A 43, 127
(1978).
69. P. Caldirola, Rivista N. Cim. 2(13), (1979).
70. P. Caldirola and L. Lugiato, Physica A 116, 248 (1982).
71. P. Caldirola, Dissipation in quantum theory (40 years of research), Hadronic
J. 6, 1400–1433 (1983).
72. P. Caldirola and E. Montaldi, Nuovo Cimento B 53, 291 (1979).
73. A.R.H. Farias and E. Recami, Introduction of a quantum of time (‘chronon’)
and its consequences for the electron in quantum and classical physics, Adv.
Imaging Electron Phys. (AIEP) 163, 33–115 (2010) [83 printed pages].
74. E. Recami and A.R.H. Farias, A simple quantum (finite-difference) equation
for dissipation and decoherence, Nuovo Cimento B 124, 765–776, and refs.
therein (2009).
75. R. Bonifacio, Lett. N. Cim. 37, 481 (1983).
76. R. Bonifacio and P. Caldirola, Lett. N. Cim. 38, 615 (1983).
77. G.C. Ghirardi and T. Weber, Lett. N. Cim. 39, 157 (1984).
78. F. Casagrande and E. Montaldi, Nuovo Cimento A 40, 369 (1977).
79. R. Mignani, Lett. N. Cim. 38, 169 (1983).
80. P. Caldirola, Nuovo Cimento 18, 393 (1941).
81. A. Janussis, et al., Lett. N. Cim. 29, 259 (1980).

85. A. Janussis, et al., Nuovo Cimento B 67, 161 (1982).
88. A. Janussis, G. Brodimas, and R. Mignani, J. Phys. A: Math. Gen. 24,
L775 (1991).
89. A. Janussis, A. Leodaris, and R. Mignani, Phys. Lett. A 197, 187 (1995).
90. P. Caldirola, Nuovo Cimento 18, 393 (1941).
91. E. Kanai, Progr. Theor. Phys. 3, 440 (1948).
92. P. Angelopoulon, et al., Int. J. Mod. Phys. B 9, 2083 (1995).
93. A. Caldeira and A. Leggett, Annals Phys. 149, 374 (1983).
94. V.S. Olkhovsky, E. Recami, and G. Salesi, Tunneling through two successive
barriers and the Hartman (Superluminal) effect, Europhys. Lett. 57, 879–
884 (2002) [e-print quant-ph/0002022].
95. E. Recami, Superluminal tunneling through successive barriers. Does QM
predict infinite group-velocities? J. Modern Opt. 51, 913–923 (2004).
96. V.S. Olkhovsky, E. Recami, and A.K. Zaichenko, Resonant and non-
resonant tunneling through a double barrier, Europhys. Lett. 70, 712–718
(2005) [e-print arXiv:quant-th/0410128].
97. Y. Aharonov, N. Erez, and B. Resnik, Phys. Rev. A 65, 052124 (2002).
98. F. Raciti and G. Salesi, J. Phys.-I (France) 4, 1783 (1994).
99. G. Nimtz, H. Spieker, and M. Brodowsky, J. Phys.-I (France) 4, 1379 (1994).
100. P. Caldirola, Lett. N. Cim. 16, 151 (1976).
103. R. Mignani, Lett. N. Cim. 38, 169 (1983).
104. R.M. Santilli, Foundations of Theoretical Mechanics, Vol. II: Birkhoffian
Generalization of Hamiltonian Mechanics, Springer, Berlin, 1983; R.M. San-
tilli, Hadronic J. 2, 1460 (1979).
105. E. Recami and G. Salesi, Deriving spin within a discrete-time theory, Found.
Phys. 37, 277–294 (2007).
106. See, e.g. P.A.M. Dirac, The classical theory of electron, Proc. Royal Soc. A
167, 148 (1938).
107. P. Caldirola, Suppl. Nuovo Cim. 3, 297 (1956); and Rivista Nuovo Cim. 2,
13 (1979), and refs. therein. Cf. also A.D. Yaghjian, Relativistic Dynamics
of a Charged Sphere, Springer, Berlin, 1992.
108. Cf., e.g. E. Recami and G. Salesi, Phys. Rev. A 57, 98 (1998), and refs.
therein.
109. E. Recami, “Relativity and Beyond”, Chap. 16 in Albert Einstein,
1879−1979: Relativity, Cosmology and Quanta, Eds. by F. de Finis and
M. Pantaleo, Johnson Rep. Co., New York, N.Y., 1979, Vol. 2, pp. 537–
597; “Teoriya Otnositelnosti i ee Obobstchenya”, in Astrofizika, Kyanti i
Teoriya Otnositelnosti, ed. by F. I. Fedorov, Mir, Moscow, 1982, Chap. 4,
pp. 53–128; both being translations, into English and Russian respectively,
of E. Recami, “Alcune Possibili Estensioni della Relatività”, Chapt. 18 in
Astrofisica e Cosmologia, Gravitazione, Quanti e Relatività — Centenario
www.ebook3000.com
di Einstein, ed. by M. Pantaleo, Giunti–Barbera, Florence, 1979, pp. 1021–

1097.
110. L. Chiatti, (2012), arXiv: 1204.6636.
111. I. Licata, Europ. Phys. J. — Web of Conferences (in press) (2013).
112. R.E. Kastner, The Transactional Interpretation of Quantum Mechanics,
Cambridge Univ. Press, UK, 2013.
113. J.G. Cramer, Int. J. Theor. Phys. 27, 227–236 (1988).
114. I. Licata, Hadronic J. 14, 225–250 (1991).
115. L. Fantappié, Collectanea Mathematica 11(2), 78–136 (1959).
116. G. Arcidiacono, Gen. Rel. Grav. 7, 885 (1976).
117. G. Arcidiacono, Gen. Rel. Grav. 7, 865 (1977).
118. I. Licata, EJTP 3(10), 211–224 (2006).
119. I. Licata and L. Chiatti, Int. J. Theor. Phys. 48, 10031018 (2009).
120. I. Licata and L. Chiatti, Int. J. Theor. Phys. 49, 23792402 (2010).
121. J. Hartle and S. Hawking, Phys. Rev. D 28, 2960 (1983).
122. V.S. Olkhovsky, M.E. Dolinska, and S.A. Omelchenko, Cent. Eur. J. Phys.
4(2), 1–18 (2006).
123. V.S. Olkhovsky and N.L. Doroshko, Europhys. Lett. 18(6), 483–486 (1992).
124. A. D’Arrigo, N.L. Doroshko, N.V. Eremin, V.S. Olkhovsky, et al., Nucl.
Phys. A549, 375–386 (1992).
125. A. D’Arrigo, N.L. Doroshko, N.V. Eremin, V.S. Olkhovsky, et al., Nucl.
Phys. A564, 217–226 (1993).
126. R. Bonifacio, Nuovo Cimento B 114, 473 (1999).
127. V.S. Olkhovsky, S.P. Maydanuyk, A. Del Popolo and E. Recami, A Fully
quantum model of Big-Bang, in Theoretical Concepts in Quantum Mechan-
ics (Intech), pp. 341–382, 2012 [ISBN 978-953-51-0088-1]; doi: 10.5772/2075.
www.ebook3000.com
Chapter 15
Emergent Space–Time
George Chapline
Lawrence Livermore National Laboratory, Livermore, CA 94551, USA
chapline1@llnl.gov
It has been shown that a nonlinear Schrödinger equation in 2 + 1 dimen-

sions equipped with an SU(N) Chern–Simons gauge field can provide an
exact description of certain self-dual Einstein spaces in the limit N −>∞.
Ricci flat Einstein spaces can then be viewed as arising from a quan-
tum pairing of the classical self-dual and anti-self-dual solutions. In this
chapter, we will outline how this theory of empty space–time might be
generalized to include matter and vacuum energy by transplanting the
nonlinear Schrödinger equation used to construct Einstein spaces to the
25+1-dimensional Lorentzian Leech lattice. If the distinguished 2 spatial
dimensions underlying the construction of Einstein spaces are identified
with a hexagonal lattice section of the Leech lattice, the wave-function
becomes an 11 × 11 matrix that can represent fermion and boson degrees
of freedom (DOF) associated with 2-form and Yang–Mills gauge symme-
tries. The resulting theory of gravity and matter in 3+1 dimensions is not
supersymmetric, which provides an entry for a vacuum energy. Indeed,
in the case of a Lemaitre cosmological model, the emergent space–time
will naturally have a vacuum energy on the order of the observed cos-
mological constant.
1. Introduction
The ultimate fate of matter undergoing gravitational collapse is a long
standing enigma. Following the seminal paper of Oppenheimer and Snyder,
it came to be widely accepted that the gravitational collapse of a sufficiently
large mass will inevitably lead to the formation of a density singularity and
an event horizon [1]. Moreover, it is generally believed that these classical
predictions will turn out to be correct even when quantum effects are taken
into account. This belief is based on the observation that if the collapsing
mass is suffiently large, then the formation of an event horizon and initiation
419
420 G. Chapline
of irreversable collapse will take place in a region of space–time where the

curvature of space is very small. However, the resulting “black hole” space–
times seem to contradict quantum mechanics in two fundamental respects.
The most famous of these puzzles concerns the question of what happens to
the quantum mechanical information carried by the collapsing matter; in
quantum mechanics information can never be lost. Another defect with the
general relativistic theory of gravitational collapse is that black hole space–
times will generically have no universal time, which is required for the usual
formulations of quantum mechanics. The most plausible resolution of these
paradoxes is that that quantum mechanics affects classical solutions of the
Einstein equations in a singular way. In fact, there are plausible arguments
that in a quantum theory of gravity, and irrespective of the local space–time
curvature, quantum fields near to where a trapped surface would form will
exhibit large fluctuations with a frequency spectrum extending up to the
Planck frequency [2, 3].
An especially intriguing idea of why in a quantum theory of space–time
large quantum fluctuations should appear near a surface, where classical
general relativity predicts that there should be an event horizon, arose in
the summer of 2,000 during a conversation between Robert Laughlin and
the author while attending a conference on high Tc superconductivity at
Los Alamos. During the conversation, Laughlin mentioned that the behav-
ior of a superfluid near a surface, where the speed of sound went to zero
might mimic the behavior of space–time near to an event horizon. As it
happened, this idea was completely synergistic with a suggestion that had
been put forward by the author in 1991 that the same kind of nonlinear
Schrödinger equation used to describe the appearance of non-classical states
of photons in nonlinear optics might plausibly provide a quantum theory
of gravitational collapse [4]. In the fall of 2000, we showed with the help of
two of his students that the idea that the nonlinear Schrödinger equation
for a superfluid actually provides a rather detailed picture of what happens
to elementary particles approaching an event horizon [5]. Indeed, there is
every reason to expect that the compact astrophysical objects studied by
astronomers are actually the superfluid objects studied in Ref. [5] for which
the author proposed the name “dark energy stars”. A similar, but slightly
different model for compact objects, known as the “gravistar” model was
introduced at about the same time by Pawel Mazur and Emil Mottola [6].
The breakdown of general relativity near the surface of a dark energy star
is associated with the weakening of the long range off-diagonal order of
a superfluid when the speed of sound approaches zero. It might also be
www.ebook3000.com
Emergent Space–time 421
noted that the model for the failure of classical general relativity discussed
in Ref. [5] is completely consistent with the claim [7] that the equivalence
principle will fail near to an event horizon if the EPR correlations of vacuum
particles across the event horizon become very weak. Indeed, the vanishing
of off-diagonal order in a superfluid is directly related to the vanishing of
EPR-like correlations between particles in the superfluid.
The scalar nonlinear Schrödinger equation for a superfluid which was
used in Ref. [5] to describe the behavior of 3 + 1-dimensional space–time
near an event horizon may appear to be much too simple to be regarded
as a realistic quantum model for space–time However, in 1991, it had been
pointed out that a nonlinear Schrödinger equation in 2 + 1 dimensions
with an SU(N) Chern–Simons gauge field can provide an exact description
of self-dual Einstein spaces, where the phase of the wave-function is the
Kahler potential for the self-dual Einstein space [8]. In 1992, the author
pointed out [9] that this result could be extened to a quantum theory of
general Einstein spaces by reinterpreting the nonlinear Schrödinger equa-
tion as the hydrodynamic equation for a superfluid where the solitons of the
self-dual and anti-self-dual solutions of the classical Einstein equations are
paired. Our purpose here is generalize this idea to space times containing
matter.
Obviously, in order to provide a quantum description for space times
containing matter — e.g. the space times needed to describe gravitational
collapse — one needs to generalize the scalar wave-function which encodes
the Kahler potential of an Einstein space to a wave-function that is capa-
ble of representing not just pure gravitational but also matter degrees of
freedom (DOF). One might naively guess that one should simply attach
internal DOF representing matter DOF to the scalar wave-function used to
construct self-dual Einstein spaces. However, superstring models guide us to
the thought that a quantum theory of gravity and Yang–Mills interactions
might have its simplest expression in a higher dimensional space–time. In
particular, there are hints that D-string models in 10 + 2 dimensions can
explain the phenomenological degrees of the standard model of elementary
partiles when these DOF are reduced to 3 + 1 dimensions [10]. Unfortu-
nately, though, there is as yet no compelling evidence that the quantum
corrections to classical general relativity provided by superstring theories
mitigate the unphysical predictions of classical general relativity.
In this chapter, we wish to draw attention to the possibility that a matrix
nonlinear Schrödinger equation living on a 25 + 1-dimensional Lorentzian
extension of the 24-dimensional Leech lattice [11] provides a framework for
422 G. Chapline
a quantum theory of gravity and elementary particles that side-steps the

unphysical predictions of classical general relativity. Our proposal is based
on the idea that the two-dimensional (2D) space used in the construction
of Einstein spaces can be identified with a hexagonal lattice section of the
Leech lattice. This identification is motivated by the observation [12] that
the motion of a charged particle on a hexagonal lattice in the presence
of an electric field provides a realization of the parity anomaly in 2 + 1
dimensions if the hopping includes hopping between next nearest neighbors
as well as nearest neighbors. As shown in Ref. [9], the currents in this model
satisfy the Hall effect equation, which can be derived from the same type
of Chern–Simons effective action in 3 + 1 that we used in our construction
of Einstein spaces.
Our proposal, outlined in Sec. 3, is also motivated by the observation
[13] that the fields representing the massless DOF of different superstring
models in 9+1 dimensions are related by symmetries related to the symme-
tries of the Leech lattice. This observation suggests in turn that the vertices
of the 24-dimensional polytope consiting of the 196,560 lattice points of the
Leech lattice closest to a fixed point (hereafter referred to as the Leech
polytope) may perhaps represent massless DOF in a fundamental quan-
tum theory of gravity and elementary particles. In fact, the automorphism
group of the Leech lattice has orbits whose size matches the number of com-
ponents of a metric tensor, 2-form gauge field, E8 × E8 Yang–Mills gauge
fields, and a gravitino field, all in 24-dimensions. Remarkably, these same
types of fields also occur in the classical unification of supergravity and
super-Yang–Mills interactions in 10-dimensions [14].
Since the solutions of the 2D nonlinear Schrödinger equation that
were used to construct Einstein spaces are either holomorphic or anti-
holomorphic functions, we will assume that it is actually the complex Leech
lattice, which is a 12-dimensional lattice whose coordinates are Eisenstein
integers, rather than the real Leech lattice that is the natural setting for
our theory. The full 12-dimensional symmetry of complex Leech lattice will
be broken by our our use of a distinguished hexagonal lattice to repre-
sent the 2D used to define the nonlinear Schrödinger equation. In Sec. 3,
we will exhibit a generalization of the scalar nonlinear Schrödinger equa-
tion where the wave-function is no longer a scalar, but a block diagonal
11 × 11 complex matrix. The blocks represent “matter” carrying the U(1)
charges of a 1-form gauge field, as well as elementary particles carrying
flavor and color charges associated with SU(3) × E6 gauge fields. If a self-
duality condition is imposed on the nonlinear wave equation, then the zero
www.ebook3000.com
energy solutions of the nonlinear matrix equation will be holomorphic or

anti-holomorphic functions describing bosonic (or fermionic) solitons with
2-form and Yang–Mills DOF living in 2D. When holomorphic and anti-
holomorphic solutions are combined to provide a superfluid desciption for
2 + 1-dimensional space–time, the residual symmetry group for the internal
DOF will be the Mathieu group M11 . This exceptional permutation group
acting on set of 11 objects was discovered in the 19th century by Emil
Mathieu, and was the first in the series of discoveries of exceptional finite
groups culminating with the discovery of the Monster sporadic group in
1973 [15].
The M11 residual symmetry of our matrix wave-function has the prop-
erty that it singles out the gauge field-like DOF associated with the Leech
lattice. The appearance of a 2-form gauge field degree of freedom associ-
ated with Yang–Mills DOF is especially intriguing from the point of view
that a Chern-Simons-like Lagrangian involving a 2-form gauge field and
Yang–Mills gauge field strengths plays an important role in cancelling the
hexagon anomaly of Yang–Mills theories in 10-dimensions [16]. In fact, the
12-dimensional Chern–Simons-like Lagrangian that we introduce in Sec. 3
as the interaction Lagrangian for the gauge fields essentially combines the
three-dimensional (3D) parity violation Lagrangian with the 12-dimensional
hexagon anomaly cancellation anomaly Lagrangian.
When time is added to the nonlinear matrix equation, the wave-function
will depend not only on position within the distinguished hexagonal lattice,
but also time. However, as in our studies of Einstein spaces, we will initially
be mainly interesed in solutions which are time independent; i.e. solutions
correponding to the ground state. Remarkably, the numbers of internal
bosonic and fermionic DOF of our theory exactly match, even though the
corresponding fields are defined in 24 dimensions. This means that at the
level of 2 + 1 dimensions the contribution of zero point fluctuations of
bosonic or fermioniic superfluid modes to the ground state energy will van-
ish. However, when time and a 3rd spatial dimension are added to construct
a theory of space–time in 3 + 1 dimensions, the contibutions of the bosonic
and fermionic DOF to the ground state energy will not cancel each other.
Consequently, in our superfluid model, the vacuum energy will in general
not be zero. In Sec. 4, we comment on the possibility that our emergent
model for space–time not only provides an explanation for the observed
value of the cosmological constant the ground state energy, but also allows
for the appearance of a vacuum energy during gravitational collapse.
424 G. Chapline
2. Superfluid Model for Einstein Spaces

The coherent state wave-function for a 2D quantum fluid of anyons inter-
acting via both a point-like interaction and Chern–Simons gauge potentials
satisfies the nonlinear Schrödinger equation:
∂ψ 1 2
i =− D ψ + eA0 ψ − g|ψ|2 ψ, (1)
∂t 2m
where Dα = ∂α − i(e/c)Aα and m is a mass parameter. The gauge fields
A0 and Aα do not satisfy Maxwell’s equations, but instead are determined
self-consistently from the equations for Chern–Simons electrodynamics in
2+1 dimensions. In the presence of a uniform 2D electric field E, the current
has the same form as the Hall current for a magnetic field perpendicular to
the plane
jαβ = σH εαβγ Eγ , (2)
where σH is the “Hall conductivity”. Neglecting spatial variations in the
electric field, the usual Guass’ law will be replaced by the Chern–Simons
equation
e
B = − ρ, (3)
κ
where B is the strength of an effective magnetic field whose direction is
perpendicular to the layer, ρ is the charge per unit area, and 1/|κ| is an
inverse length with σH = κ. In the model where charged particles hop on
a lattice, the nonlinear term with coefficient g represents the effect of spin
orbit coupling. It was shown some time ago [17] that the time indepen-
dent version of Eq. (1) in conjunction with Eqs. (1) and (2) can be solved
analytically if one assumes that
g = ±e2 /mcκ. (4)
The ground state solution of Eq. (1) contains solitons with vortex-like cur-
rents and two units of quantized magnetic flux attached to every carrier.
The two signs for g correspond to solutions where the vorticity of all the
solitons is either up or down.
In 1991, the author and Kengo Yamagishi introduced the idea of extend-
ing Eqs. (1)–(3) spatial dimensions by replacing the 2D complex plane
with a stack of N such planes and the scalar wave-function with an SU(N )
matrix:
∂Φ 1 2
i =− D Φ + e[A0 , Φ] − g[[Φ∗ , Φ], Φ], (5)
∂t 2m
where the wave-function Φ and potentials A0 and Ai are now SU(N) matri-
ces, and D ≡ ∇ − i(e/c)[A. Promoting the scalar wave-function ψ(z)
www.ebook3000.com
to an SU(N) matrix wave-function Φ(z) is a way to take into account

into account inter-layer interactions and tunneling. The magnetic field
Beff = ∂x Ay − ∂y Ax + [Ax , Ay ] seen by charged carriers in 2D is now a
diagonal matrix
e
Beff = − [Φ∗ , Φ]. (6)
κ
The in-plane electric field Eα will also be a diagonal matrix satisfying the
Hall effect equation:
1
Eα = − εαβ jβ , (7)
κ
where jα = (/2mi)([Φ∗ , Dα Φ] − [Dα Φ∗ , Φ]) is the in-plane current. Time
independent analytic solutions to Eqs. (5)–(7) can be found for any value
of N if Eq. (4) is satisfied. These analytic solutions represent zero energy
ground states and satisfy the 2D self-duality condition Dα Φ = ±iεαβ Dβ Φ.
In the limit N → ∞, the analytic solutions of Eq. (5) take a particularly
simple form such that the effective magnetic field seen by the jth soliton
has the simple form:
c
Bj = ± ∇k |Xj − Xk |, (8)
e
k
where X ≡ (z, u) is now a 3D coordinate encoding both the position z =

x + iy of a soliton within a layer and the height u of the layer. In this
solution, the vortex-like solitons present in the solution for a single layer
have become monopole-like objects, which were christened “chirons” in Ref.
[9]. The ground state corresponding to Eq. (8) has zero energy and the
wave-function has the form [10]
∞
1/2
Rjk + Ujk
Ψ = f (w) , (9)
Rjk − Ujk
k>j
2 2
where Rjk = Ujk + 4(zj − zk )(z̄j − z̄k ), Ujk = uj − uk and f is an entire
function of the {z̄i } in the self-dual case and {zi } in the anti-self-dual case.
Writing the product on the r.h.s. of Eq. (9) as exp(S) defines an effective
action for a gas of chirons:
1 Rj + u − u j
S= ln , (10)
2 j Rj − u + uj
where Rj2 = (u − uj )2 + 4(z − zj )(z̄ − z̄j ). The wave-function (9) resem-

bles in some respects Laughlin’s wave-function for the fractional quantum
426 G. Chapline
Hall effect; for example, moving the z coordinate of a chiron around the
position of another chiron in a different layer changes S by iπ [9]. However,
in contrast with the fractional quantum Hall effect, there are two distinct
degenerate ground states corresponding to the self-dual and anti-self-dual
solutions for Eq. (5), reminiscent of the Kramers pairs in systems with time
reversal symmetry. It was the motivation for the suggestion in Ref. [9] that
these two solutions can be combined to yield a model for empty space–time.
Actually, the effective action (10) for chirons already suggests a connec-
tion with the Kosterlitz–Thouless condensation of vortex and anti-vortex
pairs in the 2D XY-model [18]. It is an elementary identity that the right
hand side of Eq. (9) can be rewritten in the form of

−1 u − ui
S= ± tanh (12)
i
Ri
which is similar in form to a configuration of 2D XY vortices. In the XY

model, the phase variations in a 2D condensate can be described by a
partition function of the form
2π
K 2 ∂Θ ∂Θ
Z= DΘ exp − d ξ , (13)
2 ∂ξi ∂ξi
0
where Θ is a periodic coordinate whose period is 2π and K is a constant.

It can be shown that a discrete version of this theory interpolates between
the low and high temperature phases of the XY model. Indeed, evaluating
the exponential in (13) for a configuration of vortices yields the partition
function for a 2D Coulomb gas. On the other hand, substituting the chiron
effective action (12) into the exponential in (13) yields:
 
R ij
exp −πK  mi mj ln . (14)
|Zi − Zj |
i=j
Expression (14) illustrates why flat space–time might be viewed as a con-

densation of self-dual and anti-self-dual chirons. Although the pairing of
self-dual and anti-self-dual solitons of the Einstein equations must classicaly
be defined using two seprate 2D spaces (the “ambi-twister” construction of
Einstein spaces), it is also perfectly acceptable to regard (14) as defining
the phase of a quantum state consisting of a superpositiion of coherent sates
corresponding to self-dual and anti-self-dual solutions of Eq. (5).
www.ebook3000.com
3. Extension to Include Matter

In this section, we will outline how the nonlinear Schrödinger equation the-
ory of 3 + 1-dimensional Einstein spaces just described might be extended
to a description of space times with matter and/or vacuum fields. Our basic
idea is to replace the 2D plane used in the construction of self-dual Einstein
spaces with a certain 2D section of the 24-dimensional Leech lattice. The
Leech lattice was originally constructed using the 24-dimensional Golay
error correcting code [19]. The lattice points nearest to a fixed point of
the lattice define the Leech polytrope, a remarkable regular 24-dimensional
polytrope with 196,560 vertices. These vertices come in three different
“shapes” (±(−3), ±123 ), (±42 , 022 ), or (±28 , 016 ), where (±(−3), ±123 )
means that the coordinates of these vertices have a −3 or +3 as any two
of the 24 coordinates and +1 or −1 as the other coordinates. The other
shapes have similar interpretations. The automorphism group of the Leech
lattice, 0•, is transitive on all the vertices of the Leech polytrope, but has
a subgroup, 212 M24 , that preserves the shapes of the Leech polytrope ver-
tices, where the Mathieu group M24 is automorphism group of the Golay
code, while 212 is a group of special involutions of the Leech lattice.
From the point of view of constructing a quantum theory of space–time
containing matter perhaps the most interesting property of the Leech lat-
tice is that there is a correspondence between the shapes of the minimal
Leech lattice vectors and the types of fields that occur in the 10-dimensional
supergravity and super-Yang–Mills field theories. For example, the number
of minimal vectors with shape (42 , 022 ) is equal to 4× the number of com-
ponents of a 2-form field Bµν in 24 dimensions, while the vectors with
shape (−3, 123 ) represent 2× the number of components of a gravitino
field in 24 dimensions. A rather amazing fact is that if the Leech lattice
generators with shape (8, 023 ) — which are not vertices of the Leech poly-
tope — are added to the polytrope vertices, then the numbers of bosonic and
femionic DOF are exactly equal! Indeed, the full automorphism group of
the Leech lattice 0• includes supersymmetry-like transformations between
vectors representing bosonic and fermionic DOF. Thus, our Leech lattice
set-up parallels the unification of supergravity and super-Yang–Mills inter-
actions in 10 dimensions [15], with the exception that the gravitational DOF
associated with 22 of the 24 dimensions of the Leech lattice will be absent.
The wave-function for our model of 3 + 1-dimensional space–time lives
will initially be defined on a 2D section of Leech lattice in essentially the
same way that our wave-function for the ground state of an Einstein space
428 G. Chapline
got its start as a function defined on the 2D complex plane. Furthermore,

as in our theory of Einstein space, holomorphic and anti-holomorphic func-
tions play a central role in our theory of space–time with matter. Therefore,
the most natural setting for our theory is actually the complex Leech lat-
tice rather than the real Leech lattice, with the distinguished hexagonal
lattice representing a fixed direction of the complex Leech lattice. If the
orientation of a 2D section of the real Leech lattice is considered fixed,
then the automorphism group 0• is reduced to a subgroup 210 M22 , where
210 is a group of special reflections and M22 is a subgroup of the Golay
code automorphism group M24 . In the case of the complex Leech lattice
the subgroup of 0• which preserves the complex structure is 6.Suz, and the
subgroup which fixes one complex direction is 2 × 36 M11 This group plays
much the same role in our theory as the Weyl group for a Lie group root
lattice.
In the context of the Leech lattice, the wave-function is block diagonal
matrix, where the blocks represent internal DOF for the quantum fluid.
Following the usual prescription for an equation describing the hydrody-
namic behavior of a quantum fluid [20], the Hamiltonian for a 2D quantum
fluid with a matrix valued wave-function will the form:

1
H= d2 x (Di Ψ)∗ (Di Ψ) + g([Ψ↑ , Ψ])2 , (15)
2m
where i = 1, 2 and the Ψ is an 11 × 11 block diagonal complex matrix.
One 8 × 8 block represents fields belonging to the adjoint representa-
tion of (SU(3)×E6 )×(SU(3)×E6 ). As was the case for the scalar nonlinear
Schrödinger equation, the covariant derivative has the form
Dα = ∂α + e[Aα , (16)
where Aα is a diagonal matrix obtained by restricting one of the compo-
nents of the Leech lattice 2-form gauge field to lie in the 11 dimensions
orthogonal to the distinguished 2D section. As was the case for the nonlin-
ear Schrödinger theory of Einstein spaces, the gauge field Aµ that enters
Eq. (16) is not an independent variable, but will be entirely determined
by the matrix wave-function. Our model follows exacly Einstein spaces by
insisting that the dynamics of the gauge field Aµ be confined to the 3 + 1
dimensions of ordinary space–time. We are thus led to the use of a topo-
logical interaction Lagrangian for Aµ that distinguishes between ordinary
space–time and the 11 dimensions of the Leech lattice perpendicular to
the distinguished hexagonal lattice. Although there may be other possi-
ble choices for this interaction Lagrangian, the M11 symmetry between the
www.ebook3000.com
2-form gauge fields Aµ and the Yang–Mills gauge fields Aaµ makes the fol-
lowing choice for the Lagrangian seem most natural:
i1
LCS = −κεαβγi1 ...i9 Aiα1 Hβγ T r (Fi2 i3 Fi4 i5 Fi6 i7 Fi8 i9 ), (17)
where Hµν is the field strength for Aµ and F is the field strength for the
Aaµ . Varying this expression together with the terms obtained from the
Hamiltonian (16) with respect to Aµi leads to the following constituent
equations for the gauge fields:
ie ↑
H ∧ T rF ∧ F ∧ F ∧ F = κ [Ψ , Ψ]
1 ij (18)
T rF ∧ F ∧ F ∧ E i = ε [Dj Ψ, Ψ↑ ] − [Ψ, Dj Ψ↑ ] ,
2κ
where E and F are electric and magnetic fields associated with the SU(3)
× E6 gauge fields. It follows from Eq. (18) that the 2-form and Yang–Mills
vacuum fields will be strongly influenced by the dynamics of space–time.
Because the self-duality condition analogous to Eq. (4) will depend on the
vacuum configuration of gauge field strengths, it follows that the vacuum
energy cannot be zero everywhere if the vacuum configuration of fields
changes with time or location.
The Chern–Simons-like Lagrangian (17) is of course related to a
Pontrijagin-like form in 13 dimensions. Actually, the author anticipated
some time ago [21] that a fundamental theory of gravity and elementary
particles might involve a Pontrijagin-like form in 13 dimensions that is
closely related to the exterior derivative of 17. We also note that it was
Richard Slansky [22] who originally suggested that the Yang–Mills gauge
symmetry in a grand unified theory of elementary particles might be E6 .
4. Gravitational Collapse Solved?

We will leave for the future the question as to how our model for space–
time with matter compares with the predictions of general relativity. We
would like to point out, though, that our theory of space–time may not
only explain the origin and observed magnitude of the cosmological con-
stant, but also explain why the formation of black holes is avoided during
gravitational collapse. In particular, because the vacuum energy in our the-
ory of space–time can depend on the dynamical behavior of space–time
the continuous gravitational collapse of large masses predicted by classical
general relativity can be avoided.
430 G. Chapline
As was emphasized by Lemaitre in several seminal papers [23], the

appearance of a vacuum energy can prevent a collapsing object from collaps-
ing to a singularity. Indeed, at the classical level this is the only mechanism
known for preventing continuous collapse to a singularity. Of course, this
begs the question as to the origin of the vacuum energy that will prevent
singularities from forming. Recently, it has been suggested [24] that dur-
ing the gravitational collapse of ordinary matter a vacuum energy will be
created when the baryons in the collapsing matter encounter a trapped
surface. Analytic studies [2] suggest that a trapped surface begins to form
at the center of the collapsing object a short time before the outer radius
of the collapsing mass reaches the gravitational radius, and then rapidly
expands until it reaches the surface of the collapsing object. We believe
that changes in the quantum vacuum state associated with the expansion
of an incipient trapped surface is what is responsible for the termination of
continuous gravitational collapse.
The author and Nick Manton pointed out some time ago [25] that a
“geometric” Higgs potential could appear in 3D as a result of a topologi-
cally non-trivial configuration of vacuum gauge fields in extra dimensions.
However, the vacuum energy density associated with such a Higgs potential
would be very large unless the characteristic size of the extra dimensions
was macroscopic, which would contradicts laboratory observations that rule
out large extra dimensions. On the other hand, in our superfluid model for
space–time a change in the vacuum configuration of gauge field strengths
will in general give rise to a positive vacuum energy. The value of this energy
density would depend on the size of the 3D space. If this size is determined
by the position of a trapped surface, then explanations are at hand for both
the observed magnitude of the cosmological constant and the appearance
of a vacuum energy during gravitational collapse.
In conclusion, we believe that the equations proposed in Sec. 3 fill the
void that arises from the fact that classical general relativity cannot explain
under what circumstances a vacuum energy may appear, because general
relativity does not take into account the quantum nature of the vacuum
state.
Acknowledgments
That space–time might be described as a superfluid was independently
proposed in the 1990s by Pawel Mazur, and the author is very grateful
for many enlightening discussions with him. The author is also grateful for
www.ebook3000.com
discussions with Emil Mottola, Neal Snyderman, Samuel Braunstein, Jim

Barbieri, and Manosori Hanada.
References
1. P. Joshi, Gravitational Collapse and Spacetime Singularities, Cambridge Uni-
versity Press, Cambridge, 2007.
2. D. Boulware, Phys. Rev. D 11, 1404 (1975).
3. E. Mottola, Acta Physica Polonica B 41, 2031 (2010).
4. G. Chapline in Proc. Santa Fe Conference on the Foundations of Quantum
Mechanics, eds. T. Black et al. World Scientific, Singapore 1992.
5. G. Chapline, E. Hohlfeld, R. Laughlin, and D. Santiago, Phil. Mag. B 81,
235 (2001).
6. P. Mazur and E. Mottola, Proc. Nat. Acad. Sci. 111, 9545 (2004).
7. S. L. Braunstein, S. Pirandola, and K. Zyczkowski, Phys. Rev. Lett. 110,
101301 (2013).
8. G. Chapline and K. Yamagishi, Phys. Rev. Lett. 66, 3046 (1991).
9. G. Chapline, Mod. Phys. Lett. A 7, 1959 (1992); Proc. XXI International
Conference on Differential Geometric Methods in Theoretical Physics, eds.
C.N. Yang, M.L. Gee, and X.W. Zhou, World Scientific, Singapore, 1993.
10. C. Vafa, arXiv; hep-th/9602022v1.
11. J.H. Conway and N.J.A. Sloane, Sphere Packings, Lattices, and Groups
Springer-Verlag, New York, 1988.
12. F.D.M. Haldane, Phys. Rev Lett. 61, 2015 (1988).
13. G. Chapline, Phys. Lett. B 158, 393 (1985).
14. G, Chapline and N. Manton, Phys. Lett. 120B, 105 (1983).
15. D. Gorenstein, Finite Simple Groups, Plenum Press, New York, 1982.
16. M.B. Green and J.H. Schwarz, Phys. Lett. 149B, 117 (1984).
17. R. Jackiw and S.-Y. Pi, Phys. Rev. Lett. 64, 3230 (1990).
18. G. Chapline, Phil. Mag. B 86, 1201 (2006).
19. J. Leech, Canad. J. Math. 19, 251 (1967).
20. L. P. Pitaevskii, Sov. Phys. JETP 13, 451 (1961).
21. G. Chapline, Chaos Solitons Fract. 10, 311 (1999).
22. R. Slansky, Phys. Rep. 79 (1981).
23. G. Lemaitre, Gen. Relat. Gravit. 29, 641 (1997).
24. G. Chapline and J. Barbieri, Intl. J. Mod. Phys. D 23, 1450025 (2014).
25. G. Chapline and N. Manton, Nucl. Phys. B 184, 391 (1981).
www.ebook3000.com
Chapter 16
The Idea of a Stochastic Space–Time:

Theory and Experiments
Maurizio Consoli
INFN, Sezione di Catania, Via S. Sofia 64, I-95123 Catania, Italy
maurizio.consoli@ct.infn.it
Alessandro Pluchino
Dipartimento di Fisica e Astronomia, Università di Catania, and
Sezione INFN di Catania, Via S. Sofia 64, I-95123 Catania, Italy
alessandro.pluchino@ct.infn.it
Basic foundational aspects of both quantum physics and relativity sug-

gest that space–time may have the fundamental stochastic nature of a
turbulent fluid. After reviewing the basic theoretical motivations, we
have compared this picture with the phenomenological pattern observed
in the ether-drift experiments. To this end, we have performed numerical
simulations in which the parameters of the macroscopic Earth’s cosmic
motion are only used to fix the limiting boundaries for a microscopic
velocity field which has instead an intrinsic stochastic nature. In this
framework, both classical and modern experiments become consistent
with the type of cosmic Earth’s motion which today is used to describe
the CMB anisotropy. The need for confirmations with a new generation
of dedicated experiments is finally emphasized.
1. Introduction
There are two radically different ways to look at the origin of symmetries.
On the one hand, for esthetic reasons, a symmetry could simply be postu-
lated from scratch as, for instance, in the grand-unified scenarios of elemen-
tary particle physics. On the other hand, one could consider a symmetry as
an emergent phenomenon [1]. From this latter point of view, the symmetry
emerges from a microscopic description that, at the deepest level, does not
know about its existence. In this sense, the emergence of symmetries could
433
434 M. Consoli and A. Pluchino
also be viewed as the tendency of physical systems toward self-organization

and complexity [2].
As a definite example, one can consider the case of electromagnetism
and Lorentz symmetry. At the end of 19th century, electromagnetic waves
were described as hydrodynamic disturbances of an underlying ether repre-
sented, by Thomson, Fitzgerald and others, as an incompressible turbulent
fluid (a vortex ‘sponge’) [3]. The main point was that, due to the energy
which is locally stored in the turbulent motion, on a coarse-grained scale,
a fluid can start to behave as an elastic medium and thus support the
propagation of transverse waves whose speed cγ coincides with the average
speed c ≡ cturbulence of the chaotic internal motion of the elementary fluid
constituents.
With the advent of Einstein’s axiomatic approach, the ether started to
be considered a superfluous concept. Still, as we shall review in Sec. 2, the
concept of the vacuum as an underlying turbulent ether is re-proposed by
formal analogies with some foundational aspects of both quantum theory
and relativity. This leads to the idea that space–time may have a funda-
mental stochastic nature.
Therefore, one may ask: Beyond the simple level of a formal analogy,
could there be some definite experimental signature for this type of picture?
This possibility will be considered in Secs. 3–7, where we shall compare with
the phenomenological aspects of the ether-drift experiments. In the frame-
work of a Lorentzian form of relativity and by representing the physical
vacuum as a stochastic medium, our numerical simulations indicate that
all classical ether-drift experiments could become consistent with the aver-
age Earth’s motion which today is used to characterize the anisotropy of
the Cosmic Microwave Background (CMB). Finally, in Sec. 8, the overall
consistency of this view with the present ether-drift experiments and the
need for a new generation of dedicated experiments will also be emphasized.
2. The Physical Vacuum as a form of Turbulent Ether

In this section, we shall list a few different motivations that might induce
to represent the vacuum as a form of random medium which resembles a
turbulent fluid.
(a) At the dawn of 20th century, Lorentz symmetry was believed to
emerge from an underlying ether represented, by Thomson, Fitzgerald and
others, as an incompressible turbulent fluid (a vortex ‘sponge’) [3]. More
recently, the turbulent-ether model has been re-formulated by Troshkin [4]
www.ebook3000.com
The Idea of a Stochastic Space–Time: Theory and Experiments 435
(see also Refs. [5, 6]) in the framework of the Navier–Stokes equation. The
main point of these hydrodynamic derivations is that, due to the energy
which is locally stored in the turbulent motion, on a coarse-grained scale,
a fluid can start to behave as an elastic medium and thus support the
propagation of transverse waves whose speed cγ coincides with the average
speed c ≡ cturbulence of the chaotic internal motion of the elementary fluid
constituents.
To understand intuitively why, on coarse-grained scale, a fluid can start
to behave as a solid, one can just think of jets of water of sufficient speed.
However, this idea is also supported by the formal equivalence [7, 8] (veloc-
ity potential versus displacement, velocity versus distortion, vorticity versus
density of dislocations, . . . ) that can be established between various systems
of dislocations in an elastic solid and corresponding vortex fields in a liquid.
In this sense, the phenomenon of turbulence provides a conceptual transi-
tion from fluid dynamics to a different realm of physics, that of elasticity.
With this transition, the parameter c acquires also the meaning of a lim-
iting speed for the motion of soliton-like dislocations taken as models of
ordinary matter (see e.g. Refs. [9, 10] and references quoted therein). This
is due to the behavior of their elastic energy which increases proportionally
to (1 − v 2 /c2 )−1/2 .
This perspective is similar to starting from the basic equation that deter-
mines the mutual variations of the energy E and the linear momentum
p = M v of a body
dE d(M v)
=v· (1)
dt dt
2
and allowing for a v -dependence in M (see e.g. [11]). This gives
1
dE = M dv 2 + v 2 dM. (2)
2
The main point is that, if ordinary matter were interpreted in terms of
soliton-like excitations of an underlying turbulent ether, one now disposes
of the velocity parameter c ≡ cturbulence. Then, by setting E ≡ c2 M (v 2 /c2 ),
dE 2 dM dM 2 2 1
one has dv 2 = c dv 2 and Eq. (2) becomes dv 2 (c − v ) = 2 M Therefore,
2
for dM/dv > 0, c plays also the role of a limiting speed and one finally
obtains
M 0 c2
E = M c2 = . (3)
1 − v 2 /c2
On this basis, it becomes natural to introduce linear transformations of the
four quantities E/c and p = M v that preserve the quadratic combination
(E/c)2 − p2 = (M0 c)2 and thus, by starting from a microscopic turbulent-

ether scenario, Lorentz symmetry could also be understood as an emergent
phenomenon. In this interpretation, its ultimate origin has to be searched
in the very existence of c and thus in the deepest random fluctuations of
the fluid velocity, with time at each point and between different points at
the same instant that characterize a state of fully developed turbulence and
provide a kinetic basis for the observed space–time symmetry [12].
Note that once Lorentz symmetry is an emergent property, c is only a
limiting speed for those soliton-like, collective modes that, in an emergent
interpretation, are taken as models of ordinary matter, e.g. vortices, elastic
dislocations . . . Thus, there is nothing wrong if the internal motion of the
basic constituents takes place at an average speed c. At the same time, on
the coarse grained scale which is accessible to physical rods and clocks, the
basic constituents appear, so to speak, ‘frozen’ in the vacuum structure and
only their collective excitations are directly observable. This means that for
the elementary ether constituents, Eq. (1) is now solved by the standard
non-relativistic forms E = 12 mv 2 and p = mv, where m is the constituent
constant mass.
(b) As emphasized in Ref. [13], this qualitative picture of the vacuum,

as an underlying random medium, also arises from alternative views of the
quantum phenomena as with stochastic electrodynamics [14–18] or Nelson’s
mechanics [19]. The former is essentially the classical Lorentz-Dirac theory
with new boundary conditions, where the standard vanishing field at infin-
ity is replaced by a vacuum, random radiation field. This field, considered
in a stationary state, is assumed to permeate all space and its action on the
particles impresses upon them a stochastic motion with an intensity char-
acterized by Planck’s constant. In this way, one can get insight into basic
aspects of the quantum theory such as the wave-like properties of matter,
indeterminacy, quantization, . . . For instance, in this picture, atomic sta-
bility would originate from reaching that ‘quantum regime’ [16, 18] which
corresponds to a dynamic equilibrium between the radiation emitted in the
orbital motions and the energy absorbed in the highly irregular motions
impressed by the vacuum stochastic field. In this sense, again, Lorentz ether
should not be thought as a stagnant fluid (for an observer at rest) or as
a fluid in laminar motion (for an observer in uniform motion). Rather the
ether should resemble a fluid in a chaotic state, e.g. a fluid in a state of tur-
bulent motion. The same is true for Nelson’s mechanics. Here, the idea of a
highly turbulent fluid emerges if one uses Onsager’s original result [20] that
www.ebook3000.com
in the zero-viscosity limit, i.e. infinite Reynolds number, the fluid velocity
field does not remain a differentiable function.a This provides a basis to
expect that “the Brownian motion in the ether will not be smooth” [19]
and thus to consider the particular form of kinematics which is at the basis
of Nelson’s stochastic derivation of the Schrödinger equation.
(c) At a more elaborate level, a qualitatively similar picture is also

obtained by representing relativistic particle propagation from the super-
position, at very short time scales, of non-relativistic particle paths with
different Newtonian mass [22]. In this formulation, particles randomly prop-
agate (in the sense of Brownian motion) in a granular medium which thus
replaces the trivial empty vacuum [23]. The essential mathematical ingredi-
ent for this representation is the use of ‘superstatistics’ [24], intended as the
superposition of several statistical systems that operate at different spatio-
temporal scale, which is also known to provide a very good description of
fluid particle trajectories in high Reynolds-number turbulence [25].
(d) Finally, the idea of a fundamentally random vacuum is also moti-

vated by quantum-gravity. According to this view, space–time, when
resolved at very short distances, should exhibit quantum fluctuations and
thus appear to be ‘foamy’ or ‘spongy’ in the sense of Refs. [26,27]. This origi-
nal idea has lead to a very wide collection of ideas and intuitions including,
for instance, the holographic principle (see Ref. [28] for a review), possi-
ble deformations of Lorentz symmetry (Doubly Special Relativity) [29] or
models of dark energy and dark matter [30]. At the same time, coupling
light and matter to a fluctuating metric leads to intrinsic limitations on
the measurement of lengths [31, 32], to violations of the weak equivalence
principle [33] and to an effective decoherence of quantum systems [34].
These effects can be used to restrict the possible quantum gravity mod-
els by comparing with the results of modern gravity-wave detectors [35] or
with atomic interferometry [36] or with the beat signal of two ultrastable
optical resonators [37]. What is relevant here for our purpose is that, as in
the previous cases, the space–time foam of quantum gravity seems also to
resemble a turbulent fluid. This idea, originally due to Wheeler [26], has
been more recently exploited by Ng and collaborators [38] who have empha-
sized the close analogies between holographic models of space–time foam
a Onsager’s argument relies on the impossibility, in the zero-viscosity limit, to satisfy
the inequality |v(x + l) − v(x)| < (const.)ln , with n > 1/3. Kolmogorov’s theory [21]
corresponds to n = 1/3.
and the limit of turbulence for infinite Reynolds number. The main conclu-
sion of these rather formal derivations is that the metric fluctuations in the
2/3
holographic model, which give rise to length fluctuations ∆l ∼ l1/3 lplanck ,
when compared with those in moving fluids, can also be interpreted as a
manifestation of Kolmogorov’s scaling law for velocity ∆v ∼ l1/3 [21].
Thus, summarizing from the old ether view to the present quantum-
gravity models, there are several independent motivations to represent the
physical vacuum as an underlying turbulent fluid. This non-trivial degree
of convergence might originate from the fundamental nature of quantum
gravity (e.g. from the correspondence between the metric fluctuations in
the holographic model and Kolmogorov’s scaling law). However, one could
also adopt the complementary point of view where instead the ubiquitous
phenomenon of turbulence plays from the very beginning the most cen-
tral role. In any case, it becomes natural to wonder whether this type
of vacuum medium could represent the preferred reference frame of a
Lorentzian approach and thus to look at the ether-drift experiments for
possible experimental checks. At the same time, the non-trivial interplay
between large-scale and small-scale properties of turbulent flows may induce
one to re-consider some assumptions adopted so far in the interpretation of
the data. These issues will be analyzed in detail in the following sections.
3. Ether-Drift Experiments, the Velocity of Light

and the Lorentz Invariance of the Vacuum
Ether-drift experiments, where one attempts to measure an absolute veloc-
ity, are the only known experiments which, in principle, can distinguish
Einstein’s special relativity from the Lorentzian point of view with a pre-
ferred reference frame Σ. At the same time, by assuming the validity of
Lorentz transformations, if the velocity of light cγ propagating in the var-
ious interferometers coincides with the basic parameter c entering Lorentz
transformations, relativistic effects conspire to make undetectable the veloc-
ity parameter V associated with the motion of a given frame S with respect
to Σ. Therefore the only possibility is that cγ and c do not coincide exactly.
In this case, in fact, the existence of a small mismatch would show up
through a tiny anisotropy of the velocity of light, proportional to (c − cγ )/c,
which could be measured by rotating a Michelson interferometer.
To derive the relevant relation, we shall follow the same treatment given
in Ref. [39] which applies to light propagation in a dielectric medium when
the refractive index N = 1 + is extremely close to unity. This is the case of
www.ebook3000.com
the gaseous systems as air, helium,. . . ,which were used in the classical ether-
drift experiments (e.g. Michelson–Morley, Miller, Illingworth, Joos, . . .). For
such systems, one can find a simple theoretical framework to analyze the
experiments.
The standard assumption is that any anisotropy has to vanish when both
the observer and the container of the medium are at rest in the hypothetical
preferred frame Σ. Therefore, in the physical case where instead both the
observer and the container of the medium are at rest in the laboratory S
frame, the anisotropy should vanish identically in the two limits when either
V = 0 (i.e. S ≡ Σ) or N = 1 (i.e. when cγ ≡ c). This means that in a
power series expansion in the two small parameters β = V /c and = N − 1,
any possible anisotropy has to start to O(β) for the one-way velocity cγ (θ)
and to O(β 2 ) for the two-way velocity c̄γ (θ) (the only one that can be
measured unambiguously) which, by its very definition, is invariant under
the replacement β → −β. At the same time, for any fixed β, c̄γ (θ) is also
invariant under the replacement θ → π + θ. Therefore, to the lowest non-
trivial level O(β 2 ), one can write down the general expression
∞

2cγ (θ)cγ (π + θ) c
c̄γ (θ) = ∼ 1 − β2 ζ2n P2n (cos θ) , (4)
cγ (θ) + cγ (π + θ) N n=0
where to take into account invariance under θ → π + θ, the angular depen-

dence is given as an infinite expansion of even-order Legendre polynomials
with arbitrary coefficients ζ2n = O(1). In Einstein’s relativity, where there
is no preferred reference frame, these ζ2n coefficients vanish exactly. In a
Lorentzian relativity, consistently with Lorentz’s point of view [40] “. . . it
seems natural not to assume at starting that it can never make any dif-
ference whether a body moves through the ether or not . . . ”, there is no
reason why they should vanish a priori. Therefore, one can adopt Eq. (4)
and start to compare with experiments.
However, before analyzing its phenomenological implications, it is inter-
esting to look for a possible dynamical mechanism which can explain the
formal structure in Eq. (4). To this end, by following Refs. [39, 41], one
can explore the possible implications of those modern views where the vac-
uum state is usually represented (e.g. in the standard model) as originating
from the macroscopic condensation of some elementary quanta in the same
quantum state, say k = 0 in some reference frame Σ. This characterizes the
physically realized form of relativity and could play the role of preferred
reference frame in a modern Lorentzian approach.
This possibility is usually not considered with the motivation, perhaps,

that the average properties of the condensed phase are summarized into a
single quantity which transforms as a world scalar under the Lorentz group,
for instance, in the standard model, the vacuum expectation value Φ of the
Higgs field. However, this does not necessarily imply that the vacuum state
itself has to be Lorentz invariant. Namely, Lorentz transformation operators
U , U , . . . could transform non-trivially the reference vacuum stateb |Ψ(0)
(appropriate to an observer at rest in Σ) into |Ψ , |Ψ , . . . (appropriate to
moving observers S , S , . . .) and still, for any Lorentz-invariant operator
G, one would find
GΨ(0) = GΨ = GΨ = · · · . (5)
Now, according to general quantum field theoretical arguments, decid-

ing on its Lorentz invariance requires to consider the eigenvalues and the
algebra of the global Poincaré operators Pα , Mα,β (α, β = 0, 1, 2, 3), where
Pα are the 4 generators of the space–time translations and Mαβ = −Mβα
are the 6 generators of the Lorentzian rotations. In this framework, exact
Lorentz invariance of the vacuum requires to impose [39,41] the problematic
condition of a vanishing vacuum energy E0 = 0. However, for interacting
theories, with the exception of unbroken supersymmetric theories (which
are not phenomenologically acceptable), there is no known way to ensure
consistently the condition E0 = 0. Thus, the issue of an exact Lorentz invari-
ant vacuum remains as an open problem which, at present, cannot be solved
on purely theoretical grounds. Still, one can explore the possible observable
implications. In fact, the simplest consequence of such non-invariance of the
vacuum is an energy–momentum flow along the direction of motion with
respect to Σ. This tiny flow, acting as an effective thermal gradient, could
induce small convective currents of the molecules in weakly bound systems
as gases. In this case, refracted light would exhibit a slight anisotropy which
would produce exactly the same Eq. (4) [39].
In this scheme, one can also understand the difference [42, 43] with
experiments performed in strongly bound systems, such as solid or liq-
uid transparent media, as in the Shamir–Fox experiment [44]. Being aware
that the classical experiments might also admit a non-null interpretation
b We ignore here the problem of vacuum degeneracy by assuming that any overlapping
among equivalent vacua vanishes in the infinite-volume limit of quantum field theory (see
e.g. S. Weinberg, The Quantum Theory of Fields, Vol. II, Cambridge University press,
pp. 163–167).
www.ebook3000.com
proportional to (N − 1)β 2 , they selected a medium where the effect of the

refractive index could have been enhanced (i.e. perspex, where N ∼ 1.5).
Since this enhancement was not observed, they concluded that the exper-
imental basis of Special Relativity was strengthened. However, with the
proposed mechanism in solid and liquid dielectrics, one expects the small
energy flow generated by the motion with respect to the vacuum condensate
to mainly dissipate by heat conduction without generating any appreciable
particle motion or light anisotropy in the rest frame of the medium. Thus,
one has a physical argument to reconcile the two different behaviors.
The above non-trivial level of consistency motivates a new generation
of precise ether-drift experiments, where light propagates in weakly bound
gaseous media which seem to be the best suited to detect the tiny energy-
momentum flow associated with a Lorentz non-invariant vacuum state.
In this respect, we observe that Eq. (4), in principle, is exact to the given
accuracy and could be used for fits to the data, where the first few ζ’s are
left as free parameters. This general structure can, however, be compared
with the particular form (a pure second harmonic in θ) which is obtained
by using Lorentz transformations to connect S to the preferred framec
c
c̄γ (θ) ∼ [1 − β 2 (a + b sin2 θ)] (6)
N
with a = 2 and b = −1 which corresponds to setting ζ0 = 4/3, ζ2 = 2/3
and all ζ2n = 0 for n > 1 in Eq. (4). We can then define the anisotropy
parameter B
c̄γ (π/2 + θ) − c̄γ (θ) v2

∼ B 2 cos 2(θ − θ0 ), (7)
c̄γ c
where the pair (v, θ0 ) describes the projection of V onto the relevant plane
and
|B| ∼ . (8)
Equation (6) represents a definite realization of the general structure in

(4) and a particular case of the Robertson–Mansouri–Sexl (RMS) scheme
[45, 46] for anisotropy parameter |B| = . In this sense, it provides a partial
c We address the reader to Ref. [39] for various details concerning the derivation of Eq. (6)
(see, in particular, Appendix A) or the exact relation between the value of the refractive
index in the S frame and its value when the container of the gas is at rest in the Σ
frame.
answer to the problems posed by our limited knowledge of the electromag-

netic properties of gaseous systems and will be adopted in the following as
a tentative model for the two-way velocity of light.
To obtain an experimental check, let us adopt Eq. (6). Then, this
anisotropy of the two-way velocity of light could be measured by rotating a
Michelson interferometer. By assuming the validity of Lorentz transforma-
tions in the rest frame S of the apparatus, the length L of its arms does
not depend on their orientation so that the interference pattern between
two orthogonal beams of light depends on the time difference
2L 2L
∆T (θ) = − . (9)
c̄γ (θ) c̄γ (π/2 + θ)
In this way, by introducing the wavelength λ of the light source and the
projection v of the relative velocity in the plane of the interferometer, one
2
finds to order vc2 the fringe shift
2
∆λ(θ) c∆T (θ) L vobs
∼ ∼ cos 2(θ − θ0 ). (10)
λ Nλ λ c2
In the above equation, the angle θ0 = θ0 (t) indicates the apparent direction
of the ether-drift in the plane of the interferometer (the “azimuth”) and
the square of the observable velocity
2
vobs ∼ 2(N − 1)v 2 (11)
is re-scaled by the tiny factor 2(N − 1) with respect to the true kinematical
velocity v 2 (t).
Therefore, in this scheme, the interpretation of the experiments is trans-
parent. According to Special Relativity, there can be no fringe shift upon
rotation of the interferometer. In fact, if light propagates in a medium, the
frame of isotropic propagation is always assumed to coincide with the lab-
oratory frame S, where the container of the medium is at rest, and thus
one has vobs = v = 0. On the other hand, if there were fringe shifts, one
could try to deduce the existence of a preferred frame Σ = S provided the
following minimal requirements are fulfilled: (i) the fringe shifts exhibit an
angular dependence of the type in Eq. (10), (ii) by using gaseous media
with different refractive index one gets consistency with Eq. (11) in such a
way that different vobs correspond to the same kinematical v.
www.ebook3000.com
4. A Fresh Look at the Classical Ether-Drift

Experiments
Before considering the classical ether-drift experiments, some introductory
discussion is needed. These experiments were performed in a period when
both relativity and quantum theory were not fully developed. Therefore,
the theoretical model adopted to compare with the data was basically the
old classical physics. In this interpretative scheme, the expected effects,
although being formally O(v 2 /c2 ), were “large”, as compared to the extraor-
dinary sensitivity of the Michelson interferometer, and “smooth”, because
the only time dependence was due to slow effects such as the daily Earth’s
rotation and its annual orbital revolution.
To see this, let us first re-write the basic Eq. (10) as
∆λ(θ) 2L(N − 1) v 2 (t)
∼ cos 2(θ − θ0 (t)) ≡ 2C(t) cos 2θ + 2S(t) sin 2θ,
λ λ c2
(12)
where (x–y denotes the plane of the interferometer)
L(N − 1) v 2 (t) L(N − 1) vx2 (t) − vy2 (t)
C(t) = cos 2θ 0 (t) = , (13)
λ c2 λ c2
L(N − 1) v 2 (t) L(N − 1) 2vx (t)vy (t)
S(t) = sin 2θ0 (t) = . (14)
λ c2 λ c2
Then, the standard classical assumption is to consider a cosmic Earth’s
velocity with well defined magnitude V , right ascension α and angular dec-
lination γ that can be considered constant for short-time observations of
a few days, where there are no appreciable changes due to the Earth’s
orbital velocity around the Sun. In this framework, where the only time
dependence is due to the Earth’s rotation, one identifies v(t) ≡ ṽ(t) and
θ0 (t) ≡ θ̃0 (t), where ṽ(t) and θ̃0 (t) derive from the simple application of
spherical trigonometry
cos z(t) = sin γ sin φ + cos γ cos φ cos(τ − α), (15)
ṽx (t)
≡ sin z(t) cos θ̃0 (t) = sin γ cos φ − cos γ sin φ cos(τ − α), (16)
V
ṽy (t)
≡ sin z(t) sin θ̃0 (t) = cos γ sin(τ − α), (17)
V

ṽ(t) ≡ ṽx2 (t) + ṽy2 (t) = V sin z(t). (18)
Here, z = z(t) is the zenithal distance of V, φ is the latitude of the
observatory, τ = ωsid t is the sidereal time of the observation in degrees
(ωsid ∼ 232π
h 56 ) and the angle θ0 is counted conventionally from North
through East so that North is θ0 = 0 and East is θ0 = 90◦ . In this way, one
finds
S(t) ≡ S̃(t) = Ss1 sin τ + Sc1 cos τ + Ss2 sin(2τ ) + Sc2 cos(2τ ), (19)
C(t) ≡ C̃(t) = C0 + Cs1 sin τ + Cc1 cos τ + Cs2 sin(2τ ) + Cc2 cos(2τ ).
(20)
In this picture, the Ck and Sk Fourier coefficients depend on the three

parameters (V, α, γ) (see [39]) and, to very good approximations, should be
time-independent for short-time observations.
However, this simple theoretical framework did not fit with the observa-
tions. In fact, the experimental data, even though slightly larger than the
experimental resolution, were always much smaller than the expected size
O(β 2 ). Also, the observed pattern was highly irregular because observations
performed at the same time on consecutive days could differ sizeably. This
has always represented a strong argument to interpret the data as pure
instrumental effects, i.e. “null results”.
However, from time to time, the greatest experts have seriously ques-
tioned this traditional null interpretation. Thus, one may ask if there could
be some theoretical framework in which these “small” and “irregular”
effects can acquire a definite physical meaning. For instance, we have seen
in Sec. 2, see Eqs. (10) and (11), that, by assuming the existence of a pre-
ferred reference frame Σ and using Lorentz transformations (rather than
Galileo’s transformations), the expected effects would be proportional to
2(N − 1)β 2 and not simply to β 2 . Therefore, for instance, for air, where the
refractive index N ∼ 1.00029, the fringe shifts for V ∼ 300 km/s would be
about 17 times smaller than those classical expected for V ∼ 30 km/s. For
gaseous helium, where N ∼ 1.000035, the effect would be even 140 times
smaller.
In addition, there is another important aspect. By comparing the
Earth’s cosmic motion with that of a body in a fluid, the standard picture
Eqs. (15)–(20) amounts to the condition of a pure laminar flow where global
and local velocity fields coincide. Here, there is a logical gap. The relation
between the macroscopic Earth’s motions and the ether-drift experiments
depends on the physical nature of the vacuum. If we consider the vacuum
as a form of quantum ether, the fringe shifts will likely exhibit the typi-
cal irregular (non-deterministic) pattern which characterizes any quantum
measurement. Therefore, from the theoretical arguments of Sec. 2, rather
www.ebook3000.com
Fig. 1. The two possible ways to relate Earth’s classical motion and fringe shifts.
than adopting the simple classical model of a laminar flow, one could try to
compare the experimental data with models of a turbulent flow, see Fig. 1.
In this case, due to the typical irregular behavior, vectorial quantities
(such as the fringe shifts) might easily average to zero. But, now, this does
not mean that there is no ether-drift.
A complete analysis of all classical experiments was presented in
Ref. [39]. Here, we shall only restrict to the first, and most famous, exper-
iment performed in 1887 by Michelson and Morley in Cleveland, and to
the last, and most precise, version which was performed in 1930 by Joos
in Jena. Due to the accuracy of this latter experiment, we shall explicitly
compare the data with numerical simulations of turbulent flows.
4.1. Michelson–Morley
Michelson and Morley performed their six observations in 1887, on July 8th,
9th, 11th, and 12th, at noon and in the evening, in the basement of the Case
Western University of Cleveland [47]. As well summarized by Miller in 1933
[48], “The brief series of observations was sufficient to show clearly that the
effect did not have the anticipated magnitude. However, and this fact must
be emphasized, the indicated effect was not zero”. The same conclusion had
already been obtained by Hicks in 1902 [49]: “. . . the data published by the
Michelson and Morley, instead of giving a null result, show distinct evidence
for an effect of the kind to be expected”. Quantitatively, the situation can
be summarized in Fig. 2, taken from Miller [48], where the values of the
effective velocity measured in various ether-drift experiments are reported
Fig. 2. The magnitude of the observable velocity measured in various experi-

ments as reported by Miller [48].
and compared with a smooth curve fitted by Miller to his own results as
function of the sidereal time.
In the framework of Eq. (10), the fringe shift is a second harmonic effect,
i.e. periodic in the range [0, π], whose amplitude A2 is predicted differently
by using the classical formulas or Lorentz transformations (10)
L v2 L vobs 2
Aclass
2 = Arel
2 = ∼ 2(N − 1)Aclass
2 (21)
λ c2 λ c2
Now, for the Michelson–Morley interferometer the whole effective optical
path was about L = 11 m, or about 2 · 107 in units of light wavelengths, so
for a velocity v ∼ 30 km/s (the Earth’s orbital velocity about the Sun, and
consequently the minimum anticipated drift velocity) the expected classical
second harmonic amplitude was Aclass
2 ∼ 0.2. This value can thus be used
as a reference point to obtain an observable velocity, in the plane of the
interferometer, from the actual measured value of A2 through the relation

A2
vobs ∼ 30 km/s. (22)
0.2
www.ebook3000.com
For the Michelson–Morley experiment, the average observable velocity

reported by Miller is about 8.4 km/s. Compared with the classical predic-
tion for a velocity of 30 km/s, this means an experimental second harmonic
amplitude
2
8.4
AEXP
2 ∼ 0.2 ∼ 0.016, (23)
30
which is about twelve times smaller than the expected result.
Neither Hicks nor Miller reported an estimate of the error on the 2nd
harmonic extracted from the Michelson–Morley data. To understand the
precision of their readings, we can look at the original paper [47] where
one finds the following statement: “The readings are divisions of the screw-
heads. The width of the fringes varied from 40 to 60 divisions, the mean
value being near 50, so that one division means 0.02 wavelength”. Now, in
their tables, Michelson and Morley reported the readings with an accuracy
of 1/10 of a division (example 44.7, 44.0, 43.5, . . .). This means that the
nominal accuracy of the readings was ±0.002 wavelengths. In fact, in units
of wavelengths, they reported values such as 0.862, 0.832, 0.824, . . . Fur-
thermore, this estimate of the error agrees well with Born’s book [50]. In
fact, Born, when discussing the classically expected fractional fringe shift
upon rotation of the apparatus by 90◦ , about 0.37, explicitly says: “Michel-
son was certain that the 100th part of this displacement would still be
observable” (i.e. 0.0037). Therefore, to be consistent with both the origi-
nal Michelson–Morley article and Born’s quotation of Michelson’s thought,
the estimate ±0.004 for the error was adopted in Refs. [39, 43]. In these
papers, many other details and all numerical values for the fringe shifts are
reported.
The fringe shifts are given as a periodic function, with vanishing mean,
in the range 0 ≤ θ ≤ 2π, so that they can be reproduced in a Fourier
expansion. One can thus extract the amplitude and the phase of the 2nd-
harmonic component by fitting the even combination of fringe shifts
∆λ(θ) + ∆λ(π + θ)
B(θ) = (24)
2λ
(see Fig. 4). This is essential to cancel the 1st-harmonic contribution orig-
inally pointed out by Hicks [49]. Its theoretical interpretation is in terms
of the arrangements of the mirrors and, as such, this effect has to show up
in the outcome of real experiments. The second-harmonic amplitudes from
the six individual sessions are reported in Table 2. One can then compute
the mean and variance of the six determinations reported in Table 2 by
Fig. 3. The Michelson–Morley fringe shifts as reported by Hicks [49]. Solid and
dashed lines refer respectively to noon and evening observations.
obtaining AEXP
2 ∼ 0.016 ± 0.006. This value is consistent with an observ-
+1.5
able velocity vobs ∼ 8.4−1.7 km/s. Then, by using Eq. (11), which connects
the observable velocity to the projection of the kinematical velocity in the
plane of the interferometer through the refractive index of the medium
where light propagation takes place (in our case air where N ∼ 1.00029),
we can deduce the average value
v ∼ 349+62
−70 km/s. (25)
www.ebook3000.com
July 11 noon
0.04
0.02
Β(θ )
−0.02
−0.04
0 π /2 π
Fig. 4. A fit to the even combination B(θ) Eq. (24). The second-harmonic ampli-
tude is AEXP
2 = 0.025 ± 0.005 and the fourth-harmonic is AEXP
4 = 0.004 ± 0.005.
The figure is taken from Ref. [43]. Compare the data with the solid curve of July
11th shown in Fig. 3.
While the individual values of A2 show a reasonable consistency, there

are substantial changes in the apparent direction θ0 of the ether-drift effect
in the plane of the interferometer. This is the reason for the strong can-
celations obtained when fitting together all noon sessions or all evening
sessions [51]. According to the usual interpretation, the large spread of the
azimuths is taken as indication that any non-zero fringe shift is due to pure
instrumental effects. However, as anticipated, this type of discrepancy could
also indicate an unconventional form of ether-drift, where there are substan-
tial deviations from Eq. (6) and/or from the smooth trend in Eqs. (15)–(18).
For instance, in agreement with the general structure Eq. (4), and differ-
ently from July 11 noon, which represents a very clean indication, there are
sizeable fourth-harmonic contributions (here, AEXP4 = 0.019 ± 0.005 and
AEXP
4 = 0.008 ± 0.005 for the noon sessions of July 8 and July 9 respec-
tively). In any case, the observed strong variations of θ0 are in qualitative
agreement with the analogous values reported by Miller. To this end, com-
pare with Fig. 22 of Ref. [48] and in particular with the large scatter of the
data taken around August 1st, as this represents the epoch of the year which
is closer to the period of July when the Michelson–Morley observations were
Table 1. The amplitude of the

fitted second-harmonic component
AEXP
2 for the six experimental
sessions of the Michelson–Morley
experiment.
Session AEXP
2
July 8 (noon) 0.010 ± 0.005

July 9 (noon) 0.015 ± 0.005
July 11 (noon) 0.025 ± 0.005
July 8 (evening) 0.014 ± 0.005
July 9 (evening) 0.011 ± 0.005
July 12 (evening) 0.024 ± 0.005
actually performed. Thus, one could also conclude that individual exper-
imental sessions indicate a definite non-zero ether-drift but the azimuth
does not exhibit the smooth trend expected from the conventional picture
Eqs. (15)–(18).
We emphasize that the large spread of the θ0 values might also reflect
a particular systematic effect pointed out by Hicks [49]. As described by
Miller [48], “before beginning observations the end mirror on the telescope
arm is very carefully adjusted to secure vertical fringes of suitable width.
There are two adjustments of the angle of this mirror which will give fringes
of the same width but which produce opposite displacements of the fringes
for the same change in one of the light-paths”. Since the relevant shifts
are extremely small, “. . . the adjustments of the mirrors can easily change
from one type to the other on consecutive days. It follows that averaging the
results of different days in the usual manner is not allowable unless the types
are all the same. If this is not attended to, the average displacement may be
expected to come out zero — at least if a large number are averaged” [49].
Therefore, averaging the fringe shifts from various sessions represents
a delicate issue and can introduce uncontrolled errors. In fact, an overall
change of sign of the fringe shifts at all θ values is equivalent to replacing
the azimuth θ0 → θ0 ± π/2. However, this relative sign does not affect the
values of A2 and this is why averaging the second-harmonic amplitudes
in Table 1, as we have done, is a safer procedure. From these amplitudes,
one obtains the average kinematical velocity Eq. (25) which is completely
consistent with the average value of 369 km/s associated with the Earth’s
motion with respect to the CMB.
www.ebook3000.com
5. Numerical Simulations in a Turbulent-Ether Model

As anticipated at the end of Sec. 4, identifying the local velocity
field (vx (t), vy (t)) in Eqs. (12)–(14) with the smooth, global quantities
(ṽx (t), ṽy (t)), which describe the cosmic Earth’s motion, is equivalent to
adopt the model of a laminar flow. Instead, by adopting the different model
of a turbulent flow, the situation changes completely.
As mentioned in the introduction, in the limit of vanishing viscosity,
the local velocity field becomes non-differentiable. In these conditions, the
ordinary formulation in terms of differential equations becomes inadequate
and must be replaced by some other description such as a formulation in
terms of random Fourier series [20,52]. In this other approach, the parame-
ters of the macroscopic motion are only used to fix the limiting boundaries
[53] for a microscopic velocity field which has instead an intrinsic stochastic
nature.
The simplest choice, which represents a zeroth-order approximation,
corresponds to a turbulence which, at small scales, appears statistically
isotropic and homogeneous.d In spite of its simplicity, it is a useful example
to illustrate basic phenomenological features associated with an underlying
stochastic vacuum. The perspective is that of an observer moving in the
turbulent fluid who wants to simulate the two components of the velocity
in his x–y plane at a given fixed location in his laboratory. In a statistically
isotropic and homogeneous turbulence, one finds the general expressions
∞

vx (t) = [xn (1) cos ωn t + xn (2) sin ωn t], (26)
n=1
∞
vy (t) = [yn (1) cos ωn t + yn (2) sin ωn t], (27)
n=1
where ωn = 2nπ/T , T being a time scale which represents a common period

of all stochastic components. For our simulations, we have adopted the
typical value T = Tday = 24 h. However, we have also checked with a few
runs that the statistical distributions of the various quantities do not change
substantially by varying T in the rather wide range 0.1 Tday ≤ T ≤ 10 Tday .
The coefficients xn (i = 1, 2) and yn (i = 1, 2) are random variables
with zero mean and have the physical dimension of a velocity. By assuming
statistical isotropy, we shall denote by [−ṽ, ṽ] the common interval for these
d This picture reflects the basic Kolmogorov theory [21] of a fluid with vanishingly small
viscosity.
four parameters. In terms of ṽ, the statistical average of the quadratic values
can be expressed as
ṽ 2
x2n (i = 1, 2)stat = yn2 (i = 1, 2)stat = (28)
3 n2η
for the uniform probability model (within the interval [−ṽ, ṽ]) which we
have chosen for our simulations. Finally, the exponent η controls the power
spectrum of the fluctuating components. For the simulations, between the
two values η = 5/6 and η = 1 reported in Ref. [53], we have chosen η = 1
which corresponds to the point of view of an observer moving in the fluid.
We observe that one could further improve the stochastic model by
introducing time modulations and/or slight deviations from isotropy. For
instance, ṽ could become a function of time ṽ = ṽ(t). By still retaining
statistical isotropy, this could be used to simulate the possible modulations
of the projection of the Earth’s velocity in the plane of the interferometer.
Or, one could fix a range, say [−ṽx , ṽx ], for the two random parameters
xn (1) and xn (2), which is different from the range [−ṽy , ṽy ] for the other
two parameters yn (1) and yn (2). Finally, ṽx and ṽy could also become given
functions of time, for instance ṽx (t) ≡ ṽ(t) cos θ̃0 (t) ṽy (t) ≡ ṽ(t) sin θ̃0 (t),
ṽ(t) and θ̃0 (t) being defined in Eqs. (15)–(18). In this way, for each time t,
Eq. (28) now become
ṽx2 (t) ṽy2 (t)

x2n (i = 1, 2)stat = , yn2 (i = 1, 2)stat = . (29)
3 n2η 3 n2η
For most classical experiments, these further refinements are unnecessary.
In fact, in most cases, only observations at few selected hours were per-
formed so that, in view of the strong fluctuations of the data, one can just
extract the average magnitude of the observed velocity and, within the
errors, a macroscopic kinematical velocity. A notable exception is Joos’s
1930 experiment [54]. Its accuracy was incomparable among the classical
experiments since the observations were performed each hour to cover the
whole sidereal day and the data were recorded by photocamera. As we shall
see in the next section, Joos’s data are sensitive to the details of the Earth’s
cosmic motion and require to adopt the most refined framework Eq. (29).
5.1. Joos
Joos’s optical system [54] was enclosed in a hermetic housing and, tradi-
tionally, it was always assumed that the fringe shifts were recorded in a
partial vacuum. On the other hand, Swenson [55] explicitly reports that
www.ebook3000.com
fringe shifts were finally recorded with optical paths placed in a helium
bath. In spite of the fact that this important aspect is never mentioned in
Joos’s papers, we have followed Swenson by assuming that during the mea-
surements the interferometer was filled by gaseous helium at atmospheric
pressure.
The observations were performed in Jena in 1930, starting at 2 PM of
May 10th and ending at 1 PM of May 11th. Two measurements, the 1st and
the 5th, were finally deleted by Joos with the motivation that there were
spurious disturbances. The data were combined symmetrically, in order to
eliminate the presence of odd harmonics, and the magnitude of the fringe
shifts was typically of the order of a few thousandths of a wavelength.
Fig. 5. The selected set of data reported by Joos [54]. The yardstick corresponds
to 1/1000 of a wavelength so that the experimental dots have a size of about
0.4 · 10−3 . This corresponds to an uncertainty ±0.2 · 10−3 in the extraction of the
fringe shifts.
To this end, one can look at Joos’s picture (reported here as our Fig. 5)
and compare with the shown size of 1/1000 of a wavelength. From this
picture, Joos decided to adopt 1/1000 of a wavelength as an upper limit
and deduced an observable velocity vobs 1.5 km/s. To derive this value,
he used the fact that, for his apparatus, an observable velocity of 30 km/s
would have produced a second-harmonic amplitude of 0.375 wavelengths.
Still, since it is apparent from Fig. 5 that some fringe displacements
were definitely larger than 1/1000 of a wavelength, the values of the second-
harmonic amplitude A2 were extracted [39] from the 22 pictures. Different
from the values of the azimuth, this can be done unambiguously. The point
is that, due to the camera effect, it is not clear how to fix the reference
angular values θk in Fig. 4 for the fringe shifts. In addition, there is a small
misalignment angle, between the dots of Joos’s fringe shifts and the N, W,
and S marks, which cannot be deduced from the articles. Since clearly there
is only one correct choice for the reference angles θk , we have preferred not
to quote theoretical uncertainties on the azimuth and just concentrate on
the amplitudes whose values, instead, do not depend on the angles θk and
thus can be extracted unambiguously. Their values are reported in Fig. 6.
The accuracy of each determination is about ±0.2·10−3 as given by the size
A2
2
0
0 5 10 15 20 25
Picture
Fig. 6. Joos’s second-harmonic amplitudes, in units 10−3 . The vertical band

between the two lines corresponds to the range (1.4 ± 0.8) · 10−3 . The figure is
taken from Ref. [39].
www.ebook3000.com
of Joos’s experimental dots in Fig. 5. This uncertainty is about one order of

magnitude better than for Michelson-Morley and a factor of 3 better than
the 1/1500 reading error in the Illingworth experiment [56].
By computing mean and variance of the individual values, we obtain an
average second harmonic amplitude
Ajoos
2 = (1.4 ± 0.8) · 10
−3
(30)
+0.5
and a corresponding observable velocity vobs ∼ 1.8−0.6 km/s. By correct-
ing with the helium refractive index, Eq. (11) would then imply a true
kinematical velocity v ∼ 217+66
−79 km/s.
However, this is only a first and very partial view of Joos’s experiment.
In fact, we have compared Joos’s amplitudes with theoretical models of
cosmic motion. To this end, after transforming the civil times of Joos’s
measurements into sidereal times, by using Eqs. (15)–(18), one can com-
pare Joos’s amplitudes with theoretical predictions which, for the given
latitude φ = 50.94◦ of Jena, depend on the right ascension α and the angu-
lar declination γ. To this end, it is convenient to first re-write the theoretical
forms as
A2 (t) cos 2θ0 (t) = 2C(t)
2L(N − 1) vx2 (t) − vy2 (t) v 2 (t) − vy2 (t)
−3 x
= ∼ 2.6 · 10
λ c2 (300 km/s)2
(31)
and
A2 (t) sin 2θ0 (t) = 2S(t)
2L(N − 1) 2vx (t) vy (t) 2vx (t) vy (t)
= 2
∼ 2.6 · 10−3 ,
λ c (300 km/s)2
(32)
where we have used the numerical relation for Joos’s experiment
2
L (30 km/s)
λ c2 ∼ 0.375 and the value of the helium refractive index. Then,
by approximating vx (t) ∼ ṽx (t), vy (t) ∼ ṽy (t) and using Eq. (18) for the
scalar combination ṽ(t) ≡ ṽx2 (t) + ṽy2 (t), we have fitted the data of Fig. 6
to the smooth form
Asmooth
2 (t) = const · sin2 z(t), (33)
where cos z(t) is defined in Eq. (15). The results of the fit
α = 168◦ ± 30◦ , γ = −13◦ ± 14◦ , (34)
confirm that, as found in connection with the Michelson–Morley experi-

ment, the Earth’s motion with respect to the CMB (which has α ∼ 168◦
and γ ∼ −6◦ ) could serve as a useful model to describe the ether-drift data.
Still, in spite of the good agreement with the CMB α- and γ-values
obtained from the fit Eq. (34), the nature of the strong fluctuations in
Fig. 6 remains unclear. Apart from this, there is also a sizeable discrepancy
in the absolute normalization of the amplitude. In fact, by assuming the
standard picture of smooth time modulations, the mean amplitude over
all sidereal times can trivially be obtained from the mean squared velocity
Eq. (18)

1
ṽ 2 (t) = V 2 1 − sin2 γ sin2 φ − cos2 γ cos2 φ . (35)
2

For the CMB and Jena, this gives ṽ 2 ∼ 330 km/s so that one would
naively predict from Eqs. (31) and (32)
ṽ 2 (t)
Asmooth
2 (t) ∼ 2.6 · 10−3 ∼ 3.2 · 10−3 (36)
(300 km/s)2
to be compared with Joos mean value Ajoos 2 = (1.4 ± 0.8) · 10

−3
. In the
standard
picture, this experimental
value leads to the previous estimate
ṽ 2 ∼ 217 km/s and not to ṽ 2 ∼ 330 km/s so that it is necessary to
change the theoretical model to try to make Joos’ experiment completely
consistent with the Earth’s motion with respect to the CMB.
To try to solve this problem, and understand the origin of the observed
strong fluctuations, we have used the model Eqs. (26) and (27) of Sec. 6,
to simulate stochastic variations of the velocity field. As anticipated, due
to the high accuracy of the Joos experiment, the two random parame-
ters xn (1) and xn (2) were allowed to vary in the range [−ṽx (t), ṽx (t)] and
the other two parameters yn (1) and yn (2) to vary in the different range
[−ṽy (t), ṽy (t)], where ṽx (t) and ṽy (t) are defined in Eqs. (15)−(17). Also,
the quadratic values were fixed as in Eq. (29). It is understood that the
latitude corresponds to Joos’s experiment while V , α and γ describe the
Earth’s motion with respect to the CMB.
In this model, there will be a substantial reduction of the amplitude
with respect to its smooth prediction. To estimate the order of magnitude
of the reduction, one can perform a full statistical average (as for an infinite
number of measurements) and use Eq. (29) in Eqs. (31) and (32) for our
www.ebook3000.com
Joos Data
5 Poly Fit
Simulation
Poly Fit
A2 3
0
0 5 10 15 20 25
Picture
Fig. 7. Joos’s experimental amplitudes in Fig. 6 are compared with a single sim-
ulation of 22 instantaneous measurements. By changing the random sequence,
the typical variation of each simulated entry is (1/4) · 10−3 depending on the
sidereal time. The stochastic velocity components are controlled by the kinemat-
ical parameters (V, α, γ)CMB as explained in the text. We also show two 5th-
order polynomial fits to the two different sets of values. The figure is taken from
Ref. [39].
case η = 1. This gives

∞
ṽ 2 (t) 1 1 π 2 smooth
A2 (t)stat ∼ 2.6 · 10−3 = A (t). (37)
(300 km/s)2 3 n=1 n2 18 2
By also averaging over all sidereal times, for the CMB and Jena, one would
now predict a mean amplitude of about 1.7 · 10−3 and not of 3.2 · 10−3 .
To have an idea of the agreement between Joos’s 22 amplitude data
and a single numerical simulation of instantaneous measurements, we show
a graphical comparison in Fig. 7. We emphasize that one should not com-
pare each individual entry with the corresponding data since, by changing
the random sequence, the simulated instantaneous entries vary typically
of about (1/4) · 10−3 depending on the sidereal time. Instead, one should
compare the overall trend of data and simulation. To this end, we show two
5th-order polynomial fits to the two different sets of values.
A more conventional comparison with the data consists in quoting for
the various 22 entries simulated average values and uncertainties. To this
Joos Data
5 Simulation
A2 3
0
0 5 10 15 20 25
Picture
Fig. 8. Joos’s experimental amplitudes in Fig. 6 are compared with the result of
simulating the averaging process over 10 measurements performed, at each Joos’s
time, on 10 consecutive days. The stochastic velocity components are controlled
by the kinematical parameters (V, α, γ)CMB as explained in the text. The effect
of varying the random sequence has been approximated into a central value and
a symmetric error. The figure is taken from Ref. [39].
end, we have considered the mean amplitudes Asimul 2 (ti ) defined by aver-
aging, for each Joos’s time ti , over 10 hypothetical measurements performed
on 10 consecutive days. For each ti , the observed effect of varying the ran-
dom sequence has been summarized into a central value and a symmetric
error. The simulated values and the comparison with Joos’s amplitudes is
shown in Fig. 8.
The spread of the various entries is larger at the sidereal times, where
the projection at Jena of the cosmic Earth’s velocity becomes larger. The
tendency of Joos’s data to lie in the lower part of the simulated range mostly
depends on our use of symmetric errors. In fact, by comparing in some case
with the histograms of the basic generated configurations Asimul 2 (ti ), we have
seen that our sampling method of Asimul 2 (t i ) typically underestimates the
weight of the low-amplitude region in a prediction at the 70% C.L. For
this reason, one could improve the evaluation of the probability content.
However, in view of the good agreement already found in Fig. 8 (χ2 =
13/22), we did not attempt to carry out this more refined analysis.
www.ebook3000.com
In conclusion, after the first indication obtained from the fit equa-

tion (34), the link between Joos’ data and the Earth’s motion with respect
to the CMB gets reinforced by our simulations. In fact, by inspection of
Figs. 7 and 8, the values of the amplitudes and the characteristic scatter
of the data are correctly reproduced. From this agreement, we then deduce
that the previous kinematical value v ∼ 217+66−79 km/s has to be consider-
ably increased if one allows for stochastic variations of the velocity field.
In fact, the magnitude of the fluctuations
in vx and vy is controlled by the
same scalar parameter ṽ(t) ≡ ṽx2 (t) + ṽy2 (t) of Eq. (18). We thus con-
clude that Joos’s data are consistent with a range of kinematical velocity
v = 330+40 ◦
−70 km/s which corresponds to Eq. (18) for φ = 50.94 , V = 370
◦ ◦
km/s, α = 168 and γ = −6 .
6. Summary and Conclusions

Traditionally, the interpretation of the ether-drift experiments has been
based on a theoretical model, where all type of signals that are not syn-
chronous with the Earth’s rotation tend to be considered as spurious instru-
mental noise. However, there is a logical gap. The link between the two
concepts depends on the adopted model for the vacuum. The point of view
adopted so far corresponds to consider the vacuum as some kind of fluid
in a state of regular, laminar motion. In these conditions, global and local
properties of the flow coincide.
We believe that, without fully understanding the nature of that substra-
tum that we call physical vacuum, one should instead keep a more open
mind. As discussed in Sec. 2, the physical vacuum might be similar to a
form of turbulent ether, an idea which is deep rooted in basic foundational
aspects of both quantum theory and relativity and finds additional moti-
vations in those representations of the vacuum as a form of “space–time
foam” which indeed resembles a turbulent fluid. In this case, global and
local velocity fields might be very different and there could be forms of
random signals that have a genuine physical origin.
To explore this idea, we have re-considered from scratch the classical
experiments. These were performed in gaseous media, where the refractive
index N is extremely close to unity. In this case, in the framework of a
Lorentzian view of relativity, by expanding around N = 1 and to leading
order in v/c, one formally finds the same classical formulas with the only
replacement
v 2 → 2(N − 1)v 2 ≡ vobs

2
. (38)
As discussed in detail in Sec. 3, this replacement can be understood with

simple symmetry arguments but also admits a dynamical interpretation
in terms of the energy–momentum flow associated with a Lorentz non-
invariant vacuum. As emphasized in Sec. 3, this dynamical mechanism is
not unexpected on the basis of the present views of the vacuum as a parti-
cle condensate and is also useful to reconcile the different phenomenological
pattern between ether-drift experiments in gaseous media and those per-
formed in strongly bound systems such as solid or liquid transparent media.
Now testing the scheme is very simple: one should just check the consis-
tency of the true kinematical v s obtained in different experiments. In this
alternative interpretation, the indications of the various experiments are
summarized in Table 2 which is taken from Ref. [39] (to which we address
the reader for many details). Here, we just emphasize the following points:
(i) An analysis of the individual sessions of the original Michelson–Morley

experiment, in agreement with Hicks [49] and Miller [48] (see our
Figs. 1 and 2), gives no justification to its standard null interpreta-
tion. As discussed in Sec. 5, this type of analysis is more reliable.
In fact, averaging directly the fringe displacements of different sessions
requires two additional assumptions, on the nature of the ether-drift
as a smooth periodic effect and on the absence of systematic errors
introduced by the re-adjustment of the mirrors on consecutive days,
that in the end may turn out to be wrong.
(ii) From the Michelson–Morley, Morley–Miller, Miller and Illingworth-
Kennedy experiments, one gets average kinematical velocities which
are well consistent with the value 370 km/s which today is used to
Table 2. The average velocity observed (or the limits placed) by the clas-
sical ether-drift experiments in the alternative interpretation of Eqs. (6),
(10) and (11). The table is taken from Ref. [39].
Gas in the
Experiment interferometer vobs (km/s) v (km/s)
Michelson–Morley (1887) Air 8.4+1.5

−1.7 349+62
−70
Morley–Miller (1902–1905) Air 8.5 ± 1.5 353 ± 62
Kennedy (1926) Helium <5 <600
Illingworth (1927) Helium 3.1 ± 1.0 370 ± 120
Miller (1925–1926) Air 8.4+1.9
−2.5 349+79
−104
Michelson–Pease–Pearson (1929) Air 4.5 ± · · · 185 ± · · ·
Joos (1930) Helium 1.8+0.5
−0.6 330+40
−70
www.ebook3000.com
describe the CMB anisotropy. In view of this consistency, the stan-

dard interpretation of Miller’s observations in terms of a temperature
gradient [57] is only acceptable provided this gradient represents a
non-local effect as in the picture of Refs. [39, 41] where the ether-drift
is the consequence of a fundamental vacuum energy–momentum flow.
(iii) Some discrepancy is found with the experiment performed by Michel-
son, Pease and Pearson (MPP). At the same time, as emphasized in
Ref. [39], the uncertainty cannot be easily estimated since only a single
basic MPP observation is explicitly reported in the literature. There-
fore, since Miller’s extensive observations (see Ref. [48], Fig. 22), within
their errors, gave fluctuations of the observable velocity in the wide
range 4−14 km/s, a single observation giving vobs ∼ 4.5 km/s cannot
be interpreted as a refutation. This becomes even more true by notic-
ing that the single MPP session explicitly reported, within a period
of several months, was chosen to represent an example of extremely
small ether-drift effect.
(iv) Joos’s experiment is particularly important since the data were col-
lected at steps of 1 h to cover the full sidereal day and were recorded
by photocamera. For this reason, it is not comparable with other exper-
iments (e.g. Michelson–Morley, Illingworth) where only observations at
few selected hours were performed and for which, in view of the strong
fluctuations of the data, one can just quote the average magnitude
of the observed velocity. In fact, by fitting the experimental ampli-
tudes in Fig. 6 to various forms of cosmic motion (see Eq. (34)) we
have obtained angular parameters which are very close to those that
describe the CMB anisotropy (right ascension αCMB ∼ 168◦ and angu-
lar declination γCMB ∼ −6◦ ). Still, to get a complete agreement, one
should explain the absolute normalization of the amplitudes and the
strong fluctuations of the data. Thus, we have sharpened our analysis
by performing various numerical simulations where the velocity com-
ponents vx (t) and vy (t) are not smooth functions but are represented
as turbulent fluctuations. Their Fourier components in Eqs. (26) and
(27) vary within time-dependent ranges Eqs. (16)–(17), [−ṽx (t), ṽx (t)]
and [−ṽy (t), ṽy (t)], respectively, controlled by the macroscopic param-
eters (V, α, γ)CMB . Taking into account these stochastic fluctuations
of the velocity field tends to increase the fitted average Earth’s veloc-
ity, see Eq. (37), and can reproduce correctly Joos’ second-harmonic
amplitudes and the characteristic scatter of the data, see Figs. 7 and 8.
These results give a strong motivation to repeat these crucial mea-

surements with today’s much greater accuracy. To this end, let us now
briefly consider the modern ether-drift experiments. In this case, the test
of the isotropy of the velocity of light consists in measuring the relative fre-
quency shift ∆ν of two orthogonal optical resonators [58]. Here, the analog
of Eq. (10) is
∆ν phys (θ) c̄γ (π/2 + θ) − c̄γ (θ) v2
= = Bmedium 2 cos 2(θ − θ0 ), (39)
ν0 c c
where θ0 is the direction of the ether-drift. This can be interpreted within
Eq. (8), where
|Bmedium| ∼ Nmedium − 1. (40)
Nmedium being the refractive index of the gaseous medium filling the opti-
cal resonators. Testing this prediction requires replacing the high vacuum
usually adopted within the optical resonators with a gaseous medium and
studying the substantially larger frequency shift introduced with respect to
the vacuum experiments.
As a rough check, a comparison was made [42, 43] with the results
obtained by Jaseja et al. [59] in 1963, when looking at the frequency shift
of two orthogonal He–Ne masers placed on a rotating platform. To this
end, one has to preliminarily subtract a large systematic effect that was
present in the data and interpreted by the authors as probably due to mag-
netostriction in the Invar spacers induced by the Earth’s magnetic field.
As suggested by the same authors, this spurious effect, which was only
affecting the normalization of the experimental ∆ν, can be subtracted by
looking at the variations of the data. As discussed in Refs. [42, 43], the
measured variations of a few kHz are roughly consistent with the refractive
index NHe−Ne ∼ 1.00004 and the typical variations of an Earth’s velocity
as in Eq. (25).
More recent experiments [60–64] have always been performed in a very
high vacuum, where as emphasized in the Introduction, the differences
between Special Relativity and the Lorentzian interpretation are at the
limit of visibility. In fact, in a perfect vacuum by definition Nvacuum = 1 so
that Bvacuum will vanish.e Thus, one should switch to the new generation of
e Strictly speaking, modern experiments in vacuum are also consistent with an instan-
taneous ether-drift effect of order 10−15 . In the framework of Eq. (7), for values
v2 /c2 ∼ 10−6 , this could indicate that the velocity of light in the vacuum, as measured
on the Earth’s surface, differs from the parameter c entering Lorentz transformations at
the level O(10−9 ). A possible theoretical scenario for this difference, after incorporating
the idea of vacuum turbulence [65, 66], is completely consistent with the present data.
www.ebook3000.com
dedicated ether-drift experiments in gaseous systems. Our conclusion is that

these new experiments should just confirm Joos’s remarkable observations
of 80 years ago.
References
1. C.D. Froggatt and H.B. Nielsen, Origin of Symmetries, World Scientific,
1991.
2. Physics of Emergence and Organization, I. Licata and A. Sakaji Eds., World
Scientific, 2008.
3. E.T. Whittaker, A history of the Theories of Aether and Electricity, Dover
Publ., 1989.
4. O.V. Troshkin, Physica A 168, 881 (1990).
5. H.E. Puthoff, Linearized turbulent flow as an analog model for linearized
General Relativity, 2008, Unpublished.
6. T.D. Tsankov, Classical Electrodynamics and the Turbulent Aether Hypothe-
sis, 2009, Unpublished.
7. M.J. Marcinkowski, Physica Status Solidi. 152B, 9 (1989).
8. A.M. Kosevic, The Crystal Lattice: Phonons, Solitons, Superlattices, Wiley-
VCH Verlag Gmbh and Co. KGaA, Weinheim, 2005.
9. H. Günther, Physica Status Solidi. 149, 104 (1988).
10. C.I. Christov, Math. Comput. Simul. 74, 93 (2007).
11. R.P. Feynman, R.B. Leighton and M. Sands, The Feynman Lectures on
Physics, Vol. I, Sec. 15.9, Addison Wesley Publ. Co. 1963.
12. M. Consoli, Phys. Lett. A 376, 3377 (2012).
13. M. Consoli, A. Pluchino and A. Rapisarda, Chaos Solitons Fract. 44, 1089
(2011).
14. T.W. Marshall, Proc. R. Soc. A 276, 475 (1963).
15. T.H. Boyer, Phys. Rev. 182, 1374 (1969); ibidem 186, 1304 (1969).
16. H.E. Puthoff, Phys. Rev. D 35, 3266 (1987).
17. L. de la Peña and A. M. Cetto, The Quantum Dice — An Introduction to
Stochastic Electrodynamics, Kluwer Academic Publ., Dordrecht, 1996.
18. D.C. Cole and Y. Zou, Phys. Lett. A 317, 14 (2003).
19. E. Nelson, Phys. Rev. 150, 1079 (1966).
20. L. Onsager, Nuovo Cimento, Supplement 6, 279 (1949).
21. A.N. Kolmogorov, Dokl. Akad. Nauk SSSR 30, 4 (1940).
22. P. Jizba and H. Kleinert, Phys. Rev. D 82, 085016 (2010).
23. P. Jizba and F. Scardigli, Special relativity induced by granular space, Eur.
Phys. J. C 73, 2491 (2013).
24. C. Beck and E.G.D. Cohen, Phys. A 322, 267 (2003).
25. C. Beck, Phys. Rev. Lett. 98, 064502 (2007).
26. J.A. Wheeler, in Relativity, Groups and Topology, B.S. DeWitt and
C.M. DeWitt Eds., Gordon and Breach, New York, 1963, p. 315.
27. S. Hawking, Nucl. Phys. B 144, 349 (1978).
28. R. Bousso, Rev. Mod. Phys. 74, 825 (2002).
29. G. Amelino-Camelia, Nature 418, 34 (2002).
30. Y.J. Ng, Various Facets of Spacetime Foam, in Proceedings of the Third Con-
ference on Time and Matter, Budva, Montenegro (2010).
31. Y.J. Ng and H. van Dam, Mod. Phys. Lett. A 9, 335 (1994).
32. M.T. Jaekel and S. Reynaud, Phys. Lett. A 185, 143 (1994).
33. E. Göklü and C. Lämmerzahl, Class. Quant. Grav. 25, 105012 (2008).
34. E. Göklü et al., Class. Quant. Grav. 26, 225010 (2009).
35. G. Amelino-Camelia, Phys. Rev. D 62, 024015 (2000).
36. E. Göklü and C. Lämmerzahl, Gen. Rel. Grav. 43, 2065 (2011).
37. S. Schiller et al., Phys. Rev. D 69, 027504 (2004).
38. V. Jejjala, D. Minic, Y. J. Ng and C. H. Tze, Int. J. Mod. Phys. D 19, 2311
(2010).
39. M. Consoli, C. Matheson and A. Pluchino, Eur. Phys. J. Plus 128, 71 (2013).
40. H.A. Lorentz, The Theory of Electrons, B.G. Teubner ed., Leipzig, 1909.
41. M. Consoli and E. Costanzo, Eur. Phys. J. C 54, 285 (2008).
42. M. Consoli and E. Costanzo, Phys. Lett. A 333, 355 (2004).
43. M. Consoli and E. Costanzo, N. Cim. 119B, 393 (2004).
44. J. Shamir and R. Fox, N. Cim. 62B, 258 (1969).
45. H.P. Robertson, Rev. Mod. Phys. 21, 378 (1949).
46. R.M. Mansouri and R.U. Sexl, Gen. Rel. Grav. 8, 497 (1977).
47. A.A. Michelson and E.W. Morley, Am. J. Sci. 34, 333 (1887).
48. D.C. Miller, Rev. Mod. Phys. 5, 203 (1933).
49. W.M. Hicks, Phil. Mag. 3, 9 (1902).
50. M. Born, Einstein’s Theory of Relativity, Dover Publ., New York, 1962.
51. M.A. Handshy, Am. J. of Phys. 50, 987 (1982).
52. L.D. Landau and E.M. Lifshitz, Fluid Mechanics, Chapt. III, Pergamon
Press, 1959.
53. J.C.H. Fung et al., J. Fluid Mech. 236, 281 (1992).
54. G. Joos, Ann. d. Physik 7, 385 (1930).
55. Loyd S. Swenson Jr., J. Hist. Astro. 1, 56 (1970).
56. K.K. Illingworth, Phys. Rev. 30, 692 (1927).
57. R.S. Shankland et al., Rev. Mod. Phys. 27, 167 (1955).
58. J. Ehlers and C. Lämmerzahl (eds.), Special relativity, Lectures Notes in
Physics, Springer, 2006.
59. T.S. Jaseja, et al., Phys. Rev. 133, A1221 (1964).
60. A. Brillet and J.L. Hall, Phys. Rev. Lett. 42, 549 (1979).
61. P. Antonini, et al., Phys. Rev. A 71, 050101(R) (2005).
62. S. Herrmann, et al., Phys. Rev. Lett. 95, 150401 (2005).
63. S. Herrmann, et al., Phys. Rev. D 80, 105011 (2009).
64. Ch. Eisele, A. Newski and S. Schiller, Phys. Rev. Lett. 103, 090401 (2009).
65. M. Consoli and L. Pappalardo, Gen. Rel. and Grav. 42, 2585 (2010).
66. M. Consoli, A. Pluchino, A. Rapisarda and S. Tudisco, Physica A 394, 61
(2014).
www.ebook3000.com
Chapter 17
. . .And Kronos Ate His Sons

Giuseppe Vitiello
Dipartimento di Fisica “E. R. Caianiello” Universitá di Salerno
and Istituto Nazionale di Fisica Nucleare
I-84084 Fisciano (Salerno), Italy
vitiello@sa.infn.it
In closed systems, energy is conserved. The origin of the time axis

is completely arbitrary due to the invariance under continuous time-
translations. The flowing of time swallows those fictitious origins one
might assign on its axis, as Kronos ate his sons. Dissipation breaks such
a scenario. It implies a non-forgettable origin of time. Open systems
need their complement (their “double”) in order to become, together, a
closed system. Time emerges as an observable measured by the evolution
of the open system complement, which acts as a clock. The conservation
of the energy–momentum tensor in electrodynamics is considered and
its relation with dissipative systems and self-similar fractal structures is
discussed. The isomorphism with coherent states in quantum field the-
ory (QFT) is established and the generator of transitions among unitar-
ily inequivalent representations of the canonical commutation relations
(CCR) is shown to provide sequences in time of phases, which defines
the arrow of time. Merging properties of electrodynamics, fractal self-
similarity, dissipation and coherent states point to an integrated vision
of Nature.
1. Prelude
1.1. Pre-history. Energy conservation and time
as a dummy variable
In theories invariant under time-translation transformations, the origin of
the time axis is completely arbitrary. The time-origin can be freely trans-
lated without producing any observable effect. All the origins are alike.
There is no special or singular origin of time. All quantities describing time-
translational invariant systems must depend on time differences. Energy is
465
466 G. Vitiello
conserved. Systems are closed systems. There is no Now, no true time-

origin can be assigned. There is no preferred reference clock to synchronize
with. In this sense, absolute time does not exist. All clocks, each of them
with its own time, are equally legitimate as an effect of the continuous time-
translational symmetry. Some of them may even go backwards in time since
there is no restriction on the direction of continuous translations of time:
time-reversal symmetry is also there. Time, in its flowing, swallows those
fictitious or false origins one might assign on its axis, as Kronos ate his sons.
1.2. History. Dissipation and the origin

of time and of forms
Dissipation modifies such a scenario. Dissipative systems have a lifetime,
they are aging systems, the memory of the age of the system is now pos-
sible, their “history” can be now recorded, the arrow of time arises. Due
to dissipation the breakdown of both symmetries, time-reversal symmetry
and continuous time-translational symmetry, occurs at once. Use of the
canonical formalism requires to consider the closed system made of the
open dissipative system under study and of the environment in which it is
embedded, its complement or “Double”. Time dependent coherent states
are generated. Time evolution over their manifold is described by classical
chaotic trajectories. Freedom from unitary time evolution slavery (US) is
obtained. The true, non-forgettable origin of time can be assigned, which
can no longer be eaten by Kronos. Time stops being a dummy variable. It
emerges as an observable measured by the open system Double, which acts
as a clock, or vice versa. Dissipative time evolution causes deformations
in coherent states which manifest themselves in fractal self-similar forms
(morphogenesis). A chance to life is given.
2. Electrodynamics, Fractal Self-Similarity

and Coherence
In the frame depicted above, I consider the conservation of the energy-
momentum tensor Tµν in electrodynamics. I show that the matter field
and the electromagnetic (EM) field, when separately considered, each of
them, undergo, dissipative (non-unitary) time evolution. The unitary time
evolution of electrodynamics actually arises when both fields are considered
in their intimate entanglement.
I will also show how fractal self-similarity enters in the picture and
how crucial is the role of coherent states in all that. The isomorphism
www.ebook3000.com
between fractal self-similarity and squeezed coherent states in quantum

field theory (QFT) is indeed established [1–3]. The generator of transitions
among unitarily inequivalent representations of the canonical commutation
relations (CCR) (phase transitions) is shown to provide ordered sequences
of phases (the system history), describing classical chaotic trajectories [4]
in the space of the representations of the CCR, which defines the arrow of
time.
In the framework of the theory of entire analytical functions, the func-
tional realization of deterministic fractals has been discussed in terms of
the q-deformed Hopf algebra of coherent states. The dynamical process
controlling coherent boson condensation at a microscopic level generates
the self-similarity properties characterizing fractals (including logarithmic
spirals). Quantum dissipation is at the origin of the deformation, or squeez-
ing, of coherent states which manifests itself in the self-similarity properties
at the macroscopic level.
The discussion involves couples of damped and amplified harmonic
oscillators. The amplified oscillators are the time-reversed image, or the
“Double”, of the damped oscillators [5–7]. These last ones constitute a pro-
totype of a dissipative system and their double oscillators represent the
environment, so that the couple of the damped oscillator and its double
constitute a closed system.
The results are consistent with the ones which show that the so-called
“defects”, or “extended objects” in condensed matter physics and high
energy physics (such as kinks, vortices, monopoles, crystal dislocations,
domain walls, etc.) are generated by coherent quantum condensation at
the microscopic level. Extended objects appear to be macroscopic quantum
systems, in the sense that their macroscopic properties cannot be derived
without recurring to the underlying quantum dynamics [8, 9].
The damped harmonic oscillator and its doubled image turn out to be
equivalent to the system of a matter field interacting with the EM field
when the magnetic field may be considered to be constant and the electric
field is derivable from a harmonic potential (Sec. 3). This sheds some light
on the possibility [2, 9] of considering the EM field behaving as a reservoir
able to compensate the energy dissipated by the matter field (or vice versa).
The quantization of the damped/amplified oscillator system is summarized
in Sec. 4 and its relation with self-similar fractal-like structure and general-
ized squeezed coherent states is considered in Sec. 5. There, the connection
with non-commutative geometry is also briefly recalled. The emerging pic-
ture is the one of the isomorphism formally unifying fractal self-similarity,
468 G. Vitiello
dissipation, and electrodynamics in the frame of coherence in QFT. The

crucial algebraic ingredient is the non-commutativity of deformed Hopf
algebra.
The coherent dynamics at the basic quantum field level leads us to
an integrated vision of Nature [1, 10]. Coherence and dissipation provide
the universal morphogenetic paradigm. This is commented upon in Sec. 6.
In the next section, as a preliminary to the discussion to follow, the conser-
vation of the energy–momentum tensor in electrodynamics is considered.
3. The Conservation of the Energy–Momentum

Tensor in Electrodynamics
As a preliminary observation, let me recall [11, 12] how the conservation
of the EM energy–momentum tensor, T µν , ∂µ T µν = 0, arises from the
µν
compensation of the variations of the matter part Tm and the em part
µν µν µν µν
Tγ of the total T : ∂µ Tm = −∂µ Tγ . In a standard fashion, by closely
µν
following, e.g. Ref. [12], consider the general expression of Tm and Tγµν :
µν ∂L
Tm = Dν ψ α + h.c. − Lm g µν, (1)
∂(Dµ ψ α )
∂L βν 1
Tγµν = F gαβ + F αβ Fαβ g µν
∂ Fαµ 4
1
= −F αµ F βν gαβ + F αβ Fαβ g µν, (2)
4
where L and ψ α (x), with x ≡ (r, t), r = (x1 , x2 , x3 ), denote the Lagrangian
density and the matter field, respectively. Dµ is the covariant derivative,
Dµ = ∂ µ + i e Aµ , g µν = (1, −1, −1, −1), µ = 0, 1, 2, 3, h.c. stays for her-
mitian conjugate and Lm ≡ L + 14 F αβ Fαβ , with F αβ = ∂ β Aα − ∂ α Aβ . I
use = 1 = c. Note that Eqs. (1) and (2) hold for classical and quantum
electrodynamics (QED) (in this last case with some care on the field (oper-
ator) ordering [11]. For our task, there is no need to specify the boson or
fermion nature of ψ α (x). I only assume that the Lagrangian L is invariant
under the gauge transformations:

ψ α (x) → ψ α (x, t) = ei e Λ(x) ψ α (x, t)

Aµ → Aµ = Aµ − ∂ µ Λ(x). (3)
µ
Let J denote the current density

µ α† ∂L ∂L α
J =i ψ − ψ . (4)
∂(Dµ ψ α )† ∂(Dµ ψ α )
www.ebook3000.com
The charge density and current are e ρ and e J = e ρ v, respectively, with

µν
v = dr/dt the velocity. One finds that ∂µ Tm and ∂µ Tγµν have non-vanishing
opposite values, i.e.
µν
∂µ Tm = e F αν Jα , (5)
∂µ Tγµν = −e F αν
Jα ,
so that the total T µν for the closed system {ψ, Aµ } is conserved: ∂µ T µν =

µν
∂µ (Tm + Tγµν ) = 0. Thus, the non-vanishing divergences of Tm µν
and Tγµν
compensate each other, or, in different words, the em field acts as the reser-
voir for the dissipative matter field system (or vice versa). I remark that
the set of Eq. (5) fully characterizes classical electrodynamics and QED,
they express the whole dynamical content of Maxwell equations and the
constant of motion properties. Recall that the energy–momentum vector
P µ is

P µ = T µ0 d3 x, µ = 0, 1, 2, 3 (6)
and
Ei = F 0 i, B i = ijk F k j , i, j, k = 1, 2, 3, (7)
where E i and B i are the i-component, i = 1, 2, 3, of the electric and mag-

netic fields, E and B, respectively. Volume integration of the set of Eq. (5)
gives for ν = 0 the rate of changes in time of the energy of the matter field
and em field, Em and Eγ , respectively:
∂0 Em = e E · v = −∂0 Eγ . (8)
For ν = i = 1, 2, 3, integration of the set of Eq. (5) over the volume gives
i
∂0 Pm = e E i + e (v × B)i , (9)
∂0 Pγi = −e E i − e (v × B)i (10)
which show that the Lorentz forces Fm and Fγ , acting on two opposite
charges with same velocity v in the same E and B fields, are equal and oppo-
site, component by component, as it should be. The meaning of Eqs. (9)
and (10) (and the one of Eq. (8)) is that the conservation of the energy-
momentum vector is fulfilled only provided the matter field is considered
together with the em field (and vice versa). Each of these fields, separately
considered, behaves as an open system. Only the whole system, made of
both fields, is non-dissipative. This result leads us to the relation of the
470 G. Vitiello
(classical and quantum) electrodynamics with a damped harmonic oscilla-

tor system and its time-reversed image. This is discussed in the following
section.
4. The Damped Harmonic Oscillator,

its Time-Reversed Image and Electrodynamics
Suppose that at least in some limited space–time region the magnetic field
B can be approximated to be a constant vector. It is described by the vector
potential
1
A= B×r (11)
2
with r = (x1 , x2 , x3 ), which gives, indeed, B = ∇ × A. Moreover, ∇ · A = 0.
Without loss of generality, one may choose the reference frame in such a
way that B is anti-collinear to the third axis, B = ∇ × A = −B 3̂. Then,
from Eq. (11), we have A3 = 0 and
B
Ai = ij xj , i, j = 1, 2 (12)
2
with 12 = −21 = 1; ii = 0. With such a choice, the third component,
i = 3, of (v × B) vanishes. We also assume that the electric field E is given
k
by the gradient of the harmonic potential Φ ≡ 2e (x1 2 − x2 2 ) ≡ Φ1 − Φ2 ,
E = −∇Φ; E3 = 0 and we may thus limit our analysis to the i = 1, 2
components in Eqs. (9) and (10). Consider then i = 1 in Eq. (9). Use of
Eq. (12) and putting B ≡ γ/e gives [2]
mẍ1 + γ ẋ2 + kx1 = 0, (13)
where the force in the first member of Eq. (9) has been put equal to mẍ1
(and is equal and opposite to the i = 1 component of the force in Eq. (10)).
m, γ, and k are time independent quantities. In order to separate the x1
and x2 variables in (13) it is convenient to consider i = 2 in Eq. (10), which
gives:
mẍ2 + γ ẋ1 + kx2 = 0. (14)
The fact that Eqs. (13) and (14) describe opposite charges in a constant
magnetic field and a harmonic scalar potential is also evident since they
can be derived from the Lagrangian
m
L = (ẋ21 − ẋ22 ) + e(ẋ1 A1 + ẋ2 A2 ) − eΦ (15)
2
www.ebook3000.com
which exhibits the correct coupling between the current and the vector
potential field and can be written also as
1 1 e2
L= (mẋ1 + e1 A1 )2 − (mẋ2 + e2 A2 )2 − (A1 2 − A2 2 ) − eΦ.
2m 2m 2m
(16)
Equations (13) and (14) are derived from Eq. (16) in the familiar form
d
(mẋi + ei Ai ) = −e∂i Φi + ei ∂i vj Aj , (17)
dt
k 2 ∂
where i, j = 1, 2, i = j, no sum on i, j, Φi ≡ 2e xi , e1 = e = −e2 , ∂i ≡ ∂xi
d
and it is dt Ai = vj ∂j Ai . The Hamiltonian is
1 1
H = H1 − H2 = (p1 − e1 A1 )2 + e1 Φ1 − (p2 + e2 A2 )2 + e2 Φ2
2m 2m
(18)
which explicitly shows that x2 behaves as the em field for x1 in H1 , and vice
versa in H2 [2, 9]. The respective contributions to the energy compensate
each other in the least energy state (where H = 0, H1 = H2 ). The x1 and
x2 variable separation in Eqs. (13) and (14) is now obtained by using the
canonical transformations [5]
x1 (t) + x2 (t) x1 (t) − x2 (t)
x(t) = √ , y(t) = √ . (19)
2 2
We thus obtain the couple of damped and amplified harmonic oscillators
(dho):
mẍ + γ ẋ + kx = 0, (20)
mÿ − γ ẏ + ky = 0, (21)
respectively. Note that Eq. (21) is the time-reversed image (γ → −γ) of (20)
and the global system (x−y) is a closed system. One usually refers to the y
oscillator as to the double of the x oscillator. Consistently with the remark
on the (em field) reservoir in the previous section, also in the present case
the y oscillator may be considered as the reservoir (or the environment)
for the x oscillator (or vice versa). The y oscillator, in other words, acts
as the bath to which the x oscillator is coupled or the sink into which the
energy dissipated by the x oscillator flows. It is indeed easy to show that
by introducing the pseudo-euclidean metrics
2 2
r(t)2 ≡ x1 (t) − x2 (t) , (22)
472 G. Vitiello
namely,
1 1
x(t) = √ r(t)eu(t) , y(t) = √ r(t)e−u(t) , (23)
2 2
Equations (20) and (21) are formally equivalent to the equation for the
harmonic oscillator r(t) representing the global (x−y)-system:
mr̈ + Kr = 0 (24)
2
2
1 γ γ γ
with K ≡ m k − 4m ≡ mΩ2 , assuming k > 4m , provided u(t) ≡ − 2m t,
as required by the time independence of the coefficients m, γ, and k. Vice
versa, the oscillator (24) is decomposed into two damped/amplified oscilla-
tors (20) and (21) when the pseudo-euclidean metrics is adopted [5]. If the
euclidean metrics, r2 ≡ x21 + x22 , x1 = r cos α, x2 = r sin α, is chosen, the
r-oscillator is decomposed into two undamped oscillators.
The Lagrangian (15) and the Hamiltonian (18), rewritten in terms of
the (x − y)-system, become
γ
L = mẋẏ + (xẏ − ẋy) − kx y, (25)
2

1 1 γ2
H = px py + γ (ypy − xpx ) + k − x y, (26)
m 2m 4m
respectively, with conjugate momenta px = mẏ − γ2 y, and py = mẋ + γ2 x
12
1 γ2
and the common frequency of the two oscillators Ω ≡ m k − 4m ,
2
γ
k > 4m (i.e. assuming no overdamping). Note that similar conclusion may
be reached considering i = 2 in Eq. (9) and i = 1 in Eq. (10). Note also
that conjugate momenta cannot be defined without introducing the doubled
mode y.
In summary, in dissipative systems the time-reversal symmetry is broken
and a partition on the time axis is induced, implying that positive and
negative time directions are associated with separate modes, describing
different physical evolutions (damping and amplification). The system made
of the two separate modes is a closed system. The canonical formalism is
not able to describe separately each one of the modes. It can only describe
the closed system. It is interesting that the two separate non-conserving
modes Eqs. (9) and (10), (and (8)) out of which electrodynamics is made,
are associated with charge conjugation (e ↔ −e).
I will briefly comment in the following section on the quantization of
the damped/amplified oscillator system (20) and (21).
www.ebook3000.com
5. Quantum Dissipation
Canonical quantization of the system of oscillators Eqs. (20) and (21)
has been considered in the literature [5]. Here, I will summarize some
aspects which may be useful to clarify the general setting of the discus-
sion presented in this chapter. One starts by introducing the commutators
[ x, px ] = i = [ y, py ], [ x, y ] = 0 = [ px , py ] and the sets of annihilation
and creation operators
12 12
1 px √ 1 px √
a≡ √ − i mΩx ; a† ≡ √ + i mΩx ,
2Ω m 2Ω m
(27)
12 12
1 p √ 1 p √
b≡ √ y − i mΩy ; b† ≡ √ y + i mΩy
2Ω m 2Ω m
(28)
with [ a, a† ] = 1 = [ b, b† ], [ a, b ] = 0 = [ a, b† ]. One then considers

the canonical linear transformations A ≡ √12 (a + b), B ≡ √12 (a − b), and
γ
defines Γ ≡ 2m . The Hamiltonian H is given in terms of A and B by [5]
H = H0 + HI , (29)
† † † †
H0 = Ω(A A − B B), HI = iΓ(A B − AB). (30)
†
Note that, J+ = A† B † , J− = J+ = AB, J3 = 12 (A† A + B † B + 1),
[ J+ , J− ] = −2J3 , [ J3 , J± ] = ±J± provide the two-mode realization of
the algebra SU(1, 1). The SU (1, 1) Casimir operator C is given by C 2 =
1 † † 2
4 (A A − B B) , so that [ H0 , HI ] = 0. The vacuum state is |0 ≡ |nA =
0, nB = 0 = |0 ⊗ |0, with nA and nB the number of A and B’s and
(A ⊗ 1)|0 ⊗ |0 ≡ A|0 = 0; (1 ⊗ B)|0 ⊗ |0 ≡ B|0 = 0. Its time
evolution is controlled by HI and given by
H HI 1 † †
|0(t) = e−i t |0 = e−i t
|0 = etanh (Γt)A B |0, (31)
cosh (Γt)
0(t)|0(t) = 1 ∀t, (32)
lim 0(t)|0 ∝ lim exp (−tΓ) = 0. (33)

t→∞ t→∞
Once the initial condition of positiveness for the eigenvalues of H0 is

set, there is no danger of transitions to negative energy states since such
a condition is preserved by the time evolution. H0 is indeed the Casimir
operator and commutes with HI .
474 G. Vitiello
The creation of B modes turns out to be equivalent, up to a statistical

factor, to the destruction of A modes; states generated by B † represent
the sink where the energy dissipated by the quantum damped oscillator
flows: the B-oscillator represents the reservoir or heat bath coupled to the
A-oscillator [5].
The instability (decay) of the vacuum is expressed by Eq. (33). Thus,
time evolution leads out of the Hilbert space of the states, which means
that the framework of quantum mechanics is not suitable for the canonical
quantization of the damped harmonic oscillator. It has been shown that the
proper framework is the one of QFT [5], where the time evolution operator
U(t) and the vacuum are formally (at finite volume) given by

U(t) = exp Γκ t A†κ Bκ† − Aκ Bκ , (34)
κ
1
|0(t) = exp tanh (Γκ t)A†κ Bκ† |0, (35)

κ
cosh (Γκ t)
respectively,
with 0(t)|0(t) = 1, ∀t. In the infinite volume limit, we have
(for d3 κ Γκ finite and positive)
0(t)|0 → 0 as V → ∞ ∀ t, (36)
V

where use of the relation κ → (2π) 3 d3 κ has been made. In general,
0(t)|0(t ) → 0 as V → ∞ ∀ t and t , t = t: at each time t a represen-

tation {|0(t)} of the CCR is defined and is unitarily inequivalent to any

other representation {|0(t ), ∀t = t} in the infinite volume limit. The sys-
tem thus evolves in time through unitarily inequivalent representations of
CCR along trajectories which can be shown to be classical chaotic trajecto-
ries [4]. Note that |0(t) is a two-mode time dependent generalized SU (1, 1)
coherent state [5, 18]. A and B are entangled modes and
NAκ (t) = 0(t)|A†κ Aκ |0(t) = sinh2 Γκ t (37)
which gives the number of Aκ -modes condensed in the vacuum |0(t) at time
t. It has to be stressed that in computing the value at time t of NAκ (t) in
Eq. (37) the only non-vanishing contributions are given by the Bκ -modes:
0(t)|A†κ Aκ |0(t) = 0(t)|Bκ (t)Bκ† (t)|0(t) = sinh2 Γκ t. (38)
This means that the B-modes act as the clock measuring the flow of time
of the A-modes. In this way, time emerges for the A-mode as a significant
variable measured by its double B-mode.
www.ebook3000.com
It can be shown that |0(t) is a squeezed coherent state whose algebraic

and dynamical properties are described by the q-deformation of Lie–Hopf
algebra [9]. As a matter of fact, the non-commutative q-deformed Lie–Hopf
algebra turns out to be a characterizing feature of the structure of QFT.
Moreover, it provides a representation of the CCR at finite temperature
which is equivalent [5] to the thermo field dynamics (TFD) representation
{|0(β)} [13]. QFT thus appears to be intrinsically a thermal field theory
(TFT). I do not insist more on these aspects which have been throughout
discussed elsewhere [9].
In the present case, we have considered boson modes A and B. It is how-
ever possible to consider TFT representation {|0(β)} also for the fermion
case.
It is interesting to consider the time evolution of the one-particle state,
H HI HI
say e−it A†q |0 = e−iΩt A†q (t)|0(t), with A†q (t) = e−i t A†q ei t . We obtain
the probability amplitude 0|Aq e−iΩt A†q (t)|0(t) ∝ e−iΩt e−Γ̃t , with Γ̃ ≡

κ=q Γ, which leads in a standard fashion to the Breit–Wigner formula for
the probability density
Γ̃ 1
P (E) = (39)
2π (E − E0 )2 + (Γ̃/2)2
with E ≡ Ω and E0 the energy at the maximum of the Breit–Wigner
curve.
One also finds that time evolution is controlled by the entropy varia-
tions [5], which is consistent with the fact that dissipation implies break-
ing of time-reversal invariance, namely, the choice of a privileged direc-
tion in time evolution (the arrow of time). Heat dissipation dQ = β1 dS is
expressed by the variations in time of the number of particles condensed in
the vacuum.
As said, the vacuum |0(t) is an entangled state of the A and B modes,
which is consistent with the entanglement between the (charged) matter
field and the em field in QED. I remark that in QFT the entanglement
notion enters in a natural way through the coherent state structure of the
vacuum state (here, the coherent SU (1, 1) boson condensation of the couple
AB in |0(t)) and that entanglement cannot be destroyed by any unitary
operator action since it characterizes unitarily inequivalent representations.
Such a feature is absent in quantum mechanics.
I will not discuss further the quantization procedure. I only observe that
the classical system of oscillators considered above belongs to the class of
deterministic systems à la ’t Hooft [14–16], i.e. those systems that remain
476 G. Vitiello
deterministic even when described by means of Hilbert space techniques.

The quantum harmonic oscillator emerges from the classical (dissipative)
system when one imposes a constraint on the Hilbert space of the form
J2 |0 = 0. See Refs. [14–16] for further details on ’t Hooft analysis.
In the following section, I will show the relation between self-similarity
properties of fractal structures and the generalized coherent states entering
the quantum dissipation formalism. In view of the result presented earlier,
this establishes a link between fractal self-similarity and QED.
6. Fractal Self-Similarity, Dissipation and Coherent

States
In Refs. [1–3, 8], I have shown that self-similarity properties of determin-
istic fractals can be studied in the framework of the theory of entire ana-
lytical functions and their functional realization can be given in terms of
the q-deformed algebra of squeezed coherent states. Moreover, self-similar
structures are related to quantum dissipation and non-commutative geom-
etry [3,8]. According to these results, fractals appear as “extended objects”
or macroscopic quantum systems generated by coherent quantum conden-
sation processes at the microscopic level [1, 2, 8, 13]. The relation between
coherence, dissipation, and electrodynamics shown in the previous sections
thus also includes the self-similarity properties of fractal-like phenomena.
This leads to an integrated vision of Nature resting, in its essence, on the
paradigm of coherence and dissipation.
Let me discuss the examples of the Koch curve and the logarithmic
spiral [17]. Details can be found in Refs. [1, 8]. The conclusions can be
extended to other examples of deterministic fractals (which are generated
iteratively according to a prescribed recipe), such as the Sierpinski gasket
and carpet, the Cantor set, etc.
I will closely follow the presentation of Ref. [1]. In the construction of
the Koch curve (Fig. 1), put u0 = 1 for the starting stage u0 . Let the nth
step or stage be denoted by un,q (α), with α = 4 and q = 1/3d. One has
un,q (α) = (q α)n = 1, for any n, (40)
from which the self-similarity, or fractal dimension [17] d = ln 4/ ln 3 ≈

1.2619 is obtained. It needs to be stressed that self-similarity is properly
defined only in the n → ∞ limit. Consider now in full generality the complex
α-plane, and put q = e−d θ . Equation (40) √ is then written as d θ = ln α
and, apart from the normalization factor 1/ n!, the functions un,q (α) are
www.ebook3000.com
Fig. 1. The first five stages of Koch curve.
recognized to be the restriction to real q α of the functions

(q α)n
un,q (α) = √ , n ∈ N+ , qα ∈ C (41)
n!
which form a basis in the space F of the entire analytic functions. This
means that the fractal properties can be studied in F , by restricting, at
the end, the results to real q α, q α → Re(q α). In order to establish the
relation with coherent states it is enough to realize that F provides the
Fock–Bargmann representation of the Weyl–Heisenberg algebra [18], which
is the frame where (Glauber) coherent states are described. By setting q =
eζ , ζ ∈ C one may obtain the q-deformed algebraic structure [19], which we
do not discuss here. We only observe that by applying q N to the coherent
state |α, N ≡ α d/dα, one finds that
∞
N |qα|2 (qα)n
q |α = |qα = exp − √ |n. (42)
2 n=0 n!
478 G. Vitiello
Here, a|α = α|α, with a the annihilator operator. By applying (a)n to

|qα and restricting to real qα
qα|(a)n |qα = (qα)n = un,q (α), qα → Re(qα) (43)
the nth iteration stage of the fractal is obtained: the operator (a)n acts
as a “magnifying” lens. Thus, the one-to-one correspondence is established
between the fractal nth stage of iteration, n = 0, 1, 2, . . . , ∞, and the n-
th term in the q-deformed coherent state series Eq. (42). |qα turns out
to be a squeezed coherent state. ζ = ln q is the squeezing parameter and
q N , called the fractal operator, is the squeezing operator in F. In summary,
the coherent state squeezing transformation describes the self-similarity
properties of the Koch curve (and other fractals).
A similar result can be established for the logarithmic spiral (Fig. 2),
which is given in polar coordinates (r, θ) by [17]:
r = r0 ed θ , (44)
where r0 and d are arbitrary real constants and r0 > 0. Equation (44) is
represented by the straight line of slope d in a log–log plot with abscissa
θ = ln eθ :
r
d θ = ln . (45)
r0
The self-similarity property of the logarithmic spiral is represented by the
constancy of the angular coefficient tan−1 d. Rescaling θ → n θ affects r/r0
by the power (r/r0 )n . Consider the parametric equations of the spiral:
ξ = r(θ) cos θ = r0 ed θ cos θ, (46)
η = r(θ) sin θ = r0 ed θ sin θ. (47)
Fig. 2. The anti-clockwise and the clockwise logarithmic spiral.
www.ebook3000.com
Once the sign of d θ is assigned, in the complex z-plane the point z = ξ +

i η = r0 ed θ ei θ on the spiral is fully specified. The points z1 = r0 e− d θ e− i θ
and z2 = r0 e+ d θ e+ i θ are both considered since the completeness of the
(hyperbolic) basis {e− d θ , e+ d θ } requires that both elements q = e± d θ
must be taken into account. For convenience, opposite signs for the imagi-
nary exponent ±i θ also have been considered. Use now the parametrization
θ = θ(t). z1 and z2 can be shown to solve the equations (“dot” denotes
derivative with respect to t)
m z̈1 + γ ż1 + κ z1 = 0, (48)
m z̈2 − γ ż2 + κ z2 = 0, (49)
respectively, provided that
γ Γ
θ(t) = t = t, (50)
2md d
up to an arbitrary additive constant. m, γ, and κ are positive real constants.
Thus, the logarithmic spirals are described by z1 (t) = r0 e− i Ω t e−Γt and
z2 (t) = r0 e+ i Ω t e+Γ t solutions of Eqs. (48) and (49). The notations and
the quantities Γ ≡ γ/2m and Ω2 = (1/m)(κ − γ 2 /4m) = Γ2 /d2 , with κ >
γ 2 /4m, are the same as in Sec. 2. Also, by putting [z1 (t) + z2∗ (−t)]/2 = x(t)
and [z1∗ (−t) + z2 (t)]/2 = y(t), Eqs. (48) and (49) reduce to Eqs. (20) and
(21) (namely, they provide an equivalent representation of Eqs. (9) and
(10)). Note that θ(T ) = 2 π at T = 2 π d/Γ. At t = n T , z1 = r0 (e− 2 π d )n ,
z2 = r0 (e2 π d )n , with integer n = 1, 2, 3, . . . . We see that we can inter-
pret the parameter t as the time parameter. Note that the so-called direct
(q > 1) and indirect (q < 1) spirals are sometimes both realized in the same
system (examples are found in phyllotaxis studies). They are described by
the system of Eqs. (48) and (49) for the damped and amplified harmonic
oscillator. The spiral “angular velocity” is given by | d θ/dt | = | Γ/d |.
The spiral Lagrangian is given by Eq. (15) where x1 and x2 need to be
substituted by z1 and z2 , respectively.
By proceeding in a similar way as done in Sec. 4 we arrive at the Hamil-
tonian Eqs. (29) and (30) and also the other formulas for the evolution
operator, the ground state, etc., Eqs. (34)–(37) are obtained when working
in the proper frame of QFT.
Also, in the present case, the breakdown of time-reversal symmetry is
associated with the choice of a privileged direction in time evolution and
the entropy operator S may be defined. The indirect spiral (right-handed
480 G. Vitiello
chirality) is the time-reversed, but distinct, image of the direct spiral (left-
handed chirality). The Hamiltonian H is actually the fractal free energy for
the coherent boson condensation process out of which the fractal is formed.
By identifying H0 = 2 Ω C with the “internal energy” U and 2 J2 with the
entropy S, from Eqs. (30) and the defining equation for the temperature T
(putting kB = 1), we have ∂ S/∂ U = 1/T and obtain T = Γ. Thus, H
represents the free energy F = U − T S. The heat contribution in F is given
by 2 Γ J2 and (∂ F /∂ T )|Ω = −2 J2 . The temperature T = Γ is found to
be proportional to the background zero point energy: Γ ∝ Ω/2 [9, 16].
d (π/2)
√and assume r/r0 = e
In Eq. (44), let θ = π/2 ≡ φ, with φ denoting
the golden ratio, φ = (1+ 5)/2. Put dg ≡ (ln φ)/(π/2), where the subscript
g stays for golden. The logarithmic spiral is then called the golden spiral [17]
and its polar equation is rg (θ) = r0 edg θ . As θ grows of π/2, the radius
of the golden spiral grows in geometrical progression of ratio φ: rg (θ +
n π/2) = r0 edg (θ+n π/2) = r0 edg θ φn and rg,n ≡ rg (θ = n π/2) = r0 φn ,
n = 0, 1, 2, 3, . . . .
A good “approximate” construction of the golden spiral is obtained by
drawing in a proper way (Fibonacci tiling) [17] squares whose sides are in
the Fibonacci progression, 0, 1, 1, 2, 3, 5, 8, 13, . . . . (the Fibonacci generic
number is Fn = Fn−1 + Fn−2 , with F0 = 0; F1 = 1). The Fibonacci spiral
is then obtained from quarter-circles tangent to the interior of each square.
It does not perfectly overlap with the golden spiral since Fn /Fn−1 → φ in
the n → ∞ limit, but is not equal to φ for given √ finite n. The golden ratio
φ and its “conjugate” ψ = 1 − φ = −1/φ = (1 − 5)/2 are solutions of the
“quadratic formula”:
x2 − x − 1 = 0 (51)
and of the recurrence equation xn −xn−1 −xn−2 = 0, which, for n = 2, is the

relation (51). This is satisfied also by the geometric progression of ratio φ of
the radii rg,n = r0 φn of the golden spiral. Equation (51) is the characteristic
equation of the differential equation r̈√+ ṙ − r = 0, which admits as solution
r(t) = r0 ei ω t e+d θ(t) with ω = ± i 5/2 and θ = √ −t/(2 d) + c, with c,
r0 , and d constants. By setting c = 0, r(t) = r0 e∓ 5 t/2 e−t/2 , i.e. rφ (t) =
r0 e−φ t and rψ (t) = r0 e−ψ t .
Going back to the evolution operator U(t), note that, when written in
terms of the a and b operators (see Sec. 4) it becomes

Γt 2 †2

2 †2

U(t) = exp − a −a − b −b (52)

2
www.ebook3000.com
and it is recognized to be the two-mode squeezing generator with squeezing

parameter ζ = −Γ t. As previously said, the SU (1, 1) generalized coherent
state (35) is thus a squeezed state.
In the case of the Koch curve, by doubling the degrees of freedom (DoF)
in the operator q N one also obtains the “doubled” exponential operator
(c2 − c†2 ) − (c̃2 − c̃†2 ) = −2 C † D† − CD , (53)
where c̃ and c̃† denote the doubled DoF and C ≡ √12 (c + c̃), D ≡ √12 (c − c̃).
The fractal operator has thus the same form of the dissipative time evolution
operator exp(−itHI /) and the description in terms of generalized SU (1, 1)
coherent state is recovered also in the case of the Koch curve [1].
One may reach the same result by using q = e−d θ , with d the fractal
dimension, in Eq. (40). Then the self-similarity equation q α = 1 is written
in polar coordinates as u = u0 α ed θ , which is similar to Eq. (44). As done
in the case of the logarithmic spiral, the parametric equations for the fractal
in the z-plane can be written, and so on to obtain the relation (53), the
fractal Hamiltonian and free energy and the SU (1,1) generalized coherent
state.
Let me recall that the oscillator z1 is an open (non-Hamiltonian) sys-
tem and in order to set up the canonical formalism one needs to double the
DoF by introducing its time-reversed image z2 and consider then the closed
system (z1 , z2 ) [5]. This justifies on a physical ground the mentioned math-
ematical necessity to consider both the elements of the basis {e− d θ , e+ d θ }.
Now, I briefly mention about non-commutative geometry which arises
as an effect of dissipation [1, 2, 8]. I will use the notation + ≡ 1 and − ≡ 2
in the (z1 , z2 ) plane. pz± denote the momenta and v± = ż± the forward in
time and backward in time velocities. These are given by
1 1 γ
v± = (pz∓ ∓ γz± ), with [v+ , v− ] = −i (54)
m 2 m2
which in fact immediately provides a relation between dissipation and non-

commutative geometry in the plane. A canonical set of conjugate position
coordinates (ξ+ , ξ− ) may be defined indeed by putting ξ± = ∓(m/γ)v± , so
that
1
[ξ+ , ξ− ] = i (55)
γ
482 G. Vitiello
which characterizes the non-commutative geometry in the plane (z+ , z− ).

The quantum dissipative interference phase ϑ = S γ is associated with the
two paths P1 and P2 enclosing the area S in the non-commutative plane,
provided z+ = z− .
Also associated with non-commutativity is the algebraic structure of
the doubling of the DoF above considered in order to close the system. The
map A → A1 ⊗ A2 , which duplicates the algebra, is the Hopf coproduct
map A → A ⊗ 1 + 1 ⊗ A, and the Bogoliubov transformations of “angle”
Γ t are obtained by convenient combinations of the deformed coproducts
in the q-deformed Hopf algebra [2, 9]. These deformed coproduct maps are
non-commutative. The q-deformation parameter is related to the coherent
condensate content of the state |0(t). This sheds some light on the physical
meaning of the relation between dissipation (which is at the origin of q-
deformation) and non-commutative geometry. The non-trivial topology of
paths in the phase space provides the physical meaning of the deformed
Hopf algebraic structure into play.
As discussed in Sec. 3, the system of damped/amplified harmonic oscil-
lators Eqs. (48) and (49) is isomorph to QED under proper conditions, or,
equivalently, one of the oscillator may be considered to represent the em
field in which the other one is embedded. In the case of fractals, this tells
us that, e.g. z1 evolves in the em field represented by z2 (and vice versa).
The conclusion is that the macroscopic appearances (forms) of the frac-
tals seem to emerge out of a process of morphogenesis as the macroscopic
manifestation of the underlying dissipative, coherent quantum dynamics at
the elementary level. A much interesting aspect of such a conclusion is its
strict relation with the very same structure of electrodynamics through the
isomorphism discussed in previous sections.
7. Concluding Remarks. Towards an Integrated

Vision of Nature
In this chapter, it has been shown that the formalism of classical and QED
is isomorph to the one describing a system of damped/amplified oscilla-
tors when the magnetic field may be approximated to be constant (which,
at least in a limited space–time region, excluding critical singular behav-
iors, is always possible) and the electric field is derivable from a harmonic
potential. Such a system of oscillators is equivalent to TFD [13] and is char-
acterized by the q-deformation of Lie–Hopf algebra and squeezed SU (1,1)
www.ebook3000.com
coherent states. Moreover, dissipation induces non-commutative geometry

at a quantum level.
The reverse statement is also true: a system of damped harmonic
oscillators and the environment in which it is embedded, formally repre-
sented as the time-reversed copy of the damped oscillator system, may
be described in terms of the formalism of EM where charge densities and
currents (the matter field) represent the dissipative system and the EM
field the bath or reservoir in which it is embedded. Such a conclusion may
also be reached by considering the algebraic structure of TFD. Dissipa-
tion seems thus to imply always an associated EM field which behaves as
a reservoir compensating energy and momentum dissipated by the matter
field.
In such a scenario, one may also include the isomorphism existing
between self-similar fractal-like structures, including logarithmic spiral and
golden spiral, and the system of damped/amplified oscillators in QFT [1].
The quantum dynamics underlying the formation and the evolution of the
squeezed coherent vacuum manifests itself at a macroscopic level in the
fractal self-similar form (morphogenesis). Fractals thus appear as macro-
scopic quantum systems originated from the process of boson condensation
in the squeezed coherent state. The squeezing or deformation q-parameter
controls the fractal dimension d, q = ed θ . Quantum dissipation and non-
equilibrium characterize the fractal formation (growth). Entropy controls
the system time evolution, consistently with the breakdown of time-reversal
symmetry (the arrow of time).
On the other hand, by inverting the reasoning, the isomorphism also
implies that coherent states have fractal properties, namely, there is a “self-
similar geometry” characterizing coherent states [1, 2, 8].
The proofs of the isomorphisms discussed in the present chapter are
based on algebraic arguments and are model independent. Therefore, pro-
vided that some conditions are satisfied, they guarantee the general valid-
ity of the isomorphisms, beyond specific aspects of the considered systems,
and account for the universality of the recurrence in space and in time of
self-similar patterns. These results rest on the QFT mechanism of sponta-
neous breakdown of symmetry (SBS). Such an SBS mechanism describes
in terms of coherent boson condensation [9, 13] topologically non-trivial
“extended objects”, so-called “defects”, in condensed matter physics, high
energy physics, and cosmology and has been tested successfully in many
experimental observations, from the discovery of the Higgs particle to super-
conductors, ferromagnets, water under specific conditions, etc. A further
484 G. Vitiello
example relating fractal self-similarity to deformed coherent states is the

one of the “emergence of fractal dislocation structures” in the lattice of crys-
tals submitted to stress actions, which is indeed the macroscopic outcome
of non-homogeneous coherent phonon condensation.
The above results can be now collected in a unified perspective. The gen-
eration of coherence through SBS is indeed a very general dynamic process.
A unifying vision of natural phenomena is then generated by recognizing
that the transition from the microscopic scale of the elementary components
of a system to its macroscopic scale may be derived through a dynamic
process. The key point is the formation of coherent structures described by
QFT. The emerging conception is then not the one of Nature divided in
separated domains, each one closed in itself, constrained by the US, with
time being a dummy variable, but the vision of Nature free from US, uni-
fied by the laws of the underlying quantum field dynamics of the coherent
vacuum, with time being a true dynamical variable, an integrated vision of
Nature [1, 2].
One more suggestion pointing to a unifying vision comes from the ubiq-
uity of fractals in Nature, namely of scale-free, self-similar processes (Nature
“loves” fractals). Many observations and studies show that self-similar frac-
tal structures occur in solid state physics, earth science, medical sciences,
biology, clustering of galaxies, etc. [17, 20], and are also observed in brain
studies [1, 21, 22]. Self-similar structures persistently occur during the bil-
lions of years of the Universe evolutionary life. Strikingly self-similar pat-
terns appear in living and non-living systems in far apart sites in the world
and thousands years apart, which suggests that nonlinear, dissipative laws
of form are at work at a basic level accounting for morphogenetic pro-
cesses. Some EM properties of dilute aqueous solutions of DNA fragments
of viruses and bacteria also point to the crucial role of nonlinear, fractal-like
dynamics ruling the DNA duplication in polymerase chain reaction (PCR)
processes [23, 24]. The DNA genetic code appears to be the result, not the
cause of coherent behavior in living matter, the DNA macromolecule being
the vehicle allowing the propagation of the coherent dynamics in biolog-
ical systems [1, 2]. Coherence provides the translation into meanings, the
nested long range correlations among the system components, each com-
ponent acquiring indeed its meaningful dynamical function as a member of
the correlated component assembly. The (Shannon) syntactic level of pure
information is thus promoted by coherence to the (semantic) level of mean-
ings. Deformations (squeezing), due to dissipation, of the coherent state
may then be responsible of epigenetic modifications, in a similar way as the
www.ebook3000.com
one discussed earlier, by which fractals appear to emerge as the macroscopic

result of microscopic coherent local deformation processes [1, 8]. Perhaps,
one might conclude with Darwin that “[. . .] in this view of life, with its sev-
eral powers, having been originally breathed into a few form or into one; [. . .]
from so simple a beginning endless forms most beautiful and most wonderful
have been, and are being, evolved.” [25].
References
1. G. Vitiello, Fractals, coherent states and self-similarity induced noncommu-
tative geometry, Phys. Lett. A 376, 2527–2532 (2012).
2. G. Vitiello, On the isomorphism between dissipative systems, fractal self-
similarity and electrodynamics. Toward an integrated vision of Nature, Sys-
tems 2, 203–216 (2014).
3. G. Vitiello, Fractals and the Fock–Bargmann representation of coherent
states, in Quantum Interaction, P. Bruza, D. Sofge et al. (eds), Lecture
Notes in Artificial Intelligence, R. Goebel, J. Siekmann, W. Wahlster (eds.),
Springer-Verlag, Berlin, Heidelberg, 2009, pp. 6–16.
4. G. Vitiello, Classical chaotic trajectories in quantum field theory, Int. J. Mod.
Phys. B 18, 785–792 (2004).
5. E. Celeghini, M. Rasetti, and G. Vitiello, Quantum dissipation, Annals. Phys.
215, 156–170 (1992).
6. G. Vitiello, Dissipation and memory capacity in the quantum brain model,
Int. J. Mod. Phys. B 9, 973–989 (1995).
7. G. Vitiello, My Double Unveiled, John Benjamins, Amsterdam, 2001.
8. G. Vitiello, (2011). Topological defects, fractals and the structure of quantum
field theory, in Vision of Oneness, I. Licata and A. J. Sakaji (eds.), Aracne
Edizioni, Roma 2011, pp. 155–180.
9. M. Blasone, P. Jizba, and G. Vitiello, Quantum Field Theory and its Macro-
scopic Manifestations, Imperial College Press, London, 2011.
10. G. Vitiello, Struttura e funzione Una visione ecologica integrata, Rivista di
Filosofia Neo-Scolastica 4, 625–637 (2012).
11. S.S. Schweber, An Introduction to Relativistic Quantum Field Theory, Harper
and Row Publ. Inc., New York, 1961.
12. J. Leite Lopes, Gauge Field Theories, Pergamon Press, Oxford, 1983.
13. H. Umezawa, Advanced Field Theory: Micro, Macro and Thermal Concepts
American Institute of Physics, N.Y., 1993.
14. G. ’t Hooft, Quantum gravity as a dissipative deterministic system, Class.
Quant. Grav. 16, 3263–3279 (1999).
15. G. ’t Hooft, A mathematical theory for deterministic quantum mechanics,
J. Phys.: Conf. Series 67, 012015 (1–15) (2007).
16. M. Blasone, P. Jizba, and G. Vitiello, Dissipation and quantization, Phys.
Lett. A 287, 205–210 (2001).
486 G. Vitiello
17. H.O. Peitgen, H. Jürgens, and D. Saupe, Chaos and Fractals. New Frontiers
of Science, Springer-Verlag, Berlin, 1986.
18. A. Perelomov, Generalized Coherent States and Their Applications, Springer-
Verlag, Berlin, 1986.
19. L.C. Biedenharn, M.A. Lohe, An extension of the Borel–Weil construction
to the quantum group Uq (n), Comm. Math. Phys. 146, 483–504 (1992).
20. M.A. Selvam, Quasicrystalline pattern formation in fluid substrates and phyl-
lotaxis, in Symmetry in Plants. World Scientific Series No 4 in Mathematical
Biology and Medicine D. Barabe and R.V. Jean, (eds.), World Scientific, Sin-
gapore, 1998, pp. 795–809.
21. W.J. Freeman, and J. Zhai, Simulated power spectral density (PSD) of back-
ground electrocorticogram (ECoG), Cogn. Neurodyn. 3(1), 97–103 (2009).
22. G. Vitiello, Coherent states, fractals and brain waves, New Mathematics and
Natural Computation 5, 245–264 (2009).
23. L. Montagnier, J. Aı̈ssa, E. Del Giudice, C. Lavallee, A. Tedeschi, G. Vitiello,
DNA waves and water, J. Phys. Conf. Ser. 306, 012007 (1–10) (2011).
24. L. Montagnier, E. Del Giudice, J. Aı̈ssa, C. Lavallee, S. Motschwiller, A.
Capolupo, A. Polcari P. Romano, A. Tedeschi, G. Vitiello, Electromagnetic
Biology and Medicine, 2015, in press.
25. C. Darwin, On the Origin of Species, John Murray, London, 1860, p. 490.
www.ebook3000.com
Chapter 18
The Emergence of Space–Time:

Transactions and Causal Sets
Ruth E. Kastner
University of Maryland, College Park, MD USA
rkastner@umd.edu
A transactional account of the emergence of space–time events from a

quantum substratum is presented. In this account, space–time is not a
substantive manifold that becomes occupied with events; rather, space–
time itself exists only in virtue of specific actualized events. This implies
that space–time is discrete rather than continuous, and that properties
attributed to space–time based on the notion of a continuum are ideal-
izations that do not apply to the real physical world. It is further noted
that the transactional picture of the emergence of space–time can pro-
vide the quantum dynamics that underlie the causal set approach as
proposed by Sorkin and others.
1. Introduction and Background

The transactional interpretation (TI) of quantum mechanics was first pro-
posed by John G. Cramer [1]. Cramer showed how the interpretation gives
rise to a physical basis for the Born Rule for probabilities of measurement
outcomes. TI was originally inspired by the Wheeler–Feynman (WF) time-
symmetric, “direct action” theory of classical electrodynamics [2]. The WF
theory proposed that radiation is a time-symmetric process, in which a
charge emits a field in the form of half-retarded, half-advanced solutions
to the wave equation, and the response of absorbers combines with that
primary field to create a radiative process that transfers energy from an
emitter to an absorber. Davies later developed a quantum relativistic ver-
sion of the WF theory [3]. The present author has extended Cramer’s TI
into the relativistic domain based on the Davies theory [4]. An additional
element of this extension is to take quantum states and their interactions
487
488 R. E. Kastner
as describing pre-space–time possibilities, rather than as process occurring

in space–time. This new version of TI is called “Possibilist Transactional
Interpretation” or PTI.
It should perhaps be noted that the direct action picture of fields has
historically been somewhat neglected. This has been due not only to its
counterintuitive time symmetric character, but also on the basis that the
fields are not quantized, and therefore the direct action formalism is gen-
erally not convenient for practical computations. But it is also well known
that quantum field theory (QFT) is beset with serious mathematical con-
sistency and conceptual problems; notably Haag’s Theorema (as well as the
divergences requiring renormalization). It is therefore certainly possible that
Nature’s actual behavior is accurately described by the direct-action theory.
The basic entities of TI are the “offer wave” (OW), the retarded solution
that corresponds to the usual quantum state |Ψ > emitted by a source, and
the “confirmation wave” (CW), the advanced solution < X|. The CW is the
response of absorber X to the component of the OW |Ψ > projected onto
the state |X >. As discussed in Ref. [4], Chapter 3, the response of a set
of absorbers (A,B,C. . . ) to an OW |Ψ > yields a physical referent for von
Neumann’s “Projection Postulate,” which specifies that under measurement
a pure state |Ψ > is transformed into a mixed state, i.e.:
|Ψ >< Ψ| → Σi | < Ψ|Xi > |2 |Xi >< Xi |, (1)
where the weight of each projection operator corresponding to outcome Xi
is just the Born Rule. This process is illustrated in Fig. 1.
PTI adopts this basic formulation and extends the transactional pic-
ture into the relativistic domain by identifying the coupling amplitudes
between fields as the basic amplitude for an offer (or confirmation) to be
generated (see Kastner, 2012, Chapter 6, and Kastner, 2014). In addition,
PTI proposes a growing universe picture, in which actualized transactions
are the processes by which space–time events are created from a substra-
tum of quantum possibilities. The latter are taken as the entities described
by quantum states (and their advanced confirmations); and, at a subtler
relativistic level, the virtual quanta.
In PTI, what we call “space–time” is no more and no less than the
causally connected set of emission and absorption events corresponding to
actualized transactions. Each actualized transaction defines a time-like (or
null) space–time interval whose endpoints are the emission and absorption.
a An instructive discussion of Haag’s Theorem and the challenge it poses for QFT is
found in Earman and Fraser [5].
www.ebook3000.com
The Emergence of Space–Time 489
〈X2|Ψ〉 | X2〉
| Ψ〉 2
E
〈Ψ| X2〉 〈 X2|
Fig. 1. An OW |Ψ can be resolved into to various components corresponding

to the properties of absorbers 1, 2, 3, . . . . The product of a particular OW com-
ponent Xi |Ψ|Xi with its corresponding CW component Ψ|Xi Xi | reflects the
Born Rule which tells us that the probability of the result corresponding to the
projection operator |Xi Xi | is equal to Xi |Ψ Ψ|Xi = |Xi |Ψ|2 .
The emission is always in the past with respect to the absorption; the
relationship between these two events corresponds directly to the “link” in
the causal set picture (described further below).
If a transaction involves a photon, the interval is null; if it involves a
quantum with finite rest mass, the interval is time-like. The intervals have
a causal relationship in that an absorption event A can, and generally does,
serve as the site of a new emission event B. Thus, the set of intervals created
by actualized transactions establish a causal network with a partial order,
much like the causal set structure proposed by Sorkin [6]. (The term “causal
set” is often abbreviated as “causet”.) We address the specifics of the causet
picture in the next section, but at this point, it is interesting to note the
similar antisubstantival picture in Sorkin’s presentation:
A basic tenet of causet theory is that space–time does not exist at the
most fundamental level, that it is an “emergent” concept which is rel-
evant only to the extent that some manifold-with-Lorentzian-metric M
furnishes a good approximation to the physical causet C.
(Sorkin, 2003, p. 9, preprint version)
An important feature of PTI is its relativistic extension of the basic

transactional picture. This extension gives an account of the generation of
490 R. E. Kastner
OWs, as an inherently stochastic process, from the direct action theory of

quantum fields (cf. Davies [3]). This author has proposed, independently
of the Sorkin’s work on the causet picture, that this process is inherently
Poissonian (i.e. based on decay rates). The basic idea is that offers and con-
firmations are spontaneously elevated forms of virtual quanta, where the
probability of elevation is given by the decay rate for the process in question.
In the direct action picture of PTI, an excited atom decays because one of
the virtual processes ongoing between the excited electron and an external
absorber (e.g. a ground state atom) is spontaneously transformed into an
OW that generates a confirming response. The probability for this occur-
rence is the product of the QED coupling constant α and the associated
transition probability (see Ref. [4]). In QFT terms, the OW corresponds
to a “free photon” or excited state of the field, instantiating a Fock space
state.b
When this process occurs, a set of incipient transactions is generally set
up, as more than one absorber is generally available to any emitted pho-
ton OW. Each incipient transaction represents a choice of momentum direc-
tion for the emitted photon, which is emitted as a spherical (isotropic) wave.
The Born Rule gives the probability that any particular incipient transac-
tion will be actualized, but with certainty one of them will be actualized.
Thus, when decay occurs, a new space–time interval will be created. This
corresponds to a new causally related pair of space–time events; the emis-
sion event is the ancestor, and absorption event is the descendant. Thus, the
Poissonian decay rates directly give rise to space–time events of the kind
envisioned in space–time causal sets. We now turn to that formulation.
2. Causal Sets
The motivation for the causal set program as an approach to the vexed
problem of quantum gravity is described by Sorkin as follows:
The causal set idea is, in essence, nothing more than an attempt to
combine the twin ideas of discreteness and order to produce a structure
on which a theory of quantum gravity can be based. That such a step
was almost inevitable is indicated by the fact that very similar formula-
tions were put forward independently in G. ’t Hooft [7], J. Myrheim [8],
and L. Bombelli et al. [9], after having been adumbrated in D. Finkel-
stein [10]. The insight underlying these proposals is that, in passing from
b However, the direct action theory does not assume an independently existing, infinite
set of field oscillators, which allows it to escape the problems associated with Haag’s
theorem; this issue is explored in a separate work.
www.ebook3000.com
the continuous to the discrete, one actually gains certain information,

because “volume” can now be assessed (as Riemann said) by counting;
and with both order and volume information present, we have enough
to recover geometry.
(Sorkin, 2003, p. 5)
A causal set (causet) C is a locally finite partially ordered set of ele-

ments, together with a binary relation ≺. It has the following properties:
(i) Transitivity: (∀ x, y, z ∈ C)(x ≺ y ≺ z ⇒ x ≺ z),
(ii) Irreflexivity: (∀x ∈ C)(x¬ ≺ x),
(iii) Local finiteness: (∀ x, z ∈ C) (cardinality {y ∈ C |x ≺ y ≺ z} < ∞).
Properties (i) and (ii) together imply that the elements are acyclic, while
(iii) specifies that the set is discrete rather than continuous. This naturally
leads to a well-defined causal order of distinct events, which can be asso-
ciated with the unidirectionality of temporal becoming. Again, in Sorkin’s
terms:
the relationship x ≺ y . . . is variously described by saying that x precedes
y, that x is an ancestor of y, that y is a descendant of x, or that x lies
to the past of y (or y to the future of x). Similarly, if x is an immediate
ancestor of y (meaning that there exists no intervening z such that x ≺
z ≺ y) then one says that x is a parent of y, or y a child of x, . . . or that
x ≺ y is a link.
(Sorkin, 2003, p. 7)
Again, as noted earlier, an actualized transaction defines a “par-

ent/child” relationship or link. Elements connected by such links are said
to be comparable, or members of a chain.
Sorkin discusses how to create a causal set structure as a “coarse-
graining” of a continuous space–time manifold M. The fundamental vol-
ume element of M corresponds to a single causal set element of C, so the
basic correspondence between a causet C and a continuous manifold M is
that N=V (where N is the number of causet elements approximating the
volume V). In this context, he further notes:
Given a manifold M with Lorentzian metric gab (which is, say, globally
hyperbolic) we can obtain a causal set C(M) by selecting points of M
and endowing them with the order induced from that of M (where in M,
x ≺ y if there is a future causal curve from x to y). In order to realize
the equality N = V, the selected points must be distributed with unit
density in M. One way to accomplish this (and conjecturally the only
way!) is to generate the points of C(M) by a Poisson process.
(Sorkin, 2003, p. 9)
492 R. E. Kastner
As noted in the previous section, it was independently argued (Kastner,

2014) that transactions are generated via decays, either of atomic excited
states (which generate photon offers and confirmations) or of unstable nuclei
(which generate offers and confirmations of quanta with non-vanishing rest
mass). Such decay processes are always Poissonian. We return to the com-
parison between causets and the possibilist transactional process in Sec. 4.
3. Time-like and Space-Like Relations in the Causet

A time-like relationship (i.e. either of ancestry or descendancy) obtains
between elements of the causet that are comparable; that is, they are mem-
bers of a single chain. On the other hand, a space-like relationship obtains
among elements that are all mutually incomparable; such elements are said
to constitute an antichain. These relations between elements of a causet
can be represented in a Hasse diagram, an example is shown in Fig. 2.
In the causet formulation, one cannot define spatial measure in terms of
the structure “orthogonal” to the chain; i.e. the antichain. The elements of
an antichain by definition have no relationship to each other at all, and
of course, there is no way to measure any aspect of a relationship where
none exists. This rather strange feature is actually harmonious with the
PTI account, in the following sense. In PTI (just as in relativity), only
the space–time interval has invariant physical content. On the other hand,
temporal and spatial relationships are secondary, frame-dependent notions.
These are only definable with respect to a specific actualized transaction,
as described in a particular frame.
B E
Fig. 2. A simple example of a causet. Events are represented by dots and links
by lines. The relation of descendance is indicated by the upward direction. Events
A, B, and C are members of a chain, while events B, D, and E are members of
an antichain.
www.ebook3000.com
Since an actualized transaction is a necessary condition for definition

of a spatial relationship between emitter and absorber, and an actualized
transaction necessarily implies a temporal relationship (emission being the
ancestor of the absorption), spatial displacement only obtains where there is
also temporal displacement. That is, a temporal relationship must hold for
any spatial relationship to be defined, even a frame-dependent one. Thus,
space only exists when time exists; the concept of space has no physical
meaning without a temporal relationship. On the other hand, a tempo-
ral displacement can be defined without any spatial displacement — the
latter corresponds to a transaction viewed from the reference frame of a
transferred quantum with finite rest mass.
The basic point is that we should not be surprised if it is difficult to
define a purely “spacelike” entity in the causet model. This should not be
viewed as a weakness of the model but rather as a reflection of the fact that
spatial relationships are supervenient both on temporal relationships and
on frames of reference. Another way to put this is that no two events are
ever truly “simultaneous.” If they are not related by a chain (i.e. if they
have no temporal relationship) then they cannot be regarded as having any
spatial relationship either, including that implied by simultaneity.
4. Dynamics and Growth of the Causet

In the PTI picture, the growth of the causet is dictated by the underly-
ing quantum dynamics. This, of course, presents a difficulty if one assumes
that the time arguments in evolving quantum states |Ψ(t) necessarily refer
to space-like hypersurfaces. The latter correspond to antichains in causet
theory, and we just noted that one cannot define a spatial measure on
these entities. However, the assumption that time indices refer to space-like
hypersurfaces is not in fact a necessary one. In what follows, we explore an
alternative approach to the understanding of references to time in time-
dependent quantum states, in which it is argued that in fact it is not
appropriate to assume an absolute temporal reference for the argument
t in |Ψ(t) = exp(−iHt /) |Ψ(0).
The first point is that the Hamiltonian H governing such evolving states
is a “stand-in” for the net effect of scattering processes, which are mediated
by quantum fields at the relativistic level. The Hamiltonian formulation is
not fully relativistically covariant, since it singles out a preferred time coor-
dinate. Thus, we should not be surprised if the usual non-relativistic time-
dependent quantum state |Ψ(t) seems incompatible with the relativistic
494 R. E. Kastner
causal set space–time model; it is already incompatible with ordinary rela-

tivistic space–time. Henson further comments that “. . . Even the Feynman
path integral crucially refers to states on spacelike hypersurfaces” ([10,
p. 9]). However, the path integral formulation of non-relativistic quantum
theory also singles out a particular frame, and we should therefore not
expect it to apply to a fully relativistic model of space–time.
The way to address this issue is to view the time index in |Ψ(t) as play-
ing a conditional and relational role rather than an absolute one. Specifi-
cally, given the relevant potentials, |Ψ(t) would describe the OW responded
to by an absorber, if the absorption event in a transaction actualized
between the emitter of the OW |Ψ(0) and that absorber were recorded
at time t on a clock in the absorber’s frame.
To understand this conditional nature of the time index, recall that the
Hamiltonian describes the overall effect of relativistic scattering processes.
Suppose it is projected that the emitted OW will reach a given macro-
scopic absorber when the laboratory clock reads t = ta . The value of the
time evolution operator at ta is a measure of the interactions of the appli-
cable forces via scattering with the offer, and thus their net effects on the
offer, with respect to that proper time interval. While such interactions
are often assumed to be taking place in space–time, that is not a necessary
assumption.c It is rejected in PTI, which takes such processes as pre-space–
time and sub-empirical. Indeed these processes are what underlie and give
rise to the space–time manifold which is the causet itself.
How does this work? Consider again the Hasse diagram of Fig. 2, which
illustrates a particular stage of growth of the causet. We also have to con-
sider the causet as being embedded in a quantum substratum of interacting
emitters and absorbers (e.g. excited and unexcited atoms); this substratum
is represented in Fig. 3 by a patterned background. (Some of these atoms
have very high probabilities of emitting to other atoms, and vice versa;
such groups of mutually emitting and absorbing atoms comprise macro-
scopic objects.) A later stage of growth can be represented by the addition
of a new additional event F, which arises from the actualization of a trans-
action between C (as emitter) and F (as absorber):
At the microscopic level, an object/event actualized as an absorber in
one transaction, such as an atom labeled C in Fig. 2, becomes reactualized
c It has been noted by Beretstetskii et al. ([12, p. 3]) and Auyang ([13, p. 48]) that
processes mediated by quantum fields are not appropriately viewed as space–time pro-
cesses. Specifically, Auyang notes that space–time indices refer to points on the field, not
space–time points.
www.ebook3000.com
F G (clock pulse)
Fig. 3. A new event F is added to the causet. Its temporal relationship to the
earlier event C can be inferred by reference to a clock pulse, shown as another
new event G. These events must be causally connected at a later event H in order
to infer the time interval between C and F.
as an emitter in a succeeding transaction (as in Fig. 3, where C emits to the

absorber at F). The emission occurs as it decays from its previous excited
state and emits a photon OW to the next absorber (i.e. another atom)
actualized at F. Note again that this is a Poissonian process, which fulfills
the requirement that event “sprinkling” into the causet must be Poissonian
to preserve relativistic covariance.
The time interval between events C and F can only be defined relative
to a clock — i.e. relative to some pre-established periodic process.d This is
indicated in Fig. 3 by the chain segment from C to G, which counts one
unit of time as measured by a relevant clock. If an identical transaction
(i.e. conveying the same amount of energy) then takes place between F
and a later event H, which serves also as a direct descendant of G, then
we can infer that the time interval between C and F was one unit. This
is not strictly possible at the microscopic level, since an absorber can only
participate in one transaction at any instant. Thus, the definition of a time
interval at the microscopic level can only be approximate.e
d An example is an atomic clock, which allows one to relate an atomic transition frequency
to a unit of time by counting oscillations (as in those of the microwave oscillator driving
a Cesium clock in resonance with the principal transition frequency). Such oscillations
would constitute a causally connected set of transacted events — a “chain” in the causet
with well-defined time intervals. (See Kastner 2012 [4, Chapters 3 and 6], for details on
how the transactional picture enables definition of the macroscopic realm, which would
include objects such as a microwave oscillator.)
e The Planck time is an appropriate lower bound for the error involved in establishing a
time for event C.

496 R. E. Kastner
f G1 (clock pulse at t=1)

F
g
C
G0 (clock pulse at t=0)
B
Fig. 4. A macroscopic emitter C, macroscopic absorber F , and laboratory clock

G (world tubes indicated by dashed rectangles) are all causally connected via
ongoing transactions with the laboratory equipment. (Only those between emit-
ter/clock and absorber/clock at emission and absorption events are shown explic-
itly in this projection on one spatial dimension.) The clock measures the proper
time for the absorption at F. Note that there is an inherent limit to the accuracy
of the measurement, since the absorptions are never strictly simultaneous.
Nevertheless, in order to establish an empirical space–time structure at

the macroscopic level, it is not really required that the same atom absorb
and then re-emit. It is sufficient that a macroscopic object absorbs and
then re-emits, in which case the absorption and emission may be carried
out by different atoms or molecules comprising the macroscopic object.
As noted earlier, collections of atoms with high probabilities of repeatedly
emitting and absorbing to one another comprise macroscopic objects. (A
simple example of this sort of absorption and re-emission process is a small
macroscopic sample of gas whose molecules are undergoing continual ther-
mal interactions; the latter are transactions.)
Figure 4 is a “bare bones” model of a macroscopic absorber F with a
laboratory clock G attached to it and causally connected to the macroscopic
emitter C as well:
Returning now to the issue of the “time dependence” of an offer: imagine
an offer |Ψ(0) being emitted at C, subject to the action of a potential
describable by a Hamiltonian H. Upon receipt of the offer by the absorber
at F, what is received and confirmed is not the same as what was emitted;
it is what we could call |Ψ(t), where in this case t = 1.
Thus, the temporal reference which appears as a challenge in develop-
ing the causet picture turns out to be an asset in the PTI model: we do
not need to refer to a space-like structure in the causet in order to apply
www.ebook3000.com
quantum theory to the growth of the causet. Rather, we can understand

time-dependent quantum theory as involving a conditional reference to an
evolving entity (the changing OW) in the quantum substratum, which is
not contained in the causet itself. The temporal reference is conditional
on absorption of the OW, where the time of the absorption is defined by
reference to a co-moving clock.
Again, it should be noted that the offer’s “evolution” does not imply
temporal evolution. Entities in the quantum substratum can undergo
change without necessary reference to time, which applies only at the actu-
alized space–time level.f The relevant time interval is then defined locally
and relationally, with respect to the actualizing absorber and its interactions
with other components (such as clocks). It is only through an actualized
transaction that the evolving OW gains a well-defined temporal reference.
An absolute time reference is inappropriate for the quantum object, since
(1) the quantum object is a pre-space–time (pre-causet) entity, and in any
case (2) that would inevitably involve a hyperplane of simultaneity that
cannot be reconciled with relativistic covariance.
5. Conclusion
The possibilist transactional picture can be viewed as a physical basis
for the emergence of the partially ordered set of events in the causal set
formalism. This formalism is currently being explored as a means to con-
structing a satisfactory theory of quantum gravity, and it has much promise
in that regard. However, even apart from general relativistic considera-
tions, the formalism breaks new ground in showing that, contrary to a well-
entrenched belief, a block world ontology is not required for consistency with
relativity. The causal set structure is a “growing universe” ontology which
nevertheless preserves the relativistic prohibition on a preferred frame.
Likewise, the transactional ontology proposed here is a variation on the
“growing universe” picture. The account is consistent with relativity theory
in that the set of events is amenable to a covariant description: no preferred
frame is required. This is because the transactional process is inherently
Poissonian, and therefore preserves the relativistic covariance of the causal
set model.
f If it seems hard to understand how something could “change” without reference to time,
one can think of a sequence of numbers generated by a particular mathematical process.

The numbers change in a clearly defined way, yet this change need not be defined with
respect to any external parameter; there is no “rate” at which the numbers change.
498 R. E. Kastner
References
1. J.G. Cramer, The transactional interpretation of quantum mechanics, Rev.
Mod. Phys. 58, 647–688 (1986).
2. J.A. Wheeler and R.P. Feynman, Interaction with the absorber as the mecha-
nism of radiation, Rev. Mod. Phys. 17, 157–161 (1945); J.A. Wheeler and R.P.
Feynman, Classical electrodynamics in terms of direct interparticle action,
Rev. Mod. Phys. 21, 425–433 (1949).
3. P.C.W. Davies, Extension of Wheeler–Feynman quantum theory to the rela-
tivistic domain I. Scattering processes, J. Phys. A: Gen. Phys. 6, 836 (1971);
P.C.W. Davies, Extension of Wheeler–Feynman quantum theory to the rela-
tivistic domain II. Emission processes, J. Phys. A: Gen. Phys. 5, 1025–1036
(1972).
4. R.E. Kastner, The Transactional Interpretation of Quantum Mechanics: The
Reality of Possibility, Cambridge University Press, Cambridge, 2012; R.E.
Kastner, On real and virtual photons in the davies theory of time-symmetric
quantum electrodynamics, Elect. J. Theor. Phys. 11, 75–86 (2014). Preprint
version: http://arxiv.org/abs/1312.4007.
5. J. Earman and D. Fraser, Haag’s theorem and its implications for the foun-
dations of quantum field theory, Erkenntnis 64(3), 305–344 (2006).
6. R.D. Sorkin, Causal Sets: Discrete Gravity (Notes for the Valdivia Summer
School). In Proc. Valdivia Summer School, A. Gomberoff (ed.), 2003.
7. G. ’t Hooft, Quantum gravity: A fundamental problem and some radical
ideas”, in Recent Developments in Gravitation (Proceedings of the 1978
Cargese Summer Institute) M. Levy and S. Deser (eds.), Plenum, New York,
1979; D. Marolf and R.D. Sorkin, Geometry from order: Causal sets in Ein-
stein Online 02, 007, 2006.
8. J. Myrheim, Statistical geometry, CERN preprint TH-2538, 1978.
9. L. Bombelli, J. Lee, D. Meyer and R.D. Sorkin, Spacetime as a causal set,
Phys. Rev. Lett. 59, 521–524 (1987).
10. D. Finkelstein, The spacetime code, Phys. Rev. 184, 1261 (1969).
11. J. Henson, The causal set approach to quantum gravity, in Approaches
to Quantum Gravity: Towards a New Understanding of Space and Time.
D. Oriti (ed.), Cambridge University Press, Cambridge, 2006. Preprint ver-
sion: arxiv:gr-qc/0601121.
12. L. Beretstetskii and L.P. Petaevskii, Quantum Electrodynamics. Landau and
Lifshitz Course of Theoretical Physics, Vol. 4, Elsevier, Amsterdam, 1971.
13. S. Auyang, How is Quantum Field Theory Possible? Oxford, New York, 1995.
www.ebook3000.com
Chapter 19
An Adynamical, Graphical Approach

to Quantum Gravity and Unification
W. M. Stuckey
Department of Physics, Elizabethtown College,
Elizabethtown, PA 17022 USA
stuckeym@etown.edu
Michael Silberstein
Department of Philosophy, Elizabethtown College,
silbermd@etown.edu
Department of Philosophy, University of Maryland,
College Park, MD 20742 USA
Timothy McDevitt
Department of Mathematics, Elizabethtown College,
mcdevittt@etown.edu
We use graphical field gradients in an adynamical, background indepen-

dent fashion to propose a new approach to quantum gravity (QG) and
unification. Our proposed reconciliation of general relativity (GR) and
quantum field theory (QFT) is based on a modification of their graph-
ical instantiations, i.e. Regge calculus and lattice gauge theory (LGT),
respectively, which we assume are fundamental to their continuum coun-
terparts. Accordingly, the fundamental structure is a graphical amalgam
of space, time, and sources (in parlance of QFT) called a “space–time
source element”. These are fundamental elements of space, time, and
sources, not source elements in space and time. The transition ampli-
tude for a space–time source element is computed using a path integral
with discrete graphical action. The action for a space–time source ele-
ment is constructed from a difference matrix K and source vector J
on the graph, as in lattice gauge theory. K is constructed from graph-
ical field gradients so that it contains a non-trivial null space and J is
then restricted to the row space of K, so that it is divergence-free and
499
500 W. M. Stuckey, M. Silberstein and T. McDevitt
represents a conserved exchange of energy–momentum. This construct

of K and J represents an adynamical global constraint (AGC) between
sources, the space–time metric, and the energy–momentum content of
the element, rather than a dynamical law for time-evolved entities. In
this view, one manifestation of quantum gravity becomes evident when,
for example, a single space–time source element spans adjoining simplices
of the Regge calculus graph. Thus, energy conservation for the space–
time source element includes contributions to the deficit angles between
simplices. This idea is used to correct proper distance in the Einstein–de
Sitter (EdS) cosmology model yielding a fit of the Union2 Compilation
supernova data that matches ΛCDM without having to invoke acceler-
ating expansion or dark energy. A similar modification to LGT results
in an adynamical account of quantum interference.
1. Introduction
1.1. Overview
In this chapter, we introduce our relational, adynamical, background inde-
pendent approach to quantum gravity (QG) and the unification of physics.a
This approach is based on and motivated by our foundations-driven account
of quantum physics called Relational Blockworld [1,2] (RBWb ), and employs
methods from general relativity (GR) (background independence and vari-
able geometry), particle physics (path integral formalism), and lattice
gauge theory (LGT) (graphical construction of transition amplitude). More
specifically, we propose a reconciliation of GR and quantum field theory
(QFT) via modification of their graphical instantiations, i.e. Regge calcu-
lus (Sec. 5.1) and LGT, respectively, which we assume are fundamental to
their continuum counterparts. The modifications we propose deal with our
fundamental ontological elements of quantum physics, i.e. graphical amal-
gams of space, time, and sourcesc that we call “space–time source elements”
a We have relegated many of the calculations to an arXiv paper http://arxiv.
org/abs/0908.4348 in order to keep this chapter focused on conceptual issues.

b RBW as a realist psi-epistemic retrocausal interpretation of quantum mechanics is
published elsewhere [1], so this chapter will focus on our “modified lattice gauge theory”
approach to QG and unification.
c We use the word “source” in formal analogy to QFT where it means “particle sources” or
“particle sinks” (creation or annihilation events, respectively). However, our “sources”

are not always equivalent to the sources in QFT, just as our “fields” are not always
equivalent to those of quantum physics. When we want to specify “a source of energy–
momentum” we will use “Source”. As we will explain in mathematical detail in Sec. 3.1,
our “source” represents a divergence-free property of the space–time source element
determined contextually by “classical objects”, i.e. objects with a worldline/tube in
space–time.
www.ebook3000.com
An Adynamical, Graphical Approach to QG and Unification 501
ABY
RG
ABY
RG
ABY
RG
AB
RGY
AB
RGY
AB
RGY
Worldtube 1 Worldtube 2
Fig. 1. Quantum Exchange of Energy–Momentum — The property Y is associ-

ated with the source on the space–time source element (rectangle) shared by the
worldtubes. As a result, property Y disappears from Worldtube 1 (Y Source) and
reappears later at Worldtube 2 (Y detector) without mediation. That is, there
is no third worldtube/line needed to explain the exchange of energy–momentum
associated with property Y between Worldtube 1 and Worldtube 2. While these
properties are depicted as residing in the worldtubes, they do not represent some-
thing truly intrinsic to the worldtubes, but are ultimately contextual and rela-
tional, i.e. being the Source of Y only makes sense in the context of (in relation
to) a “Y detector”, and vice versa. The A, B, R, and G properties shown might
be established with respect to COs not shown in this Figure, for example.
(Fig. 1). Accordingly, these are elements of space, time, and sources, not
source elements in space and time. The source of a space–time source ele-
ment is divergence-free and represents an unmediated, conserved exchange
of energy–momentum. Our approach constitutes a modification of LGT and
Regge calculus in three respects.
First, were assuming QFT and GR are continuum approximations of
LGT and Regge calculus, respectively, which is the opposite of conventional
thinking. Second, we are underwriting the fundamental computational ele-
ment of LGT, i.e. the transition amplitude, in a relational and adynamical
fashion. More significantly, third, we are assuming that the size of space–
time source elements of LGT and simplices of Regge calculus (Fig. 9) can be
as small or large as the situation requires. We find these changes discharge
the technical and conceptual difficulties of QFT and quantum mechan-
ics (QM) while leaving their computational structures and empirical suc-
cesses intact, for all practical purposes [1–4]. For example, the flexibility in
element/simplex size provides an adynamical explanation of twin-slit inter-

ference (Sec. 3.2) and a novel solution to the dark energy problem (Sec. 5.2).
We expect of course that this view of fundamental physics will suggest new
experiments in other areas as well. We will only briefly touch on such issues
here, but some consequences will be obvious to the reader familiar with
quantum physics. The focus of this chapter will be on explaining the math-
ematical and conceptual structure of the space–time source element of our
“modified lattice gauge theory” (MLGT) in the Schrödinger, Klein–Gordon
(KG), Dirac, Maxwell, and Einstein–Hilbert actions, with extension to the
Standard Model of particle physicsd and consequences for QG and unifica-
tion.
We finish Sec. 1 with an overview of our proposed formalism, a dis-
cussion of locality issues associated with our beable (the space–time source
element), and our proposed method of adynamical explanation. In Sec. 2, we
articulate more of our ontology, i.e. ontic structural realism (OSR) in a block
universe. Section 3 contains formal details associated with the construct of
the space–time source element in the Schrödinger, KG Dirac, Maxwell, and
Einstein–Hilbert actions. The space–time source element of the Schrödinger
action is used to provide an adynamical explanation of twin-slit interfer-
ence. The approach is extended to the space–time source elements for the
Standard Model in Sec. 4, where we also explain our view of particle physics
to include ultraviolet (UV) and infrared (IR) regularization. In Sec. 5, we
outline consequences for astrophysics and cosmology, to include an overview
of our resolution of the dark energy problem via modified Regge calculus
(MORC). We find that correcting proper distance in the Einstein–de Sit-
ter (EdS) cosmology model according to our MORC yields a fit [5] of the
Union2 Compilation supernova data that matches ΛCDM without having
to invoke accelerating expansion or dark energy [6]. We conclude in Sec. 6
with a list of outstanding questions for our program and consequences for
unification and QG.
In the language of dynamism, our space–time source elements represent
quantum exchanges of energy–momentum (in the form of mass, charge, or
other property) between objects with worldlines/tubese in space–time, e.g.
Sources, beam splitters, mirrors, detectors, etc. Let us call such objects with
worldlines/tubes in space–time “classical objects” (COs). In a quantum
d Hereafter simply “the Standard Model”.

e We follow convention and use “worldtubes” for spatially extended objects and “world-
lines” for objects treated as points.
www.ebook3000.com
exchange between COs per RBW, energy–momentum disappears from one

CO and reappears later in another (Fig. 1), i.e. the quantum exchange is
unmediated. The energy–momentum loss and gain are two distinct events
involving two distinct COs separated in space and time where the spa-
tial separation, temporal duration, and energy–momentum exchanged are
all associated with one space–time source element. Obviously, mediated
exchanges can occur between COs as well. For example, a buckyball that
decoheres between Source, diffraction grating, and detector such that it
does not contribute to an interference pattern at the detector could be rep-
resented by a worldline. Such a buckyball would therefore constitute a CO
mediating a (non-quantum) exchange of mass between Source and detector.
However, a buckyball Source and diffraction grating creating an interfer-
ence pattern at the detector [7] would represent a set of quantum exchanges
which would not have worldlines (Sec. 3.2). This is in stark contrast to a
view like de Broglie–Bohm where a buckyball interference pattern would
be created by objects with wordlines guided by waves. As we have written
elsewhere [2], a dynamical, algebraic counterpart to the space–time source
element in the path integral of RBW might be “elementary process” in the
Clifford algebra of Hiley’s Implicate Order.
The notion that energy might disappear from one location and reappear
at another without traversing the space between has also been claimed by
the authors of the DFBV experiment [8], Danan, Farfurnik, Bar-Ad, and
Vaidman.f Their conclusion is based on the so-called “weak measurements”
of photons in an interferometer that contains a “nested interferometer”.
They explain their weak measurement results assuming photons exist inside
the nested interferometer even though those photons “never entered and
never left the nested interferometer”. They therefore conclude that, “The
photons do not always follow continuous trajectories”, just as we assert
concerning quantum exchange in general [1].
So just to be clear, in the parlance of RBW, the properties associ-
ated with space–time source elements — spatial and temporal extent,
mass, charge, etc. — are determined relationally between COs, they
are not intrinsic properties of individual space–time source elements.
f For another so-called “direct-action” approach to QG see D. Wesley, and J.A. Wheeler,
Towards an action-at-a-distance concept of space–time, In A. Ashtekar et al. (eds.),

Revisiting the Foundations of Relativistic Physics: Festschrift in Honor of John Stachel.
Boston Studies in the Philosophy and History of Science (Book 234), pp. 421–436, Kluwer
Academic Publishers, Dordrecht, the Netherlands, 2003.
That is, the existence of these properties requires a collection of connected

space–time source elements in the context of COs, e.g. mass and charge
(properties) associated with tracks (connected space–time source elements)
in a particle detector (CO) as we explain in Sec. 4.2. Essentially, we are
claiming that the worldtube of any particular CO in space and time (defined
relationally by its surrounding COs) can be decomposed into space–time
source elements of space, time, and sources organized per an adynamical
global constraint (AGC) using the context of those surrounding COs. Herein
we articulate the part of that constraint dealing with individual space–time
source elements.
To do so, we relate gauge invariance, gauge fixing, divergence-free
sources, and relationally defined interacting COs in an adynamic, graphical

fashion. Specifically, each row of our difference matrix K for field gradi-
ents in the graphical action for our space–time source element is a vector
constructed relationally via the connectivity of some graphical entity, i.e.
nodes connected by links, links connected by plaquettes, or plaquettes con-
nected by cubes. Since each row vector is relationally defined, its compo-

nents sum to zero, which means [111 . . .]T is a null eigenvector of K. Our

AGC then demands that the source vector J in the action for our space–

time source element reside in the row space of K, so that it is orthogonal to
[111 . . .]T which means its components sum to zero, i.e. it is divergence-free.
A divergence-free source in each space–time source element then under-
writes relationally defined, spatially distributed, trans-temporally identi-
fied (conserved) properties exchanged between interacting COs, i.e. it pro-
vides the fundamental ontological element for relationally and contextually

defined COs per OSR (Sec. 2). That K possesses a non-trivial null space

is the graphical equivalent of gauge invariance and restricting J to the row

space of K provides a natural gauge fixing, i.e. restricting the path inte-

gral of the transition amplitude to the row space of K. That K possesses

a non-trivial null space also means the determinant of K is zero, so the

set of vectors constituting the rows of K is not linearly independent. That
some subset of these row vectors is determined by its complement follows
from having the graphical set relationally constructed. Thus, divergence-

free J follows from relationally defined K as a direct result of our AGC.
Consequently, we agree with Rovelli that [9], “Gauge is ubiquitous. It is
www.ebook3000.com
not unphysical redundancy of our mathematics. It reveals the relational

structure of our world”.
So these fundamental elements of space–time source are our beables.
Are such beables local?
1.2. Locality
Concerning the locality of beables, Einstein writes [10]
. . .if one asks what is characteristic of the realm of physical ideas inde-
pendently of the quantum theory, then above all the following attracts
our attention: the concepts of physics refer to a real external world, i.e.
ideas are posited of things that claim a ‘real existence’ independent of the
perceiving subject (bodies, fields, etc.), and these ideas are, on the other
hand, brought into as secure a relationship as possible with sense impres-
sions. Moreover, it is characteristic of these physical things that they are
conceived of as being arranged in a space–time continuum. Further, it
appears to be essential for this arrangement of the things introduced in
physics that, at a specific time, these things claim an existence inde-
pendent of one another, insofar as these things ‘lie in different parts of
space’. Without such an assumption of mutually independent existence
(the ‘being-thus’) of spatially distant things, an assumption which orig-
inates in everyday thought, physical thought in the sense familiar to us
would not be possible . . . .
There has been a great deal of handwringing lately in the foundations lit-
erature on QG as to whether the most fundamental unifying theory from
which space–time emerges, must have local beables to be empirically coher-
ent and make full correspondence with higher-level physical theories and
the experienced world [11]. Maudlin notes that [11] “local beables do not
merely exist: they exist somewhere,” or as Bell puts it [12], beables are
“definitely associated with particular space–time regions”. Of course, there
is less consensus about the necessary and sufficient conditions for being a
local beable, and that discussion is beyond the scope of this chapter. In any
case, we share the consensus view that a successful theory of QG need not
have local beables [13]. To return to the main question about the status of
space–time source elements, local beables are thought of as being separate
from, but located somewhere in space–time, whereas, again, space–time
sources are of space, time, and sources.
Einstein appears to conflate (or at least highlight) several different
notions of “local” in the passage above, including, (1) local as localized
in space–time, (2) local as possessing primitive thisness with intrinsic
properties, (3) local as in no superluminal interactions, and (4) local as in
being otherwise independent (e.g. statistically) of entities at other points

in space–time. On the graph, directly connected nodes are “local” to each
other. However, due to the third modification of LGT and Regge calculus
described above, it can be the case (see Sec. 5.2) that neighboring points on
the graph are very far apart in the space–time manifold M, so those points
are not local/neighboring in M (this is called “disordered locality”). Thus,
our beables are local in the first sense per graphical locality, and local in
the third sense on both the graph and its continuum approximation, the
space–time manifold M (see Sec. 5.1). Again, recalling what we said above,
our beables are not local in senses 2 or 4.
The manner by which we correct EdS cosmology and explain quantum
interference is a form of disordered locality, as explained supra, similar
to the situation in quantum graphity [14, 15]. Our physical model thus
implements a suggestion made by Weinstein among others [16]:
What I want to do here is raise the possibility that there is a more fun-
damental theory possessing nonlocal constraints that underlies our cur-
rent theories. Such a theory might account for the mysterious nonlocal
effects currently described, but not explained, by quantum mechanics,
and might additionally reduce the extent to which cosmological models
depend on finely tuned initial data to explain the large scale correla-
tions we observe. The assumption that spatially separated physical sys-
tems are entirely uncorrelated is a parochial assumption borne of our
experience with the everyday objects described by classical mechanics.
Why not suppose that at certain scales or certain epochs, this indepen-
dence emerges from what is otherwise a highly structured, nonlocally
correlated microphysics?
As he says, every extant fundamental theory of physics assumes the non-

existence of such non-local constraints [16]:
Despite radical differences in their conceptions of space, time, and the
nature of matter, all of the physical theories we presently use, non-
relativistic and relativistic, classical and quantum, share one assump-
tion: the features of the world at distinct points in space are understood
to be independent. Particles may exist anywhere, independent of the
location or velocity of other particles. Classical fields may take on any
value at a given point, constrained only by local constraints like Gauss’s
law. Quantum field theories incorporate the same independence in their
demand that field operators at distinct points in space commute with
one another. The independence of physical properties at distinct points
is a theoretical assumption, albeit one that is grounded in our everyday
experience. We appear to be able to manipulate the contents of a given
region of space unrestricted by the contents of other regions. We can
www.ebook3000.com
arrange the desk in our office without concern for the location of the
couch at home in our living room.
RBW provides an exact model in which precisely this type of locality (type
2 and type 4 above) fails to obtain, thereby allowing us to explain a diverse
range of phenomena from quantum interference to so-called dark energy.
Furthermore, as will be explained, the failure of locality in question, the
way is it implemented in our theory, is consistent with and driven by an
appropriately MORC.
1.3. Adynamical explanation

Our approach also differs from common practice (even quantum graphity)
in that it is adynamical [1, 2]. Carroll sums up nicely what we mean by a
dynamical approach [17]:
Let’s talk about the actual way physics works, as we understand it. Ever
since Newton, the paradigm for fundamental physics has been the same,
and includes three pieces. First, there is the “space of states”: basically,
a list of all the possible configurations the universe could conceivably be
in. Second, there is some particular state representing the universe at
some time, typically taken to be the present. Third, there is some rule
for saying how the universe evolves with time. You give me the universe
now, the laws of physics say what it will become in the future. This way
of thinking is just as true for quantum mechanics or general relativity
or quantum field theory as it was for Newtonian mechanics or Maxwell’s
electrodynamics.
Carroll goes on to say that all extant formal models of QG, even those
attempting to recover space–time [18], are dynamical in this sense. While it
is true that integral calculus and least action principles have been around
for a long time, most assume these methods are formal tricks and not fun-
damental to dynamical equations. While our adynamical approach employs
mathematical formalism akin to dynamical theories, e.g. LGT, we redefine
what it means to “explain” something in physics. Rather than finding a
rule for time-evolved entities as per Carroll (e.g. causal dynamical triangu-
lation [19]), the AGC leads to the self-consistency of a graphical space–time
metric and its relationally defined sources. While we do talk about “con-
structing” or “building” spatio-temporal objects in this chapter, we are not
implying any sort of “evolving block universe” as in causet dynamics [20].
Our use of this terminology is merely in the context of a computational
algorithm. So, one might ask for example, “Why does link X have metric
G and stress-energy tensor T ?” A dynamical answer might be, “Because
link X − 1 has metric G − 1 and stress-energy tensor T − 1 and the law

of evolution thereby dictates that link X has metric G and stress-energy
tensor T.” Note how this answer is independent of future boundary condi-
tions; indeed, it is independent of conditions anywhere else on the graph
other than those of the 3D hypersurface in the immediate past. Contrast
this with an adynamical answer such as, “Because the values G and T on X
satisfy the AGC for the graph as a whole, given input anywhere in the past,
present, and/or future of X”. For example, we will argue in Sec. 4.2 that par-
ticle trajectories of high energy physics experiments satisfy the AGC given
the space–time configuration of colliding beams and detector. It should be
clear from what we have said thus far that explanations involving AGCs
typically involve future boundary conditions, which brings us to the next
section.
2. Quantum Physics Reconceived: Ontic Structural

Realism in a Block Universe
2.1. Dynamism denied
Our account of space–time and matter is very much in keeping with Rovelli’s
intuition that [21]:
General relativity (GR) altered the classical understanding of the con-
cepts of space and time in a way which . . . is far from being fully under-
stood yet. QM challenged the classical account of matter and causality,
to a degree which is still the subject of controversies. After the discovery
of GR we are no longer sure of what is space–time and after the discov-
ery of QM we are no longer sure of what matter is. The very distinction
between space–time and matter is likely to be ill-founded. . . . I think it is
fair to say that today we do not have a consistent picture of the physical
world. [italics added]
We agree with Rovelli and believe a current obstacle to unification is the

lack of a true marriage of space–time with matter. That is, we believe one
of the main obstacles to unification has been a form of “space–time-matter
dualism” whereby the space–time metric (or simply “metric”) is subject to
quantization distinct from the matter and gauge fields. This view is car-
ried over from QFT and GR. In QFT, although matter-energy fields are
imagined to pervade space, the metric is independent of the matter-energy
content of space–time. And, although Weyl characterized GR as providing
Raum, Zeit, Materie [22], there are vacuum solutions in GR, i.e. space–
time regions where the stress-energy tensor is zero. Thus, neither QFT nor
www.ebook3000.com
GR embody a true unity of “space–time source” and both employ a differ-

entiable manifold structure for space–time.g Herein, we propose unification
based on a true unity of space, time, and sources, finishing Einstein’s dream
so to speak.
Fundamental theories of physics (e.g. M-theory, loop QG, causets) may
deviate from the norm by employing radical new fundamental entities (e.g.
branes, loops, ordered sets), but the game is always dynamical, broadly
construed (e.g. vibrating branes, geometrodynamics, sequential growth pro-
cess). As Healey puts it [23]:
Physics proceeds by first analyzing the phenomena with which it deals
into various kinds of systems, and then ascribing states to such systems.
To classify an object as a certain kind of physical system is to ascribe
certain, relatively stable, qualitative intrinsic properties; and to further
specify the state of a physical system is to ascribe to it additional, more
transitory [time dependent], qualitative intrinsic properties . . . . A phys-
ical property of an object will then be both qualitative and intrinsic just
in case its possession by that object is wholly determined by the under-
lying physical states and physical relations of all the basic systems that
compose that object.
Dynamism then encompasses three claims: (A) the world, just as appear-
ances and the experience of time suggest, evolves or changes in time in some
objective fashion, (B) the best explanation for A will be some dynamical
law that “governs” the evolution of the system in question, and (C) the
fundamental entities in a “theory of everything” will themselves be dynam-
ical entities evolving in some space however abstract, e.g. Hilbert space.
Our model rejects not only tenets A and B of dynamism, but also C. In our
view, time-evolved entities or things are not fundamental and, in fact, it is
in accord with (OSR) [24, 25]:
Ontic structural realists argue that what we have learned from con-
temporary physics is that the nature of space, time and matter are not
compatible with standard metaphysical views about the ontological rela-
tionship between individuals, intrinsic properties and relations. On the
broadest construal OSR is any form of structural realism based on an
ontological or metaphysical thesis that inflates the ontological priority
of structure and relations.
g For an overview of problems associated with “the manifold conception of space and
time” in QG, see Butterfield, J. and Isham, C.J. Space–time and the Philosophical Chal-
lenge of Quantum Gravity, http://arxiv.org/abs/gr-qc/9903072 1999.
More specifically, our RBW version of OSR agrees that [24] “The relata of
a given relation always turn out to be relational structures themselves on
further analysis.” Note that OSR does not claim there are relations without
relata, just that the relata are not individuals (e.g. things with primitive
thisness and intrinsic properties), but always ultimately analyzable as rela-
tions as well (Fig. 1). OSR already somewhat violates the dynamical bias by
rejecting things with intrinsic properties as fundamental building blocks of
reality — the world is not fundamentally compositional — the deepest con-
ception of reality is not one in which we decompose things into other things
at ever smaller length and time scales.h Our beables (space–time source ele-
ments) are certainly a violation of a compositional picture of realty, since
their properties are inherited from their classical context. We however go
even further in rejecting dynamism, not merely because it is a block uni-
verse, but because the fundamental modal structure, the fundamental AGC,
is not a dynamical law or even space–time symmetries.
A good deal of the literature on OSR is driven by philosophical concerns
about scientific realism and intertheoretic relations, rather than motivated
by physics itself [25, 26]. There has also been much debate in the philo-
sophical literature as to whether OSR provides any real help in resolving
foundational issues of physics such as interpreting QM or in advancing
physics itself. Consider the following claims for example:
OSR is not an interpretation of QM in addition to many worlds-type
interpretations, collapse-type interpretations, or hidden variable-type
interpretations. As the discussion of the arguments for OSR from QM
in section 2 above has shown, OSR is not in the position to provide on
its own an ontology for QM, since it does not reply to the question of
what implements the structures that it poses. In conclusion, after more
than a decade of elaboration and debate on OSR about QM, it seems
that the impact that OSR can have on providing an answer to the ques-
tion of what the world is like, if QM is correct, is rather limited. From
a scientific realist perspective, the crucial issue is the assessment of the
pros and cons of the various detailed proposals for an ontology of QM,
as it was before the appearance of OSR on the scene [27].
While the basic idea defended here (a fundamental ontology of brute
relations) can be found elsewhere in the philosophical literature on ‘struc-
tural realism’, we have yet to see the idea used as an argument for
advancing physics, nor have we seen a truly convincing argument, involv-
ing a real construction based in modern physics, that successfully evades
h Thisis an ontological claim. Computationally, of course, the space–time lattice of LGT

is “composed of” hypercubes with fields on nodes and links.
www.ebook3000.com
the objection that there can be no relations without first (in logical order)
having things so related [28].
As this chapter will attest, RBW is a counterexample to Esfeld’s claim

and it provides exactly the physical model that Rickles and Bloom are
looking for. As they say in the following passage, OSR has the potential to
re-ground physics, dissolve current quagmires and lead to new physics [28]:
Viewing the world as structurally constituted by primitive relations has
the potential to lead to new kinds of research in physics, and knowledge of
a more stable sort. Indeed, in the past those theories that have adopted
a broadly similar approach (along the lines of what Einstein labeled
‘principle theories’) have led to just the kinds of advances that this essay
competition seeks to capture: areas “where thinkers were ‘stuck’ and had
to let go of some cherished assumptions to make progress.” Principle
theory approaches often look to general ‘structural aspects’ of physical
behaviour over ‘thing aspects’ (what Einstein labeled ‘constructive’),
promoting invariances of world-structure to general principles.
Rickles and Bloom lament the fact that OSR has yet to be so motivated
and further anticipate our theory almost perfectly when they say [28]:
The position I have described involves the idea that physical systems
(which I take to be characterized by the values for their observables)
are exhausted by extrinsic or relational properties: they have no intrin-
sic, local properties at all! This is a curious consequence of background
independence coupled with gauge invariance and leads to a rather odd
picture in which objects and [space–time] structure are deeply entan-
gled. Inasmuch as there are objects at all, any properties they possess
are structurally conferred: they have no reality outside some correla-
tion. What this means is that the objects don’t ground structure, they
are nothing independently of the structure, which takes the form of a
(gauge invariant) correlation between (non-gauge invariant) field values.
With this view one can both evade the standard ‘no relations without
relata’ objection and the problem of accounting for the appearance of
time (in a timeless structure) in the same way.
In this chapter, we provide physics that embodies their suggestion.
2.2. Block universe

As stated, we must further exacerbate this violation of dynamism by apply-
ing OSR to a block universe. The block universe perspective (the reality
of all events past, present, and future including the outcomes of quantum
experiments) is suggested, for example, by the relativity of simultaneity in
special relativity or, more generally, the lack of a preferred spatial foliation
of space–time M in GR, and even by quantum entanglement according to
some of us [4]. Geroch writes [29]:
There is no dynamics within space–time itself: nothing ever moves
therein; nothing happens; nothing changes. In particular, one does
not think of particles as moving through space–time, or as following
along their world-lines. Rather, particles are just in space–time, once
and for all, and the world-line represents, all at once, the complete life
history of the particle.
When Geroch says that “there is no dynamics within space–time itself”,

he is not denying that the mosaic of the block universe possesses pat-
terns that can be described with dynamical laws. Nor is he denying the
predictive and explanatory value of such laws. Rather, given the reality
of all events in a block universe, dynamics are not “event factories” that
bring heretofore non-existent events (such as measurement outcomes) into
being; fundamental dynamical laws that are allegedly responsible for dis-
charging fundamental “why” questions in physics are not brute unexplained
explainers that “produce” events on our view.
In addition, there is the problem of time in canonical GR. That is, in a
particular Hamiltonian formulation of GR, the reparametrization of space–
time is a gauge symmetry. Therefore, all genuinely physical magnitudes are
constants of motion, i.e. they do not change over time. In short, as Rovelli
stated in an earlier quote, gauge invariance merely “reveals the relational
structure of our world”.
Finally, the problem of frozen time in canonical QG is that if the canon-
ical variables of the theory to be quantized transform as scalars under time
reparametrizations, which is true in practice because they have a simple
geometrical meaning, then [30] “the Hamiltonian is (weakly) zero for a
generally covariant system”. The result upon canonical quantization is the
famous Wheeler–DeWitt equation, void of time evolution. While it is too
strong to say a generally covariant theory must have H = 0, there is no well-
developed theory of QG that has avoided it to date [31]. It is supremely
ironic that the dynamism and unificationism historically driving physics
led us directly to block universe and frozen time, but RBW discharges the
irony.
Just as people rarely take seriously the possibility that the path inte-
gral or Lagrangian approach (with its future boundary conditions) is fun-
damental [32], they rarely take seriously the block universe even when
they embrace it. That is, the overriding assumption as noted above is that
www.ebook3000.com
dynamical explanation is fundamental. But taking the block universe seri-

ously opens up the possibility that AGCs might be fundamental both for
individual systems within the universe, e.g. the delayed-choice experiment,
and for the universe as a whole, e.g. the Big Bang (the allegedly special
and unlikely initial conditions).
Back to the problem of time, Rickles notes that the problem can be
solved by [26], “(1) global quantities defined over the whole space–time
and (2) ‘relational’ quantities built out of correlations between field values
and/or invariants. There seems to be some consensus forming that the latter
type are the way to go, and these will serve as the appropriate vehicle for
defining time in an unchanging mathematical structure, as well as defining
the structures themselves.” Our theory, it will become clear, provides a
solution precisely in terms of number 2.
We think therefore that both QM, e.g. delayed-choice experiments, and
relativity are telling us that Nature is a block universe, so it is time to
promote this idea from mere metaphysics to physics. This is what RBW
does.
2.3. OSR in a block universe

Putting it all together, reality is a block universe best characterized as
space–time source, as opposed to the “space–time + sources” picture of
current physics. In the foundations literature on the eternalism debate and
the structural realism debate respectively, the biggest complaint is that the
fate of these topics makes no real difference for physics itself, i.e. it does
not lead to new models, new insights, or new predictions and it does not
resolve conceptual problems. In short, the complaint is that such debates
are nothing but pure metaphysics. We, however, actually do provide a new
formal model for fundamental physics based on the block universe with
relationally defined sources that has all the aforementioned virtues. Our
approach employs an AGC.
2.4. AGC
Our use of an AGC is not without precedent, as we already have an ideal
example in Einstein’s equations of GR
1 8πG
Rαβ − gαβ R = 4 Tαβ .
2 c
Momentum, force, and energy all depend on spatio-temporal measurements

(tacit or explicit), so the stress-energy tensor Tαβ cannot be constructed
without tacit or explicit knowledge of the space–time metric gαβ (techni-
cally, Tαβ can be written as the functional derivative of the matter-energy
Lagrangian with respect to g αβ ). But, if one wants a “dynamic space–time”
in the parlance of GR, gαβ must depend on the matter-energy distribution
in space–time. GR solves this dilemma by constraining Tαβ to be “con-
sistent” with gαβ everywhere on the space–time manifold M per Einstein’s
equations.i This AGC hinges on divergence-free sources, which finds a math-
ematical counterpart in ∂∂ = 0, i.e. the boundary of a boundary principle
[33]. So, Einstein’s equations of GR are a mathematical articulation of the
boundary of a boundary principle in classical physics, i.e. they constitute an
AGC in classical physics.j In fact, our AGC at the level of the space–time
source element is based on the same topological maxim (∂∂ = 0) for the
same reason, as is the case with quantum and classical electromagnetism
[33, 34].
3. Underwriting the Free Field Transition Amplitude

Graphical Harmonic Oscillator and the AGC
The Lagrangian for the coupled masses of Fig. 2 is
1 2 1 2 1 2
L= mq̇ + mq̇ − k (q1 − q2 ) , (1)
2 1 2 2 2
q1 q2
Fig. 2. Coupled masses.
i Concerningthe stress-energy tensor, Hamber and Williams write, “In general its covari-
ant divergence is not zero, but consistency of the Einstein field equations demands
∇α Tαβ = 0,” Hamber, H.W., and Williams, R. Nonlocal Effective Gravitational Field
Equations and the Running of Newton’s G http://arxiv.org/pdf/hep-th/0507017.pdf,
2005.
j The fundamental ontological entities of GR are described via worldlines/tubes, so it
admits a dynamical interpretation, of course. The adynamical and global nature of a

GR explanation is more evident in its discrete graphical counterpart, Regge calculus
(Sec. 5.1).
www.ebook3000.com
so our transition amplitude is ( = 1)

Z(J) = Dq(t)
 T 

1 1 1 1
expi dt mq̇12 + mq̇22 − kq12 − kq22 + kq1 q2 + J1 q1 + J2 q2 ,
2 2 2 2
0
(2)
giving
“ m ” −m 
− k∆t 0 k∆t 0 0
 ∆t „ ∆t « 
 −m 2m −m 
 − k∆t 0 k∆t 0 
 ∆t ∆t ∆t 
 −m “m ” 
 0 − k∆t 0 0 k∆t 

 ∆t ∆t 
K = “m ” −m ,
 k∆t − k∆t 
 0 0 0 
 ∆t „ ∆t « 
 −m 2m −m 
 0 k∆t 0 − k∆t 
 ∆t ∆t
“m
∆t
”
−m
0 0 k∆t 0 − k∆t
∆t ∆t
(3)
on the graph of Fig. 3. The null space (space of eigenvalues 0) is spanned
by the eigenvector [111111]T. The space orthogonal to the null space of

K is called the row spacek of K. Therefore, any source vector J in the

row space of K has components which sum to zero and this is referred to

in graphical approaches to physics as “divergence-free J ”. If J is a force,

this simply reflects Newton’s third law. If J is energy, this simply reflects
3
−k∆t
6
2
−k∆t
5
−k∆t
1 4
Fig. 3. Graph for coupled masses.

k The column space is equal to the row space here, since K is symmetric.

conservation of energy. We will use J on space–time source elements to

underwrite conserved properties defining COs, so we require that J reside

in the row space of K, as well as represent an interaction with conserved

source across a space–time source element. Thus, K must be constructed
so as to possess a non-trivial null space, which is the graphical equivalent
of gauge invariance.
Now that we have explained the AGC, our choice of gauge fixing is
obvious. The discrete, graphical counterpart to Eq. (2) is
∞ ∞
1
Z= ... dQ1 . . . dQN exp i Q ·K ·Q + iJ · Q
(4)
2
−∞ −∞
with solution

12
N
(2πi) 1
−1
Z= exp −i J·K · J .
(5)
det(K) 2

−1
However, K does not exist because K has a non-trivial null space. This
is the graphical characterization of the effect of gauge invariance on the

computation of Z(J). Because we require that J reside in the row space of

K, the graphical counterpart to Fadeev–Popov gauge fixing is clear, i.e. we

simply restrict our path integral to the row space of K. Nothing of physical

interest lies elsewhere, so this is a natural choice. In the eigenbasis of K
with our gauge fixing Eq. (4) becomes
∞ ∞ N
1
2
Z= ... dQ̃2 . . . dQ̃N exp i Q̃n an + iJñ Q̃n , (6)
n=2
2
−∞ −∞

where Q̃n are the coordinates associated with the eigenbasis of K and Q̃1

is associated with eigenvalue zero, an is the eigenvalue of K corresponding

to Q̃n , and Jñ are the components of J in the eigenbasis of K. Our gauge
independent approach revises Eq. (5) to give
  12
N

 (2πi)N −1  Jñ2
Z= N  exp −i . (7)
 2an
an n=2
n=2
www.ebook3000.com
Thus, we find that the adynamically constrained co-construction of space,

time, and divergence-free sources entails gauge invariance and gauge fix-
ing. We now apply this idea to the Schrödinger, KG, Dirac, Maxwell, and
Einstein–Hilbert actions.
3.1. Non-relativistic scalar field on nodes

and the twin-slit experiment
The non-relativistic limit of the KG equation gives the free-particle
Schrödinger equation (SE) by factoring out the rest mass contribution
to the energy E, assuming the Newtonian form for kinetic energy, and
discarding the second-order time derivative [35]. To illustrate the first
two steps, plug ϕ = Aei(px−Et)/ into the KG equation and obtain
(−E 2 + p2 c2 + m2 c4 ) = 0, which tells us E is the total relativistic energy.
p2
Now plug ψ = Aei(px−Et)/ into the free-particle SE and obtain 2m = E,
which tells us E is only the Newtonian kinetic energy. Thus, we must fac-
2
tor out the rest energy of the particle, i.e. ψ = eimc t/ ϕ, assume the
low-velocity limit of the relativistic kinetic energy, and discard the relevant
term from our Lagrangian density (leading to the second-order time deriva-
tive) in going from ϕ of the KG equation to ψ of the free-particle SE. We
will make these changes to Z(J) for the KG equation and obtain ψ(x,t),
which we will then compare to ψ(x,t) from QM (with a source) in order to
produce our probability amplitude.
For the KG equation, we have

1 2 1
Z(J) = Dϕ exp i d4 x (∂ϕ) − m̄2 ϕ2 + Jϕ , (8)
2 2
(overall factor of in exponent = 1) which in (1 + 1)D is
2
2
1 ∂ϕ c2 ∂ϕ 1 2 2
Z(J) = Dϕ exp i dxdt − − m̄ ϕ + Jϕ ,
2 ∂t 2 ∂x 2
(9)
2 √
(m̄ ≡ mc ). Making the changes described above with ψ = e im̄t
m̄ϕ,
Eq. (9) gives the non-relativistic KG transition amplitude corresponding to
the free-particle SE [35]
2
∗ ∂ψ c2 ∂ψ
Z(J) = Dψ exp i dxdt iψ − + Jψ . (10)
∂t 2m̄ ∂x
Now integrate the second term by parts and obtain

∗ ∂ψ ∗ ∂2ψ
Z(J) = Dψ exp i dxdt iψ + ψ + Jψ . (11)
∂t 2m ∂x2
This gives

1 ∗
Z(J) = Dψ exp i dxdt ψ Kψ + Jψ , (12)
2
where

∂ ∂2
K= 2i + . (13)
∂t m ∂x2
The solution to this is

i
Z(J) = Z(0) exp − dxdx J(x)D(x − x )J(x ) ≡ Z(0) exp (iW (J)),
2
(14)
where x and x are each shorthand for both a spatial dimension and a
temporal dimension,

1
W (J) = − dxdx J(x)D(x − x )J(x ) (15)
2
and

∂ ∂2
2i + D(x − x ) = δ(x − x ), (16)
∂t m ∂x2
that is, D(x − x ) is the Green’s function, aka the QFT propagator. A
solution is
ik(x−x ) iω(t−t )
−1 e e dωdk
D(x − x ) = , (17)
(2π)2
2
2ω + km
(where x and x are just spatial on the RHS). Note from Eq. (14) that
D(x − x ) is worthless in the absence of a source. This is important in an
RBW approach, since Nature is COs defined relationally/contextually via
“quantum interactions” there is no truly “sourceless” physics.
That D(x − x ) is worthless without sources is significant because the
QM free-particle propagator [36] with ψ(x, 0) = δ(x) gives

m imx2
ψo (x, t) = exp (18)
2πit 2t
www.ebook3000.com
and this gives

∂ ∂2
2i + ψo (x, t) = 0. (19)
∂t m ∂x2
Thus, ψo (x, t) obtained from the QM free-particle propagator is a solution
of the SE without a source, i.e. J = 0 in Eq. (14). So, QM’s ψo (x, t) in
Eq. (18) is not physically relevant in RBW per QFT’s Z(J). That is, since
Z(J) = Z(0)eiW (J) , the sourceless solutions ψo (x, t) would appear in the
exponent of Z(0) which Zee describes as [35], “often of no interest to us.”
In order to obtain a physically relevant “free-particle amplitude” related
to non-relativistic QFT, the SE must have a “source.” Essentially, in our
QFT approach, we want a particle of mass m created at the Source and
annihilated at the sink (detector) — with no worldline connecting them —
and this happens at sources J. We can write the SE [37]

∂ ∂2
2i + ψ(x) = 2V (x)ψ(x), (20)
∂t m ∂x2
so that

ψ(x) = dx D(x − x )2V (x )ψ(x ). (21)
With 2V (x )ψ(x ) = δ(x − xi ), we have
ψ(x) = D(x − xi ). (22)
We could still add solutions ψo (x) of the sourceless equation, but again they
are associated with Z(0) and therefore of “no interest to us”.
To find the QFT counterpart to Eq. (22), we use Eq. (15) with point
sources J(x ) at xi (Source) and J(x) at xf (sink/detector) to obtain

1 1
W (J) = − dxdx δ(x − xf )D(x − x )δ(x − xi ) = − D(xf − xi ).
2 2
(23)
So, with D(x − x ) given by Eq. (17) we have our QFT derivation of the
“free-particle” QM probability amplitude, i.e. ψ(x) = −2W (J), which is

−1
J · K · J on the graph of MLGT. That we must always supply J(x), and
that J(x) is always coupled to J(x ) via D(x − x ) in Z(J), is consistent
with the relational ontology of RBW. Now we formulate our discrete MLGT
counterpart to this result.
Since ψ∗ appears undifferentiated in Eq. (10), we do not have a fully
relational form. We imagine this is because ψ needs to be underwritten by a
“coordinate” field that reveals the underlying relational form of the action.
For example, if one writes the spring potential of Sec. 3.1 in terms of the
displacement x from equilibrium, one obtains the term 12 kx2 in the action,
but this obscures the relational structure revealed using coordinates q, i.e.
1 2 ∗ ∗ ∗
2 k(q1 − q2 ) . So, we replace ψ∗ with a relational structure ψ → (ψ2 − ψ1 )
in the following discretizations (with extrapolations):

∂ψ ψ2 − ψ1
iψ ∗ → i (ψ2∗ − ψ1∗ )
∂t ∆t
2
c2 ∂ψ ψ3∗ − ψ1∗ ψ3 − ψ1
− →− ,
2m̄ ∂x 2m ∆x ∆x
where ψ2 is at node ψ1 + ∆t, ψ3 is at node ψ1 + ∆x, and ψ4 is at node

ψ1 + ∆x + ∆t (Fig. 4). We obtain for K in 12 Ψ∗ · K · Ψ:
 
2i 2i
−
 ∆t m∆x2 − 0 
 ∆t m∆x2 
 
 2i 2i 
 − − 0 
 ∆t ∆t m∆x 2 m∆x 2 

 
K = ,
 2i 2i 
 0 − − 
 m∆x2 ∆t m∆x2 
 ∆t 
 
 2i 2i 
0 2
− − 2
m∆x ∆t ∆t m∆x
(24)
ignoring the volume element ∆x∆t. The eigenvalues are {0, 4it , − mx 2 4i
2, t −
2
mx2 }, where we have dropped the ∆ for simplicity, and the corresponding
eigenvectors are {(1, 1, 1, 1), (−1, 1, −1, 1), (−1, −1, 1, 1), (1, −1, −1, 1)}, i.e.
Fig. 4. Wave-function on the graph.
www.ebook3000.com
a Hadamard structure we see repeated in both the KG and Dirac equations.

These eigenvectors correspond to the following four modes, respectively:
Mode 1

There is no spatial or temporal variation in J , so J is not divergence-free

and therefore does not reside in the row space of K. This source does not
satisfy the AGC.
Mode 2

There is only temporal variation in J . While J resides in the row space

of K and is therefore divergence-free in the mathematical sense, it is not
conserved within the element. Therefore, this source does not satisfy the
AGC.
Mode 3

There is only spatial variation in J . While J resides in the row space of K
and is conserved within the element, it does not represent an interaction.
Therefore, this source does not satisfy the AGC.
Mode 4

There is both spatial and temporal variation in J , which resides in the row

space of K, is conserved in the element, and represents an interaction. This
source satisfies the AGC.

In the eigenspace of K, the source associated with mode 4 is J =
(0, 0, 0, Jo ), where Jo is complex in general, so it is easily seen that (with
our gauge fixing)

−1
Jo2
J ·K · J = 4i 2
. (25)
t − mx 2
Equations (22) and (23) tell us that Eq. (25) is the MLGT counterpart to
Eq. (17), i.e.

Jo2 −1 eikx eiωt dωdk
4i 2
= 2 2ω+k2
, (26)
t − mx 2 (2π) m
where t and x represent the temporal and spatial extent of the element,
respectively, and Jo2 = Jo · Jo (not Jo∗ · Jo ). The LHS of Eq. (26) simply
explains the graphical origin of the RHS which gives

1 m m x2 m x2 i m x2
− iC +S exp ,
4 π t πt πt 2t
z z
where C(z) = 0 cos π2 u2 du and S(z) = 0 sin π2 u2 du are Fresnel inte-
grals. Let us denote this A(x, t, m). Now to construct the amplitude Atotal
for a space–time source element for an outcome in the twin-slit experiment,
we have (Fig. 5):
Atotal = A(x1 , t1 , m)A(x3 , t3 , m) + A(x2 , t2 , m)A(x4 , t4 , m),
www.ebook3000.com
time
space space
Slit 2
Source
Detection Event
Slit 1
Fig. 5. Twin-slit interference. The boxes are the components of the space–time
source element depicting mass m loss at the Source emission event and mass
m gain at the Detection Event contributing to an interference pattern at the
detector.
where x1 and t1 are the distance and time from Source to Slit 1, x2 and t2
are the distance and time from Source to Slit 2, x3 and t3 are the distance
and time from Slit 1 to the Detection Event, and x4 and t4 are the distance
and time from Slit 2 to the Detection Event. For an electron traveling at
1.00 m/s through the device (dynamic language), we obtain the following
plots. [Note: The amplitudes of Eqs. (18) and (26) were computed for the
properties of space, time, and mass. In order to model the data for their
twin-slit experiment with electrons, Bach et al. [38] had to modify the “free
space” amplitude to include other properties. Modifications included an
electromagnetic potential at the double slits, an image charge potential at
the collimation slit, and incoherent sources associated with the electron
gun. Therefore, differences in the plots below are not expected to be exper-
imentally observable for electrons (the angles shown below exceed ±π, for
example). The point of this exercise is only to illustrate the manner by

which RBW underwrites QM via the AGC.]
Intensity versus angular displacement in radians for electrons with λ =

728 µm, slit separation of 1.00 mm, screen-to-detector distance of 50.0 cm,
and Source-to-slits distance of 50.0 cm. This is the RBW result. There is
an oscillatory substructure that is suppressed by the horizontal scale (see
inset).
www.ebook3000.com
Fig. 6. Twin-slit interference viewed as waves.
Intensity versus angular displacement in radians for electrons with λ =

728 µm, slit separation of 1.00 mm, screen-to-detector distance of 50.0 cm,
and Source-to-slits distance of 50.0 cm. This is the free-particle SE result
without a source given by Eq. (18). Large maxima align with those in the
RBW result above at this very low energy.
In the standard account of twin-slit interference, quantum waves
emanate from a pair of coherent Sources and strike all along the detec-
tor surface (Fig. 6). Exactly how these waves produce a single point
on the detector surface is then left to interpretation, e.g. they actually
“guide” quantum particles per the de Broglie–Bohm interpretation. The
standard account is a very dynamical view of twin-slit interference. By
contrast, the 4-component space–time source element of Fig. 5 is obtained
in spatio-temporally global fashion, as we described in Sec. 1, i.e. the entire
4-component space–time source element is a fundamental, indivisible, onto-
logical subset of the interacting Source, screen, and detector (again, the
properties of the space–time source element are relational). There is no
worldline connecting the emission event at the Source and the Detection
Event. You can think of this space–time source element as just one of many
responsible for the Source, screen, and detector, to include their spatio-
temporal properties. This understanding of quantum exchanges, interfer-
ence, and space–time source elements, combined with our view of particle
physics (Sec. 4.2), will explain why the Standard Model Lagrangian density
is so complicated (Fig. 8) and why it is not fundamental. We next study the
KG action and find it shares the Hadamard structure of the Schrödinger

result.
3.2. Scalar field on nodes

We now consider Eq. (9). The 4-node graph of Fig. 4 with ψ replaced
by ϕ depicts our space–time source element for this case. Again, we
have an undifferentiated field in the action so we will have to replace
it with a relational form. To mirror the SE discretization we choose
(ϕ∗4 − ϕ∗3 )(ϕ4 − ϕ3 ) + (ϕ∗2 − ϕ∗1 )(ϕ2 − ϕ1 ) for the discretization of ϕ2 in our
space–time source element. This gives (having factored out 12 and ignored
∆x∆t)

1 c2 2 1
 ∆t2 − 2
− m̄ 2
− m̄2
 ∆x ∆t

 1 1 c2
 − m̄ 2
− − m̄ 2
 ∆t2 ∆t2 ∆x2


K=
 c2

 0
 ∆x2

 c2
0
∆x2

c2
0 
∆x2 

c2 
0 
∆x2 

. (27)
1 c 2
1 
− − m̄ 2
− m̄ 2 
∆t2 ∆x2 ∆t2 


1 1 c 2 
2 2
2
− m̄ 2
− 2
− m̄
∆t ∆t ∆x
2 4 2 2 2 4
The eigenvalues are {0, t22 − 2m2c , − 2c 2c 2
x2 , − x2 + t2 −
2m c
2 }, where we
have dropped the ∆ for simplicity, and the corresponding eigenvectors
are Hadamard just as with the SE {(1, 1, 1, 1), (−1, 1, −1, 1), (−1, −1, 1, 1),

(1, −1, −1, 1)}. Again, mode 4 is the J that satisfies our fundamental AGC
so our amplitude is

−1
Jo2
J ·K ·J = 2 2m2 c4
. (28)
− 2c
x2 +
2
t2 − 2
www.ebook3000.com
We next study the Dirac action and find it also shares the Hadamard
structure.
3.3. Vector field on nodes

We apply this approach to vector fields on nodes and note that the KG
operator for scalar fields is the square of the Dirac operator for vector fields,

i.e. (−iγ µ ∂µ − m)(iγ µ ∂µ − m) = (∂ 2 + m2 ). The Dirac K has a nested
Hadamard forml similar to that found in the Schrödinger and KG actions.
That is, reading across the rows for each node one simply has a collection
of the link weights relating the nodes which are connected. The first four

columns of the eigenbasis matrix for K are (column entries top to bottom
read left to right here):

J being orthogonal to each of these vectors simply means that the global

sum over each space–time component of J at each node gives zero, as
required for vector addition over all 16 nodes. We next study the Maxwell
action.
3.4. Scalar field on links

We now apply this approach to gauge fields, for the exchange of energy
via photons. In order to model the construct of action for the exchange
of energy via photons, we use the Maxwell Lagrangian density L for free
electromagnetic radiation
1 αβ
L=− F Fαβ (29)
4µo
with the field strength tensor given by

(Aβ (n + α̂) − Aβ (n)) (Aα (n + β̂) − Aα (n))
Fαβ = ∂α Aβ − ∂β Aα = − ,
α β
(30)
l Again, computational details are in http://arxiv.org/abs/0908.4348.

on the graph [39] where n is the node number, i the lattice spacing
in the ith direction, and α̂ and β̂ are displacements to adjoining nodes

in those directions.
Applying this to the (1 + 1)D case K has eigenval-
ues 0, 0, 0, 2 x12 + t12 . The dimensionality of the row space represents the

degrees of freedom available with local conservation of J . That is, specify-

ing J on just one link dictates the other three values per conservation of J
on the

links at each node.

K for the hypercube has eigenvalues

2 2
{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} − 2 − 2 ,
t x

2 2 2 2 2 t2 − x2 2
− 2− 2 , + , + 2 ,
t y x2 y2 t2 x2 y

2 t2 + x2 2 2 2 2 2
− − , − − , + ,
t2 x2 y2 t2 z2 x2 z2

2 t2 − x2 2 2 t2 + x2 2 2 2
+ , − − , + ,
t2 x2 z2 t2 x2 z2 y2 z2

2 t2 − y 2 2 2 t2 − y 2 2
+ 2 , − − 2 ,
t2 y 2 z t2 y 2 z

2 x2 + y 2 2 2 x2 + y 2 2
+ 2, + 2 ,
x2 y 2 z x2 y 2 z

2 t2 x2 + t2 y 2 − x2 y 2 2 2 t2 x2 + t2 y 2 − x2 y 2 2
+ 2, + 2 ,
t2 x2 y 2 z t2 x2 y 2 z

2 t2 x2 + t2 y 2 + x2 y 2 2
− − 2 .
t2 x2 y 2 z
Again, the dimensionality of the row space (17) represents the degrees of

freedom available with local conservation of J . If we specify J on all 12 links
of the “inner” cube of the hypercube, all the time-like links connecting the
“inner” cube to the “outer” cube are determined by local conservation.
Then if you specify the 4 link values on one face of the “outer” cube,
www.ebook3000.com
local conservation leaves only one free link to specify on the opposite face,
12 + 4 + 1 = 17. We next study the Einstein–Hilbert action.
3.5. Scalar field on plaquettes

This is linearized GR, i.e. the harmonic terms only. We have for the
Einstein–Hilbert Lagrangian density [40]
L = −∂λ hαβ ∂ λ hαβ + 2∂λ hαβ ∂ β hαλ , (31)
omitting a trace term not relevant to our application.m To discretize this

on the hypercube we first label our scalar field on each plaquette according
to its span. For example, the front face of the “inner” cube is spanned by x
and z, so it’s labeled h13 . Of course, there are three other such plaquettes,
one displaced from the front towards the back (in y) of the “inner” cube, one
displaced in t to the front of the “outer” cube, and one displaced in t and y to
the back of the “outer” cube. There are six fields (h01 , h02 , h03 , h12 , h13 , h23 )
which generate such a quadruple, accounting for all 24 plaquettes of the
hypercube. Likewise, for the cube we have (h01 , h02 , h12 ) and their pairing
partners giving us the six plaquettes.

One null eigenbasis for K of the hypercube is
{{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1},
{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0},
{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0},
{0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
{0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
{1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}}.

J orthogonal to each of these null eigenvectors means it is conserved across
each set of four plaquettes associated with (h01 , h02 , h03 , h12 , h13 , h23 ). We
are now in position to interpret the Standard Model per MLGT, which
makes clear how we approach unification and QG.
m The missing trace term is gauge equivalent to 2∂α hα

µ which would be used for multiple,
connected graphical elements.
4. Unification and QG
4.1. The Standard Model
Strictly speaking, when finding the gradient of a vector field on the graph
as we did with the Dirac operator, we need to specify a means of parallel
transport. So, in our view and that of LGT, local gauge invariance is seen
as a modification to the matter field gradient on the graph required by
parallel transport per Uµ , i.e.

µ 0 U0 ψ̃0 − ψ 1 U1 ψ̃1 − ψ
γ Dµ ψ = γ +γ + ··· , (32)
ct x
where ψ̃i is the vector field on the node adjacent to ψ in the positive
ith direction. The Lagrangian density L = ψ̄(iγ µ ∂µ + eγ µ Aµ − m)ψ −
1 αβ
4µo F Fαβ is therefore seen as the addition of parallel transport and a
curvature !term A† (∂2 † ∂2 )A, where A generates the parallel transport, to
†
L = 12 ψ̄ ∂1 ∂1 ψ to produce a well-defined field gradient between

ψ̃i and ψ. K now has the form
  
Dirac
  
plus  

 0 
K =  parallel  ,
 
 transport 
0 (Maxwell)

where Dirac K has been modified to contain Aµ . Each row of the Dirac-plus-

parallel-transport K sums to zero, since it still has the form as the Dirac

K, so it possesses a non-trivial null space (that of Maxwell K is obviously
unaffected). The transition amplitude no longer has the simple Gaussian

form since Dirac K is now a function of one of the fields of integration. And

J in the row space of K now contains terms on links and nodes, representing
conservation of 4-momentum between the interacting fields.
www.ebook3000.com
By introducing two vectors at each node, this same standard requires

 1 
C011 C012 ψ̃01 ψ
 C021 C022 ψ̃ 2 − ψ 2 
 0 
γ µ Dµ ψ = γ 0  
 ct 
 1 
C111 C112 ψ̃11 ψ
 C121 C122 ψ̃ 2 − ψ 2 
 1 
+ γ1   + ··· , (33)
 x 
where the matrix Cµab is an element of SU (2) associated with the link in
1
ψ
the positive µth direction from . Again, we have the same form for
ψ2
our field gradients, i.e. the nodal field gradients parallel transported by
the link field, which still contributes a gradient to the Lagrangian density
"
− 4g12 a,α,β Fαβ
a a
Fαβ where g is the coupling constant and a = 1, 2, 3, since
SU (2) has three generators. You can see that now the pure gauge part
(“Maxwell” part) of the Lagrangian density contains third- and fourth-order

terms in the gauge field. Thus, Maxwell K now contains the gauge field,

just like Dirac-plus-parallel-transport K. We can symmetrize this Maxwell

+K so that the rows sum to zero and it possesses a non-trivial null space.

It is now the case that both the matter field and gauge field portions of K

contain the gauge field. Thus, we see the progression from free field K to

Abelian-interaction K to non-Abelian-interaction K is a simple progression

from K with no gauge field terms to Dirac K with gauge field terms to both

Dirac and Maxwell K with gauge field terms.
This pattern extends to SU (3)n where each link has eight different val-
ues for the gauge field (one for each generator of SU (3)) which we label Aaµ
with a = 1, 2, . . . , 8. And, each of the eight values can be different on differ-
ent links. Again, the pure gauge part (“Maxwell” part) of the Lagrangian
density contains third- and fourth-order terms in the gauge field and we can

symmetrize Maxwell +K such that the rows sum to zero and it possesses a
n In fact, this pattern extends to SU (N).

non-trivial null space. All possible mixing between U (1), SU (2), and SU (3)
forms the Standard Model. We next explain particle physics per RBW.
4.2. Particle physics

In our graphical approach, the role of the field is different than in the
continuum approach of QFT where it pervades otherwise empty, continuous
space to mediate the exchange of matter-energy between sources. In MLGT,
as in LGT, the field is a scalar or vector associated with nodes, links, or
plaquettes on the graph. One obtains QFT results from LGT by letting the
lattice spacing go to zero. In fact, one can understand QFT renormalization
through this process of lattice regularization [41]. As it turns out, however,
this limit does not always exist, so calculated values are necessarily obtained
from small, but non-zero, lattice spacing [39]. With this picture in mind,
we can say simply what we are proposing: the lattice is fundamental, not
its continuum limit. Once one accepts this premise, it is merely a matter of
degree to have large space–time source elements, which is the basis for our
explanation of the twin-slit experiment (Sec. 3.2) and dark energy (Sec. 5.2).
In this approach, there is no graphical counterpart to “quantum systems”
traveling through space as a function of time from Source to sink to “cause”
detector clicks. This implies the empirical goal at the fundamental level is
to tell a unified story about detector events to include individual clicks —
how they are distributed in space (e.g. interference patterns, interferometer
outcomes, spin measurements), how they are distributed in time (e.g. click
rates, coincidence counts), how they are distributed in space and time (e.g.
particle trajectories), and how they generate more complex phenomena (e.g.
photoelectric effect, superconductivity). Thus in our theory, particle physics
per QFT is in the business of characterizing large sets of detector data, i.e.
all the individual clicks.
As was eminently apparent from our examples in Sec. 3, it is practically
impossible to compute Z(J) in MLGT for all possible spatio-temporally
relative click locations in a particle physics “event”, which contains
“approximately 100,000 individual measurements of either energy or spa-
tial information [42]”. However, we know from theory [43] and experi-
ment that, with overwhelming probability, detector clicks will trace classical
pathso , so it makes sense to partition large click distributions into individual
o Individual detector clicks (called “hits in the tracking chamber”) are first localized spa-
tially (called “preprocessing”), then associated with a particular track (called “pattern
www.ebook3000.com
trajectories and treat these as the fundamental constituents of high energy

physics experiments.p This is exactly what QFT does for particle physics
according to our interpretation. Since the individual trajectories are them-
selves continuous, QFT uses propagators in continuous space–time which
entails an indenumerably infinite number of locations for both clicks and
interaction vertices. Thus, issues of regularization and renormalization are
simply consequences of the continuum approximation necessary to deal with
very large click distributions, having decided to parse the click distributions
into continuum trajectories.
Essentially, we are saying a particle physics detector event is one giant
interference pattern, as we previously characterized interference, and the
way to understand a particular pattern involving thousands of clicks can
only realistically be accomplished by parsing an event into smaller subsets,
and the choice of subsets is empirically obvious, i.e. space–time trajectories.
These trajectories are then characterized by mass, spin, and charge. Per
RBW’s adynamical explanation, the colliding beams in the accelerator and
the detector surrounding the collision point form the graphical input that,
in conjunction with the AGC, dictate the possible configurations of space–
time source elements responsible for particle trajectories. Each trajectory
can be deduced one click at a time in succession using ψ(x,t), as shown
by Mott for alpha particles in a cloud chamber [43]. Therefore, a proba-
bility amplitude could be computed for each trajectory using space–time
source elements detection event by detection event a la our twin-slit analysis
above. However, as shown by Mott, after the first click the remaining clicks
recognition”). The tracks must then be parameterized to obtain dynamical characteris-

tics (called “geometrical fitting”). See Fernow, R.C. Introduction to Experimental Parti-
cle Physics, Cambridge University Press, Cambridge, 1986, Secs. 1.7.1, 1.7.2, and 1.7.3,
respectively.
p Some assumptions are required, e.g. “Sometimes it is necessary to know the identity (i.e.
the mass) of at least some of the particles resulting from an interaction” (Fernow, 1986,
17), “Within the errors [for track measurements], tracks may appear to come from more
than one vertex. Thus, the physics questions under study may influence how the tracks
are assigned to vertices” (Fernow, 1986, 25), and “Now there must be some minimum
requirements for what constitutes a track. Chambers may have spurious noise hits, while
the chambers closest to the target may have many closely spaced hits. The position of
each hit is only known to the accuracy of the chamber resolution. This makes it difficult
to determine whether possible short track combinations are really tracks” (Fernow, 1986,
22). Despite these assumptions, no one disputes the inference. While we do not subscribe
to the existence of “click-causing entities” with trajectories, we agree that clicks trace
classical paths. Indeed, this is the basis for our approach and consequently, the results
and analysis of particle physics experiments are very important.
follow a classical trajectory with high probability, so the only real quan-
tum computation needed is for the probability amplitude of the space–time
source element of the set of first clicks, i.e. the first click for each trajectory
in the collection. And, the properties (mass, charge, momentum, energy,
etc.) for that space–time source element would simply be the properties of
the subsequent particles defined relationally in the context of the accelera-
tor Source and particle detector. In standard LGT → QFT, the calculated
outcomes are found by taking the limit as the lattice spacing goes to zero
via renormalization, but we need not assume the spacing goes to zero, since
our sinks are the pixel locations in the detector CCD. Likewise, assuming
the accelerator and detectors are sufficiently isolated during the brief period
of data collection, the graph size is not infinite as in QFT. This, of course,
justifies the UV and IR cutoffs in regularization, respectively.
This severely undermines the dynamical picture of perturbations moving
through a continuum medium (naı̈ve field) between sources, i.e. it under-
mines the naı̈ve notion of a particle as traditionally understood. In fact,
the typical notion of a particle is associated with the global particle state
of n-particle Fock space and per Colosi and Rovelli “the notion of global
particle state is ambiguous, ill-defined, or completely impossible to define
[44].” What we mean by “particle” is a collection of detector hits forming
a space–time trajectory resulting from a collection of adynamically con-
strained space–time source elements in the presence of colliding beams and
a detector. And this does not entail the existence of an object with intrinsic
properties, such as mass and charge, moving through the detector to cause
the hits.
Our view of particles agrees with Colosi and Rovelli [44] on two impor-
tant counts. First, that particles are best modeled by local particle states
rather than n-particle Fock states computed over infinite regions, squaring
with the fact that particle detectors are finite in size and experiments are
finite in time. The advantage to this approach is that one can unambigu-
ously define the notion of particles in curved space–time as excitations in
a local M4 region, which makes it amenable to Regge calculus. Second,
this theory of particles is much more compatible with the quantum notion
of complementary observables in that every detector has its own Hamilto-
nian (different-sized graph), and therefore its own particle basis (unlike the
unique basis of Fock space). As per Colosi and Rovelli, “In other words, we
are in a genuine quantum mechanical situation in which distinct particle
numbers are complementary observables. Different bases that diagonalize
different HR [Hamiltonian] operators have equal footing. Whether a particle
www.ebook3000.com
exists or not depends on what I decide to measure.” Thus, in our view, par-
ticles simply describe how detectors and Sources are relationally co-defined
via the AGC.
4.3. Unification and QG

With this understanding of the Standard Model and particle physics, we see
that the next logical addition to our collection of fundamental space–time
source elements would be those constructed from the gradient of vector
fields on links. The scalar field on plaquettes (basis for graviton in standard
LGT) would define parallel transport for this field gradient in the manner
scalar fields on links defines parallel transport for the vector fields on nodes.
This is the standard approach to QG in the particle physics community.
The problem with this is, of course, we simply have gravitons in M4, we
still need space–time curvature as in GR. In our view, since MLGT is
necessarily contextual, that is accomplished by understanding the context
of the properties in question a la the simplices of Regge calculus (see Sec.
5.1 and Fig. 9).
According to Regge calculus, gravity is a scalar field on the plaque-
ttes of its simplices, i.e. Newtonian gravity in M4, and space–time curva-
ture (variable space–time geometry) is accounted for via “deficit angles”
between simplices in the global structure (Fig. 9). This space–time cur-
vature is a function of the energy–momentum content of space–time (Sec.
5.1) to include all forms, not just the graviton (associated with the property
of mass). Thus, if we were able to construct experiments with individual
gravitons, a QG experiment could be the simple twin-slit experiment with
gravitons in M4, i.e. in a single simplex. Or, we could view graviton interfer-
ence patterns generated between a Source and detector in non-neighboring
simplices, in which case the space–time source element for the exchange
of gravitational energy would have simplex-to-simplex segments between
Source and Detection Event. To compute the amplitude for the exchange
of gravitational energy associated with that space–time source element, one
would pick up phase factors associated with the deficit angles between sim-
plices, just as a photon amplitude picks up phase factors associated with
reflection from mirrors and beam splitters as computed between Source and
Detection Event in an interferometer.
Since we do not yet have the technology to manipulate individual gravi-
tons, we might rather explore the effect of variable geometry on space–time
source elements passing through neighboring simplices for the exchange of
photons (as on astrophysical scales). Since different energy distributions

mean different space–time geometries per Regge calculus, an exchange of
energy, quantum or otherwise has geometric implications. And, since geom-
etry affects the value of the action in Regge calculus, one should expect that
energy lost at the Source will not only reappear at the detection event, but
will also appear in the space–time geometry, as compared to the geometry
where the photon energy exchange is not taken into account. We consider
just such an example of this “disordered locality” in Sec. 5.2 below. There
the energy lost by the photon per its cosmological redshift between Source
(supernova) and detector (telescope) in a cosmology model based on matter
alone (no electromagnetic contribution to the space–time geometry) will be
used to justify a first-order correction to the proper distance between Source
and detector in that dust-filled cosmology model. That correction is small
(scaled by a factor of (8.38 Gcy)−1 ), but we show that it does allow for a fit
of distance modulus versus redshift for the Union2 Compilation supernovae
data without accelerating expansion and, therefore, without dark energy.
At this point, the reader should appreciate that underwriting interacting
COs via space–time source elements leads to a relatively simple picture
of unification (Fig. 7) compared to that based on fundamental particles
(Fig. 8). However, while we do not view particle physics as the study of what
is ultimately fundamental in Nature, it has been essential to understanding
how the fundamental elements of space–time source are to be constructed

and combined, and what properties are represented by J , as we explained
in Secs. 3.2 and 4.2. Since high energy particle physics deals with large
energy densities, our disordered locality does not need to be taken into
account. Disordered locality becomes a concern for relatively small energy
exchanges over relatively large spatio-temporal regions. We encounter that
Scalar field on One vector each One vector each

nodes node link
Scalar field on Two vectors each Two vectors each

links node link
Scalar field on Three vectors each Three vectors each

plaquettes node link
Fig. 7. Fundamentals space–time source elements for unification via MLGT.
www.ebook3000.com
Fig. 8. The Standard Model Lagrangian Density. Credit: T.D. Gutierrez.

http://nuclear.ucdavis.edu/∼tgutierr/files/sml.pdf.
situation in astrophysics, so that is where we expect disordered locality per

our version of QG to become important.
5. Implications for Astrophysics and Cosmology

5.1. Regge calculus
In Regge calculus, the space–time manifold is replaced by a lattice geometry
where each 4D cell (simplex) is Minkowskian (flat). Curvature is represented
by “deficit angles” (Fig. 9) about any plane orthogonal to a “hinge” (tri-
angular side to a tetrahedron, which is a 3D side of a 4D simplex). The
1 "
Hilbert action for a 4D vacuum lattice is IR = 8π σi ∈L εi Ai where σi
is a triangular hinge in the lattice L, Ai is the area of σi and εi is the
deficit angle associated with σi . The counterpart to Einstein’s equations is
then obtained by demanding δI R
δ2j
= 0, where 2j is the squared length of
the jth lattice edge, i.e. the metric. To obtain equations in the presence
of matter-energy, one simply adds the appropriate term IM−E to IR and
Fig. 9. A 2-geometry with continuously varying curvature can be approximated

arbitrarily closely by a ployhedron built of triangles, provided only that the num-
ber of triangles is made sufficiently great and the size of each sufficiently small.
The geometry in each triangle is Euclidean. The curvature of the surface shows
up in the amount of deficit angle at each vertex (protion ABCD of ployhedron
laid out above on a flat surface).
Source: Reproduced from C.W. Misner, K.S. Thorne, J.A. Wheeler, W.H. Grav-
itation. Freeman, San Francisco, 1973, 1168.
www.ebook3000.com
carries out the variation as before to obtain δI R

δ2j
= − δIδ
M −E
2 . One finds the
j
stress-energy tensor is associated with lattice edges, just as the metric, and
Regge’s equations are to be satisfied for any particular choice of the two
tensors on the lattice. [Note that the AGC nature of GR is particularly
evident in the Regge calculus approach.]
5.2. Dark energy and other astrophysical implications

Since one recovers GR from Regge calculus by making the simplices small
(as in LGT → QFT), it seems that empirical evidence of the deviation
from GR phenomena posed by the large space–time source elements of dis-
ordered locality, i.e. MORC, might be found in the exchange of photons on
cosmological scales. Therefore, we modified the Regge calculus approach to
EdS cosmology [5] and compared this MORC model, EdS, and the con-
cordance model ΛCDM (EdS plus a cosmological constant Λ to account
for dark energy) using the data from the Union2 Compilation, i.e. dis-
tance moduli and redshifts for type Ia supernovae [45] (Fig. 10). We found
that a best fit line through log(DL /Gpc) versus log(z) gives a correlation of
0.9955 and a sum of squares error (SSE) of 1.95. By comparison, the best fit
ΛCDM gives SSE = 1.79 using a Hubble constant of Ho = 69.2 km/s/Mpc,
ΩM = 0.29 and ΩΛ = 0.71. The parameters for ΛCDM yielding the most
robust fit to [46] “the Wilkinson Microwave Anisotropy Probe data with
the latest distance measurements from the Baryon Acoustic Oscillations
44
42
40
38
36
34
z
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Fig. 10. Plot of Union2 data along with the best fits for EdS (dashed), ΛCDM
(gray), and MORC (dotted). The MORC curve is terminated at z = 1.4 in this
figure so that the ΛCDM curve is visible underneath.
in the distribution of galaxies and the Hubble constant measurement” are

Ho = 70.3 km/s/Mpc, ΩM = 0.27 and ΩΛ = 0.73, which are consistent
with the parameters we find for its Union2 Compilation fit. The best fit
EdS gives SSE = 2.68 using Ho = 60.9 km/s/Mpc. The best fit MORC
gives SSE = 1.77 and Ho =! 73.9 km/s/Mpc with the EdS proper distance
D
Dp corrected by a factor of 1 + Ap where A = 8.38 Gcy. A current “best
estimate” for the Hubble constant is Ho = (73.8±2.4) km/s/Mpc [47]. Thus,
MORC improves EdS as much as ΛCDM in accounting for distance mod-
uli and redshifts for type Ia supernovae even though the MORC universe
contains no dark energy and is therefore always decelerating. So, per our
theory, it is quite possible that this data does not constitute “the discovery
of the accelerating expansion of the Universe”, (Nobel citation, 2011), i.e.
there is no accelerating expansion, so there is no need of a cosmological
constant or dark energy in any form [6].
Our theory has other possible implications for astrophysics and cosmol-
ogy as well. Perhaps, MORC’s version of the Schwarzschild solution will
negate the need for dark matter as its counterpart to Eds cosmology did
with dark energy. What will MORC have to say about the event horizon
and singularity in the Schwarzschild solution, i.e. black holes? Perhaps, the
singularity will be avoided as in Regge calculus cosmology where backwards
time evolution “stops” at a time determined by the choice of lattice spac-
ing.q And, with an adynamical approach, cosmological explanation takes
on an entirely new form. No longer is one seeking explanation in the form
of a time-evolved spatial hypersurface of homogeneity — an explanation
that cannot be satisfied with the Big Bang or even a non-singular “stop
point”. Thus, such dynamical explanation results in contentious, mislead-
ing, or unverifiable notions about [17] “creation from nothing,” the mul-
tiverse, etc. Rather, explanation via AGC does not rest ultimately on the
Big Bang or any other region of the graph.r The reason the fields on node
X and link Y have the values they do is required by the solution for the
entire graph, i.e. it is required by the values of the fields on all the other
nodes and links. As we pointed out in Sec. 1 when we contrasted dynamical
q This is the “stop point problem” of Regge calculus cosmology. Of course it’s not a
“problem” for our approach, because Regge calculus is fundamental to GR, not the
converse, so one does not require Regge calculus reproduce the initial singularity of GR
cosmology.
r See Rovelli, C. Why do we remember the past and not the future? The “time oriented
coarse graining” hypothesis http://arxiv.org/abs/1407.3384, 2014, for an idea that is

similar in spirit.
www.ebook3000.com
explanation with our adynamical global constraint explanation, no region

of the graph is distinguished over any other in this explanatory scheme.
6. Conclusion
Our combination of MLGT and MORC is an entirely new way to reconcile
QFT with GR and unify physics:
QFT ← MLGT ↔ MORC → GR.
Many of the major questions that need to be answered in this new view of
unification and QG are clear. Is there a limit to the number of vectors that
can be (or need be) introduced on nodes and links? If so, does it have to do
with information density? Is it related to quark confinement? Or, is there
a purely mathematical fact that underwrites it? Why is there no physical
counterpart to a scalar field on cubes? Is this because it requires (4 + 1)D
to close graphically and satisfy the boundary of a boundary principle for
all graphical entities? What physical objects correspond to vector fields
on links? Are they just quarks and leptons interacting gravitationally? Or,
will this generate new fermions that only interact gravitationally, e.g. dark
matter? While these questions are not going to be answered in a trivial
fashion, we believe the RBW program of unification and QG offers a viable
alternative to the existing landscape. In fact, MORC has already produced
an empirical result, i.e. an explanation of dark energy, as we showed in
Sec. 5.
Our explanation of dark energy resulted from RBW’s modification to
GR cosmological proper distance per disordered locality (MORC). Astro-
physical data is very amenable to analysis via MORC, since it represents
low energy exchanges over large spatio-temporal extents. Thus, we expect
that dark matter is also a candidate for explanation via kinematical correc-
tions per MORC. In contrast, high energy physics deals with large energy
densities and that is precisely where we expect analytic techniques such
as those of the Standard Model to work well. Thus, we do not suggest
any sweeping changes to the formalism of particle physics as it is currently
employed. Rather, MLGT vindicates the formalism by providing rationale
for some of its questionable techniques, e.g. UV and IR cutoffs in regu-
larization. Instead, we would expect to see corrections to QFT in the low
energy regime, which is where QM takes over. In fact, it is in this regime
where experiments have vindicated some of QM’s most “mysterious” pre-
dictions, e.g. delayed-choice experiments, and it was just such phenomena
that inspired RBW. Thus, we expect to expand the list of “mysterious”

quantum predictions in the context of QFT by looking for particle physics
effects related to detector size and shape, for example. We will stop such
speculation at this point, given the incipient nature of MLGT and MORC.
References
1. W.M. Stuckey, M. Silberstein, and T. McDevitt, Relational blockworld: Pro-
viding a realist psi-epistemic account of quantum mechanics, Int. J. Quantum
Found. 1(3), 123–170 (2015), http://www.ijqf.org/archives/2087.
2. M. Silberstein, W.M. Stuckey, and T. McDevitt, Being, Becoming and the
undivided Universe: A dialogue between relational blockworld and the impli-
cate order concerning the unification of relativity and quantum theory,
Found. Phys. 43(4), 502–532 (2013), http://arxiv.org/abs/1108.2261.
3. W.M. Stuckey, M. Silberstein, and M. Cifone, Reconciling spacetime and the
quantum: Relational blockworld and the quantum liar paradox, Found. Phys.
38(4), 348–383 (2008), http://arxiv.org/abs/quant-ph/0510090.
4. M. Silberstein, W.M. Stuckey, and M. Cifone, Why quantum mechanics favors
adynamical and acausal interpretations such as relational blockworld over
backwardly causal and time-symmetric rivals, Stud. His. Phil. Mod. Phys.,
39(4), 736–751 (2008).
5. W.M. Stuckey, T. McDevitt, and M. Silberstein, Modified Regge Calculus
as an explanation of dark energy, Classical Quan. Grav. 29, 055015 (2012),
http://arxiv.org/abs/1110.3973.
6. W.M. Stuckey, T. McDevitt, and M. Silberstein, Explaining the supernova
data without accelerating expansion, Int. J. Mod. Phys. D 21(11), 1242021
(2012).
7. S. Gerlich, et al. Quantum interference of large organic molecules, Nat.
Comm. 2, 263, doi: 10.1038/ncomms1263 (2011).
8. A. Danan, D. Farfurnik, S. Bar-Ad, and L. Vaidman, Asking photons where
they have been, Phys. Rev. Lett. 111, 240402 (2013), http://arxiv.org/abs/
1304.7469.
9. C. Rovelli, Why Gauge? http://arxiv.org/pdf/1308.5599v1.pdf, 2013, p. 7.
10. A. Einstein, Quantum mechanics and reality, Dialectica 2, 320–324 (1948).
11. T. Maudlin, Completeness, supervenience, and ontology, J. Phys. A 40, 3151–
3171 (2007).
12. J.S. Bell, Speakable and Unspeakable, Cambridge University Press, Cam-
bridge, 1987, 234.
13. N. Huggett, and C. Wüthrich, Emergent spacetime and empirical
(in)coherence. Stud. His. Phil. Mod. Phys. 44, 276–285 (2013), http://arxiv.
org/pdf/1206.6290.pdf.
14. F. Caravelli, and F. Markopoulou, Disordered Locality and Lorentz Dis-
persion Relations: An Explicit Model of Quantum Foam, http://arxiv.org/
pdf/1201.3206v1.pdf, 2012.
www.ebook3000.com
15. C. Prescod-Weinstein, and L. Smolin, Disordered locality as an explana-

tion for the dark energy, Phys. Rev. D 80, 063505 (2009), http://arxiv.org/
pdf/0903.5303.pdf.
16. S. Weinstein, Patterns in the Fabric of Nature. FQXi Essay Contest http://
www.fqxi.org/community/forum/topic/1529, 2012.
17. S. Carroll, A Universe from Nothing? Discover Magazine Online http://
blogs.discovermagazine.com/cosmicvariance/2012/04/28/a-universe-from-
nothing/, 2012.
18. See examples in Oriti, D. Disappearance and Emergence of Space and Time
in Quantum Gravity, http://arxiv.org/abs/1302.2849, 2013.
19. J. Ambjorn, A. Goerlich, and R. Loll, Quantum Gravity via Causal Dynam-
ical Triangulations http://arxiv.org/abs/1302.2173, 2013.
20. R. Sorkin, Relativity theory does not imply that the future already exists: A
counterexample http://arxiv.org/pdf/gr-qc/0703098.pdf, 2007.
21. C. Rovelli, ‘Localization’ in quantum field theory: How much of QFT is com-
patible with what we know about spacetime?, in T. Cao, (ed.), Conceptual
Foundations of Quantum Field Theory, Cambridge University Press, Cam-
bridge, 1999, pp. 207–232.
22. H. Weyl, Space, Time, Matter, Dover Publications, New York, 1952.
23. R. Healey, Gauging What’s Real: The Conceptual Foundations of Gauge The-
ories. Oxford University Press, Oxford, 2007, 47.
24. J. Ladyman, Structural Realism in E.N. Zalta, (ed.), The Stanford Ency-
clopedia of Philosophy, http://plato.stanford.edu/archives/sum2009/entries/
structural-realism (2009).
25. J. Ladyman, D. Ross, D. Sturrett, and J. Collier, Everything Must Go: Meta-
physics Naturalized, Oxford University Press, Oxford, 2007.
26. D. Rickles, Time Observables and Structure, in E. Landry, and D. Rickles,
(ed.) Structural Realism: Structure, Object, and Causality, Springer, New
York, 2012a, pp. 135–148.
27. M. Esfeld, Ontic structural realism and the interpretation of quantum
mechanics, EJPS 3(1), 19–32 (2013).
28. D. Rickles, and J. Bloom, Things Ain’t What They Used to Be. FQXi Essay
Contest, http://www.fqxi.org/community/forum/topic/1537, 2012b.
29. R. Geroch, General Relativity from A to B, University of Chicago Press,
Chicago, 1978, pp. 20–21.
30. M. Henneaux, and C. Teitelboim, Quantization of Gauge Systems, Princeton
University Press, Princeton, 1992, 106.
31. C. Kiefer, Time in quantum gravity in C. Callender, (ed.) The Oxford Hand-
book of Philosophy of Time, Oxford University Press, Oxford, 2011, p. 667.
32. K. Wharton, Quantum states as ordinary information, Information 5, 190–
208 (2014).
33. C.W. Misner, K.S. Thorne, and J.A. Wheeler, Gravitation. W.H. Freeman,
San Francisco, 1973, p. 364.
34. D.K. Wise, p-Form electromagnetism on discrete spacetimes, Classical Quan.
Grav. 23, 5129–5176 (2006).
35. A. Zee, Quantum Field Theory in a Nutshell, Princeton University Press,

Princeton, 2003, p. 172.
36. R. Shankar, Principles of Quantum Mechanics, 2nd Edition Plenum Press,
New York, 1994, p. 226.
37. See for example: http://www.phy.ohiou.edu/∼elster/lectures/relqm 17.pdf.
38. R. Bach, D. Pope, L. Sy-Hwang, and H. Batelaan, Controlled double-slit
electron diffraction, New J. Phys. 15, 033018 (2013).
39. H.J. Rothe, Lattice Gauge Theories, An Introduction, World Scientific,
Singapore, 1992, p. 72.
40. T. Padmanabhan, From Gravitons to Gravity: Myths and Reality
http://arxiv.org/pdf/gr-qc/0409089v1.pdf, 2004.
41. G. ‘t Hooft, The Conceptual Basis of Quantum Field Theory in J. Butterfield,
J. Earman, (eds.), Philosophy of Physics Part A, North-Holland, Amsterdam,
2007, pp. 661–729.
42. H.J. Frisch, Pattern recognition at the Fermilab collider and Superconducting
Supercollider. Proc. Nat. Acad. Sci. 90, 9754–9757, 1993.
43. N. Mott, The wave mechanics of alpha-ray tracks. Proc. of the Royal Society
A126, 79–84, 1929.
44. D. Colosi, and R. Rovelli, What is a particle?, Classical Quan. Grav. 26,
025002 (2009), http://arxiv.org/abs/gr-qc/0409054.
45. R. Amanullah, C. Lidman, D. Rubin, G. Aldering, P. Astier, K. Barbary,
M.S. Burns, A. Conley, K.S. Dawson, S.E. Deustua, M. Doi, S. Fabbro, L.
Faccioli, H.K. Fakhouri, G. Folatelli, A.S. Fruchter, H. Furusawa, G. Gar-
avini, G. Goldhaber, A. Goobar, D.E. Groom, I. Hook, D.A. Howell, N.
Kashikawa, A.G. Kim, R.A. Knop, M. Kowalski, E. Linder, J. Meyers, T.
Morokuma, S. Nobili, J. Nordin, P.E. Nugent, L. Ostman, R. Pain, N. Pana-
gia, S. Perlmutter, J. Raux, P. Ruiz-Lapuente, A.L. Spadafora, M. Strovink,
N. Suzuki, L. Wang, W.M. Wood-Vasey, and N. Yasuda, (The Supernova
Cosmology Project): Spectra and light curves of six type Ia supernovae at
0.511 < z < 1.12 and the Union2 compilation, Astrophys. J. 716, 712–738
(2010) http://arxiv.org/abs/1004.1711.
46. E. Komatsu, K.M. Smith, J. Dunkley, C.L. Bennett, B. Gold, G. Hinshaw,
N. Jarosik, D. Larson, M.R. Nolta, L. Page, D.N. Spergel, N. Halpern, R.S.
Hill, A. Kogut, M. Limon, S.S. Meyer, N. Odegard, G.S. Tucker, J.L. Weiland,
E. Wollack, and E.L. Wright, Seven-Year Wilkinson Microwave Anisotropy
Probe (WMAP) Observations: Cosmological Interpretation, 192, 18 (2011)
http://arxiv.org/abs/1001.4538.
47. A.G. Riess, L. Macri, S. Casertano, H. Lampeitl, H.C. Ferguson, A.V.
Filippenko, S.W. Jha, W. Li, and R. Chornock, A 3% solution: Determi-
nation of the Hubble constant with the Hubble space telescope and wide
field camera 3, Astrophys. J. 730, 119 (2011).
www.ebook3000.com
Chapter 20
Is Bohr’s Challenge Still Relevant?

Leonardo Chiatti
AUSL VT Medical Physics Laboratory,
Via Enrico Fermi 15, 01100 Viterbo (Italy)
leonardo.chiatti@asl.vt.it
We argue that not all the theoretical content of the Bohr model has been
captured by the “definitive” quantum formalism currently in use. In par-
ticular, the notion of “quantum leap” seems to refer to non-dynamic fea-
tures, closely related to non-locality, which have not yet been formalized
in a satisfactory way.
1. Introduction
The Bohr–Rutherford planetary model [1] is still the general public’s
favourite image of the mystery of the atom because of its simplicity and
“visualizability”. For students of physics and chemistry, it represents a sort
of inevitable rite of passage on the path towards orbitals and quantum
mechanics (QM). As an educational tool, it allows the soft introduction of
the quantum of action h, by its appearance in a series of constraints on
the otherwise classical motion of electrons represented as classical material
points.
Even in a scientific perspective, the model has partly reemerged in
the context of semi-classical approaches to the quantization of atomic and
molecular structures [2–10]. In particular, Bucher [11–13] shows that by
removing the condition of the impenetrability of matter (this assumption
seems reasonable with reference to the time when the model was developed,
but is now known to be less significant on a microphysical scale) origi-
nally introduced by Bohr and Sommerfeld, a different counting of states is
obtained, in better agreement with that offered by QM.
On the centenary of its formulation (1913) it is appropriate to consider
whether, beyond its educational role, the Bohr model is still relevant to
545
546 L. Chiatti
current research on the foundations of QM. This short chapter argues that
it is. This personal centennial celebration does not propose any new concept,
but nevertheless covers a range of issues that are generally overlooked in
the historical and educational debate on the Bohr model.
2. Context
First, it is worth mentioning that the Bohr model was not the first atomic
model to include the quantum of action h.a After Planck’s seminal work,
Johannes Stark was probably the first physicist to understand the link
between this new constant and the micro-world [14–16], playing an impor-
tant role in the dissemination of this concept among German physicists in
the first decade of the 20th century (although, oddly enough, he did not
include this concept in the atomic model he proposed [17]).
In 1910, Arthur Erich Haas presented his quantum model of the hydro-
gen atom [18–20], probably as a result of these suggestions. In contemporary
terms, it could be said that Haas derived a semi-classical quantization of
the ground state of this atom, in the context of Thomson’s plum pudding
model. This approach provided the correct expression for the radius of the
atom in terms of the charge and mass of the electron and h (currently
known as “the Bohr radius”). However, it is not clear whether Haas con-
sidered h as a new fundamental constant. His choice of Thomson’s model
as the theoretical framework was largely justified by its classical stability
(unlike Rutherford’s planetary model). The sole purpose of quantization
was here to constrain the radius.
It is a well-known fact that Bohr chose Rutherford’s model as his frame-
work; indeed, his assiduous presence in Rutherford’s laboratory allowed
him to acquire first-hand results of well-known experiments that led to the
rejection of Thomson’s model. Further, Bohr was firmly convinced of the
fundamental nature of the quantum of action [1].
The existence, fundamentality and irreducibility of this quantum make
the analysis of physical phenomena over time, with a level of detail equiv-
alent to variations of action significantly smaller than h impossible. Thus,
it is reasonable to expect processes that cannot be causally analyzed on an
atomic scale. Bohr was aware of the need for a new mechanical approach
(based on the finiteness of h) expressing this limitation, and of the purely
a In this chapter we do not discuss J.W. Nicholson’s important contribution.
www.ebook3000.com
provisional nature of a representation of the atom based on electronic orbits

and other classical elements.
3. Quantum Leaps
Putting a part their different theoretical framework, which is irrelevant here,
Bohr’s model contributed three important innovative elements compared to
Haas. It leads:
(1) To the quantization of the motion of material bodies (quantization of

orbits).
(2) To the field quantization (the application of Planck’s law to emit-
ted/absorbed radiation: the concept of photon).
(3) To the existence of discontinuous leaps from one stationary state to
another (quantum leaps).
Haas’s proposal led uniquely to result (1), limited exclusively to the ground
state, and did not address transitions — points (2) and (3). Bohr’s model
therefore allowed predictions about the position of spectral lines that were
impossible with Haas’s approach.
The subsequent developments are discussed in textbooks on the history
of physics [21,22]. The research inspired by Bohr’s model paved the way for
the construction of a complete system of formally self-consistent quantum
physics. With regards to point (1), this led to QM and the first quantization
formalism, whereas point (2) led to quantum field theory (QFT) and the
second quantization formalism. The history books, as well as the accounts of
the legacy of the Bohr model which is currently considered entirely absorbed
in QM and QFT formalisms, normally stop here.
However, point (3) also exists. In experiments involving micro-objects,
a “quantum leap” is the event which prepares the initial quantum state, or
detects the final quantum state. In other words, quantum leaps are some-
how connected to the projection of the temporal evolution of the initial
state on the final state: the infamous “collapse” of the wave-function. This
collapse is controversial as QM formalism does not specify when and how
this event occurs. This lack of formal description, according to a minimalist
interpretation such as the Copenhagen interpretation, leads to well-known
paradoxes such as Wigner’s friend, Schrödinger’s cat, and so forth.
Therefore, it can safely be asserted that point (3) has not yet been prop-
erly developed in the context of an appropriate self-consistent formalism.
The challenge launched by Bohr is still relevant to this day.
548 L. Chiatti
4. Beyond Time
Given the above, it is reasonable to ask why Bohr decided to frame QM
in a meta-theoretical structure — the well-known “Copenhagen interpre-
tation” — designed to suppress a priori any question about the effective
location and structure of quantum leaps. Indeed, for a long time, these
issues were well beyond any permissible limits.
A possible answer is given by examining the mechanics of “leaps” in
Bohr’s original model. One problem arises immediately: given that transi-
tions are only permitted between allowed levels, how does an electron know
whether there is a free level to leap onto and thus take flight? Once the
electron has taken a leap, how does it manage to end its leap exactly on
the required level without trial and error?b
It should be emphasized that even current quantum formalism fails to
provide a comprehensive answer to this problem. In this formalism, the elec-
tronic orbital is a superposition of the initial and final orbitals, with time-
dependent coefficients obtained by solving the time-dependent Schrödinger
equation. Although the probability of the electron being located in the final
state increases with time, the electron is always — at any given moment —
either on the initial or the final orbital. Indeed, a measurement projecting
the electronic state onto these orbitals will always yield one of these two
outcomes.
Although measuring the electron’s position repeatedly on a ensemble
of identical preparations shows the gradual evolution of the probability
distribution of its location to that corresponding to the final orbital, the
transition of single atom occurs at a definite point in time, characterized
by the emission/absorption of a photon.
This implies the sudden transition (in the case of a single atom) of an
entire extended orbital and raises issue of non-locality. However, this non-
locality is confined to the atom and the moment of the leap and is not
related to the connection between different events. Consequently, it is not
the same non-locality as that of entangled states, for example. Nevertheless
the existence of this “hidden” non-locality is the modern version of the
ancient objection to the Bohr model.
This problem can be avoided by assuming that the quantum leap is not
a dynamic process. To clarify, suppose there is an a-spatial and a-temporal
physical reality, simply referred to as background. A quantum leap can then
b Rutherford’s objection to Bohr, referred for example, in Ref. [23].
www.ebook3000.com
be modeled as a double dissolution process in this background of the phys-

ical state of a micro-object, for example, an atomic state, followed by the
emergence of a new physical state. Since the background is a-temporal, this
dual process has no duration and from an observer’s (foreground) perspec-
tive, the quantum leap is instantaneous. The two physical states connected
by the leap are not causally related in the sense of temporal dynamics rep-
resented by differential equations, but rather in the sense of an eternal and
universal algebra of states. Thus, the problem of a non-local causal evolu-
tion (even though in the “hidden” and innocuous sense described earlier)
is bypassed.
The introduction of a synchronous correlation between background and
foreground restricted to quantum leaps is arguably a totally unnecessary
dialectic game, since the non-locality of quantum leaps is “internal” and
therefore quite innocuous. However, the approach adopted for atomic leaps
can also be applied to systems of entangled particles undergoing to mea-
surements by distant observers; non-locality becomes, in this case, evident.
It enables this type of situation to be comfortably addressed, allowing the
source of non-locality to be identified in the background.
5. Theories of Nothing
Thus, quantum leap structure theory requires the definition of a theoretical
framework for the physical observables emerging from the background, and
their reabsorption into it. Such a description must be non-dynamic and
therefore based on algebra and logic rather than differential equations.
It should be noted that this approach must include spatial and tem-
poral position as observables; thus, contrary to popular belief, the emer-
gence of the spatial-temporal order should be defined on an atomic/particle
scale, and not necessarily on the Planck scale. A second important observa-
tion is that this type of approach should constrain possible physical states
(and interactions) starting from non-dynamic general conditions and should
therefore be “archetypal”, in the philosophical sense. These archetypal con-
ditions should define, for example, the spectrum of elementary particles and
their interactions. Basic interactions are in fact expressed as quantum leaps
in the foreground, and particle states are specific connections between quan-
tum leaps.
Paradoxically, the outcome is a “theory of nothing” rather than a
“theory of everything”, because the inevitable starting point would be an
adequate definition of the background; and from a foreground perspective,
background is pure nothing.
550 L. Chiatti
There are several historical examples of branches of theories that tended

towards this type of approach (but remained incomplete) including, among
others: Von Weizsäcker’s qubit approach [24]; Bohm and Hiley’s [25, 26]
holoalgebra (and the holomovement); Finkelstein’s space–time code [27–30];
Rowlands’ Universal Rewrite System [31]; Pierre Noyes’ bit string physics
[32, 33]; and Stuckey and Silberstein’s “block world” QFT [34]. In a sense,
even Chew and Capra’s latest version of bootstrap [35, 36] belongs to this
group, although this approach focuses on archetypal constraints for S matrix
rather than on quantum leaps. Even Schroer’s “algebraic” QFT [37] has
many points of similarity with these research programs, although remaining
quite distinct. Finally, in Refs. [38, 39] quantum collapse is described by
means of mapping between Clifford algebras.
6. A Background Model
To better illustrate the concept of background, in this section we present a
“toy model” for the annihilation/creation of the wave function of a single
elementary particle. In quantum formalism, the quantum leap that brings
the state Ψ of a particle in the state Ψ is represented by the action of the
projector |Ψ Ψ | on the ket |Ψ, which leads to a new ket proportional
to |Ψ . The projector |Ψ Ψ | may in turn be considered as the entry
(represented by Ψ| ) of the component Ψ of Ψ in an a-temporal and a-
spatial background condition, followed by the exit (represented by |Ψ )
of the new state Ψ from this same condition. In this sense, we have the
annihilation (Ψ |) of the state Ψ followed by the creation (|Ψ ) of a new
state Ψ .
We assume that in the background condition the information associated
with the state Ψ is encoded in a kind of “internal wave function” inacces-
sible by direct observation. First, we postulate the existence of an “internal
time” variable of the background which will be denoted by τ . The internal
wave function associated with the particle will contain a factor Φ(τ ), real
and harmonic in τ , null at the boundary and outside the interval [−θ0 /2,
+θ0 /2]. The reality condition must be satisfied for the absence of a-temporal
direction, which implies Φ = Φ∗ . The following equation is a consequence
of this postulate:

−2 ∂2
Φ = (Msk c2 )2 Φ for τ ∈ [−θ0 /2, +θ0 /2]
[∂(2πτ )2 ] (1)

Φ=0 otherwise,
www.ebook3000.com
where Msk is the original (skeleton) component of the particle mass. From
Eq. (1), it follows that:

Msk c2 = n , (2)
θ0
where n = 0, 12 , 1, 3/2, . . . is an integer for odd solutions, a half-integer for
even solutions.
A second postulate states that each oscillating solution of Eq. (1) is
globally characterized by a second variable T (which is dimensionally a
temperature) such that the not normalized probability of a given value of
that variable is expressed by

exp[−(/θ0 )/kT ] for kT ≥ /θ0
(3)
0 otherwise,
where k is the Boltzmann constant and is T ≥ 0. Assuming τ = /kT ,
this probability becomes the square modulus of a factor:
Λ(τ ) = exp[−(/θ0 )/2kT ] = exp(−τ /2θ0 ). (4)
If we assume that while in the background condition the particle is in a
state of superposition of different values of τ , we have:

 ∂
−i Λ= Λ for 0 ≤ τ ≤ θ0

[∂(iτ )] 2θ0 (5)


Λ=0 otherwise.
The creation of the wave function associated with the particle state Ψ
exiting from the background can be considered as the passage from Eq. (3)
to a probability identically equal to 1 for any value t of the “external” time
accessible to the observer. This probability will be the square modulus of a
factor that we can write as:

iτ̃
Λ = exp − . (6)
2θ0
We can imagine this factor to derive from Eq. (4) by the Wick rotation
τ → iτ̃ ; simultaneously to this transformation, the factor Φ(τ ) must dis-
appear.
The third postulate states that the quantity τ̃ /2θ0 is merely the external
time t measured in units of the oscillation period of the old factor Φ, i.e.
θ0 /n . Therefore,
τ̃ t
= ± θ0 ⇒ τ̃ = ±2n t. (7)
2θ0 n
552 L. Chiatti
And thus:

−in t t
Λ → exp ± = exp ∓iMsk c2 , (8)
θ0
i.e. the de Broglie’s temporal phase factor. Annihilation is represented by

the inverse process.
It is immediately clear that (1) and (5) are two wave equations, respec-
tively, in the real component τ and the imaginary component iτ of a
complex internal time τ = τ + iτ , which thus becomes the “precursor” of
time t measured in the laboratory; this last physical quantity is thus emerg-
ing. The frequency of the “hidden” oscillation Φ becomes the frequency of
the de Broglie’s phase factor, experimentally accessible as the particle mass.
Clearly, the entire model dissolves in the limit θ0 → 0. The interval
θ0 must be a time scale characterizing the realm of elementary particles,
and this leads to the conclusion that it is cθ0 ≈ 10−13 cm, and therefore
/θ0 ≈ 50−100 MeV. Consideration should be given to the fact that this
time scale is not a minimum time interval, nor does cθ0 define a minimum
spatial interval.
A unitary time evolution process connecting the initial state
Ψin to the final (post-collapse) state Ψf in through the transition
amplitude Ψin |S|Ψf in really connects two quantum leaps (clicks)
|Ψin Ψin |e|Ψf in Ψf in |. We refer to Ref. [40] for details on this connec-
tion and the derivation of the Born Rule. We also note that the “collapse”
(|Ψ Ψ |)|Ψ = Ψ |Ψ|Ψ is generally an a-spatial process because Ψ and
Ψ are generally not position eigenstates. In the case where Ψ and Ψ are
represented on a same finite dimensional vector basis, it is also possible
to retrieve the Born Rule through a geometric collapse model, based on
a generalization of the Bloch sphere [41, 42]. This generalization is also
incompatible with spatial representation.
7. The Experimental Situation

If we accept the idea that the energy of the stationary state of a quan-
tum system is defined, and the conservation of energy is imposed, then
the energy of the single photon emitted/absorbed in a radiative transition
between two stationary states must also be defined. Therefore, the detec-
tion of the photon of energy E2 − E1 emitted by a system prepared in the
energy level E2 is an indirect proof of the occurred quantum leap of that
system on the energy level E1 . In this sense, the indirect proof of quantum
www.ebook3000.com
leaps in atoms, nuclei and molecules is historically rooted at the very ori-
gins of microphysics. It must be noted that photon emission/absorption is
instantaneous, unlike the unitary evolution of the wave function from the
initial state to the final state which is instead gradual and characterized by
the half-life of the initial state.
The direct observation of quantum leaps is rather more difficult experi-
mentally, and has only been achieved more recently. Indeed, it is necessary
to observe a single microsystem (for example, a single atom) and control its
quantum state in real-time. Historically, this has only been possible since
the mid-1980, thanks to the ion trap method.
Nowadays, the direct observation of quantum leaps has been widely
confirmed for trapped atoms and ions [43–45], single molecules [46], pho-
tons [47], single electrons in cyclotron [48], Josephson junctions [49],
nuclear [50] and electronic [51] spin, superconducting cavities [52] thus pro-
viding an impressive demonstration of the helpful Bohr’s intuition. The
initial hesitancy about the real existence of quantum leaps, in particular by
the community of quantum optics, is now only a distant memory of long
time ago [53].
In principle, the scheme of these experiments contemplates a system
with three 1, 2, 3 levels: level 1 is the ground state and is stable, the level 2
decays promptly to level 1 by spontaneous or stimulated emission; level 3 is
metastable and its half-life is much longer than that of level 2. The system
is pumped into resonance from level 1 to level 2 and then decays rapidly
to level 1 where the cycle begins again; thus the system continuously emits
photons which have a frequency equal to the resonance frequency between
levels 1 and 2. If the system is simultaneously pumped between levels 1 and
3, or its decay to level 3 is permitted in some other way, then occasionally
it will jump to that level. This will lead to the sudden interruption of the
fluorescence from level 2, and an observer will see then the system “switch
off”. Some time later, the system will decay from level 3 to level 1 and the
fluorescence will become visible again i.e. the system will “switch back on”.
The distribution of the time intervals between switching off and then on
again will have to correspond to the distribution for the level 3 decay, as
calculated starting from the half-life of this level. A particularly important
detail is the sampling of the fluorescence signal from the level 2 decay,
which is performed on time intervals which are negligible relative to the
level 3 half-life and the extension to the time delay between switching off
and back on. We therefore reaffirm that the leap is instantaneous respect
to the evolution of the wave function.
554 L. Chiatti
A variation of this scheme allows the experimental verification of the

quantum Zeno effect induced by null projective measures [54], which is a
further proof of the existence of quantum jumps.
8. Conclusions
The Bohr intransigent promotion and defence of the “Copenhagen inter-
pretation” probably resulted from the need to avoid a premature contact
of physics community with the a-temporal aspect, without adequate theo-
retical tools to assimilate it in the context of physical theory.
Without these tools, we would be dealing with a purely qualitative ref-
erence to a vague concept with “mystical” objective connotations, and this
could prevent a newly introduced theory from being accepted by the sci-
entific community. Conceptual problems were averted by relegating them
to the background, and the focus was on applications with productive out-
comes: the entire micro-world had to be discovered.
Small groups of researchers finally managed to bring unresolved issues to
the fore [55] only after the thrill of exploring new territories had passed. This
led to some important discoveries, including the non-locality of quantum
theory.
In this centenary of the Bohr model, which introduced the concept of
“quantum leap” explicitly for the first time, we may ask whether the time
has come for a more in-depth investigation of this mysterious process.
References
1. N. Bohr, On the constitution of atoms and molecules, Philos. Mag. Ser. 6(26),
1–25 (Part I), 476–502 (Part II), 857–875 (Part III) (1913).
2. R.G. Pearson, Semiclassical model for atoms, Proc. Natl. Acad. Sci. USA
78(7), 4002–4005 (1981).
3. G. Chacón-Acosta and H.H. Hernández, Effective description of the quantum
Kepler problem. Quantum Matter J. 2, 364–376 (2013).
4. F. Caruso and V. Oguri, Bohr’s atomic model revisited, (2008),
arXiv:0806.0652v1.
5. R.D. Harcourt, Bohr orbit theory revisited I. Ground state energies for the
helium isoelectronic sequence, J. Phys. B 16, 2647–2657 (1983).
6. R.D. Harcourt, Bohr orbit theory revisited II. Energies for 1S, 2P, 3D and
4F states of helium, Int. J. Quantum Chem. 31, 445–453 (1987).
7. A. Svidzinsky, M. Scully, and D. Herschbach, Bohr’s 1913 molecular
model revisited, Proc. National Acad. of Science, 102(34), 11985–11988
(2005).
www.ebook3000.com
8. A. Svidzinsky, M. Scully, and D. Herschbach, Simple and surprisingly accu-

rate approach to the chemical bond obtained from dimensionality scaling,
Phys. Rev. Lett. 95, 080401 (2005).
9. A. Svidzinsky, M. Scully, and D. Herschbach, Bohr’s molecular Model, a
Century Later, Phys. Today., 67(1), 33–39 (2014).
10. G. Chen, Z. Ding, S. Hsu, M. Kim, and J. Zhou, Mathematical analysis of a
Bohr atom model, J. Math. Phys. 47, 022107 (2006).
11. M. Bucher, Coulomb oscillation in the hydrogen atom and molecule ion
(2006), arXiv:physics/0605258 [physics.hist-ph].
12. M. Bucher, Coulomb oscillations as a remedy for the helium atom (2007),
arXiv:0705.4321 [physics.hist-ph].
13. M. Bucher, Rise and fall of the old quantum theory (2008), arXiv:0802.1366
[physics.hist-ph].
14. J. Stark, Elementarquantum der Energie, Modell der negativen und der pos-
itiven Elektrizität, Physikalische Zeitschrift 8, 881–884 (1907a).
15. J. Stark, Beziehung des Dopplers-Effektes bei Kanalstrahlen zur Planckschen
Strahlungstheorie, Physikalische Zeitschrift 8, 913–919 (1907b).
16. J. Stark, Neue Beobachtungen an Kanalstrahlen in Beziehung zur Lichtquan-
tenhypothese, Physikalische Zeitschrift 9, 767–773 (1908).
17. J. Stark, Prinzipien der Atomdynamik, Teil I: Die Elektrischen Quanten.
Hirzel, Leipzig, 1910.
18. A.E. Haas, Über eine neue theoretische Methode zur Bestimmung des elek-
trischen Elementar-quantums und des Halbemessers des Wasserstoffatoms,
Physikalische Zeitschrift 11, 537–538 (1910).
19. A.E. Haas, Der Zusammenhang der Planckschen elementaren Wirkungsquan-
tums mit den Grundgrössen der Elektronentheorie, Jahrbuch der Radioak-
tivität und Elektronik 7, 261–268 (1910).
20. A.E. Haas, Über Gleichgewichtlagen von Elektronengruppen in einer
äquivalenten Kugel von homogener positiver Elektrizität, Sitzungsberichte
der Kaiserliche Akademie der Wissenschaften (Wien) IIa, 1111–1171 (1912).
21. F. Hund, Geschichte der Quantentheorie, Bibliographisches Institut AG,
Zurich, 1975.
22. O. Darrigol, From c-Numbers to q-Numbers: The Classical Analogy in the
History of Quantum Theory, University of California Press, Oakland, 1992.
23. E. Segrè, Personaggi e Scoperte Nella Fisica Contemporanea EST Mondadori,
Milano (in Italian), 1976.
24. L. Castell, M. Drieschner, and C.F. von Weizsäcker Quantum Theory and the
Structures of Space and Time, Vols. 1–4. Hanser, Munich, 1975, 1977, 1979,
1981.
25. D. Bohm, Wholeness and the Implicate Order, Routledge & Kegan Paul,
London, 1980.
26. B.J. Hiley and M. Fernandes, Process and time Time, Temporality, Now,
H. Atmanspacher and E. Runhau (eds), Springer, Berlin, 1977.
27. D. Finkelstein, Space-time code, Phys. Rev. 184(5), 1261–1271 (1969).
28. D. Finkelstein, Space-time code 2, Phys. Rev. D 5(2), 320–328 (1972).
556 L. Chiatti

31. P. Rowlands, Zero to Infinity: The Foundations of Physics World Scientific,
Singapore, 2007.
32. P. Noyes, Why discrete physics? ANPA West 1(1), 3–8 (1988).
33. P. Noyes, Bit-string physics: A novel “Theory of Everything”, SLAC-PUB-
6509, 1994.
34. M. Silberstein, W.M. Stuckey, and T. McDevitt, Being, becoming and the
undivided universe: A dialogue between relational blockworld and the impli-
cate order concerning the unification of relativity and quantum theory,
Found. Phys., 43(4), 502–532 (2013).
35. F. Capra, Quark physics without quarks: A review of recent developments in
S-matrix theory, Am. J. Phys. 47, 11–23 (1979).
36. G.F. Chew, Bootstrapping quarks and gluons, Proc. XIVth Annual Rencontre
de Moriond, Savoie, France, 11–23 March, 1979, pp. 1–47.
37. B. Schroer, The holistic structure of causal quantum theory, its implemen-
tation in the Einstein–Jordan conundrum and its violation in some particle
theories, (2012), arXiv:1107.1374v4 [math-ph].
38. E. Conte, A proof of Von Neumann’s postulate in quantum mechanics, Quan-
tum Theory Reconsideration of Foundations 5, American Institute of Physics,
Maryland, 2010a, pp. 201–205.
39. E. Conte, A reformulation of von Neumann’s postulates on quantum mea-
surements by using two theorems in clifford algebra, Int. J. Theor. Phys., 49
587–614. (2010b), doi:10.1007/s10773-009-0239-z.
40. L. Chiatti and I. Licata, Relativity with respect to measurement: Collapse
and quantum events from fock to Cramer, Systems, 2, 576–589 (2014),
doi:10.3390/systems2040576.
41. D. Aerts and M. Sassoli de Bianchi, Many-measurements or many-worlds? A
dialogue, Found. of Sci. 20(4), 399–427 (2015).
42. D. Aerts, and M. Sassoli de Bianchi, The extended Poincaré–Bloch represen-
tation of quantum mechanics and the hidden-measurement solution to the
measurement problem, Ann. Phys. 351, 975–1025 (2014).
43. J.C. Bergquist, R.G. Hulet, W.M. Itano, and D.J. Wineland, Observation of
quantum jumps in a single atom, Phys. Rev. Lett. 57, 1699–1702 (1986).
44. W. Nagourney, J. Sandberg, and H. Demhelt, Shelved optical electron ampli-
fier: observation of quantum jumps, Phys. Rev. Lett. 56, 2797–2799 (1986).
45. T. Sauter, W. Neuhauser, R. Blatt, and P.E. Toschek, Observation of quan-
tum jumps, Phys. Rev. Lett. 57(14), 1696–1698 (1986).
46. T. Basche, S. Kummer, and C. Brauchle, Direct spectroscopic observa-
tion of quantum jumps of a single molecule, Nature 373, 132–134 (1995),
doi:10.1038/373132a0.
47. S. Gleyzes, S. Kuhr, C. Guerlin, J. Bernu, S. Deléglise, U.B. Hoff, M. Brune,
J.-M. Raimond, and S. Haroche, Quantum jumps of light recording the birth
and death of a photon in a cavity, Nature 446, 297–300 (2007).
48. S. Peil and G. Gabrielse, Observing the quantum limit of an electron
cyclotron: QND measurements of quantum jumps between fock states, Phys.
Rev. Lett. 83, 1287–1290 (1999).
www.ebook3000.com
49. Y. Yu, S.-L. Zhu, G. Sun X. Wen, N. Dong, J. Chen, P. Wu, and S. Han,
Quantum jumps between macroscopic quantum states of a superconducting
qubit coupled to a microscopic two-level system, Phys. Rev. Lett. 101, 157001
(2008).
50. P. Neumann, J. Beck, M. Steiner, F. Rempp, H. Fedder, P.R. Hemmer,
J. Wrachtrup, and F. Jelezko, Single-shot readout of a single nuclear spin,
Science 329, 542–544 (2010).
51. A.N. Vamivakas, C.-Y. Lu, C. Matthiesen, Y. Zhao, S. Fält, A. Badolato, and
M. Atatüre, Observation of spin-dependent quantum jumps via quantum dot
resonance fluorescence, Nature, 467, 297–300 (2010).
52. R. Vijai, D.H. Slichter, and I. Siddiqi, Observation of quantum jumps in a
superconducting artificial atom, Phys. Rev. Lett. 106, 110502 (2011).
53. W.M. Itano, J.C. Bergquist, and D.J. Wineland, (in press). Early observa-
tions of macroscopic quantum jumps in single atoms, Int. Journ. Mass Spectr.
377, 403–409 (2015).
54. W.M. Itano, D.J. Heinzen, J.J. Bollinger, and D.J. Wineland, Quantum zeno
effect, Phys. Rev. A 41(5), 2295–2300 (1990).
55. D. Kaiser, How the Hippies Saved Physics, W.W. Norton, 2011.
www.ebook3000.com
Chapter 21
In and Out of the Screen.

On Some New Considerations
About Localization
and Delocalization
in Archaic Theory
Ignazio Licata
ISEM, Institute for Scientific Methodology, 90146 Palermo
and
School of Advanced International Studies on Applied Theoretical
and Non Linear Methodologies of Physics, 70132 Bari
Ignazio.licata@ejtp.info
The Theory of Archaic Universe, based on the de Sitter Projective

Relativity, and the transactional interpretation of Quantum Mechan-
ics merge into a new Archaic Holographic Principle which provides a
unitary description of physical processes whose central idea consists of
the finite exchange of information in microinteractions.
1. Introduction
The development of Theoretical Physics can be regarded as a progres-
sive refining process of the notions of space–time-matter and vacuum.
That occurred through the interpretations and paradigms that each time
guided the development of theories and experimental investigation. Fur-
ther achievements thus require control and innovation of our conceptual
equipment as well as new paradigms. In this chapter, we will focus on the
foundational problems of quantum mechanics (QM) with a short reference
to Cosmology and their deep interconnection.
559
560 I. Licata
Lately, the idea of an informational description is making its way; it

has found its natural theoretical context in the different formulations of
the holographic principle [1]. An approach that has very little in common
with the like-Computer Universe idea [2]: in Physics, it makes no sense to
draw a distinction between “hardware” and “software”, the only existing
information is that materialized in actual interactions. A very important
question, strongly put forward by ’t Hooft, is the more or less fundamental
role given to Quantum information [3]. We will see that also in the theory
here outlined Quantum Physics “emerges” from a networks of R events (R
as Reduction, following Penrose terminology), but the constructive role of
locality and non-locality stays equal, only the actual observable processes
are defined in strictly local terms. We think that the most interesting aspect
of our theoretical proposal is not to consider “atoms” of space–time, but
processes which involve events in a scale-time going from a microphysical
horizon (connected at least to baryogenesis) to the cosmological horizon
implicitly fixed by Λ, the cosmological constant. Finally, we think that the
link between matter and Shannon–Turing Information does not pertain to
any tiling of the physical world, but to the “counting” of the R processes [4].
The Theory of Archaic Vacuum was born within a program of Quantum
Cosmology and has also developed a new approach to Particle Physics [5].
For our purpose, it suffices here to define Vacuum as a Universal Action
Reservoir [6] placed on a 4D surface of an Euclidean five-dimensional
(5D) hypersphere. This surface can be converted into a 4D hyperboloid
which represents a de Sitter space–time by a Wick rotation. The Bel-
trami projective representation of this de Sitter space–time on a 4D hyper-
plane tangent to the hyperboloid in the point-event of observation is
known as “Castelnuovo chronotope.” It is important to remark that the
group approach [7] makes possible to individuate the de Sitter Universe
as a framework of physical processes without referring to any “local”
Physics.
We note that it is not necessary to imagine the hypersphere as an
“enlarged” space–time. The space–time labels of observable events do not
belong to the archaic phase, represented on the hypersphere: it is populated
with virtual processes only. We called such theory “Archaic” because the
role of the hypersphere and that of the observed Universe are not related
by a “before” and an “after”. The hypersphere is rather like a highly non-
local, a-temporal phase. The space–time positions or “labels” are, instead,
related to the description of physical phenomena performed by classes of
observers on their private Castelnuovo chronotope [7]. The passage from a
www.ebook3000.com
In and Out of the Screen 561
description to the other one is defined by the Wick rotation: before jumping
out from vacuum, particles are in a virtual status described by the imag-
inary time of pre-space; during the observable existence lag, the real time
comes into play. The manifestation of particles from the vacuum and their
disappearance into the vacuum are the real microinteractions described in
quantum theory through the wave-function“collapse” or reduction (R pro-
cesses). To all effects, a“localization” at the level of individual event in
microphysics, a “nucleation” in the case of the Big Bang. Such strong unity
between macro and microphysics justifies the consideration of an archaic
holography ruled by a Wick rotation and projectivity. In Sec. 2, we will
explain the salient features of the cosmological scenario, in Sec. 3 will point
our attention on the microphysical aspects of the R processes, and we will
conclude in Sec. 4 with a critical reflection on the concept of holography.
Let us explore now the relation between time and temperature information.
2. Time and Information; The Meaning

of Wick Rotation
We can think of an axis x0 of the hypersphere representing inverse temper-
atures, and imagine that below a critical value of this coordinate the phys-
ical processes are constrained to remain virtual. This constraint is removed
when x0 exceeds the critical value, thus permitting the emergence of real
processes in real time, at a rate completely analogous to the exponential
one of the radioactive decay. This axis can be considered as an “archaic
precursor” of time. The 5-sphere is:
(x0 )2 + (x1 )2 + (x2 )2 + (x3 )2 + (x5 )2 = r2 . (1)
It is reasonable to assume the critical value as corresponding to baryogenesis

temperature:
TC = /kθ0 ≈ 1013 K, (2)
that is θ0 ∼ 10−23 s. We will come back later on this time interval (chronon)
and its relations with timescales.
One can believe, without too much effort, that even in the archaic phase
the state of matter could still be described by means of macroscopic vari-
ables. A set of values of these variables can be produced with many different
microstates, and the number of these microstates will define the probabil-
ity P of the macrostate in question. At this point, an entropy S and a
562 I. Licata
temperature T can be introduced, in purely formal terms, by means of the

definitions:
S = k ln P, (3)
where k is the customary Boltzmann constant and
dS −1
= , (4)
dF T
where F is the energy that the system would liberate if all the particles
and fields which it is made of become real. By combining the two relations,
one has:

−F
P = exp . (5)
kT
Combining (2) and (5), we have:

−F x0 −p0 x0 −Σ
P = exp = exp = exp , (6)
c
where p0 = F/c and Σ is the total action held by the Universe “before” the
Big Bang. It is interesting to note that the following relation exists between
the action and the entropy of the pre-Big Bang Universe:
Σ −S
= , (7)
k
as it can be seen by direct comparison with Eq. (3). In other words, Σ is

a negative entropy or, one might say, a sort of information whose bit is
ln(2). From Eq. (7), one has −Σ = ln(P ) and thus, for P = 12 (binary
choice), Σ = ln(2). In general, a dimensionless amount of information
I = Σ/[ ln(2)] can be introduced.
From the relation x0 ≤ cθ0 , which is valid in the “pre-Big Bang” era, if
one puts cθ0 = 2πR one has p0 x0 ≤ 2πp0 R, i.e. Σ ≤ 2πF R/c. Thus:
2πF R
I≤ (8)
[c ln(2)]
and this is a form of the Bekenstein relation which is valid for the “pre-Big
Bang” phase.
www.ebook3000.com
Let us consider now the “unfolding” process of information, and operate

a Wick rotation on the (1):
(x0 )2 − (x1 )2 − (x2 )2 − (x3 )2 + (x5 )2 = r2 . (9)
The space expansion is described by the canonical extension of Eq. (9):

(x0 )2 − R2 (τ )[(x1 )2 + (x2 )2 + (x3 )2 ] + (x5 )2 = r2 . (10)
An important difference with respect to Feynman cosmology is that

while it admits a multiplicity of possible models, to be subsequently selected
based on observation, the approach described here leads to a single cosmo-
logical model. It corresponds to the Friedmann model having null spatial
curvature (k = 0) and a positive cosmological term λ = 4/3t20 . The reduc-
tion of arbitrariness is a first mark of the power of this approach based
on group theory. Fixing the cosmological constant, we also fix a new nat-
ural constant t0 which has the dimensions of time; this time is related
to the de Sitter radius r through the relation r = ct 0 , with c the speed
of light in a vacuum. At the start of the expansion [R(τ ) = 0], Eq. (10)
becomes:
(x0 )2 + (x5 )2 = r2 . (11)
If it is therefore assumed that the start of the expansion coincides with

the origin of x0 , i.e. that the Big Bang occurs on the equator x0 = 0 of the
hypersphere (1), the value ±r is obtained for the variable x5 . In geometrical
terms, this corresponds to a point-like Big Bang associated with a point on
the equator of the 5-sphere. However, the x5 -axis can be rotated on this
equator giving rise to ∞3 different (and equivalent) intersections. One thus
has ∞3 different (and equivalent) big bangs or, to be more precise, ∞3
different (and equivalent) views of the same Big Bang, which are pertinent
to distinct fundamental (inertial) observers. In individual observer’s coordi-
nates, the metric is consistent with Eq. (10) and therefore all the observers
see a Universe in expansion. At a certain value of cosmic time τ , all the
observers see the Universe under the same conditions and the cosmolog-
ical principle thus applies, provided that the conditions of matter on the
equator x0 = 0 are homogeneous.
The dimensionless vacuum starting from which the Big Bang develops is
therefore substituted, in this approach, by a pre-existing space: the equator
of the 5-sphere (1). The passage from condition (1) to condition (10) takes
place at a critical value θ0 of the variable x0 /c for which processes of quan-
tum localization of elementary particles on space–time become possible.
564 I. Licata
The emergence of all the elementary particles on the space–time domain

is the true essence of the Big Bang. Starting from this nucleation by 5-
sphere vacuum, the propagation of particles is described by wave-functions
in which coordinates satisfying condition (10) and no longer condition (1)
appear as an argument. The “archaic phase” governed by condition (1)
comes to an end and the actual history of the Universe governed by condi-
tion (10) begins. And time “flows”.
It must be noted that the contraction resulting from the scale dis-
tance operates on the private space–times of the individual fundamental
observers, not on the public space–time, which remains unchanged. As one
approaches the big bang proceeding backwards in cosmic time, the pri-
vate contemporaneousness space of each observer contracts in one point;
but the uncontracted public space will be identical for all observers. Apart
from fluctuations, the final mass-energy density will be the same every-
where and will be equal to the ratio between F (the energy released in the
transition) and the volume of the section x0 = cθ0 , which is finite.a Thus,
there is never a singular density value; in other words, in public space–time
the Big Bang is not truly a singularity. Therefore, the origin of Universe
(and time) is a nucleation process implying a passage from information to
energy. Given the initial homogeneity, all the fundamental observers will
see the same physical cosmic conditions, despite the absence of causal cor-
relations between their respective positions.
Two difficulties with the standard model are bypassed in this way, i.e.
the justification of the initial homogeneity (which is here the natural aspect
of a pre-vacuum) and the appearance of a singularity. Space–time isotropy
and homogeneity are the consequences of the decay of an isotropic and
homogeneous archaic (pre-)vacuum; this line of reasoning agrees with some
prominent features of other contemporary approaches [8].
Finally, as for cosmology, it is interesting to notice that the wave-
function in archaic space (1)
√
2mE E
Ψ = Ψ0 exp ±j x−
kT
after the “Big Bang” becomes:

√
2mE Et
Ψ = Ψ0 exp ±j x−j .

a This volume is practically that of the equator of the 5-sphere, i.e. 2π 2 r 3 .
www.ebook3000.com
This solution is very similar to that of Hartle–Hawking Universe without

boundaries, but also to the one proposed by Bohm et al. for the processes
of “spontaneous localization” [10]. If we assume as valid the Born rela-
tion in the archaic phase, we can easily see the meaning of the archaic
wave-function. As a matter of fact, if we put:

2 2E
P = dV Ψ = exp − ,
kT
we will see that the probability of existence of an energy particle E is not

conserved at varying of temperature because the production of a particle
implies a clear exchange of energy with the thermal bath: archaic QM is a
form of thermostatics.
3. The Quantum Weirdness: Fock, Feynman,

and that’s all
In developing the Theory of Archaic Vacuum, we have used a so-called
“minimal ontology”, i.e. the only things having a physical meaning are the
events R of creation and destruction of a physical quality (position, energy,
impulse, spin, etc.). As can be seen, that is the language of the interaction
vertices. We are convinced that the foundational problems of QM become
clearer if they are framed as an effective pre-QFT and not as a still back
in 1927 conceptual isle which receives non-locality like an unexpected host.
The methodological suggestion comes from Fock, and the QM version we
adopt, which meets the requirements of a pre-QFT, is the powerful and
suggestive one of the Feynman Paths [11]. These ones — and their formal
correlates, the Bohm trajectories [12] — suggest us to take very seriously
the Feynman Variations as the QM “phenomenic heart”:
A physical system is not observed in any physically possible state.
An emergent interpretation must be able to connect the observable R events

to the Feynman’s “possible states”. It is easy to see that this is possible if
we consider the Archaic Vacuum as a beable [13]. Moreover, it is necessary
to remove the semiclassical concept of “particle” as a “permanent object”.
Our starting principles are:
(a) The physical world is described as a discrete net of interaction vertices

where some properties (space–time position, quadri-impulse, spin, etc.)
are destroyed and created. The measurement of such properties is all
566 I. Licata
that we know of the physical world from an operational view point. Any
other construction in Physics — like the continuous space–time notion
itself or the evolution operators — has the role to connect causally the
measured properties. So we can say they are “emergent” with respect
to the network of events;
(b) Heisenberg uncertainty principle in its more general form, phase for
number of quanta, ∆n ∆ϕ ≥ 2π — does not indicate the limits of
measurement between classical variables, but the applicability limit of
the continuous space–time concept itself. In phenomena involving a few
number of potential “impacts” (interaction vertices) the representation
of the field as a continuous propagation in a space–time environment
is no more applicable;
(c) Motion is no more a continuous phenomenon, but a discontinuous pro-
cess in the space–time coordinates. There are no more “objects” as
exclusive bearers of permanent “qualities”. The propagation of physical
quantities in the space–time appears more similar to the phenomenon
of switching on a line of blinking bulbs, or dislocation in a crystal.
A quantum leap where the quantum state “i” is destroyed and the new
quantum state “j” is created involves the element X(i, j) of the physical
quantity. Only when the matrix X(i, j) is diagonal, the interaction will
leave the X value unchanged and that leap can then be considered as a
measurement of X. So the quantum laws have generally a matrix-like form
and the existence of not simultaneously defined quantities arises in a natural
way. This frame modifies quite widely the semiclassical vision associated to
QM. Actually, if we consider interaction vertices as finite and fixed by the
introduction of a chronon θ0 , we can imagine that the “dogma motion” of
an electron, for instance, dissolves in a complex network of events between
delocalization (in the Archaic Vacuum) and localization (in space–time)!
This recalls a Bohm reflection during his duel with M. Pryce broadcasted
by BBC in 1952:
We wondered what actually an electron does. What would it do while it
is passing from the source to the slit? That’s the point. Well, I could propose,
for example, that the electron is not a particle in the sense it is currently
meant, but an event. I assume such event happens in a generic medium —
a “field ” — we can suppose in this field there’s an impulse, a wave moving
forward and converges in a point so producing a very strong impulse and
then diverges and scatters away. Let’s imagine these impulses in a series all
reaching a line there producing a series of intense pulse. The impulses will
www.ebook3000.com
be so close one to the other that they will look like a particle. In most cases,
all that will behave just like a particle and it will behave differently when
goes through the two slits, because each impulse will come out according to
the way the incident wave passes the slits. The result is that we are looking
at something that’s neither a wave nor a particle. If you wonder how the
electron has actually passed the slit and if it has really passed one slit or the
other, I would reply that probably is not that kind of thing which can pass
a slit or the other one. Actually, it is something which forms and dissolves
continuously and that can be the way it really acts [14].
There starts emerging a picture able to conciliate the finiteness of R
events with the wideness “at the World bottom” required by a vision a la
Bohm–Feynman. Speaking in a more general way, we shall have at t = t1
the event of the creation–destruction of a quality Q(|Q Q|) and at t = t2
the event of the creation–destruction of a quality R(|R R|). These two
processes will be linked by a time evolution operator S according to the
transactional ring:
|Q Q| t = t1 ,
S ↓ ↑ S+
|R R| t = t2 .
In other words, |Q is transported from S into |Q and projected into
R|, |R is transported by S + into |R and projected onto Q|. The ampli-
tudes product:
R|S|Q Q|S + |R = | R|S|Q|2
is immediately obtained, which is the probability of the entire process.

If quality Q is constituted by a complete set of constants of motion then
R = Q and this is the type of process which can describe the propagation of
a photon-type quantum, otherwise it is the generic process of the creation
of a quality Q causally linked (by means of S) to the destruction of a quality
R. Moving to the representation of the coordinates, by substituting bras
and kets with wave-functions, we once again obtain as a particular case
the result already seen with the well-known Schrödinger non-relativistic
expressions.
From an algebraic point of view, the transactional ring is a sort of iden-
tity operator, because SS + = S + S = 1 and the qualities Q, R are simulta-
neously created and destroyed. This is exactly the case of EPR and GHZ
phenomena. It must be said that the S and S + transaction amplitudes (for-
ward and backward) are an event symmetric loop, but once we choose a
568 I. Licata
convention on the energy variation ∆E with the Archaic Vacuum, the time
arrow for the R processes is fixed. The emergence of a time order of micro-
scopic events from a totally symmetric scenario is one of the interesting
aspects of the theory; macroscopic systems in turn emerge as quasi-stable
aspects of extended, enormous networks of transactions. We can say that
we do not observe the backward side because it is stored in the Archaic
Vacuum. That is a further sign of the emergent trait of Quantum Physics.
All the traditional equipments of QM (Born Rule, Quantum Potential, etc.)
can be obtained by the formalism here outlined. We have to underline that
there is no more randomness than any other emergent approach can admit;
in other words, any “randomness” comes from the impossibility to follow in
detail the connections of “micro-event”, which is — in our case — the trans-
actional loop. When the two extreme Rs are fixed, the process is fixed and
totally causal! If one of the extremes is unfixed, we find the full meaning of
the statement ascribed to Bragg: everything in the future is a wave, every-
thing in the past is a particle. A statement we could paraphrase in terms of
localization/delocalization. The Born Rule is a statistics of transactions to
all intents and purposes.
The probability of a transaction connecting the two extremes given by
the two distinct R processes, a fixed one and a variable one on a set of sup-
port, is expressed by the number of “loops” which connect these extremes.
Thus, it is a Kolmogorov probability. Starting from the loops, it is also
possible to derive the concept of transaction amplitude (forward and back-
ward) associated to the transaction. This amplitude varies in continuous
way according to the instant when the R process takes place, and such
instant is a real continuous variable, too. Anyway, we have to observe that
different instants correspond to different and distinct transactions, each one
corresponding to the genesis of a different complex of loops; and it is the
single transaction to be physically realized as a “total experiment” in Fock
sense, or not. So, there does not exist a physical process of evolution from a
transaction to the other one. The time evolution of the transaction ampli-
tude at varying the instant of one of the extremes (which is to say, in the
end, the time evolution of the state vector) is a coarse grain description typ-
ical of the emergent nature of QM on an enormous number of transactions.
Thus, even considering the single realized transaction, we have this sit-
uation: decomposing the state vector entering/exiting one of the extremes
of this transaction on the basis of a suitable Hilbert space and defining a
measure of information on the coefficients of the superposition, no physical
process contained in space–time elaborates such information. Information
www.ebook3000.com
which, furthermore, could be infinite if the coefficients were continuous vari-

ables. An R process at the extreme of a transaction is physically realized if
the quality under observation realizes an effective physical process of local-
ization in the time domain. It is reasonable to ask whether we can define
a measure of the positional information gained by this localization. If we
can, it should clearly be a normal, bit measurable Shannon information,
because the time localization is not described by any “quantum ampli-
tude”. These are the premises to generalize the holographic features of the
Archaic Theory.
4. Holographies in Comparison: Chronon

and Planck Scale
The Archaic Universe Theory has got a natural holographic structure, with
the five-dimensional pre-space as screen and its projections on the tangent
plane as bulk, i.e. the manifestation of the physical world observed. Two
event horizons are defined there — microphysical and cosmological; they
will be the basis for the comparison to the holographic conjecture derived
from the black holes’ thermodynamics.
If the version of the Projective General Relativity (PGR) proposed
in Ref. [5] is correct, then exists a de Sitter cosmological horizon, whose
chronological distanceb from any “here, now” is a new fundamental con-
stant of Nature equal to t0 . Such constant is invariant in cosmic time and
we have ct 0 ≈ 1028 cm, where c is the velocity limit.
Let us admit, in agreement with a recent proposal [15], that the localiza-
tion of an R process is associated to the genesis of a de Sitter microhorizon
with center O and radius cθ0 ≈ 10−13 cm, with O generally delocalized in
accordance with the wave-function coming in/out of the process. The con-
stant θ0 is independent of the cosmic time, so also the relation t0 /θ0 ≈ 1041
is independent of the cosmic time. This relation expresses the number of
totally distinct, temporal localizations the R process can access within of
the de Sitter Cosmological horizon.
Basically, the length of the temporal line on which an observer located
here and now places the R process is t0 , whereas the duration of the R
process is in the order of magnitude of θ0 ; the segment is so divided in
t0 /θ0 ≈ 1041 different “cells”. Each cell can be in two states: “on” or “off”.
The temporal localization of a single R process corresponds to the situation
b We mean this distance in Beltrami sense, not in the sense of the observer’s local clock.
570 I. Licata
when all the cells are off except one. Configurations where many cells are
on will correspond to a localization of many different R processes on the
same temporal line. If we accept the idea that each cell is independent, we
41
will have 210 different configurations in total. The positional information
associated to the localization of 0, 1, 2, . . . , 1041 R processes thus amounts
to 1041 bits, the binary logarithm of the number of configurations. This is
a sort of codified information on the time axis contained in the observer’s
de Sitter horizon.
The R processes are actually real interactions between real particles.
During the interactions, an amount of action in the order of the Planck
quantum h is exchanged. So, in terms of phase space, the manifestation of
one of these processes is equal to the switching on of an elementary cell
whose volume is h3 . The number of switched on cells in the phase space
of a given macroscopic physical system is the estimator of the volume it
fills in this space, and thus of its entropy. We can so hypothesize that the
localization information of R processes is linked to entropy through the
Uncertainty Principle. This possibility presupposes the “objective” nature
of R processes.
Thus, it is natural to ask if a sort of Bekenstein limit on entropy could
be applied, in a way or other, to the above-mentioned two horizons. If we
assume that the information on the temporal localization of R processes,
I = 1041 bits, can be connected to the area of the microhorizon A = (cθ0 )2
≈ 10−26 cm2 through the holographic relation:
A
= I, (12)
4 l2
then the l spatial extension of the “cells” associated to a bit of informa-
tion will be ≈ 10−33 cm, the Planck scale! It has to be underlined that in
this way the Planck scale shows to be a consequence of the holographic
conjecture (12), combined with the two horizons, and so of the finiteness
of the I information as well. In no way it represents a limit for the conti-
nuity of space–time, nor for the spatial or temporal distance between two
events (that is a continuous variable). Moreover, being I = t0 /θ0 and t0
correlated to the λ by the relation λ = 4/3t20 , the (12) is basically a defini-
tion of the Planck scale as a function of the cosmological constant. There
appears a global–local relation that is exactly what we expect by a theory
on “information-endowed” Vacuum.
Let us note that the number of the distinct spatial localizations of a R
process within the cosmological de Sitter horizon is ≈ (ct0 /cθ0 )3 ≈ 10123
or, within few orders of magnitude, the Bekenstein limit. The inner space
www.ebook3000.com
of the de Sitter visible horizon is divided in cells with volume (cθ0 )3 , each
cell can be on or off. The spatial localization of a R process is equivalent
to all the cells off except one. Configurations with many cells correspond
to the simultaneous manifestation of many R processes. Considering that
123
we have 210 distinct configurations in total, the positional information
associated to localization amounts to 10123 bits, the binary logarithm of the
number of configurations. Patiently, we can repeat what we have already
said for the temporal position and its connection to entropy also for the
spatial position.
If we put, in (12), I ≈ 10123 and we leave l ≈ 10−33 cm, we will have
A ≈ (1028 cm)2 , i.e. the cosmological horizon area. In other words, the
cosmological horizon satisfies the Beckenstein limit.
We might ask what would happen if in (12) we put the cosmological
horizon area at nominator, and l = cθ0 — as the dimension of the cell asso-
ciated with a single bit — at denominator instead of the Planck length. The
result is 1082 , namely — within some order of magnitude — the number of
particles in Dirac cosmology. This result can be interpreted by saying that
among all the 10123 available cells at a precise instant, only 1082 can be
filled. The percentage of filling — let call it so — of the de Sitter cosmo-
logical horizon is thus 1082 /10123 ≈ 10−41 .
Now, the lighter stable particle with a definite, not oscillating mass is
the electron. In addition, cθ0 approximatively coincides with the classical
radius of electron. So, by admitting that the 1082 cells are all filled with
electrons with mass 10−27 g and density equal to:
10−27 g 3
3 ≈ 1012 g/cm , (13)
(cθ0 )
we have the following estimate of the maximum electronic density in the
Universe, within some orders of magnitude:
3 3
10−41 1012 g/cm = 10−29 g/cm . (14)
This estimate, very rough, is of the same order of the critical density
(≈0.95 × 10−29 g cm−3 )! Even though this value holds for today’s Uni-
verse, we remark that according to PGR approach the current Universe
radius is of the same order than r, so that the agreement is good.
We can reasonably suppose that the holographic conjecture is a property
of both temporal horizons: the cosmological one (t0 ) and the “particle” one
(θ0 ); it cannot be generalized for other physical systems (except the black
holes which get their own event horizon). The information associated to
572 I. Licata
the localization of a physical event (R process) along time is finite, and

the same thing goes for its spatial localization. This happens because the
extension of a transaction in the Beltrami time of an observer is bounded
by the “particle” horizon below and by the cosmological one above. This
suggests an interesting interpretation of the cosmological constant as the
index of the information unfolding. We cannot but ask, like Hawking did,
if the Universe wave-function could “rewind” [16].
Actually, the holographic relation is already implicit in Dirac’s cosmol-
ogy, as it is shown by the (14) which indicates a connection to the cosmic
density of matter. The Planck scale emerges when we take into considera-
tion an elementary cell, associated to the codification of the temporal line,
on the particle horizon. The same scale appears when we take into con-
sideration an elementary cell, associated to the space codification, on the
cosmological horizon. Whatever the process underlying such codifications
may be, it is clear that it (if any exists) is trans-temporal and non-local.
5. Tentative Conclusions and Open Problems

In his masterpiece The Physical Principles of the Quantum Theory [17],
Werner Heisenberg clearly identifies two possible lines of development of
the new theory, by using as a milestone the Uncertainty Principle. On
the one hand, to use spatiotemporal descriptions and then “to save” the
classical concepts through a statistical description; or to use a mathematical
pre-spatial scheme changing the classical notion of causality dynamics. In
the 1930s, there were no compelling reasons for the second way. Today,
holographic approaches provide a strong suggestion in this direction. The
aim is to overcome the distinction between “dynamic laws” and “boundary
conditions” that runs through all the Physics. On a fundamental level, this
dichotomy between “Heraclitian” and “Parmenidean” aspects is no longer
sustainable.
The idea of the Archaic Vacuum connects global and local describing the
localization as an emerging process. The old wave-function can be placed
on the“FAPP tools” shelf, and the quantum amplitudes find their meaning
in transactional logic, which plays here the role of an effective beable. The
price to pay is an unavoidable non-locality, whose roots are not in space
and time. So, the Einstein–Bohr debate would end in a draw, once we intro-
duce the new construction of the informational vacuum as pre-QFT. This
should not surprise us that much. The vacuum is not only the state of min-
imum energy, but something more. For example, it is written there why
www.ebook3000.com
we always observe one electron, and not half. In addition, it has not been
sufficiently taken into consideration that the existence of invariant“laws”
is, in some way, a sign of non-locality. This could be a new, genuine sense
for the QM “instant interpretation” quoted by Feynman: I received a tele-
phone call one day at the graduate college at Princeton from Professor
Wheeler, in which he said, “Feynman, I know why all electrons have the
same charge and the same mass” “Why? ” “Because, they are all the same
electron!” [18].
We still know too little about a global holographic picture of the phys-
ical world. If space–time is no longer a primary entity what can we replace
it with? An option, suggested by Chew and Hiley, is to restart from the
energy–momentum phase space [19], and a precious hint, shared by many
holographic approaches, is the close relationship between Planck’s constant
and the Boltzmann one [20]. Particularly, in the Archaic Theory, a local-
ization is defined with a maximum uncertainty cθ0 , within a horizon with
radius ct0 . Thus, it is clear that N = t0 /θ0 is an a-dimensional constant
of Nature with a precise informational meaning. In fact, the plane wave-
front codifies N 2 possible equivalent “corpuscles” and only one of them will
manifest actually. N (the Dirac number) shows to be very big (≈1040 ), but
finite. It suggests, moreover, the possibility to connect the statistics to a
more general principle of “counting” (see for example, Ref. [21]).
The thorniest problem is surely the one of emergence of space. Among
all the physical “labels”, position is absolutely that which should emerge
from a deeper level. A sort of basic level in relational dynamics. In the PGR,
using classical terms, we say that the group approach let us identify the de
Sitter geometry of a pre-space. In the quantum domain, we add that it is
populated with only virtual processes. What is the physical meaning of all
that? The entire setup seems to suggest that the space is materialized by
transactions. To understand this reasoning we consider the process (trans-
action) consisting in emission/absorption of a quantum of a given field with
energy E.
In the absence of an R process, the virtual process |E) (E| (B.
d’Espagnat notation) is a closed loop in the Archaic Vacuum. From the
point of view of the vacuum, therefore, this cycle is an elementary (pre-
)event. We can hypothesize that an occurrence probability is associated
with this elementary event:

−2E
P = exp ,
kT
574 I. Licata
with T > 0. We can subdivide the cycle into the two processes forward (f ),
backward (b), which are not true events (they cannot occur by themselves)
and factorize P according to the symmetrical expression:
P = Pf × Pb ,

−E
Pf = Pb = exp .
kT
Obviously, the factors Pf , Pb are not probabilities because, as we have
said, the processes f , b are not events in space–time. When the transaction
occurs, the half-processes f, b become true distinct processes on space–time,
instead. Process f consists of event |E) (E| which occurs at t = t0 and of
the forward connection with the event |E) (E| which occurs at t = t1 > t0 .
Process b consists of the event |E) (E| which occurs at t = t1 and of the
backward connection with event |E) (E| which occurs at t = t0 . A true
transaction is manifested, having two distinct R processes as its ends. The
space–time unfolding corresponds to a Wick rotation. For the first process,
the Wick rotation takes the form:
1 −i(t1 − t0 )
→ ,
kT
while for the second process, it takes the form:
1 −i(t0 − t1 )
→ .
kT
In these equations, the variable t indicates the proper time of the quan-
tum of energy E exchanged between the two events. Consequently, the
factors Pf , Pb become:

iE(t1 − t0 )
Pf → Πf = exp ,

−iE (t1 − t0 )
Pb → Πb = exp .

Thus, the product of these factors is the transformation of P according
to the Wick rotation; this product is equal to 1. It is still a probability, but
its meaning now is completely different. This is the probability that, given
the emission (absorption) of a quantum of energy E, the energy of the suc-
cessively absorbed (previously emitted) quantum is E. Since the propaga-
tion of the quantum is free (we are considering a single transaction without
intermediate vertices) this probability is certainly 1. Alternatively, one can
say that 1 is the number of quanta of energy E propagated between the
www.ebook3000.com
emission and absorption events. So, the Wick rotation connects the prob-
ability of a virtual process to the probability of the same process once it
has become real (see Chiatti, The Transaction as a Quantum Concept [11]).
This line of reasoning is similar to the one of the graphs KN , whose expli-
cation is expressed as a function of a parameter which is a time precursor,
proposed by Fotini Markopoulou Quantum Graphity [22]. It is possible to
implement both lines of research with quantum superpositions of binary
alternatives, and to reconnect to space through the (local) isomorphism
SU(2)→SO(3), as showed by Weizsäcker and Görnitz [23]. Thus, each tiling
is not in the space, but in information.
In the Archaic Holography, the localization and delocalization processes
appear as complementary features of the World, and the Indeterminacy
Principle is the door between two levels of description. To cross such door
is less dramatic than it was believed in the past, indeed, and the scenario so
revealed is even more comprehensible than the classical one. The ratio —
cosmological constant, chronon, Planck scale — are strongly interconnected
one to the other and globally indicate the finiteness of information in the
physical Universe as well as the emergent nature of QM. At this point,
non-locality should not surprise us that much: by means of the localiza-
tion/delocalization process, each node of the transactional network “sees”
all the other ones with a finite exchange of holographic information. In
other words, it simply indicates that the manufacturing of the World is
just one.
This chapter is the fruit of a long collaboration with Leonardo Chiatti
and the synthesis of our burning confrontations about Chronon and Planck
scale since 2004.
It is dedicated to my cats. Each one in its own eigenvalue.
References
1. P.C.W. Davies, The implications of a cosmological information bound for
complexity, quantum information and the nature of physical law, in Ran-
domness & Complexity, from Leibniz to Chaitin, Calude, C. (ed.), World
Scientific, Singapore, 69–87, 2007. See also: A. Dobado, An Elementary Intro-
duction to the Holographic Principle, (2005), arXiv:hep-ph/0506027.
2. K. Zuse, The computing universe, Int. J. Theor. Phys. 21(6), 589–600 (1982);
E. Fredkin, An introduction to digital philosophy, Int. J. Theor. Phys.
42(2), 189–247; I. Licata, The big computer complexity and computabil-
ity in Physical Universe, in Determinism, Holism, and Complexity, Kluwer
Academic/Plenum Publishers, Berlin, 117–123, 2003.
576 I. Licata
3. G. ‘t Hooft, The fate of the quantum, (2013), arXiv:1308.1007 [quant-ph];

Dimensional reduction in quantum gravity, SalamFest, (1993), arXiv:gr-qc/
9310026.
4. I. Licata, Emergence and computation at the edge of classical and quantum
systems, in Physics of Emergence and Organization, I. Licata and A. Sakaji
(eds.), World Scientific, Singapore, 1–25, 2008.
5. I. Licata, Universe without singularities. A group approach to de Sitter cos-
mology, EJTP 3(10), 211–224 (2006); I. Licata and L. Chiatti, The archaic
universe: Big Bang, cosmological term and the quantum origin of time in
projective cosmology, Int. J. Theor. Phys. 48(4), 1003–1018, 2009; I. Licata
and L. Chiatti, Archaic universe and cosmological model: “Big-Bang” as
nucleation by vacuum, Int. J. Theor. Phys. 49(10), 2379–2402 (2010); L.
Chiatti, A possible mechanism for the origin of inertia in de Sitter-Fantappie-
Arcidiacono projective relativity, EJTP 9(26), 11–26 (2012).
6. A. Garrett Lisi, Quantum Mechanics from a Universal Action Reservoir,
arXiv:physics/0605068, 2006.
7. G. Arcidiacono, A new projective relativity based on the de Sitter universe,
Gen. Rel. Grav. 7, 885–889 (1976); G. Arcidiacono, The de Sitter universe
and the mechanics, Gen. Rel. Grav. 8, 865–870 (1977).
8. Pourhasan, R.N. Afshordi, and R.B. Mann, Out of the White Hole; A Holo-
graphic Origin for the Big Bang, JCAP04:005 (2014).
9. J.B. Hartle and S.W. Hawking, Wave function of the Universe, Phys. Rev. D
28, 2960–2975 (1983).
10. A. Baracca, D. Bohm, B.J. Hiley, and A.E.G. Stuart, On some new notions
concerning locality and nonlocality in the quantum theory, Nuov. Cim.
28B(2), 453–465 (1975).
11. L. Chiatti, Path integral and transactional interpretation, Found. Phys.
25(3), 481–490 (1995); L. Chiatti, The transaction as a quantum concept, in
Space-Time Geometry and Quantum Events, I. Licata (ed.), Nova Publ., NY,
11–43, 2014; I. Licata, Transaction and non locality in quantum field theory,
EPJ Web Conf. 70, 00039 (2014); L. Chiatti and I. Licata, Relativity with
respect to measurement: collapse and quantum events from Fock to Cramer,
Systems 2, 576–589 (2014).
12. I. Licata and D. Fiscaletti, Bohm trajectories and Feynman paths in light of
quantum entropy, Acta Phys. Pol. B 45(4), 885–904 (2014).
13. J.S. Bell, Speakable and Unspeakable in Quantum Mechanics. Collected
Papers on Quantum Philosophy, 2nd ed, intr. Alain Aspect, Cambridge Univ.
14. D.-M. Bohm, Pryce debate, in Quanta and Reality. A Symposium, S. Toul-
min, (ed.), Hutchinson & Co., Camberley, 1962.
15. I. Licata and L. Chiatti, Timeless approach to quantum jumps, Quanta 4(1),
10–26 (2015).
16. S.W. Hawking, Arrow of time in cosmology, Phys. Rev. D 32, 2489 (1985).
17. W. Heisenberg, The Physical Principles of the Quantum Theory, Dover Pub-
lications, New York, 1949.
18. R. Feynman, Nobel Lecture, Elsevier Publishing Company, Amsterdam, 1965.
www.ebook3000.com
19. B.J. Hiley, Process, Distinction, Groupoids and Clifford Algebras: An Alter-
native View of the Quantum Formalism, in New Structures for Physics,
B. Coecke, (ed.), Springer, New York, 2011.
20. D. Acosta, P. Fernandez de Cordoba, J.M. Isidro, and J.L.G. Santander,
A holographic map of action onto entropy, Jour. Phys. Conf. Ser. 361, 012027
(2012).
21. O.W. Greenberg, Example of infinite statistics, Phys. Rev. Lett. 64, 705
(1990); P. Davies, A New Link Between Boson and Fermion Algebras: An
Alternative to Supersymmetry, arXiv:quant-ph/0610051, 2006.
22. F. Markopoulou, Space Does Not Exist, So Time Can, arXiv:0909.1861 [gr-
qc], 2009; The computing spacetime, in How the World Computes Turing
Centenary Conference Proceedings, S. Barry Cooper, A. Dawar and B. Löwe
(eds.), 472–484, 2012; G. Chew, Gentle quantum events as the source of
explicate order, in Quantum Implications, B.J. Hiley and F.D. Peat (eds.),
Routledge & Kegan Paul, London, 249–254, 1987; G. Chew and H.P. Stapp,
Three-space from quantum mechanics, Found. Phys. 18(8), 809–831 (1988).
23. C. Weizsacker, The Unity of Nature, Farrar, Straus & Giroux, New York,
1981; T. Görnitz, Abstract quantum theory and spacetime structure. I. Ur
theory and Bekenstein–Hawking entropy, Int. J. Theor. Phys. 27(5), 527–542
(1988).
www.ebook3000.com
Chapter 22
Schrödinger–Milne
Big Bang — Creating
a “Universe of Threeness”
Geoffrey F. Chew
Theoretical Physics Group
Physics Division
Lawrence Berkeley National Laboratory
Berkeley, CA 94720, USA
A Schrödinger-evolving forward-lightcone-interior “Milne” universe

(“SMU”) is governed by “centered-Lorentz” (CL) symmetry — that of
a 9-parameter Lie group with a 6-parameter SL(2,c) “exterior” and a
3-parameter “quality-space” center. “Reality” resides in current densi-
ties of electric charge and energy–momentum — the Dalembertian of an
SMU-ray-specified classical retarded Lorentz-tensor field with 22 electro-
magnetic and 32 gravitational components.
Nine conserved Dirac momenta comprise the CL algebra. We here
propose a Dirac self-adjoint CL-invariant Hamiltonian — kinetic energy
plus electromagnetic and gravitational potential energy — to evolve
the SMU ray from a featureless beginning. Illuminated here via dis-
creteness of electric charge are baryon number (“nuclear forces” arising
from “almost-screened” electric charge), Bohm’s “hidden variables” —
dark matter and dark energy — and three generations of “elementary”
fermions.
1. Introduction
Dirac’s non-relativistic quantum theory [1] was based on Hilbert-space self-
adjoint operators — Dirac coordinates and Dirac momenta plus a Hamilto-
nian which commuted with the Dirac-momentum-represented algebra of a
Euclidean Lie symmetry group. Dirac coordinates represent the symmetry-
group’s manifold. Ray evolution was according to a Schrödinger first-order
differential equation. Such quantum theory has for “relativistic” physics
579
580 G. F. Chew
been obstructed by absence of unitary finite-dimensional Lorentz-group

representations.
The Gelfand–Naimark (GN) unitary Hilbert-space infinite-dimensional
representations, [2] although inapplicable to a physics focused on “objec-
tive reality” — i.e. on particles — apply quantum-cosmologically to a
Schrödinger–Milne Planck scale-originated — “big-bang” — a continuously-
evolving universe (SMU) whose galaxies presently contain more “dark mat-
ter” than particulate. Non-galactic non-particulate unphysical energy con-
tinues at present to exceed that of galaxies.
Milne’s universe [3] situates inside a Lorentz–Minkowski forward light-
cone — a four-dimensional (4D) manifold whose boundary locates outside
Milne’s universe-occupied submanifold. A positive Lorentz-invariant “age”,
τ ≥ τ0 > 0, of any location within Milne’s universe is defined to be its
“Minkowski distance” (not a “Riemann geometrical distance”) from the
lightcone vertex. Universe age at Big Bang was τ0 .
Milne’s geometrical 3-space at any fixed age is hyperbolic — negatively
curved in Riemann sense as suggested by the surprising astronomically-
observed Nobel-acknowledged correlation between luminosity distance
and redshift [4]. The present chapter will “embellish” GN’s unitary
Hilbert-space Lorentz-group representation with discrete electric charge.
A 9-parameter SMU Lie symmetry group, here denoted “CL”, then enriches
Milne’s 6D SL(2,c) “universe-exterior” by a 3D group center that, in span-
ning the universe’s “quality space”, recognizes discreteness both of electric
charge and of energy.
Through an approximate relation between redshift and the (geometri-
cal) distance light travels through Milne’s negatively-curved non-compact
3-space, age is astronomically estimate-able. Present age is approximated
by the reciprocal of astronomy’s “Hubble constant”[4].
“Cosmological photons” (γc ), first created at extremely early SMU ages,
enjoy a “semi-foundational” status in SMU’s unification of electromag-
netism and gravity — a GN-dependent unification without “gravitons”
that attributes the strength and “short range” of (present-age) “nuclear
forces” to any nucleon (of unit baryon number) being a composite of nine
electrically-charged “quantum-universe constituents” — qucs. One of many
important outcomes of discrete-charge-screening is “short-range nuclear
forces”. A principal task of the present chapter is to define “quc”. An elec-
tron comprises three electrically-charged qucs, and a photon or neutrino
two. Dark-matter-composing qucs are chargeless.
www.ebook3000.com
Schrödinger–Milne Big Bang 581
Decades of thought about alternative possibilities, plus attention to

Occam’s principle, have led the author here to propose SMU birth with-
out any “particles” at a Planck scale age. But there was, at SMU’s start, a
huge (although finite) energy which has diminished subsequently by inverse
age-proportionality of each quc’s energy. The here-proposed set of SMU-
composing qucs is finite and fixed, although ginormous.
The SMU ray at age τ ≥ τ0 > 0 — a sum of (“tensor”) products
of single-quc Hilbert vectors — is an indefinitely-differentiable function of
τ — without singularities at any age greater than or equal to the positive
starting-age τ0 whose value, together with the (permanently-fixed) number
of (different) qucs and an electric-charge unit, provides SMU’s foundation.
At any age (above τ0 ), ray expectations of certain here-defined self-
adjoint Hilbert-space operators specify a “reality” consistent at present and
recent ages with Karl Popper’s humanistic classical meaning for “physical
measurement”. But SMU’s age-varying 3-space curvature requires that an
inherently-approximate Euclidean meaning for “quantum physics” be dis-
tinguished from the meaning of “quantum-cosmology”.
Specified here is a (Dirac-sense) self-adjoint (even though “cosmologi-
cal”) Hilbert-space Hamiltonian operator whose potential energy generates
“creative” ray dynamics (crd). Kinetic energy “perpetuates” the creativity
of evolution. Appendix A proposes a creation-bereft (particle-absent) initial
ray all of whose qucs carried positive energy.
SMU’s attribution, through mathematical language, of cosmological
meaning to Hamiltonian kinetic and potential energy as well as to “momen-
tum”, “angular momentum’, “electric charge”, and “energy” — all notions
uncovered by human physics — has amazed the author. Is it conceivable
that the human-uncovered mathematical notions of “Riemannian geom-
etry”, “Lie-group algebra”, “Hilbert space”, “fiber bundle”, “differential
equation”, and “Mersenne prime” play “cosmological” roles that transcend
humanity? This chapter supposes such to be the case.
Milne’s hyperbolic 3D space at fixed universe age is CL-invariantly
metricized and thereby, although (negatively) curved, endowed with
an unambiguous (positive, Lorentz-frame-independent) shortest path
(geodesic) between any pair of spatial locations [4]. Completely age-
determined (without dependence on 3-space location), Milne’s spatial cur-
vature ignores energy distribution — curvature being initially at Planck
scale and diminishing thereafter — paralleling diminishment with advanc-
ing age of any and all quc energies. Hubble’s “constant” approximates both
582 G. F. Chew
the inverse of present-universe age and the current magnitude of (negative)

3-space curvature [4].
Above emphasized is SMU governance by the symmetry of a 9-
parameter CL Lie group whose (below-compactified) center is 3-parameter
U(1) × SL(2,c,D) [4]. Here, the symbol D denotes 2-parameter left-acting
diagonal complex unimodular 2 × 2 matrices. The (non-compact) CL exte-
rior is 6-parameter right-acting SL(2,c). Formula (8) below mentioned,
within our main text, shows how “left–right” distinction is achieved by GN’s
remarkable Hilbert space (never associated by its discoverers to cosmology).
CL extends SL(2,c) by a trio of single-parameter CL-commuting
subgroups. Algebra extension from 6 to 9 elements defines, firstly, additively-
conserved (ac) discrete and superselected electric charge with units propor-
1
tional to (c) 2 , secondly, a discrete “semi-superselected” attribute dubbed
“chirality” with /2 units and, finally, discrete ac energy with /2τ units.
As later elaborated, the universe’s 3D “quality space” is spanned by CL’s
center.
Energy discreteness does not mean Hamiltonian diagonalization. (Spac-
ing between successive possible single-quc energies is h/2τ .) The quantum-
physics notion of “stationary state” enjoys no cosmological meaning. SMU
displays “onflow” — never-ending continuous development of “newness”.
The 6-element (“acting from the right”) exterior CL algebra
defines angular momentum and momentum. Although the former (when
Stone–Dirac represented) is discrete, the latter is not. Non-compactness of
CL exterior associates in SMU to continuous spectra for a (3-vector) trio
of Hilbert-space self-adjoint (Dirac) quc-momentum operators, whose non-
Euclidean (cosmological) failure to mutually commute lacks physics prece-
dent. Physics meaning for the term “boost” is absent from SMU’s cos-
mological vocabulary. An SMU Stone–Dirac momentum 3-vector operator
generates displacements in Milne’s curved 3-space.
Conservation of electric charge and chirality, as well as of angular
momentum, is SMU (Noether) sustained by unitary Hilbert-space CL
representation and CL-invariance of Hamiltonian. Despite inverse age-
proportionality of all quc energies and momenta (not of “particle masses”,
whose significance is physical — not cosmological), the Cl-central algebra
adjoins a positive ac energy-integer to charge and chirality integers —
thereby defining a (permanent) “complete set” of different SMU con-
stituents.
Wigner’s flat 3-space (10-parameter) “Poincaré group”, founda-
tion for quantum field theory (QFT) and the S matrix, fails to
www.ebook3000.com
admit (Dirac-required) unitary Hilbert-space representation. Having been

apprised by Finkelstein (private communication) that the Lorentz group
“Inonu-contracts” to the Euclidean group in a τ → ∞ limit, we have
come to regard QFT as a “3-space-flattened” micro-macro-scale approxi-
mation which, for human-history values of τ /τ0 (∼1060 ), is adequate for
human physics purposes (FHPP, imitating John Bell’s acronym) although
not for all purposes of Schrödinger–Milne cosmology. (The author believes
Bell, philosopher as well as physicist but not cosmologist, to have regarded
humanity’s 3-space as FAPP Euclidean.)
Physics, able to “Popper-define” particles but not qucs, is unable
to describe “dark matter”. Bewilderingly (to the author), Maxwell
classical-electromagnetic theoretical physics, through 4-vector electric-
charge current density with discrete electric charge, plus symmetric-tensor
energy–momentum current density, seems capable of classical Popper-
physics (discrete) “particle” definition, regardless of 3-space curvature. But
QFT requires flatness for its 3-space. There is no SMU meaning for “quan-
tum radiation field”.
QFT had become an accurate “local” approximation for micro to macro
spatial scales by the macro-scale ages, τ /τ0 ∼ 1038 , when galaxy and star-
building commenced — well after the micro-scale ages, τ /τ0 ∼ 1019 , when
SMU’s massive-elementary-particle-building got underway. The Poincaré
group and the associated QFT identical-elementary-particle micro–macro
physics approximation will be addressed by other chapters. Appendices here
make a start.
QFT’s “parity-reflection” (not a Lie-group generator) parallels sign-
reversal of SMU’s self-adjoint “chirality” — one of the three central CL
generators. (All nine generators are “Noether conserved”). Dirac’s writ-
ings never mention “parity” but he proposed a “doubling” of electron spin
through a velocity direction (not momentum direction) that was either par-
allel or antiparallel to spin direction.
Dirac’s “relativistic-electron (quantum-physics) wave function” (frus-
tratingly for Dirac, not a “Hilbert vector”) satisfied a first-order
(Schrödinger) equation of motion via his doubling of 2-valued electron spin.
Later this doubling became represented by the notation (0, 12 ) and ( 12 , 0) for
a pair of inequivalent nonunitary finite-dimensional SL(2,c) representations.
The three eigenvalues, 0, ±1, of SMU’s self-adjoint quc “chirality” amount
to a “Dirac tripling” relevant to all particles, not only spin- 12 fermions.
A pair of main-text 4-vector fiber-bundle quc Dirac-coordinates, here
employed to achieve a retarded Lorentz-tensor-field classical-cosmological
584 G. F. Chew
definition of “reality”, represents the “Maxwell–Lorentz” group SO(3,1)

while also providing ( 12 , 12 ) representation of SL(2,c)R . Classical electro-
magnetic fields provide (0, 1) and (1, 0) representations of SL(2,c)R , but
strikingly-absent from the definition of these fields is any reference to chi-
rality — a term here accorded cosmological “Dirac-tripling” meaning.
Although no finite-dimensional representation of any “Lorentz” group
is unitary, a classical bridge between “particulate Popper physics” and cos-
mology is provided by a below-specified Lorentz-tensor “reality” that does
not require 3-space to be Euclidean and makes no reference to chirality.
Cosmological absence of QFT’s quantum-theoretic (S-matrix) meaning for
“particle” is alleviated by a classical cosmological “Popper” meaning that
is based on charge discreteness together with Maxwell’s equations and an
energy–momentum tensor.
Formula (8) here shows how, with GN’s unitary representation, Dirac
might, in the spirit of later-appreciated “supersymmetry”, has extended the
Hilbert-space-representable U(1) group generated by a self-adjoint oper-
ator representing electric charge. The 1D compact manifold spanned by
chirality-generated GN Hilbert-vector argument displacements covers an
interval twice that spanned by charge-generated displacements of a (com-
plex) Hilbert vector’s phase.
Formula (8) formalizes the foregoing. Unitary SL(2,c) representations
remained undiscovered for more than a decade after Dirac’s attempt to
“relativize” the electron. Dirac’s opinion, either about Milne cosmology or
about GN’s unitary SL(2,c) representations, is unknown to the author.
The “exterior” 6-element non-Abelian SL(2,c)R algebra, a subalgebra of
the CL group [4], defines SMU conserved momentum and angular momen-
tum — the generators, respectively, of infinitesimal displacements of loca-
tion and orientation within Milne’s negatively-curved metricized 3-space.
Milne seems not to have represented either energy or electric charge; he
almost certainly did not represent chirality.
The 9-member CL algebra provides a complete set of conservation laws
to govern SMU evolution of a reality that encompasses Hubble redshift
together with non-particle extra-galactic dark energy, galactic “dark mat-
ter” and QFT’s set of macro-scale-observable micro-scale “identical” ele-
mentary particles within our galaxy. (Certain other galaxies may be found
to contain an alternative set of elementary particles — with QFT’s “left-
handed weak-vector bosons” replaced by “right-handed” counterparts.)
The terms “particle” and “identical particles” are FHPP-meaningful at
present and recent SMU ages. But Milne–Lorentz cosmological symmetry
www.ebook3000.com
transcends physics by attending to dark matter and dark energy as well as

to “early” SMU history when 3-space curvature was huge on macro scale.
[Note: Our adjective “macro” applies both to the (kilometer) “lab” spatial
scale at which “measurements” are performed by “conscious beings” and
to the temporal scale of those SMU ages when galaxy-clumping began.]
SMU recognizes, while not depending on, approximate scale-dependent
human-language meaning for “free-will measurement by an observer” — a
macro-scale Galilean notion on which human (Popper) physics is founded.
The author believes human language is incapable of “exactly true sen-
tences”. Any human-language “truth” we believe is a scale-dependent
approximation, which sometimes may enjoy high accuracy because of ginor-
mous differences between five different spatial scales currently recognizable
in SMU — Great Unification Theories scale, micro scale, macro scale, galac-
tic scale and Hubble scale. Human language is macro. Far above macro while
still far below the (present) scale of Hubble is the galactic scale of dark
matter. All five scales play roles in the present chapter’s content, which to
the author appears consistent with the belief, by an increasing number of
philosophers, that “free will is a (macro-scale) human illusion”.
Both universe age (approximately measurable by redshift) and the pro-
posed SMU Hamiltonian — Formula (21) below — which “analytically”
generates universe-ray evolution with increasing age, are CL invariant,
with the CL algebra (Stone) representable by self-adjoint GN–Hilbert-space
(Dirac-momentum) operators. Once again: CL algebra comprises (Noether-
conserved) continuous-momentum times age, angular momentum, electric-
charge, chirality, and energy times age — the latter sextet all Dirac-discrete.
2. SMU Elemental Constituents

The unitary CL representation by Formula (8) has led the author to rec-
ognize a set of SMU elemental constituents, each here bearing the (pro-
nounceable) name “quc” (quantum-universe constituent), which we here
suggest compose the entire universe in a sense evocative of that accorded
by nuclear physicists to Gell-Mann’s acronym “quark”. No single quc, indi-
viduated by three central integers specifying its electric charge, its chirality
and its energy, is a “particle”. Nevertheless, we propose SMU to be com-
pletely “quc-composed”.
A subscript q, equivalent to a trio of integers, Qq , Nq , and Mq , here
distinguishes any quc from all other qucs. It can be seen that the total
number of different qucs is 21 Mmax , with Mmax a ginormous but finite
586 G. F. Chew
(age-independent) positive integer. Dependence of any single-quc wave

function on one of its six “Dirac” quc coordinates — that which is
“canonically-conjugate” to Mq — “spreads” this quc over the entire Milne
3-space. Any individual particle “location” similarly is spread. All “mean-
ing” is relative. Application of any element of the CL symmetry group to a
(multi-quc) SMU ray “changes nothing”.
An SMU ray, at some fixed age, is a sum of products with 21 Mmax
factors — each of these a wave-function of a different quc. An “elementary
particle” is Newton–Maxwell-theoretically (Popper) a classical “clump” of
conserved energy, momentum and angular momentum, with an (approxi-
mate) “FHPP mass”, a “spin” and an electric charge equal to some integral
multiple of a universal charge unit.
Such “fixed and settled” Popper reality is later prescribed by
(mathematically-defined) expectations of certain self-adjoint operators for
the SMU ray at the age in question. Any particle, whether or not considered
“elementary”, associates to electro-dynamically-correlated wave-functions of
many different charged qucs.
3. Two-Quc “Cores” of Massive Elementary Particles

Within the individual terms of an SMU ray’s tensor-product summation,
a zero-chirality zero-charge 2-charged-quc “core” factor associates to any
elementary particle except photons and neutrinos. A core factor depends
on the quc-pair’s “relative” coordinate, which later in this chapter will be
seen essential in a more general context to the SMU Hamiltonian’s potential
energy. A particle’s mass reflects its core.
Not only does the net charge and chirality of a “particle-core” quc-pair
vanish but so does its below-specified baryon number. (A photon or neutrino
is a neutral zero-baryon-number but chirality-bearing quc pair.) When the
elementary particle in question is charged, the charge is provided by a
“valence” quc. Also established by valence is particle chirality and baryon
number.
What about elementary-particle energy, momentum and angular
momentum? A valence quc adds its contribution of the foregoing conserved
attributes to that of particle core. A Hamiltonian-prescribed superposition
of products with variable values of total energy, momentum and angular
momentum, as well as different distributions thereof between the three qucs,
represents the particle. Different energy distributions require different qucs
to appear within the 2-quc or 3-quc sets that quantum-theoretically repre-
sent a single elementary particle.
www.ebook3000.com
Each quc carries a positive energy integer with one of a ginormous

although finite set of possible values, an electric-charge integer with one
of the seven values 0, ±1, ±2, ±3 and a “chirality” integer with one of the
three values, 0, ±1. (All particle-building qucs have non-vanishing electric
charge!) The foregoing options became author-appreciated after decades
of Occam-guided contemplation that included many discussions with col-
leagues. Also additively-carried by a particle’s constituent-qucs are (“famil-
iar”) momentum and angular momentum.
4. Baryon Number and Chirality

Temporal microscale stability of a particle requires any particle-composing
quc to have non-vanishing electric charge. The electric-charge integer Qq
defines Quc-q’s baryon number, Bq , which vanishes when Qq takes any one
of the three values 0, ±3. If Qq is either +2 or — 1, Bq is + 31 . If Qq is either
−2 or +1, Bq is − 13 . Individual qucs thereby may be categorized as either
“baryonic” or “non-baryonic”; total-universe baryon number vanishes. (See
Table I in Appendix E).
Despite absence of “particle” status for single qucs, the nine conserved
Dirac-momentum quc attributes as well as baryon number are each addi-
tively manifested by particles “built” from qucs. “Particles” — (spatial)
“clumps of reality” each with integer net charge and one-third-integer
baryon number, odd or even chirality and an “FHPP mass” — are, in
a flat 3-space (QFT) approximation, fermions (bosons). Not addressed
by this chapter is the error in the S-matrix concept of “identical”
particles.
Particles of common FHPP mass, spin, charge, baryon number and
momentum but with differing energy-distribution among constituent qucs,
are “physically identical” — FHPP identical — despite not being “cos-
mologically identical” (Appendix C). Pedagogically-useful quc segrega-
tion, both by angular momentum and by chirality, into “fermionic” and
“bosonic” categories is unaccompanied by any meaning for “quc statistics”.
The physics word “particle” lacks precise SMU meaning! SMU classi-
cal cosmology is a “continuous onflow of electro-gravitational reality”, with
discretely-conserved (because of quantum superselection) electric charge
and baryon number and continuously-conserved energy and momentum
times age. The author believes the term “plasma” to be useful, both phys-
ically and cosmologically, but there is no cosmological usefulness for the
term “vacuum”.
588 G. F. Chew
[Energy integer, whose definition in Ref. [4] is repeated below in this

chapter’s main text, was suggested to the author not by QFT but by Fermi
momentum notions useful in condensed-matter physics. At present-universe
age the difference in energy associating to successive SMU energy integers —
a difference determined by SMU age — is “ginormously tiny.”]
The SMU Hamiltonian, specified by formula (21) to be a CL invariant
(not a 4-vector component), comprises a sum of single-quc kinetic-energy
and quc-pair electro-gravitational (EG) potential-energy self-adjoint Dirac
operators. The finite although ginormous set of SMU qucs is age-independent.
5. Cosmological vs. Physical Photons

Other chapters, through “recent-age” flat 3-space (Euclidean) micro–macro
approximations, will depict as Schrödinger–Dirac charged-quc composites
not only QFT quarks, charged leptons and W bosons but neutrinos, Z0 ’s,
and Higgs bosons. Transcending Euclidean 3-space QFT (whose accuracy
derives from recent macro-scale hugeness of SMU age) are net electrically-
neutral but chirality-carrying zero-mass cosmological photons (γc ) — each a
pair of non-baryonic qucs oppositely charged but of same non-zero chirality
(total γc chirality being ±2).
Equality, between the value of its chirality and that of twice its helic-
ity, accompanies physical-photon unique masslessness — equality of energy
and kinetic energy. The quc structure of a propagating γc might spatio-
temporally be described as an “electro-gravitationally-stabilized double
helix” — a “perfect quc marriage” — remarkably enjoying the same num-
ber (6) of “Dirac degrees of freedom” (Ddof) as a single chargeless quc.
(Appendix C) All other quc marriages are “imperfect”.
Two of a γc ’s six Ddof are “relative” (internal and fluctuating) coordi-
nates on which a unique γc “ground-state” internal wave-function depends.
Four of the six Ddof are external “Dirac momenta” (3-vector momentum
plus helicity). Future investigation we expect to expose not only the “shape”
of the γc ’s internal ground state but absence of other stable internal states.
(The double-helix characterization involves both external and internal Ddof
of a γc .)
Pedagogically, the term “mass” helps to distinguish the “particle” con-
cept from that of “quc”. There is no meaning for “quc mass” — in con-
trast to the long-appreciated physics meaning for zero photon mass, which
Appendix C, via the 2-quc γc , accords to FHPP “identical photons” obeying
Bose statistics.
www.ebook3000.com
6. Miscellany, Perhaps Helpful to Reader Thinking

Throughout this chapter, reference to a single quc may for convenience of
reader thinking be understood as in an “SMU local frame” whose definition
relates to later-defined quc “Dirac coordinates”. FHPP meaning for local
frame relates to Milne’s celebrated “cosmological principle”. In local frame,
present-universe cosmic background radiation (CMB) is isotropic. (QFT’s
meaning for “lab frame” roughly matches — with error ∼10−3 — that of
SMU local frame!) In local frame, time change and age change are equal —
“time interval since big bang” being equal to τ − τ0 .
“Local frame” associates to the CL-invariant meaning of “quc energy”.
Wherever the latter term here is used it refers to the quc’s energy in a space
where a (3D) spherically-symmetric big bang occurred — all clumping of
spatial energy being a consequence of crd after universe birth. (Appendix
A) Because of clumpings developed before (“recent”) CMB decoupling from
atoms (at τ /τ0 ∼ 1057 ), the local frame is only approximately establishable
by CMB astronomy.
The quc-composed Schrödinger–Milne universe may not be described
as “composed of elementary particles”. Additivity of quc “Dirac momenta”
nevertheless allows an “elementary-particle set” of two or three (differ-
ent) charged qucs to represent the CL symmetry group through a Dirac-
momentum unirrep csco — a complete set of seven commuting self-adjoint
operators that adjoins two CL Casimirs to the direction (two operators) of
(exterior) conserved momentum and the three central momenta. SMU’s
multi-quc “external-momentum csco” parallels the “asymptotic Hilbert
space” of the particle-physics S matrix. Appendix B addresses the single-quc
“Dirac-momentum” csco.
The SMU Hamiltonian — to which this chapter’s main text leads —
is expressed through Dirac quc coordinates and Dirac quc momenta, but
not through particles. The author nevertheless expects SMU’s Hamiltonian
(eventually) to explain (approximately) the observed values of elementary-
particle masses and other arbitrary QFT parameters (Appendix E).
Each of nine conserved elementary-particle attributes is the sum
of (corresponding) constituent-quc attributes. QFT elementary-particle
“asymptotic Hilbert space” enjoys useful approximate (flat 3-space) S-
matrix meaning because the macro scale of a human laboratory, although
huge on micro scale, is tiny on Hubble scale. Sums of quc (Dirac) momenta
approximate particle (S-matrix) momenta. No such feature attaches to
(non-conserved) quc (Dirac) coordinates.
590 G. F. Chew
Even with quantum fluctuation of the energy integer, Mq (essential

to single-quc identification), it is useful to think of any QFT elementary
particle as a macro-stable (“Popper-measurement-accessible”) potential-
energy-induced marriage (after universe birth and of GUT or micro internal
spatial scale) between two or three electrically-charged chirality-carrying
qucs whose wave-functions at SMU birth were uncorrelated (Appendix A).
An “early” such marriage — at GUT or micro-scale ages — might
dynamically be described as Hamiltonian-generated “collapse” of a product
of uncorrelated charged single-quc wave-functions into a macro-temporally-
stable micro or GUT-sized “object” wave-function. (The photon is the only
“elementary” particle where, because of GUT-scale internal extension, grav-
ity significantly contributes to stability.) In any present-day S-matrix “con-
nected part”, it is electromagnetic potential energy that causes charged
married qucs to “change partners”.
Dark matter comprises gravitationally-sustained galactic-scale “col-
onies” of individually-chargeless bachelor qucs. (A dark-matter bachelor-quc
wave-function spatially spans its galaxy.) Inter-galactic universe-spanning
bachelor qucs, both charged and uncharged but without previous “marriage
history” because of energy too high for electromagnetic macro-clumping,
constitute “dark energy”.
Particle mass, a physics word without (onflow) cosmological meaning,
reflects Hilbert-space collaboration between the discrete ca trio (electric
charge, chirality, energy) and continuous zitterbewegung (zbw). The lat-
ter (Schrödinger-coined) term refers to fluctuation of light-like velocity
direction — a non-conserved Dirac coordinate — at fixed conserved Dirac
momenta. As above noted, the CL unirrep csco provides a Hilbert-space
FHPP elementary-particle-momentum basis that, although without parti-
cle mass among its labels, resembles “in” → “out” S-matrix “asymptotic
Hilbert space”.
Physics-foundational, however, are conserved (and with commut-
ing components) energy–momentum 4-vector Poincaré-group generators.
Essential SMU roles are played in this chapter’s main text by two non-
conserved positive-4-vector quc Dirac coordinates, specifying the quc’s
spatio-temporal location and its light-like velocity direction, but there are no
SMU quc-energy–momentum 4-vectors. SMU momentum-basis portrayal of
a photon’s external Ddof extends to massive elementary particles — via CL-
invariant particle energy and a smaller invariant, although not conserved,
momentum magnitude (“relativistic kinetic energy”). But neither portrayal
is 4-vector!
www.ebook3000.com
For pedagogical reasons the CL unirrep (Dirac momentum basis) is

not addressed by this chapter’s main text. Appendix B exposes the quc
Dirac-momentum csco that complements the main-text-explored quc Dirac-
coordinate csco.
QFT is a spatially-restricted scale-dependent approximation that
accords meaning to “vacuum”, ignores (photon-redshifting) universe expan-
sion and relies on a S matrix with a priori elementary-particle masses
and electric-charge screening. Electron mass is treated by QFT as non-
fluctuating and age independent.
SMU (approximate) age-independence of electron mass we associate to
ongoing interchange between electron-attached qucs and charged intergalac-
tic bachelor (un-clumped “dark-energy”) qucs of, on average, slightly higher
Mq than those of the electron qucs they replace. The author counts on even-
tual verification of such interchange by Hamiltonian-based computation.
The “vacuum” concept is absent from SMU, where the set of qucs is fixed.
Incompleteness of electric-charge screening is of major SMU importance.
Any quc subset represents CL. The total set of universe-comprising qucs,
although ginormous, is finite and constant (τ -independent). Qucs are never
created or annihilated. We repeat: Despite each of eight CL generators — all
except chirality — representing a conserved quc attribute to which a physics-
familiar name may be attached, while all nine ca are particle-carried, no
quc may ever individually be called “particle”.
Six conserved “exterior” attributes of a quc — its momentum and
its angular momentum — may change from crd interaction with other
qucs as universe-age increases. Three integer-specified conserved “central”
attributes — electric charge, chirality and energy times age — remain
unchanged. Continuous momentum-basis variability of a quc’s momentum-
magnitude times age, while its energy integer remains fixed, renders impos-
sible any definition of “single-quc mass”.
Objective reality — spatially-localized and temporally-stable current
density — involves at the very least two qucs. A single quc cannot represent
an “object” — the definition of which requires a self-sustaining relationship
between different qucs. Dark matter, despite its gravitationally-sustained
localization being at galactic scale, is “objectively real”. A galaxy, although
not built entirely from “particles”, may astronomically be described
(approximately) as a “spatially-localized object”.
Any future SMU (of age greater than present age) is determined
by the quantum state of SMU now (as the reader is seeing this sen-
tence). SMU’s quc foundation transcends measurement-based physics.
592 G. F. Chew
A self-adjoint Hamiltonian operator is a Schrödinger–Dirac-imitating sum

of single-quc kinetic-energy and quc-pair potential-energy operators.
Each quc carries a discrete energy that permanently remains posi-
tive despite perpetual decrease by its inverse proportionality to positive-
increasing universe age. Accompanying perpetual photon redshift is ongoing
(never complete) flattening of hyperbolic 3-space [4]. These SMU features
(in profound contrast to general relativity) involve no reference to “local”
energy density!
Current densities of (classical) energy–momentum and electric charge —
specified by expectations of EG field-operators — define at every age an
SMU “reality”. Only a portion of this reality, nevertheless, is “objective” —
expressible through the “stable object” notion (which includes a “galaxy
of stars with attached dark matter”). Most of the present universe’s energy
remains non-objective. Bohm hidden-reality comprises both dark energy
and, w.r.t. “Copenhagen (S-matrix) quantum physics”, dark matter. Dis-
regard of SMU’s hidden reality has required probabilistic interpretation for
“Copenhagen” quantum theory.
At any universe age, τ ≥ τ0 , an SMU Hilbert-space ray, “regularly”
representing CL, is a complex normed multiply-differentiable function of
the six Dirac coordinates of each member of a ginormous but finite age-
independent set of qucs. We shall see how the six coordinates of any quc,
equivalent to three complex coordinates, specify via a complex unimod-
ular 2 × 2 coordinate matrix the quc’s location within a 6D manifold. A
unit such matrix locates the quc at an SMU “oriented center”. Any “exte-
rior quc location” — later shown to include a 5D fiber bundle — is ren-
dered “central” by that right-SL(2,c) transformation (an exterior element
of the CL group) which transforms this quc’s Dirac coordinate to a unit
matrix.
Fixed Popper reality — current densities of energy, momentum, angu-
lar momentum and electric charge (not chirality) — is prescribed by the
Dalembertian of a 13-component (22 + 32 ) retarded classical EG Lorentz-
tensor field. The present chapter’s main text specifies this field by SMU-
ray expectations of self-adjoint Lorentz-tensor field (not radiation-field)
operators.
At SMU birth all reality was hidden and full reality fails (at any age) to
specify the SMU ray that is Schrödinger-determined by the “immediately-
preceding” ray. Present reality fails to determine future reality! Hilbert
space with Schrödinger Hamiltonian dynamics is SMU essential.
www.ebook3000.com
7. Unification
This chapter “unifies” gravity and electromagnetism by “bundling” classi-
cal Newton–Maxwell (G − c), quantum Planck–Schrödinger–Dirac () and
classical Hubble–Milne (τ ). The foregoing brackets associate to natural
philosophers symbols for four positive dimensionful real parameters — three
constant (G, c, ) — and one perpetually increasing (τ ) — that underpin
the present chapter.
Any objective reality, such as a photon, a proton, a molecule, a planet,
a star, or a galaxy, associates to exceptional temporally-stable spatially-
localized multi-quc wave-functions where the “expansion” tendency of pos-
itive quc kinetic energy — to increase spatial separation between different
qucs — is opposable by negative gravitational and (or) electromagnetic
potential energy that tends to decrease separation.
“Strong interactions” (“nuclear forces”) arise from Formula (21) Hamil-
tonian’s kinetic plus electromagnetic potential energy, applied to systems
with baryon-number-carrying valence qucs. Gravitational potential energy,
together with electromagnetic, we believe essential at GUT scale to the pho-
ton double helix as well as, at ginormously larger scales, to planets, stars,
and galaxies.
8. “Quality” 3-Space: Natural Units

SMU displays a 3D “quality space”. The central SMU “momentum dimen-
sionalities” — those of energy, electric charge, and chirality — span all uni-
verse dimensionality. SMU chirality shares the dimensionality of angular
momentum. SMU associates G (Newton) to energy, c (Maxwell) to electric
charge, and (Planck) to chirality.
Dimensionless non-conserved quc attributes prominently include (as will
later be seen) 2D positive light-like velocity direction — an SMU “Dirac quc
coordinate”. Mathematics distinction between number-theoretic “topology”
and analysis-theoretic “geometry” provides some analog to the distinction
between “dimensionless” and “dimensionful” quc attributes, but neither
mathematics nor theoretical physics has so far dignified through a symbol
the 3D “quality space” displayed by the universe. Such a symbol would
“legitimatize” SMU’s Lie symmetry-group center — currently puzzling to
mathematicians as well as physicists.
Quality-space 3-dimensionality dovetails with the trio of indepen-
dent universal dimensionful constants with independent dimensionali-
ties. To these constants the symbols G, c and have become attached.
594 G. F. Chew
Occam-inclined natural philosophers expect no universal dimension-ful con-

stant (udc) beyond this trio.
Notational economy has long been appreciated for units that assign
the value 1 to each member of the udc trio. This chapter will henceforth
employ such units, thereby attaching unambiguous numerical value to the
dimension-ful SMU age symbol τ (present-universe age in “natural” units
being ∼1060 ), to the “big-bang” birth age τ0 (1 or ten to power zero), and
to dimension-ful particle-physics parameters such as Higgs mass (∼10−17 )
and electron mass (∼10−22 ).
Values for all the many arbitrary QFT parameters we expect to be
shown SMU-Hamiltonian-determined by the (macro) temporal-stability
spatial-localizability requirement implicit in objectivity. [Our guess for
“diameter of photon double helix” — a notion unrecognized by QFT —
locates in the neighborhood of QFT’s “GUT” space–time scale — ∼104 in
natural units (Appendix C)].
9. Mysterious Physics-Enabling Macro-Scale;

Avogadro Number; Feynman’s Perturbative
S Matrix
This chapter assigns “foundational” status neither to the micro spatial
scale, ∼1019 , set (inversely) by elementary-particle masses nor to the much
larger (∼1038 ) “macro” spatial scale that characterizes at once the (kilome-
ter) scale of human “laboratories” and the Schwartzschild radius of stellar
mass. These scales we portray here as crd outcomes. Far below micro spatial
scale is SMU-foundational GUT scale — that we associate to the electric-
charge unit g. Far above macro are galactic and (present) Hubble scales.
“Macro” locates in the “logarithmic middle”. Two sections below we men-
tion a huge Mersenne prime that, via Mmax , might relate to “macro”.
QFT’s spatial parameters are “micro” — in a neighborhood below macro
by a factor larger than the cube root of Avogadro’s mysteriously-huge
number. Long appreciated is dependence of physical chemistry, statistical
mechanics and condensed-matter physics on the latter’s largeness.
Macro scale understood as “lab scale” — the “scale of measurement” —
is recognized by the QFT S-matrix as well as by Avogadro-number-based
atomic and molecular physics, statistical mechanics and condensed-matter
physics. How does QFT recognize macro scale? Within each denominator
of Feynman’s perturbative series for an S-matrix element there appears the
symbol ε — representing an energy which, although “vanishingly-small” for
www.ebook3000.com
S-matrix purposes, cannot be ignored. S-matrix definition regards “macro

time” as “ginormously larger” than the times associating to elementary-
particle masses.
The inverse of Feynman’s ε may be regarded QFT’s definition of
“macro” time scale — huge for particle-physics purposes, as well as for
those of any science relying on Avogadro-number hugeness, while “tiny”
compared to galactic scale and yet tinier when compared to that of Hubble.
“Macro scale” includes that extremely narrow scale range where “free-will
measurement by conscious life” enjoys meaning. Alfred North Whitehead
acknowledged mysterious macro scale not by a number but through use of
the term, “God”. Schwartzschild was thinking “stellar black-hole radius”
when recognizing macro scale.
10. Age-Independent Huge Finite Set

of Different Qucs
SMU is populated by a finite set of different qucs — distinguished by an
electric-charge integer, Qq , a chirality integer, Nq , and a positive energy
integer, Mq . Each of the corresponding attributes enjoys a separate dimen-
sionful unit. QMU has exactly one quc for any Qq , Nq , Mq integer-trio, with
Qq allowed the seven values 0, ±1, ±2, ±3, and Nq allowed the three values
0, ±1 while Mq is allowed Mmax possible values, 1, 2, . . . , Mmax . The total
number of SMU qucs is thus 21 Mmax . Our choice of allowed values for Qq
and Nq has been influenced by Occam, by Mersenne and by QFT’s set of
elementary particles (Appendix E). For present-chapter purposes we elect
to leave still unspecified the ginormous value of Mmax .
Any DMU ray is a sum of products of 21 Mmax single-quc normed
functions, any quc appearing exactly once in each such “tensor” prod-
uct. It will below be seen that GN unitary Hilbert-space representation
of SL(2,c) requires each single-quc Dirac-coordinate-basis wave-function to
depend on that quc’s location within a non-compact 4D manifold — not
a space–time manifold but a product of two complex-variable-coordinated
manifolds.
An approximately temporally-stable multi-quc although single-particle
wave-function, such as that of an electron, correlates constituent-quc
energy-integers with constituent-quc chiralities and charges. In “reactions”
that annihilate this particle while creating other particles, quc “creation”
or “annihilation” never occurs. Instead there is quc “reallocation”.
Sub-product clusters of two or three qucs represent elementary particles
of sharply (integer)-specified electric charge and baryon number. Chirality
596 G. F. Chew
evenness or oddness of any particle is unambiguous. Always, within any

elementary-particle wave-function there is superposition of different ener-
gies whose spacing, (2τ )−1 , is presently smaller than any “soft-photon”
energy by a factor of order 1022 .
The author anticipates eventual number-theoretic specification of Mmax ,
the maximum quc local-frame energy in units of 1/2τ . Perhaps Mmax , is
a ginormous Mersenne prime. Mmax , is one of several SMU “foundational
numbers”. Another is the dimensionless coefficient g, related to the “fine-
structure constant” of particle physics, that determines GUT scale and
appears below in our formulas for electric-current density and Hamiltonian
electromagnetic potential energy. The “large” Mersenne prime 27 − 1 = 127
has often been conjectured to set the value of g −2 . Our number–theoretic
Occam-choice for a third foundational parameter — the SMU birth age τ0
in “natural” units — is 1. The huge Mersenne prime, 2127 − 1, may relate
to Mmax and to mysterious macro scale.
As age τ (>τ0 ) continuously increases, the value of every quc’s energy —
some fixed-integer multiple of 1/2τ — decreases in inverse proportion to
age: Milne redshift. The value of (continuous, positive, dimensionful and
global) age establishes an SMU scale (presently “Hubble”) not only for
quc total and kinetic energies [see formula (21)], but for quc momentum
(main-text section on EL Casimirs) and for distance (at same age) between
different quc locations in the hyperbolic Milne 3-space [3] whose Riemann
(negative) curvature is τ -determined [4].
Classical (although quc-sourced) retarded EG tensor fields whose
Dalembertians prescribe SMU reality, will be seen below to have 13 com-
ponents. The electromagnetic and gravitational fields sourced by Quc q are
proportional, respectively, to gQq and to Mq /2τ . The former is the quc’s
electric charge while the latter is its energy. Total universe energy at age τ
is, by a simple computation, 21(Mmax/2)2 τ −1 , while total charge and total
chirality vanish, together with total baryon number.
11. Unitary Hilbert-Space Dirac-Coordinate

Representation of 9-Parameter CL
We now reproduce, with minor notation adjustment, certain formalism from
Ref. [4]. An SMU ray at Age τ is a sum of (“tensor”) products, each with
21Mmax factors, of single-quc Hilbert vectors. In the “Dirac-coordinate”
Hilbert-space basis, each of the latter is a normed complex differentiable
function, ψqτ (aq ), of Quc q’s location aq in a 6D manifold.
www.ebook3000.com
The symbol aq denotes the spectra of a complete set of six commut-

ing self-adjoint operators — a “Dirac-coordinate” csco that complements
a 6-element, Casimir-based, “Dirac-momentum” csco which our main text
ignores, apart from references to CL Casimirs. Appendix B attends to the
momentum basis.
Reference [4] has shown how aq comprises a 3D metricized base-space
location and a (non-metricized) fiber-space location whose dimensionality
is 2 + 1 = 3. The coordinate of the 1D fiber subpace will be seen Dirac-
conjugate to chirality. The complete 6D “quc-locating” coordinate amounts
to a 2 × 2 complex unimodular matrix. Henceforth in this chapter’s main
text any boldface symbol is to be understood as denoting a 2 × 2 matrix.
The single-quc Dirac-coordinate-basis wave function, ψqτ (aq ), unitarily
representing CL at each age τ ≥ τ0 [see formula (8)], is a Hilbert vector
with the invariant (finite) norm,

daq |ψqτ (aq )|2 . (1)
The CL-invariant 6D volume element (Haar measure) daq we below express

through a trio of complex Dirac coordinates equivalent to the matrix aq .
Because the present section and that following refer to a single quc and
a single age, we shall in these sections omit both the superscript τ and
the subscript q. Also ignored, except in Eq. (8), is the charge integer Qq ;
U(1) transformation does not affect the wave-function’s (Dirac-coordinate)
argument — merely shifting complex-wave function phase (in any basis) by
an increment proportional to Qq .
The unimodular 2×2 complex (Dirac) quc-coordinate matrix a is equiv-
alent to three complex variables: s, y, z (six real variables), according to the
following product of three unimodular 2 × 2 matrices, each of which coor-
dinates the manifold of a 2-parameter Abelian CL subgroup:
a(s, y, z) = exp(−σ 3 s) × exp(σ + y) × exp(σ − z). (2)
The s subgroup lies within the CL (diagonal-matrix) center. The complex

variables s and y together coordinate the manifold of a 4-parameter non-
Abelian CL subgroup. The latter feature is essential both to GN’s unitary
transformation between Dirac-coordinate and Dirac-momentum bases [2]
and to coordination of base and fiber spaces.
598 G. F. Chew
The 2 × 2 real-matrix pair σ ± is defined as 1/2(σ1 ± iσ2 ). The (familiar

to physicists) Pauli-matrix symbols σ 3 and σ 1 represent Hermitian real self-
inverse traceless 2×2 matrices with determinant −1, σ 3 being diagonal and
σ 1 off-diagonal, while the symbol σ 2 represents an imaginary such matrix
equal to −iσ3 σ 1 . The 6D Haar measure,
da = ds dy dz (3)
is invariant under a → a Γ ≡ aΓ−1 , with Γ a 2 × 2 unimodular complex

matrix representing a right SL(2,c) transformation of the coordinate a.
The measure (3) is also invariant under analogous left transformation. Any
“volume-element” symbol dξ in (3), with ξ complex, means d Re ξ × d Im ξ.
The Hilbert-vector norm (and inner-product)-defining integration (1) is,
wrt Im s, over any continuous 2π interval of Im s. Interpreting 2τ Re s as
periodic “quc time” we shall below shrink the Hilbert space so that Re s and
Im s enjoy similar status in vector-norm (and inner-product) regular-basis
definition. Full real lines for Re y, Im y, Re z, and Im z remain spanned by
the shrunken space’s vector-norm definition.
A transformation specified by the 2×2 complex unimodular right-acting
matrix Γ is unitarily Hilbert-space represented by
Ψ(a) → Ψ(aΓ−1 ). (4)
Straightforward calculation shows aΓ−1 to be equivalent to
z Γ = (Γ22 z − Γ21 )/(Γ11 − Γ12 z), (5)
y Γ = (Γ11 − Γ12 z)[(Γ11 − Γ12 z)y − Γ12 ], (6)
sΓ = s + ln(Γ11 − Γ12 z). (7)
Under the 9-parameter CL symmetry group the two-dimensional volume

element ds within the Haar measure (3) is seen from (7) to be invariant.
Also invariant is the 4D volume element dy dz.
We now explicitly display CL representation by the single-quc Dirac-
coordinate Hilbert-space basis. A CL element (a location within the 9D CL
manifold) is specified by a U(1)-representing angle, ω, with 0 ≤ ω < 2π (1
parameter), by a left-acting SL(2,c,D) complex argument displacement s →
s + ∆ (2-parameter) and, finally, by the Formula (4) SL(2·c)R -representing
argument displacement (6-parameter). Under a so-specified (9-parameter)
CL element,
ΨQ (s, y, z)(ω,∆,Γ) → eiQω Ψ(sΓ + ∆, y Γ , z Γ ). (8)
www.ebook3000.com
Essential is commutativity of the two s-displacements. The 1D Haar-

measure volume elements d(Re s) and d(Im s) are separately CL invariant,
together with the (4D) dy dz volume element.
12. Periodicity in “Quc-Time” — a Hilbert-Space

Reduction
Displacement in the coordinate Re s, at fixed Im s, y, z, and τ , displaces
what we choose to call “quc local time” at fixed values of global age and
the quc’s five other coordinates. Quc energy — a self-adjoint Hilbert-space
operator representing a member of the CL center subalgebra and the source
of SMU gravity — is canonically-conjugate in Dirac sense to 2τ Re s.
Although positive lightlikeness of a quc velocity 4-vector (defined in the
following section) invites confusion between “temporal” and “spatial” quc
displacement, the group algebra unambiguously distinguishes CL-invariant
quc energy from any non-invariant quc-momentum 3-vector component of a
right 6-vector — an algebra member that generates infinitesimal quc spa-
tial displacement (at fixed age) in some (arbitrarily-specifiable) direction
through curved metricized 3D base-space [4].
[Because the infinitesimal-displacement direction is specified in some
fixed right-Lorentz frame, whereas a geodesic follows a curved path requiring
parallel transport of direction-defining axes, the later-defined invariant self-
adjoint quc kinetic energy — a function of Casimir geodesic-associated
second derivatives — is not proportional to the 3-vector inner product with
itself of quc-momentum.]
Already noted has been the explicit indication by formula (7) that
(fixed-τ, y, z) displacements in s are right-Lorentz invariant (both real and
imaginary parts). They further are invariant under the 3-parameter sym-
metry (central) subgroup (with energy, chirality, and electric charge as
generators) that defines quc type, despite failure to be invariant under
the full (6-parameter) left-Lorentz group. A quc Hilbert-space shrinkage
that requires ray periodicity in regular-basis dependence on Re s (location
of a “quc-timepiece hand”) maintains quc capacity to represent 9-para-
meter CL.
We therefore diminish quc Hilbert space by the Dirac-coordinate-basis
periodicity constraint,
Ψ(s, y, z) = Ψ(s + 2π, y, z), (9)
modifying the Hilbert-vector norm to integration in (1) over any (single,

continuous) 2π interval of Re s, as well as of Im s. The constraint (9)
600 G. F. Chew
specifies integer eigenvalues for the self-adjoint operator that is canonically-

conjugate to Re s. (The quc-energy operator has eigenvalues M/2τ .) We fur-
ther reduce the Hilbert space by requiring energy to be positive and upper
bounded (M a finite positive integer).
For each quc there are nine “conserved-momentum” self-adjoint oper-
ators, although they do not all commute with each other. A 6-element
Dirac-momentum csco that includes the operators M and N is identified
in Appendix B with an “irreducible” CL representation. The operators M
and N are, respectively, canonically-conjugate to the operators Re s and
Im s.
13. “Dirac-Coordinate” 4-Vector Operators that

Locate a Quc in a Five-Dimensional Fiber Bundle
Reference [4] defines a (classical, “exterior”) positive 4-vector as a
(4-parameter) positive-Hermitian 2 × 2 matrix that transforms under an
exterior Lorentz transformation Γ through right multiplication by Γ−1 and
left multiplication by the Hermitian conjugate of Γ−1 . A 4-vector’s invariant
“squared magnitude” is the Hermitian-matrix’s determinant. (The time-like
component is half the Hermitian-matrix trace.) Time-like, light-like, and
space-like 4-vector matrices have, respectively, positive, zero, and negative
determinants.
A pair of commuting (exterior) 4-vector Dirac-coordinate self-adjoint
operators, one positive time-like and one positive light-like, are equivalent
to a quintet of real quc coordinates that specifies the following Quc-q 2 × 2
unimodular matrix [4],
bq ≡ exp(iσ 3 Im sq ) × aq , (10)
= exp(−σ 3 Re sq ) × exp(σ + yq ) × exp(σ − zq ). (11)
Henceforth, any symbol with a q subscript, whether attached to a unit

3-vector (see below), a 2 × 2 matrix (boldface indicated, as in the foregoing
formulas), to a Lorentz tensor with 4-valued indices or to an EL Casimir,
is to be understood as a self-adjoint (Dirac) Quc-q operator on the SMU
Hilbert space.
The left multiplication (10) of the quc-coordinate matrix a q by the diag-
onal unitary unimodular matrix, exp (iσ3 Im sq ), has deleted the coordinate
Im sq from the coordinate-sextet sq , yq , zq (a csco). The eliminated coordi-
nate, Im sq (understood as a self-adjoint operator), is canonically-conjugate
www.ebook3000.com
in Dirac sense to that conserved integer-Nq -eigenvalued Quc-q momentum

(an element of the CL algebra) which we have called “chirality”.
The present section ignores chirality. Quc-q’s 5D “fiber-bundle location”,
either in 3D metricized base space or in a 2D (unmetricized) “velocity-
direction” fiber space, fails to depend on Im sq . We now show that base
3-space location and velocity-direction fiber 2-space location are equivalent
to the five coordinates, Re sq , yq , zq — the latter symbols representing a
set of five commuting self-adjoint Dirac-coordinate operators collectively
representable by the (single) 2 × 2 unimodular-matrix symbol b q .
Either coordination associates in Dirac sense to a quintet of commuting
self-adjoint coordinate operators on a single-quc Hilbert space. Throughout
the remainder of this chapter’s main text, all quc-coordinate symbols are to
be understood as referring to self-adjoint Dirac operators, either on single-
quc or multi-quc Hilbert spaces. Such operators commute neither with CL’s
momentum algebra nor with Casimir quadratic functions of that algebra.
Formulas (2) and (9) together expose as positive-Hermitian the unimod-
ular (single-quc) 2 × 2 coordinate matrix,
B q ≡ b †q b q = a †q a q (12)
= exp(−βq σ · nq ) (13)
the (non-matrix) symbol βq in (13) denoting a rotationally-invariant non-

negative continuous-spectrum self-adjoint operator while the (non-matrix)
symbol nq denotes a unit 3-vector self-adjoint operator that commutes
with βq . The symbol σ · nq denotes the inner product of two 3-vectors —
one a Hilbert-space operator and the other a 2 × 2 Hermitian matrix [4].
The Hermitian unimodular 2 × 2 matrix (and self-adjoint Dirac-operator)
symbol B q denotes a positive time-like dimensionless (exterior) 4-vector of
unit “Minkowski magnitude”.
Through the dimension-ful positive-real factor τ (age) the positive
time-like 4-vector (operator) symbol x q ≡ τ B q locates the quc within
(non-Riemannian) Milne (Minkowski) space–time — prescribing (in Dirac-
operator sense) its displacement from the vertex of the universal forward
lightcone. In a (to physicists) more familiar notation, the four (operator)
components of x q are τ cosh βq , τ nq sinh βq .
[Warning to physicists familiar with the Copenhagen statistical inter-
pretation of Dirac quantum theory: SMU reality does not associate to the
expectation of single-quc self-adjoint operators, such as x q — attaching
602 G. F. Chew
rather to the expectations of self-adjoint EG potential-field operators on the

multi-quc Hilbert space — operators defined in the section which follows.
The foregoing operators do not include zero-Dalembertian “radiation” fields
of the kind that QFT associates to its elementary particles.]
Complementing dimensionless B q , which coordinates a quc-fiber-
bundle’s metricized base space, is a second dimensionless positive
4-vector — this one light-like — to be denoted by the symbol v q and coor-
dinating a 2D unmetricized fiber space. The pair of 4-vectors, B q , v q , is
equivalent to b q and thereby to Re sq , yq , zq .
The latter equivalence is below exhibited via invariant 4-vector inner
products. The inner product of two 4-vectors will be denoted by the
symbol •. The inner product of two positive 4-vectors is non-negative.
[Because the inner product of any two right (exterior) 4-vector operators
may be shown equal to the inner product of a unitarily-equivalent left 4-
vector pair, [4] either product is invariant under the 12-parameter group
SL(2c)L × SL(2c)R and thereby is CL invariant.]
The quc-velocity-direction positive light-like 4-vector self-adjoint Dirac-
coordinate operator, v q , is defined to be the dimensionless zero-determinant
positive-Hermitian matrix
v q ≡ b †q (σ 0 − σ 3 )bq (14)
the symbol σ 0 here denoting the unit 2 × 2 matrix. Equivalence of the

coordinate matrix b q to the dimensionless positive 4-vector pair B q , v q
follows from the inner-product trio, B q ·v q = 1, B q ·B q = 1 and v q ·v q = 0,
deducible by going to the special (“local”) frame where B q = σ 0 (i.e. where
βq = 0).
In the physicist-familiar 4-component notation, the 4-vector self-adjoint
operator v q is equal to
(1, uq )
, (15)
(cos h βq − uq · nq sin h βq )
where uq is a zq -equivalent (2D, see Appendix B) unit 3-vector (a self-

adjoint Dirac-coordinate operator accompanying the unit 3-vector nq but
independent thereof) that admits the name, “direction of quc light-like-
velocity”. A 5D quc fiber bundle is thus coordinated by the (3D) base-
space coordinates, βq , nq , together with the (2D) velocity-direction fiber
coordinate uq . An equivalent Dirac-coordinate set is Re sq , yq , zq .
www.ebook3000.com
The positive light-like quc 4-velocity v q will be seen in the follow-

ing section, together with Qq and Mq , “almost completely” to specify
“electromagnetism-gravity from Quc-q source”.
14. Classical Retarded EG (Non-Radiation) Fields

As emphasized in the preceding section, throughout this chapter’s remain-
ing main text any symbol with a quc-designating subscript is to be under-
stood as representing a self-adjoint age-independent Dirac operator. The
only age-dependent self-adjoint SMU operator is the Hamiltonian. (SMU
does not admit the “Heisenberg picture”. No operator represents “quc
acceleration”.)
The present section deals with classical retarded electromagnetic
and gravitational Lorentz-tensor fields (not zero-Dalembertian “radiation
fields”) whose Dalembertian prescribes SMU’s “reality”. At any age τ ≥ τ0 ,
such fields are prescribed by expectations, with respect to the SMU ray at
that age, of certain self-adjoint retarded Hilbert-space operators which sum
over all quc sources of the field in question. These retarded-field tensor
operators we now define through the previous section’s pair of 4-vector
single-quc Dirac-coordinate operators.
Four electromagnetic SMU field components are complemented by
nine gravitational-field components. The preceding section’s quc-fiber-
bundle coordinating 2-parameter positive light-like quc-velocity 4-vector
and 3-parameter quc space–time-location 4-vector, together with quc elec-
tric charge and energy, “co-variantly” prescribe a “quc-source” for 13 EG
retarded-field operators. Positivity of quc-velocity 4-vector associates to the
source’s retarded nature; this 4-vector also prescribes the “direction” of the
generated tensor field. The source-location 4-vector, through the denomi-
nator of the Lienard–Wiechert (LW) formula, joins quc charge and energy
to determine magnitude of the quc-generated field.
We define retarded EG field operators — not “radiation” quantum
fields — by applying the LW formula to Quc-q source of gravity and elec-
tromagnetism. We begin with the latter, for which the 4-vector retarded
potential at some “field point” x , (x · x = τ 2 ), generated by Source q, is
gQq θret (bq , x)vqµ

Πq µ (x) ≡ (16)
(vq · x − τ )
the retardation step function, θret (bq, x ), being defined below via the light-
cone whose vertex locates at x .
604 G. F. Chew
The symbol Oret (bq, x ) in (16) denotes an operator function equal to 1

iff the space–time straight line of direction v q , that passes at age τ through
the “source” space–time location x q , intersects the x backward lightcone
(which does not include this lightcone’s vertex). Otherwise Oret (bq, x ) van-
ishes. (Any light-like straight line not passing through x intersects the x
lightcone exactly once.)
The Age-τ expectation, of the self-adjoint operator Π µ (x ) that sums
(16) over all qucs, prescribes the reality-defining classical electromagnetic
4-vector field Aµ (x ). The Dalembertian of Aµ (x ) is the electric-charge cur-
rent density — an aspect of reality that, despite “classical” status, manifests
electric-charge discreteness.
Discretization of quc energy renders almost straightforward an exten-
sion of the foregoing to gravity. In place of (16) the 9-component “traceless”
symmetric Lorentz-tensor retarded self-adjoint gravitational potential oper-
ator, that Source-q generates, is defined by
−Mq vqµ vqυ

þµυ
q (x ) ≡ θret (bq, x ) (17)
2τ (v q · x − τ )
with the symbol denoting the self-adjoint “symmetrized product”,

F G ≡ 1/2(FG + GF), of two individually self-adjoint factors that may
not commute. (Mq fails to commute with b q because the latter operator
quintet — equivalent to x q /τ and v q -depends on Re sq .)
The Age-τ expectation, of the self-adjoint operator þµυ (x ) that sums
(17) over all qucs, prescribes the classical gravitational traceless-symmetric-
tensor field Φµυ (x ). The Dalembertian of −Φµυ (x ) is the energy–
momentum tensor.
There is no retarded SMU-LW “acceleration field”. Photons are rep-
resented, within SMU reality, not by a zero-Dalembertian electromagnetic
field but via the energy–momentum tensor — which comprises current den-
sities of all energy, momentum and angular momentum.
Photon annihilations or creations — aspects of “objective reality” —
become (Popper) physically inferrable from positive-energy–momentum
current density together with classically-discrete electric-charge current den-
sity — the Dalembertian of the classical but discretely-sourced electromag-
netic vector field. Charge discreteness together with energy positivity allows
experimenters to infer discrete photons from observed current densities by
application of (classical) Newton–Maxwell theory to current densities of
electric charge and energy–momentum.
www.ebook3000.com
15. Self-Adjoint Single-Quc Kinetic-Energy

Operator — a Positive Function of CL Casimirs
The algebra of the 6-parameter semi-simple non-Abelian exterior-SL(2,c)
CL subgroup, comprises the conserved components of a 6-vector — a
second-rank antisymmetric exterior-Lorentz tensor. Three algebra mem-
bers associate to quc angular momentum, Jq (a 3-vector), and three to quc
momentum, Kq /τ (also a 3-vector). Each sextet-algebra member is repre-
sented by a self-adjoint operator on the quc Hilbert space. In the Dirac-
coordinate (sq , yq , zq ) basis each of these momentum operators linearly and
homogeneously superposes first (partial) derivatives [2]. As emphasized by
Appendix B, each algebra member is a Dirac momentum — not a Dirac
coordinate.
The two (invariant) CL-group Casimirs (commuting with all nine of the
conserved CL generators) are the 3-vector operator inner products Kq · Jq
and Kq · Kq − Jq · Jq — homogeneous in regular-basis (partial) second
derivatives [2]. Neither of the foregoing forms is positive, but Ref. [2] dis-
plays algebraic equivalence to another pair of invariant self-adjoint oper-
ators, one of which has (positive–negative) integral eigenvalues while its
companion enjoys a continuous positive spectrum. Denoting the former by
the symbol mq and the latter by the symbol ρq , the algebraic relation is
ρ m ρ 2 m 2
q q q q
Kq · Jq = , Kq · Kq − Jq · Jq = − + 1. (18)
2 2 2 2
The positive continuous-spectrum SMU quc-kinetic-energy operator is

ρq /2τ , joining in the 6-element unirrep csco (see Appendix B) the posi-
tive discrete-spectrum quc-energy operator, Mq /2τ .
16. Self-Adjoint Quc-Pair Potential -Energy

Operators
The SMU Hamiltonian potential-energy operator is a sum over 21Mmax
(21Mmax − 1)/2 quc pairs of CL-invariant electromagnetic-gravitational
potential energies, Vqq (τ ) = Vq q (τ ), whose individual status parallels that
of the Euclidean-group-invariant “Coulomb-gravity” potential energy in a
Hamiltonian for two (slowly-moving) charged, massive particles. We pos-
tulate an SMU Hamiltonian potential-energy operator, for the quc pair,
qq , that depends on the exterior-invariant “relative Dirac coordinate”
a qq ≡ a q a −1 −1
q . Note that a qq = a q q .
606 G. F. Chew
Guided by Ref. [4] and the LW denominator in formulas (16) and (17),

we further postulate inverse proportionality to eβqq − 1. Here the positive
symbol, βqq = βq q , stands for (EL-invariant) shortest distance in (curved)
relative base-space between the locations of Quc q and Quc q . This distance
equals cos h−1 [1/2tr(a †qq a qq ] — the same function of a qq as that which
in Formula (13) above specified the single-quc coordinate βq in terms of a q .
Beginning with electromagnetism, as we did above when defining clas-
sical EG fields via the LW formulas (16) and (17) for field operators, we
postulate

el 2 −1
Vqq (τ ) = g τ Qq Qq (eβqq − 1)−1 . (19)
The corresponding CL-invariant gravitational potential-energy operator is

gr −1 Mq Mq
Vqq (τ ) = −τ (eβqq − 1)−1 (20)
2τ 2τ
the complete quc-pair potential-energy operator being the sum, Vqq (τ ) ≡
el gr
Vqq (τ ) + Vqq (τ ). (The absence, anticipated earlier, of a separate “nuclear-
force” potential-energy Hamiltonian component will be reviewed in our con-

cluding section.) Notice how the rightmost factor in (19) and (20) exhibits
“Newton–Coulomb” dependence on βqq for βqq 1 while “Yukawa” expo-
nential dependence for βqq 1.
17. SMU Hamiltonian and Schrödinger Equation

As was the case for Schrödinger, our Hamiltonian sums symmetry-group-
invariant self-adjoint kinetic-energy and potential-energy operators that do
not commute. SMU dynamics proceeds through a multi-quc Schrödinger
(first-order) differential equation where, at each post-big-bang age, a CL-
invariant although age-dependent self-adjoint Hamiltonian operator (not
a CL-algebra member) generates an infinitesimal wave-function change
that prescribes the “immediately-subsequent” universe wave-function.
Schrödinger’s 1927 equation was similar although based on a 7-parameter
extended-Euclidean group with flat 3-space translations (instead of the
9-parameter CL group with curved 3-space translations).
The invariant age-dependent self-adjoint Hamiltonian operator is
Σ q ρq
H(τ ) = + Σq=q Vqq (τ ), (21)
2τ
while the evolution equation for the universe ray is
i∂Ψ(τ )
= H(τ )Ψ(τ ). (22)
∂τ
www.ebook3000.com
An initial ray of uncorrelated (bachelor) qucs, at τ = τ0 , is proposed in

Appendix A. Note that our Schrödinger equation, in absence of gravita-
tional potential energy, is conformally (“scale” ) invariant — dependent
only on age ratios and thereby paralleling a QFT feature important to
renormalization. Related is our conjecture that Maxwell’s equations are
satisfied by SMU classical electromagnetic fields.
18. Conclusion
A 9-parameter CL Lie symmetry group collaborates with a Schrödinger
equation that prescribes Schrödinger–Milne quantum-universe electro-
gravitational evolution with increasing universe age. SMU resides inside
a forward lightcone, the age of any location its “Minkowski distance”
from lightcone vertex. Age is a CL-invariant non-geometrical perpetually-
increasing parameter approximately equal at present to the reciprocal of
Hubble’s “constant”.
At each age greater than or equal to a starting (big-bang) age (1, in
units where G = = c = 1; in seconds, big bang age is ∼10−43 ), the Dirac-
coordinate-basis universe ray is a sum of (tensor) products of single-quc
wave-functions. A “quc” is an SMU constituent. The argument of a quc’s
wave-function in the latter’s “Dirac-coordinate” basis specifies the quc’s
location within a CL-dictated 6D manifold.
The number of qucs is ginormous but finite and unchanging. Each quc
represents CL [formula (8)] through displacements of its 6 Dirac coordinates
and of its wave function phase. The total number of qucs and the unit of
electric-charge remain to be specified. An estimate of the latter is provided
by the (incomplete) “Standard Model of particle-physics”.
Appendix A proposes an initial ray of “bachelor” qucs-devoid, at the
beginning, of mutual correlations. We suppose the first “marriages” of
electrically-charged qucs to have emerged at GUT-scale ages (∼10−39 s)
with creation of 2-quc “double-helix cosmological photons”. Later, at micro-
scale universe ages (∼10−24 s), there emerged “massive elementary parti-
cles — electrically-neutral 2-quc neutrinos, Higgs bosons and Z0 ’s, together
with charged 3-quc quarks, leptons, and W bosons. (Charged-quc compo-
sition of elementary particles, broached in Appendices C and E, will be
addressed in separate chapters.) At macro-scale universe ages (∼10−5 s),
we believe stellar construction began. All such conjectures are in principle
verifiable by computation.
608 G. F. Chew
Present-age (∼1017 s) “dark matter” comprises galactic-scale colonies

of electrically-neutral “bachelor” qucs that gravitationally attach to entire
stellar galaxies. A (still larger) universe component, remaining today
galaxy-unassociated, comprises quc bachelors that so far have maintained
their independence.
Quc chirality (conjugate to one of six quc Dirac coordinates) we have
(“Occam”) limited to the values 0, ±1 — a “Dirac-tripling” that accom-
panies limitation on any quc’s electric-charge integer to the seven values
0, ±1, ±2, ±3. In early-universe dynamics we believe the combination of
chirality and electric charge at particle-physics micro scale to have (dynami-
cally) distinguished baryon-number-carrying quarks and associated “strong
interactions” from elementary bosons and leptons with zero baryon number
(Appendix Table E.1).
We expect the SMU Schrödinger equation (22) to reveal “nuclear
forces”, along with other particle-physics, as an approximate notion — use-
ful at micro scale but not at all SMU scales and not a foundational feature
of a quantum universe — all of whose “forces” our Hamiltonian proposes
to be EG. The “short range” of nuclear forces manifests electric-charge
screening — important whenever the number of qucs in some charged-quc
set exceeds the sum of this set’s charge integers.
All physics measurements “Popper-rely” for interpretation of observed
objective reality, on classical-physics EG theory. The SMU ray speci-
fies, through expectations of self-adjoint electromagnetic and gravitational
field operators, a physics-enabling “fixed and settled reality” that includes
locally-unobservable energy (Bohm hidden reality) together with macro-
scale “observable objectivity”.
Acknowledgments
Decades of discussions with Henry Stapp have been invaluable to this chap-
ter. Also contributing to the ideas here have been Eyvind Wichmann,
David Finkelstein, Jerry Finkelstein, Dave Jackson, Stanley Mandelstam,
Ralph Pred, Bruno Zumino, Ramamurti Shankar, Don Lichtenberg, Ling-
Lie Chau, Ivan Muzinich, Korkut Bardacki, Bob Cahn, Lawrence Hall, and
Nicolai Reshetikhin.
Essential has been support and encouragement, especially during the
final years of my life, from my five children, Pauline, Frank, Pierre, Beverly
and Berkeley. Without their participation, this creation would never have
occurred.
www.ebook3000.com
Appendix A. Initial-Universe (Planck Scale)

Wave Function
The initial (τ = τ0 = 1) SMU wave-function we propose to have been a
single product of 21 Mmax single-quc wave-functions that represent uncor-
related “initially-bachelor” qucs. No (GUT-scale or micro-scale) marriages
between two or three qucs — electro-gravitationally-stabilized elementary
particles — were a priori. All particles and dark matter — all objective real-
ity — we presume to have resulted from Schrödinger-equation Hamiltonian-
generated evolution. Present-universe Hubble scale — the space–time scale
set by our universe’s age — is larger than the scale of τ0 by a factor of order
1060 .
Von Neumann ideas have led us to propose, as initial ray in the regu-
lar CL-representation (“Dirac-coordinate”) basis, the following product of
21Mmax “Gaussian” factors, with no dependence on the arguments of the
complex coordinates yq and zq — dependence only on their magnitudes,

Ψ(τ0 ) = exp(−iN q Im sq ) exp(−iM q Re sq )|yq zq |−1
q

1 2 1 2
× exp − ln |yq | − ln |zq | . (A.1)
2 2
With the starting ray (A.1), which recognizes electric charge by the sub-
script q on any GN–Dirac coordinate being defined as synonymous with the
integer trio Qq , Nq , Mq , the 3-vector momentum operator of each quc has
vanishing expectation. This ray, further, is an eigenvector of total 3-vector
angular momentum with zero eigenvalues for all components thereof —
thereby satisfying classical Mach-Milne principles perpetuated by (main-
text) Eq. (22). Total electric charge and chirality have zero eigenvalues in
the initial ray (A.1) as well as in all subsequent rays. Total starting energy
has the value, 21(Mmax /2)2 τ0−1 . Subsequent total SMU energies replace τ0
(=1) in the foregoing formula by τ .
Appendix B. CL Unirrep-Csco as “Dirac Momentum”

Basis
GN’s unitary SL(2,c)R Hilbert-space representation via normed complex
differentiable functions, Ψ(a), of (single-quc) location within the 6D group
manifold, was called by these authors the “regular” Lorentz-group represen-
tation [2]. Our main text has characterized the corresponding complete set
610 G. F. Chew
of six commuting self-adjoint operators (“csco”) as a “Dirac-coordinate”

Hilbert-space basis. Not GN-emphasized is representability, by this same
csco, of the 12-parameter group SL(2,c)L × SL(2,c)R . An 8-parameter sub-
group of the latter — keeping only the diagonal left-multiplying 2 × 2
complex-unimodular matrices — is the electric-charge-ignoring subgroup
of the 9-parameter SMU-foundational CL group [4]. [The CL center aug-
ments SL(2,c,D)L with U(1).]
Exposed in detail by GN, beyond their regular SL(2,c)R represen-
tation, was the latter group’s unitary irreducible representation by a
“unirrep-csco” that also [although not noted in Ref. (2)] represents 12-
parameter SL(2,c)L × SL(2,c)R . GN’s “Lorentz unirrep” becomes a “Dirac-
momentum-basis” when two of its six csco members are “Fourier-replaced”
by CN-center generators. (The 4-member remainder of GN’s 6-member
unirrep-csco commutes with the foregoing pair of Dirac-momentum csco
members.)
Both the GN unirrep-csco and its “Dirac-momentum counterpart”
include the two Lorentz-group Casimirs appearing in the main-text para-
graph preceding Formula (18). These Casimirs commute with all 12 gen-
erators of the “left-right Lorentz group” and with all nine generators of
the CL group. The pair of self-adjoint operators appearing in the Dirac-
momentum csco (but not in GN’s unirrep csco) are the main-text energy
and chirality integer-eigenvalued operators. CL unirrep then follows from
6D unirrep-csco augmentation by electric charge.
The complete set of seven commuting self-adjoint “Dirac-momentum”
single-quc operators comprises Qq , Nq , Mq , mq , ρq , and z1q , where Qq and
Nq take (the by now familiar) possible values Qq = 0, ±1, ±2, ±3 and Nq =
0, ±1. The integer mq takes all positive–negative-integer values while ρq has
a continuous spectrum spanning the positive real line and Mq takes (main-
text, positive-integer) values 1, 2, . . . , Mmax . Evenness (oddness) of mq is
accompanied by evenness (oddness) of Nq . Earlier we have described ρq /2τ
as the “local-frame kinetic energy of Quc q”; in the present context, “local-
frame magnitude of Quc-q momentum” is a more appropriate appellation.
(Remember that “quc mass” is devoid of meaning!)
The continuous spectra of both Re z1q and Im z1q span (full) real lines;
we choose to describe the “meaning” of z1q as “local-frame-direction of Quc-
q 3-vector momentum”. Reminder of trickiness in terminology choice is the
Haar measure for the 3D (non-compact) ρq , z1q subspace: (m2q + ρ2q )dρq dz1q
[2]. [In terms of directional (real) polar angles, π ≥ θ1q ≥ 0 and 2π > ϕ1q ≥
www.ebook3000.com
0, the complex z1q may be written i tan(θ1q /2) exp(iϕ1q ). The absolute
value of z1q is then tan(θ1q /2).]
We perceive the set of four commuting self-adjoint operators, mq , ρq ,
and z1q , as “Dirac-momenta” even though not “canonically conjugate” to
the yq , zq quartet of Dirac coordinates — the complex commuting opera-
tors that, despite lack of “ordinary-language” names, have appeared promi-
nently in this chapter’s main text. We think of mq as “quc helicity” — the
“component of quc angular momentum in the direction of its momentum”.
The author is comfortable in calling the GN–Dirac coordinate zq “quc veloc-
ity direction” (related to velocity polar angles in the manner above used for
“momentum direction”) but has yet to achieve comfort with any (physics-
familiar) name for the GN–Dirac coordinate yq .
Appendix C. Photons of Differing Diameters Although

Same Momentum and Helicity
Main-text noted has been SMU’s (2-quc) 4D (3-momentum plus helicity)
“Dirac-momentum” basis for the “external” properties of a single photon.
But γc also has an “internal” Hilbert vector — a function of the 2-quc rela-
tive coordinate, whose spatial extension transverse to momentum direction
might be called “photon diameter”.
The internal γc Hilbert vector is a complex normalizable function of
location within a Dirac-relative-coordinate manifold. Among elementary
particles, photons are “special” by important dependence of their inter-
nal quc dynamics on gravitational attraction between qucs, as well as on
electromagnetic inter-quc attraction or repulsion.
The foregoing we have main-text summarized by attaching to the photon
the acronym, “double helix”. A physics-unappreciated photon attribute is
GUT-scale double-helix diameter.
With (see Table E.1) Qq = ±3 and Qq = −Qq , (local-frame) qq -
composed photon energy at age τ is Ey = (2τ )−1 (Mq + Mq ). This “exter-
nal” energy remains unchanged if Mq → Mq + σ, Mq → Mq − σ, with σ
an integer whose absolute value is smaller than either Mq or Mq . Gravi-
gr
tational potential energy — Vqq — by Formula (20), however, is changed,
with an associated change in double-helix diameter. (Plausibly the smallest

diameter associates to Mq = Mq .)
The total number of different double-helix quc pairs, with same momen-
tum and helicity (and same chirality) but differing helix diameters, is τ Eγ .
Even for “soft” photons in the present universe (those of wavelength ∼ km
612 G. F. Chew
or greater), the number of different SMU (γc ) photons sharing the same
momentum and helicity is of order 1022 . (A huge number have “almost the
same” diameter.)
It follows that, despite SMU finiteness of photon total number, high
accuracy may attach to physics coherent-state QFT representation of clas-
sical electromagnetic radiation — a τ → ∞ approximation that recognizes
indefinitely many FHPP-identical photons. Present-universe FHPP accu-
racy of Bose–Einstein identical-photon statistics is understandable even
though any γc is different from any other. Identity of all “photons with
common momentum and helicity” is one of many physics approximations
that accompany 3-space flattening.
Already at spatial micro-scales (far above GUT scale although far below
macro scale) physics notions such as Bose–Einstein and Fermi–Dirac statis-
tics become accurate. Lack of meaning for “quc statistics” accompanies
higher dimensionality of “quc space” compared to that of “particle space”.
Appendix D. Dark Matter as Non-Particulate (Bohm)

“Hidden Reality”
Via a self-adjoint Hamiltonian with kinetic- and potential-energy compo-
nents, the universe’s evolving “ray” specifies “evolving reality” through
expectations of self-adjoint operators that represent current densities of
energy, momentum, angular momentum, and electric charge. A “particle”
is a micro-scale clump of energy–momentum with some integral electric
charge, some baryon number, some integral or half-integral angular momen-
tum in units of and some approximately-determined mass. All qucs in
particle clumps are electrically charged.
Any “observer” is a macro-scale clump of particles with approximately-
zero total charge (charge screening). Distinction between “particle” and
“observer” resides not in the SMU Hamiltonian but in ray aspects that
emerge as distinct scales develop with universe expansion. “Dark matter”
resides in non-particulate galactic-scale clumps of chargeless but energy-
carrying qucs. Neither particle nor observer “contains” dark matter.
Schrödinger’s equation determines reality evolution without requiring
either that all energy density be “particulate” or that all reality be micro–
macro-scale. Galactic-scale “dark matter” is the non-particulate electrically-
neutral source of gravitational potential energy which helps determine, via
the SMU Schrödinger equation, the age at which a “radioactive” parti-
cle decays. Dark matter constitutes reality that Bohm characterized as
www.ebook3000.com
“hidden”. All SMU history is “deterministic” — with “observations” merely

one among many “onflow” (Ralph Pred’s term) aspects.
Appendix E. Three Elementary-Fermion Generations

The famously-mysterious three generations of QFT elementary fermions
associate to the three possible absolute values for non-vanishing quc elec-
tric charge. (All QFT elementary particles, whether charged or neutral,
are composed exclusively of charged qucs.) Generation mass magnitudes we
believe associate inversely to the quc-charge integers 1, 2, 3. Elementary-
fermion mass ratios are, to an accurate approximation, electrodynamically
determined. Approximate mass-ratio gravity-ignoring computation should
be possible.
Each of the three qucs (approximately) building any charged elemen-
tary fermion carries a charge integer, ±1, ±2, ±3. Two of the three indi-
vidually carry zero chirality while opposite charges — the net chirality and
charge of this “core” pair vanishing. The remaining “valence”-quc carries
the fermion’s chirality, charge and spin as well as baryon number. For a
charged lepton the charge, is –3g while the (fluctuating) chirality is ±/2.
For a quark the valence quc carries either the charge 2g or the charge — g,
together with baryon number 1/3 and ±/2 (fluctuating) chirality.
Three generations of charged elementary fermions associate to three pos-
sibilities for the electrically-neutral “core”-pair of individually-zero-chirality
charged qucs. We expect the lowest-mass generation to be that with the
Q = ±3 pair because here the negative electromagnetic potential energy
is greatest. The highest-mass generation we expect to be that with the
Q = ±1 quc pair.
Three “types” of neutrino, each a Q = ±1, ±2, or ±3, quc pair, differ
from the foregoing three charged “generations” by one member of any neu-
trino quc-pair having N = ±1, with a sign that in our galaxy agrees with
that of this quc’s electric charge; the other quc has zero chirality.
Table E.1
A pedagogically-helpful definition is possible, through quc electric charge,
of SMU “dark side” and “bright side”, separated by a “baryonic middle”.
Note that qucs composing photons, electrons, positrons and first-generation
neutrinos (Dirac’s concern) are exclusively “bright”.
614 G. F. Chew
Dark Baryonic Bright

Qq 0 ±1 ±2 ±3
Bq 0 ∓1/3 ±1/3 0
References
1. P.A.M. Dirac, Quantum Mechanics, Clarendon Press, Oxford, 1947.
2. M. Naimark, Linear Representations of the Lorentz Group, MacMillan, New
York, 1964.
3. E.A. Milne, Relativity, Gravitation and World Structure, Clarendon Press,
Oxford, 1935.
4. G.F. Chew, (2013), arXiv 1308.4366 and 1107.0492 (2011).
5. E. Schrödinger, Sitzungsb. d. Berlin Akad., in Proceedings of The Prussian
Academy of Sciences Physics-Mathematical Section, 1930. XIX, pp. 296–303.
www.ebook3000.com
Chapter 23
Quantized Fields à la Clifford

and Unification
Matej Pavšič
Jožef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia
matej.pavsic@ijs.si
It is shown that the generators of Clifford algebras behave as creation

and annihilation operators for fermions and bosons. They can create
extended objects, such as strings and branes, and can induce curved
metric of our space–time. At a fixed point, we consider the Clifford alge-
bra Cl(8) of the 8D phase space, and show that one quarter of the basis
elements of Cl(8) can represent all known particles of the first generation
of the Standard model, whereas the other three quarters are invisible to
us and can thus correspond to dark matter.
1. Introduction
Quantization of a classical theory is a procedure that appears somewhat
enigmatic. It is not a derivation in a mathematical sense. It is a recipe
of how to replace, e.g. the classical phase space variables, satisfying the
Poisson bracket relations, with the operators satisfying the corresponding
commutation relations [1]. What is a deeper meaning for replacement is usu-
ally not explained, only that it works. A quantized theory so obtained does
work and successfully describes the experimental observations of quantum
phenomena.
On the other hand, there exists a very useful tool for description of geom-
etry of a space of arbitrary dimension and signature [2–7]. This is Clifford
algebra. Its generators are the elements that satisfy the well-known rela-
tions, namely that the anticommutators of two generators are proportional
to the components of a symmetric metric tensor. The space spanned by
those generators is a vector space. It can correspond to a physical space,
615
616 M. Pavšič
for instance to our usual three-dimensional (3D) space, or to the 4D space–

time. The generators of a Clifford algebra are thus basis vectors of a physical
space. We will interpret this as a space of all possible positions that the
center of mass of a physical object can possess. A physical object has an
extension that can be described by an effective oriented area, volume, etc.
While the center of mass position is described by a vector, the oriented area
is described by a bivector, the oriented volume by a trivector, etc. In gen-
eral, an extended object is described [8–12] by a superposition of scalars,
vectors, bivectors, trivectors, etc., i.e. by an element of the Clifford algebra.
The Clifford algebra associated with an extended object is a space, called
Clifford space.a
Besides the Clifford algebras whose generators satisfy the anticommu-
tation relations, there are also the algebras whose generators satisfy com-
mutation relations, such that the commutators of two generators are equal
to the components of a metric, which is now antisymmetric. The Clifford
algebras with a symmetric metric are called orthogonal Clifford algebras,
whereas the Clifford algebras with an antisymmetric metric are called sym-
plectic Clifford algebras [13].
We will see that symplectic basis vectors are in fact quantum mechani-
cal operators of bosons [14, 15]. The Poisson brackets of two classical phase
space coordinates are equal to the commutators of two operators. This is so
because the Poisson bracket consists of the derivative and the symplectic
metric which is equal to the commutator of two symplectic basis vectors.
The derivative acting on phase space coordinates yields the Kronecker delta
and thus eliminates them from the expression. What remains is the com-
mutator of the basis vectors.
Similarly, the basis vectors of an orthogonal Clifford algebra are quan-
tum mechanical operators for fermions. This becomes evident in the new
basis, the so-called Witt basis. By using the latter basis vectors and their
products, one can construct spinors.
Orthogonal and symplectic Clifford algebras can be extended to infi-
nite dimensional spaces [14, 15]. The generators of those infinite dimen-
sional Clifford algebras are fermionic and bosonic field operators. In the
case of fermions, a possible vacuum state can be the product of an infinite
a Here, we did not go into the mathematical subtleties that become acute when the
Clifford space is not flat but curved. Then, strictly speaking, the Clifford space is a
manifold, such that the tangent space in any of its points is a Clifford algebra. If Clifford
space is flat, then it is isomorphic to a Clifford algebra.
www.ebook3000.com
sequence of the operators [14,15]. If we act on such a vacuum with an oper-

ator that does not belong to the set of operators forming that vacuum, we
obtain a “hole” in the vacuum. This hole behaves as a particle. The con-
cept of the Dirac sea, which is nowadays considered as obsolete, is revived
within the field theories based on Clifford algebras. But in the latter the-
ories we do not have only one vacuum, but many possible vacuums. This
brings new possibilities for further development of quantum field theories
and grand unification. Because the generators of Clifford algebras are basis
vectors on the one hand, and field operators on the other hand, this opens
a bridge towards quantum gravity. Namely, the expectation values of the
“flat space” operators with respect to suitable quantum states composed
of many fermions or bosons, can give “curved space” vectors, tangent to a
manifold with non-vanishing curvature. This observation paves the road to
quantum gravity.
2. Clifford Space as an Extension of Space–time

Let us consider a flat space M whose points are possible positions of the
center of mass P of a physical object O. If the object’s size is small in com-
parison to the distances to surrounding objects, then we can approximate
the object with a point particle. The squared distance between two possible
positions, with coordinates xµ and xµ + ∆xµ , is
∆s2 = ∆xµ gµν ∆xν . (1)
Here, index µ runs over dimensions of the space M , and gµν is the metric
tensor. For instance, in the case in which M is space–time, µ = 0, 1, 2, 3,
and gµν = ηµν = diag(1, −1, −1, −1) is the Minkowski metric. The object
O is then assumed to be extended in space–time, i.e. to have an extension
in a 3D space and in the direction x0 that we call “time”.
There are two possible ways of taking the square root of ∆s2 .
Case I.

∆s = ∆xµ gµν ∆xν . (2)
Case II.
∆x = ∆xµ γµ . (3)
In Case I, the square root is a scalar, i.e. the distance ∆s.

618 M. Pavšič
Fig. 1. The center of mass point P of an extended object O is described by a

vector xµ γµ .
In Case II, the square root is a vector ∆x, expanded in term of the basis
vectors γµ , satisfying the relations
1
γµ · γν ≡ (γµ γν + γν γν ) = gµν . (4)
2
If we write ∆x = ∆xµ γµ = (xµ − xµ0 )γµ and take xµ0 = 0, we obtain
x = xµ γµ , which is the position vector of the object’s O center of mass
point P (Fig. 1), with xµ being the coordinates of the point P .
In spite of being extended in space–time and having many (practically
infinitely many) degrees of freedom, we can describe our object O by only
four coordinates xµ , the components of a vector x = xµ γµ .
The γµ satisfying the anticommutation relations (4) are generators of
the Clifford algebra Cl(1, 3). A generic element of Cl(1, 3) is a superposition
1 µν 1 1
X = σ1+xµ γµ + x γµ ∧γν + xµνρ γµ ∧γν ∧γρ + xµνρσ γµ ∧γν ∧γρ ∧γσ ,
2! 3! 4!
(5)
where γµ ∧γν , γµ ∧γν ∧γρ and γµ ∧γν ∧γρ ∧γσ are the antisymmetrized prod-
ucts γµ γν , γµ γν γρ and γµ γν γρ γσ , respectively. They represent basis bivec-
tors, 3-vectors, and 4-vectors, respectively. The terms in Eq. (5) describe a
scalar, an oriented line, area, 3-volume, and 4-volume. The antisymmetrized
product of five gammas vanishes identically in four dimensions.
A question now arises as to whether the object X of Eq. (5) can describe
an extended object in space–time M4 . We have seen that x = xµ γµ describes
1 µν
the center of mass position. We anticipate that 2! x γµ ∧ γν describes an
oriented area associated with the extended object. Suppose that our object
O is a closed string. At first approximation is described just by its center of
mass coordinates (Fig. 2a). At a better approximation it is described by the
www.ebook3000.com
x3
13 X 23
M4 µ X
X
x2
µ
X (ξ) x1
X 12
(a) (b)
Fig. 2. With a closed string one can associate the center of mass coordinates
(a), and the area coordinates (b).
M4 Xµ
X 123
µ
X (ξ)
Fig. 3. Looking with a sufficient resolution one can detect eventual presence of
volume degrees of freedom.
quantities xµν , which are the projections of the oriented area, enclosed by
the string, onto the coordinate planes (Fig. 2b). If we probe the string at a
better resolution, we might find that it is not exactly a string, but a closed
membrane (Fig. 3). The oriented volume, enclosed by this 2D membrane
is described by the quantities X µνρ . At even better resolution we could
eventually see that our object O is in fact a closed 3D membrane, enclosing
a 4-volume, described by xµνρσ . Our object O has finite extension in the
4D space–time. It is like an instanton.
Let us now introduce a more compact notation by writing
4

X= xµ1 µ2 ...µr γµ1 µ2 ...µr ≡ xM γM , (6)
r=0
where γµ1 µ2 ...µr ≡ γµ1 ∧γµ2 ∧· · ·∧γµr , and where we now assume µ1 < µ2 <
· · · < µr , so that we do not need a factor 1/r!. Here, xM are interpreted
as quantities that describe an extended instantonic object in M4 . On the
other hand, xM are coordinates of a point in the 16-dimensional (16D)
space, called Clifford space C. In other words, from the point of view of C,
xM describe a point in C.
620 M. Pavšič
The coordinates xM of Clifford space can describe not only closed, but
also open branes. For instance, a vector xµ γµ can denote position of a point
event with respect to the origin (Fig. 1), or it can describe a string-like
extended object (an instantonic string in space–time). Similarly, a bivector
xµν γµ ∧ γν can describe a closed string (2a), or it can describe an open
membrane. Whether the coordinates xM ≡ xµ1 µ2 ...µr describe a closed r-
brane or an open (r + 1)-brane is determined by the value of the scalar and
pseudoscalar coordinates, i.e. by σ and σ̃ (for more details see Ref. [16]).
A continuous 1D set of points in C is a curve, a worldline, described by
the mapping
xM = X M (τ ), (7)
where τ is a monotonically increasing parameter and X M embedding func-

tions of the worldline in C. We assume that it satisfies the action principle

1
I[X M ] = M dτ (GMN Ẋ M Ẋ N ) 2 , (8)
where GMN is the metric in C and M a constant, analogous to mass. From

the point of view of space–time, the functions X M (τ ) ≡ X µ1 µ2 ...µr (τ ),
r = 0, 1, 2, 3, 4, describe evolution of an extended instantonic object in
space–time. Some examples are in Fig. 4 (see also Ref. [10]).
Fig. 4. Evolution of an instantonic cigar-like (a) and a ring-like (b) extended

object in space–time. At different values of τ , (e.g. at τ = τ1 , τ2 , τ3 ), we have dif-
ferent extended instantonic objects that correspond to different 4D slices through
Clifford space.
www.ebook3000.com
In this setup, there is no “block universe” in space–time. There do not

exist infinitely long worldlines or worldtubes in space–time. Infinitely long
worldlines exist in C-space, and in this sense a block universe exists in
C-space.
The action (8) is invariant under reparametrizations of τ . A conse-
M
the constraint among the canonical momenta PM = ∂L/∂ Ẋ =
quence is
MẊM / gJK Ẋ J Ẋ K :
PM P M − M2 = 0. (9)
The metric of Clifford space is given by the scalar product of two basis
vectors,
‡ ‡
ηMN = γM ∗ γ = γM γN 0 , (10)
where “‡” is the operation that reverses the order of vectors in the product
‡
γM = γµ1 γµ2 . . . γµr , so that γM = γµr . . . γµ2 γµ1 . The superscript “0”
denotes the scalar part of an expression. For instance,
γµ γν 0 = ηµν , γµ γν γα 0 = 0, γµ γν γα γβ 0 = ηµβ ηνα − ηµα ηνβ . (11)
So we obtain
ηMN = diag(1, 1, 1, 1, 1, 1, 1, 1, −1, −1, −1, −1, −1, −1, −1, −1) (12)
which means that the signature of C-space is (+ + + + + + + + − − − −

− − −−), or shortly, (8, 8).
The quadratic form reads
X ‡ ∗ X = ηMN xM xN
= σ 2 + ηµν xµ xν + (ηµβ ηνα − ηµα ηνβ )xµα xνβ + ηµν x̃µ x̃ν − σ̃ 2
= ηµ̂ν̂ xµ̂ xν̂ + σ 2 − σ̃ 2 , (13)
1 µ
where xµ̂ = (xµ , xµν , x̃µ ), with x̃µ ≡ 3! νρσ xνρσ being the pseudoscalar
1
coordinates, whereas σ is the scalar and σ̃ ≡ 4! µνρσ xµνρσ the pseudoscalar
coordinate in C-space.
Upon quantization, PM becomes operator PM = −i∂/∂xM , and the
constraint (9) becomes the Klein–Gordon equation in C-space:
(∂M ∂ M + M2 )Ψ(xM ) = 0. (14)

622 M. Pavšič
In the new coordinates,

1 1
s= (σ + σ̃) λ= (σ − σ̃), (15)
2 2
in which the quadratic form is
X ‡ ∗ X = ηµ̂ν̂ xµ̂ xν̂ − 2sλ (16)
the Klein–Gordon equation reads
η µ̂µ̂ ∂µ̂ ∂ν̂ φ − 2∂s ∂λ φ = 0. (17)
If we take the ansatz
φ(xµ̂ , s, λ) = eiΛλ ψ(s, xµ̂ ), (18)
then Eq. (17) becomes [17]
η µ̂µ̂ ∂µ̂ ∂ν̂ φ − 2iΛ∂s φ = 0, (19)
i.e.
∂ψ 1 µ̂µ̂
i = η ∂µ̂ ∂ν̂ ψ. (20)
∂s 2Λ
This is the generalized Stueckelberg equation. It is like the Schrödinger equa-
tion, but it describes the evolution of the wave function ψ(s, xµ̂ ) in the 14D
space whose points are described by coordinates xµ̂ . The evolution param-
eter is s.
A remarkable feature of this setup is that the evolution parameter has
a clear physical meaning: it is given in terms of the scalar, σ, and the
pseudoscalar, σ̃, coordinate according to Eq. (15). The latter quantities, as
shown before, are given by a configuration of the object, sampled in terms
of the coordinates X M of the Clifford space C.
The wave function ψ(s, xµ̂ ) is the probability amplitude that at a given
value of the evolution parameter s we will find an instantonic extended
object with coordinates xµ̂ .
This is illustrated in Fig. 5. In principle, all points of C-space are pos-
sible in the sense that we can find there an instantonic extended object.
A wave packet determines a subset of point of C that are more probable
to “host” the occurrence of an instantonic object (an event in C). The
wave function determines the probability amplitude over the points of C.
Its square determines the probability density. From the point of view of
space–time, wave function determines which instantonic extended objects
are more likely to occur. It determines the probability amplitude, and its
www.ebook3000.com
Fig. 5. Extended instantonic object in space–time (a) is represented by a

point in C-space (b). Quantum mechanically, the extended object is blurred
(c). In C-space, we have a blurred point, i.e. a “cloud” of points occurring with
probability density |ψ(s, xµ̂ )|2 .
square the probability density of occurrence of a given instantonic extended

object. The probability amplitude ψ is different at different values of the
evolution parameter s. In other words, ψ changes (evolves) with s.
Instead of one extended object, described by xM , we can consider several
or many extended objects, described by xiM , i = 1, 2, . . . , n. They form
an instantonic configuration {Oi } = {Oi }, i = 1, 2, . . . , n. The space of all
possible instantonic configurations will be called configuration space C. The
infinitesimal distance between two configurations, i.e. between two points
in C, is
dS 2 = η(iM)(jN ) dxiM dxjN , (21)

624 M. Pavšič
where η(iM)(jN ) = δij ηMN is the metric of a flat configuration space.

We will assume that the Klein–Gordon equation (14) can be general-
ized so to hold for the wave function ψ(xiM ) in the space of instantonic
configurations {Oi }:
∂
(η (iM)(jN ) ∂iM ∂jN + K2 )φ(xiM ) = 0, ∂iM ≡ . (22)
∂X iM
Let us choose a particular extended object, O1 , with coordinates
1M
x ≡ xM = (σ, xµ , xµν , x̃µ , σ̃). The coordinates of the remaining extended
objects within the configuration are x2M , x3M , . . . . Let us denote them
xīM , ī = 2, 3, . . . , N . Following the same procedure as in Eqs. (15)–(20),
we define s and λ according to (15) to the first object. We have thus split
the coordinates xiM of the configuration according to
xiM = (s, λ, xµ̂ , xīM ) = (s, λ, xM̄ ), (23)
where xM̄ = (xµ̂ , xīM ). By taking the ansatz
φ(s, λ, xM̄ ) = eiΛλ ψ(s, xM̄ ), (24)

Eq. (22) becomes
η M̄ N̄ ∂M̄ ∂N̄ ψ − 2iΛ∂s ψ = 0, (25)

i.e.
∂ 1 M̄ N̄
i ψ= η ∂M̄ ∂N̄ ψ. (26)
∂s 2Λ
Equation (26) describes evolution of a configuration composed of a system
of instantonic extended objects.
The evolution parameter s is given by the configuration itself (in the
above example by one of its parts), and it distinguishes one instantonic con-
figuration from another instantonic configuration. So we have a continuous
family of instantonic configurations, evolving with s. Here, “instantonic
configuration” or “instantonic extended object” is a generalization of the
concept of “event”, associated with a point in space–time. An event, by
definition is “instantonic” as well, because it occurs at one particular point
in space–time.
A configuration can be very complicated and self-referential, and thus
being a record of the configurations at earlier values of s. In this respect,
this approach resembles that by Barbour [18], who considered “time cap-
sules” with memory of the past. As a model, he considered a triangleland,
whose configurations are triangles. Instead of triangleland, we consider
www.ebook3000.com
here the Clifford space, in which configurations are modeled by oriented

r-volumes (r = 0, 1, 2, 3, 4) in space–time. In this respect our model dif-
fers from Barbour’s model, in which the triangles are in 3D space. Instead
of 3D space, I consider a 4D space with signature (+ − −−). During the
development of physics it was recognized that a 3D space is not suitable
for formulation of the theory describing the physical phenomena, such as
electromagnetism and moving objects. In other words, the theory of rela-
tivity requires 4D space, with an extra dimension x0 , whose signature is
opposite to the signature of three spatial dimensions. The fourth dimension
was identified with time, x0 ≡ t. Such identification, though historically
very useful, has turned out to be misleading [9, 19–21]. In fact, x0 is not
the true time, it is just a coordinate of the fourth dimension. The evolu-
tion time is something else. In the Stueckelberg theory [9, 19–21], its origin
remains unexplained. In the approach with Clifford space, the evolution
time (evolution parameter) is s = (σ + σ̃)/2, i.e. a superposition of the
scalar coordinate, σ, and the pseudoscalar coordinate, σ̃. This is the param-
eter that distinguishes configurations within a one-dimensional family. In
principle, the configurations can be very complicated and self-referential,
including conscious experiences of an observer. Thus, s distinguishes differ-
ent conscious experiences of an observer [9, 20]; it is the time experienced
by a conscious observer. A wave function φ(xiM ) = eiΛ ψ(s, xM̄ ) “selects”
in the vast space C of all possible configurations xiM a subspace S ∈ C
of configurations. More precisely, φ assigns a probability density over the
points of C, so that some points are more likely to be experienced by an
observer than the other points. In particular, φ(xiM ) can be a localized wave
packet evolving along s. For instance, such a wave packet can be localized
around a worldline xiM = X0iM (s) in C, which from the point of view of
M4 , is a succession (evolution) of configurations X0M̄ at different values of
the parameter s. If configurations are complicated and include the exter-
nal world and an observer’s brain, such wave packet ψ(s, xM̄ ) determines
the evolution of conscious experiences of an observer coupled by his sense
organs to the external world.
The distinction between the evolution time τ and the coordinates x0
in the wave function ψ(s, xM̄ ), can help in clarifying the well-known Libet
experiment [22]. The latter experiment seemingly demonstrates that we
have no “free will”, because shortly before we are conscious of a decision,
our brain already made the decision. This experiment is not in conflict
with free will, if besides the theory described above, we as well invoke the
Everett many-worlds interpretation of quantum mechanics [23], and the
626 M. Pavšič
considerations exposed in Ref. [9]. Further elaboration of this important

implication of the Stueckelberg and Everett theory is beyond the scope of
this chapter. But anyone with a background in those theories can do it after
some thinking. An interested reader can do it as an exercise.
3. Generators of Clifford Algebras as Quantum

Mechanical Operators
3.1. Orthogonal and symplectic Clifford algebras
After having exposed a broader context of the role of Clifford algebras in
physics, let me now turn to a specific case and consider the role of Clifford
algebras in quantization. The inner product of generators of Clifford algebra
gives the metric. We distinguish two cases:
(i) If metric is symmetric, then the inner product is given by the anti-
commutator of generators; this is the case of an orthogonal Clifford algebra:
1
{γa , γb } ≡ γa · γb = gab . (27)
2
(ii) If metric is antisymmetric, then the inner product is given by the
commutator of generators; this is the case of a symplectic Clifford algebra:
1
[qa , qb ] ≡ qa ∧ qb = Jab . (28)
2
Here, qa are the symplectic basis vectors that span a symplectic space,
whose points are associated with symplectic vectors [14]
z = z a qa . (29)
Here, z a are commuting phase space coordinates,
z a z b − z b z a = 0. (30)
An example of symplectic space in physics is phase space, whose points are
coordinates and momenta of a particle:
z a = (xµ , pµ ) ≡ (xµ , x̄µ ) ≡ (xµ , xµ̄ ). (31)
The corresponding basis vectors then split according to
qa = (qµ(x) , qµ(p) ) ≡ (qµ , q̄µ ) ≡ (qµ , qµ̄ ), µ = 1, 2, . . . , n (32)
and the relation (28) becomes
1 (x) (p) 1
q , qν ≡ [qµ , qν̄ ] = Jµν̄ = gµν ,
2 µ 2

1 (x) (x) 1 (p) (p)
qµ , qν = 0, q , qν = 0, (33)
2 2 µ
www.ebook3000.com
where we have set

0 gµν
Jab = . (34)
−gµν 0
Here, depending on the case considered, gµν is the euclidean, gµν = δµν ,
µ, ν = 1, 2, . . . , n, or the Minkowski metric, gµν = ηµν . In the latter case
we have µ, ν = 0, 1, 2, . . . , n − 1.
We see that (33) are just the Heisenberg commutation relations for
coordinate and momentum operators, identified asb
1 i
x̂µ = √ qµ(x) ; p̂µ = √ qµ(p) . (35)
2 2
Then we have
[x̂µ , p̂ν ] = igµν , [x̂µ , x̂ν ] = 0, [p̂µ , p̂ν ] = 0. (36)
Instead of a symplectic vector z = z a qa , let us now consider another
symplectic vector, namely
∂f a
q ,
F = (37)
∂z a
where f = f (z) is a function of position in phase space. The wedge product
of two such vectors is
∂f ∂g ∂f ab ∂g
F ∧ G = a qa ∧ qb b = J , (38)
∂z ∂z ∂z a ∂z b
where in the last step we used the analog of Eq. (28) for the reciprocal
quantities q a = J ab qb , where J ab is the inverse of Jab .
Equation (38) is equal to the Poisson bracket of two phase space func-
tions. Namely, using (31) and (34), we have
∂f ab ∂g ∂f µν ∂g ∂f ∂g
J = η − µ η µν ν ≡ {f, g}P B . (39)
∂z a ∂z b ∂xµ ∂pν ∂p ∂x
In particular, if
f = z c, g = zd, (40)
Eqs. (38) and (39) give
q a ∧ q b = J ab = {z a , z b }P B . (41)
We see that the Heisenberg commutation relations for operators x̂µ , p̂µ
are obtained automatically, if we express the Poisson bracket relations in
b We insert factor i in order to make the operator p̂µ Hermitian.

628 M. Pavšič
terms of the wedge product of the symplectic vectors

∂f a ∂f µ ∂f µ ∂g ∂g µ ∂f µ
F = a
q = µ
q(x) + µ q(p) and G = a
= µ
q(x) + µ q(p) .
∂z ∂x ∂p ∂z ∂x ∂p
(42)
By having taken into account not only the coordinates and functions in a
symplectic space, but also corresponding basis vectors, we have found that
basis vectors are in fact quantum mechanical operators [14]. Moreover, the
Poisson bracket between classical phase space variable, {z a , z b }P B , is equal
to the commutator, 12 [q a , q b ] = q a ∧ q b , of vectors (i.e. of operators) q a and
q b [14]. According to this picture, quantum operators are already present
in the classical symplectic form, if we write the symplectic metric as the
inner product of symplectic basis vectors. The latter vectors are just the
quantum mechanical operators.
Analogous procedure can be performed with orthogonal Clifford alge-
bras. Then a point in phase space can be described as a vector
λ = λa γa , (43)
where λa are anticommuting phase space coordinates,
λa λb + λb λa = 0 (44)
and γa basis vectors, satisfying Eq. (27). If we split the vectors γa and the
metric γab according to
γa = (γµ , γ̄µ ), µ = 0, 1, 2, . . . , n − 1, (45)

gµν 0
gab = (46)
0 gµν
and introduce a new basis, the so-called Witt basis,
1
θµ = √ (γ µ + iγ̄µ ),
2
1 µ
θ̄µ = √ (γ − iγ̄µ ), (47)
2
then the Clifford algebra relations (27) become
1
θµ · θ̄ν ≡ (θµ θ̄ν + θ̄ν θµ ) = ηµν ,
2
θµ · θν = 0, θ̄µ · θ̄ν = 0. (48)
These are the anticommutation relations for fermionic creation and annihi-
lation operators.
www.ebook3000.com
Let us now introduce functions f˜(λ) and g̃(λ), and consider the vectors
∂ f˜ a ∂g̃ a
F̃ = γ , g̃ = γ . (49)
∂λa ∂λa
The dot product of those vectors is
∂ f˜ a ∂g̃ b ∂ f˜ ab ∂g̃
F̃ · G̃ = γ · γ = g = {f˜, g̃}P B , (50)
∂λa ∂λb ∂λa ∂λb
where g ab = γ a · γ b is the inverse of gab .
Equation (50) shows that the dot product, which in the orthogonal case
corresponds to the inner product, is equal to the Poisson bracket of two
phase space functions, now composed with the symmetric metric g ab .
If
f˜ = λc , g̃ = λd , (51)
Eq. (50) gives
F̃ · G̃ = γ c · γ d = g cd (52)
which in the Witt basis reads as the fermionic anticommutation relations

(48). This means that the Poisson bracket between the (classical) phase
space variables λa , λb is equal to the anticommutator of the “operators” γ a
and γ b :
1 a b
{λa , λb }P B = {γ , γ } = g ab . (53)
2
Again, we have that the basis vectors behave as quantum mechanical oper-
ators.
3.2. Equations of motion for a particle’s coordinates

and the corresponding basis vectors
We will now consider [14] a point particle, described by the phase space
action

1
I= dτ (ẋa Jab z b + z a Kab z b ), (54)
2
where
1 a
z Kab z b = H (55)
2
is the Hamiltonian, the quantity Kab being a symmetric 2n × 2n matrix.
630 M. Pavšič
Variation of the action (54) with respect to z a gives

∂H
ż a = J ab (56)
∂z b
which are the Hamilton equations of motion.
A solution of Eq. (56) is a trajectory z in phase space. We can consider a
trajectory as an infinite dimensional vector with components z a (τ ) ≡ z a(τ ).
Here a(τ ) is the index that denotes components; it is a double index, with
a being a discrete index, and (τ ) a continuous one. Corresponding basis
vectors are qa (τ ) ≡ qa(τ ) , and they satisfy the relations
qa(τ ) ∧ qb(τ ) = Ja(τ )b(τ ) = Jab δ(τ − τ ) (57)
which are an extension of the relations (28) to our infinite dimensional case.
A trajectory is thus described by the vector

a(τ )
z=z qa(τ ) ≡ dτ z a (τ )qa (τ ). (58)
The phase space velocity vector is

v = ż a(τ ) qa(τ ) = −z a(τ )q̇a(τ ) , (59)
where we have assumed that the “surface” term vanishes:
τ2

v = dτ ż (τ )qa (τ ) = − dτ z a (τ )q̇a (τ ) + z a (τ )qa (τ ) .
a
(60)
τ1
The last term vanishes if z a (τ2 )qa (τ2 ) = z a (τ1 )qa (τ1 ).
The action (54) can be written as
1 a(τ )
I= (ż Ja(τ )b(τ ) z b(τ ) + z a(τ )Ka(τ )b(τ ) z b(τ ) ), (61)
2
where Ja(τ )b(τ ) is given in Eq. (57), and
Ka(τ )b(τ ) = Kab δ(τ − τ ). (62)
The corresponding equations of motion are

ż a(τ ) = J a(τ )c(τ ) Kc(τ )b(τ ) z b(τ ) . (63)
Multiplying both sides of the latter equation by qa(τ ) , we obtain

ż a(τ ) qa(τ ) = −q a(τ ) Ka(τ )b(τ ) z b(τ ) . (64)

We have raised the index by J a(τ )c(τ ) and taken into account that

J a(τ )c(τ ) = −J c(τ )a(τ ) . Equation (64) is just Eq. (63), expressed in terms
of the basis vectors. Both equations are equivalent.
www.ebook3000.com
Fig. 6. If the operator equations of motion (65) hold for any path z a (τ ) this
means that coordinates and momenta are undetermined.
Using the relation (59) in Eq. (64), we obtain

z b(τ ) q̇b(τ ) = q a(τ ) Ka(τ )b(τ ) z b(τ ) . (65)
Apart from the surface term that we have neglected in Eq. (60), the last
equation, (65), is equivalent to the classical equation of motion (56), only
the τ -dependence has been switched from the components to the basis
vectors.
A curious thing happens if we assume that Eq. (65) holds for an arbi-
trary trajectory (Fig. 6).
Then, instead of (65), we can write
q̇b(τ ) = q a(τ ) Ka(τ )b(τ ) . (66)
Inserting into the latter equation the explicit expression (62) for Ka(τ )b(τ )
and writing q̇b(τ ) = q̇b (τ ), q a(τ ) = q a (τ ), we obtain
q̇a (τ ) = Kab q b (τ ). (67)
This can be written as
q̇a = [qa , Ĥ], (68)
where
1 a
q Kab q b
Ĥ = (69)
2
is the Hamilton operator, satisfying
[qa , Ĥ] = Kab q b . (70)
Starting from the classical action (54), we have arrived at the Heisenberg
equations of motion (68) for the basis vectors qa . On the way, we have made
632 M. Pavšič
a crucial assumption that the particle does not follow a trajectory z a (τ )

determined by the classical equations of motion, but that it can follow any
trajectory. By the latter assumption, we have passed from the classical to
the quantized theory. We have thus found yet another way of performing
quantization of a classical theory. Our assumption that a trajectory (a path)
can be arbitrary, corresponds to that by Feynman path integrals. In our
procedure we have shown how such an assumption of arbitrary path leads
to the Heisenberg equations of motion for operators.
3.3. Supersymmetrization of the action

The action (54) can be generalized [14] so as to contain not only a sym-
plectic, but also an orthogonal part. For this purpose, we introduce the
generalized vector space whose elements are
z = z A qA , (71)
where
z A = (z a , λa ), z a = (xµ , x̄µ ), λa = (λµ , λ̄µ ), (72)

symplectic part orthogonal part
are coordinates, and
qA = (qa , γa ), qa = (qµ , q̄µ ), γa = (γµ , γ̄µ ), (73)

symplectic part orthogonal part
are basis vectors. The metric is

Jab 0
qA qB 0 = GAB = , (74)
0 gab
where Jab = −Jba and gab = gba .

Let us consider a particle moving in such space. Its worldline is
z A = Z A (τ ). (75)
An example of a possible action is

1
I= dτ Ż A GAB Z B + interaction terms. (76)
2
www.ebook3000.com
Using (72)–(74), the latter action can be split as

1
I= dτ (ż a Jab z b + λ̇a gab λ) + interaction terms
2

1 µ ˙
= dτ (ẋµ ηµν x̄ν − x̄˙ ηµν xν + λ̇µ ηµν λν + λ̄η ν
µν λ̄ )
2
+ interaction terms. (77)
Here, z a are commuting, and λa anticommuting (Grassmann) coordinates.
The canonical momenta are
∂L 1 ∂L 1
p(x)
µ = µ
= ηµν ẋν , p(x̄)
µ = µ = − ηµν x
ν
∂ ẋ 2 ∂ x̄˙ 2
∂L 1 ∂L 1
p(λ)
µ = = ηµν λν , p(µλ̄) = ν
µ = ηµν λ̄ . (78)
∂ λ̇µ 2 ˙
∂ λ̄ 2
Instead of the coordinates λa = (λµ , λ̄µ ), we can introduce the new
coordinates
1
λa = (λµ , λ̄µ ), λµ ≡ ξ µ = √ (λµ − iλ̄µ ),
2
1
λ̄µ ≡ ξ¯µ = √ (λµ + iλ̄µ ), (79)
2
in which the metric is

0 ηµν
gab = γa · γb = . (80)
ηµν 0
In the new coordinates, we have
λ̇a gab λb = λ̇a gab

λb = ξ˙µ ηµν ξ̄ ν + ξ̄˙µ ηµν ξ ν . (81)
Now the pairs of canonically conjugate variables are (ξ µ , 12 ξ̄µ ) and (ξ¯µ , 12 ξµ ),
whereas in the old coordinates the pairs were (λµ , 12 λµ ) and (λ̄µ , 12 λ̄µ ),
which was somewhat unfortunate, because the variables in the pair were
essentially the same.
The interaction term can be included by replacing the τ -derivative in
the action (76) with the covariant derivative:
Ż A → Ż A + AA C Z C . (82)
So we obtain [14]

1
I= dτ (Ż A + AA C Z C )GAB Z B . (83)
2
This is a generalized Bars action [24], invariant under τ -dependent (local)
rotations of Z A . As discussed in Ref. [14], the gauge fields AA C (τ ) are
634 M. Pavšič
not dynamical; they have the role of Lagrange multipliers, whose choice
determines a gauge, related to the way of how the canonically conjugated
variables can be locally rotated into each other.
For a particular choice of AA C , we obtain
AA C Z C GAB Z B = α pµ pµ + β λµ pµ + γ λ̄µ pµ . (84)
Here, α, β, γ are Lagrange multipliers contained in AA C . Other choices of
AA C are possible, and they give expressions that are different from (84).
A nice theory of how its works in the bosonic subspace, was elaborated by
Bars (see, e.g. Ref. [24]).
The action (83), for the case (84), gives the constraints
pµ pµ = 0, λµ pµ = 0, λ̄µ pµ = 0 (85)
or equivalently
pµ pµ = 0, ξ µ pµ = 0, ξ¯µ pµ = 0, (86)
if we use coordinates ξ a = (ξ µ , ξ),
¯ defined in Eq. (79).
Upon quantization, we have
p̂µ p̂µ Ψ = 0, λ̂µ p̂µ Ψ = 0, ˆ µ p̂ Ψ = 0

λ̄ (87)
µ
or equivalently
µ
p̂µ p̂µ Ψ = 0, ξ̂ µ p̂µ Ψ = 0, ξˆ¯ p̂µ Ψ = 0, (88)
where the quantities with hat are operators, satisfying
[x̂µ , p̂ν ] = iη µν , [x̂µ , x̂ν ] = 0, [p̂µ , p̂ν ] = 0, (89)
µ ν
{λ̂µ , λ̂ν } = 2iη µν , ˆ , λ̄
{λ̄ ˆ } = 2iη µν , ˆ ν } = 0,
{λ̂µ , λ̄ (90)
ν µ ν
{ξˆµ , ξˆ
¯ } = η µν , {ξˆµ , ξ̂ ν } = 0, {ξˆ¯ , ξ¯ˆ } = 0, (91)
The operators can be represented as
∂ µ ∂
x̂µ → xµ , p̂µ → −i , ξˆµ → ξ µ , ξˆ¯ → µ , (92)
∂xµ ∂ξ
where
xµ xν − xν xµ = 0, ξ µ ξ ν + ξ ν ξ µ = 0. (93)
A state Ψ can be represented as a wave function ψ(xµ , ξ µ ) of commuting
coordinates xµ and anticommuting (Grassmann) coordinates ξ µ .
www.ebook3000.com
ˆµ
In Eq. (87), we have two copies of the Dirac equation, where λ̂µ and λ̄
satisfy the Clifford algebra anticommutation relations (90), and are related
to γ µ , γ̂ µ according to
λ̂µ = γ µ , ˆ µ = iγ̄ µ .
λ̄ (94)
Using (92), we find that the quantities γµ , γ̄µ , satisfying
γµ · γν = ηµν , γ̄µ · γ̄ν = ηµν (95)
can be represented according to

1 ∂ 1 ∂
γµ = √ ξµ + µ , γ̄µ = √ ξµ − µ . (96)
2 ∂ξ 2 ∂ξ
If we expand ψ(xµ , ξ µ ) in terms of the Grassmann variables ξ µ , we

obtain a finite number (i.e. 2n ) of terms:
n

ψ(xµ , ξ µ ) = ψµ1 µ2 ...µr ξ µ1 ξ µ2 . . . ξ µr . (97)
r=0
In the case of 4D space–time, n = 4, the wave function has 24 = 16

components. The state Ψ can then be represented as a column ψ α (x),
α = 1, 2, . . . , 16, and the operators γ µ , γ̄ µ as 16 × 16 matrices. Because
we have built our theory over the 8D phase space, our spinor has not only
four, but 16 components. This gives a lot of room for unified theories of
particles and fields [25–29].
4. Basis Vectors, Clifford Algebras, Spinors,

and Quantized Fields
4.1. Spinors as particular Clifford numbers
We have seen that the generators of Clifford algebras have the properties of
quantum mechanical operators. Depending on the kind of Clifford algebra,
they satisfy the commutation or anticommutation relations for bosonic or
fermionic creation and annihilation operators.
From the operators θµ and θ̄µ , defined in Eq. (45), we can build up
spinors by taking a “vacuum”

Ω= θ̄µ , which satisfies θ̄µ Ω = 0 (98)

µ
636 M. Pavšič
and acting on it by “creation” operators θµ . So we obtain a “Fock space”

basis for spinors [30–34]
sα = (1Ω, θµ Ω, θµ θν Ω, θµ θν θρ Ω, θµ θν θρ θσ Ω), (99)
in terms of which any state can be expanded as

ΨΩ = ψ α sα , α = 1, 2, . . . , 2n . (100)
Components ψ α can be space–time dependent fields. With the operators

θµ , θ̄µ we can construct spinors as the elements of a minimal left ideal of a
Clifford algebra Cl(2n). We will take the dimension of space–time n = 4,
so that our phase space will have dimension 8, and the Clifford algebra,
built over it, will be Cl(2, 6) which will simply denote Cl(8) or, in general,
Cl(2n).
Besides (98), there are other possible vacuums, e.g.

Ω= θµ , θµ Ω = 0, (101)
µ
  

Ω= θµ   θ̄µ , θµ Ω = 0, if µ ∈ R2
µ∈R1 µ∈R2
θ̄µ Ω = 0, if µ ∈ R2 (102)
where
R1 = {µ1 , µ2 , . . . , µr }, R2 = {µr+1 , µr+2 , . . . , µn }. (103)
There are 2n vacuums of such a kind. By taking all those vacuums, we

obtain the Fock space basis for the whole Cl(2n). If n = 4, the latter
algebra consists of 16 independent minimal left ideals, each belonging to a
different vacuum (102) and containing 16-component spinors (2n = 16 if
n = 4), such as (100). A generic element of Cl(8) is the sum of the spinors
ΨΩi , i = 1, 2, 3, 4, . . . , 16, belonging to the ideal associated with a vacuum
ΨΩi :

Ψ= ΨΩi = ψ αi sαi ≡ ψ Ã sÃ , Ã = 1, 2, 3, 4, . . . , 256, (104)
i
where sÃ ≡ sαi , α, i = 1, 2, . . . , 16, is the Fock space basis for Cl(8), and
ψ Ã ≡ ψ αi are space–time dependent fields. The same element Ψ ∈ Cl(8)
www.ebook3000.com
can be as well expanded in terms of the multivector basis,
Ψ = ψ A γA , A = 1, 2, 3, 4, . . . , 256, (105)
where
γA = 1, γa1 , γa1 ∧ γa2 , . . . , γa1 ∧ γa2 ∧ · · · ∧ γa2n (106)
which can be written compactly as
γA = γa1 ∧ γa2 ∧ · · · ∧ γar , r = 0, 1, 2, . . . , 2n. (107)
We see that if we construct the Clifford algebra of the eight-dimensional

phase space, then we have much more room for unificationc of elementary
particles and fields than in the case of Cl(1, 3), constructed over 4D space–
time. We have a state Ψ that can be represented by a 16 × 16 matrix,
whose elements can represent all known particles of the 1st generation of
the Standard model. Thus, 64 elements of this 16 × 16 matrix include the
left- and right-handed (L, R) versions of the states (e, νe ), (u, d)r , (u, d)b ,
(u, d)g , and their antiparticles, times factor two, because all those states,
satisfying the generalized Dirac equation [27, 28] (see also Sec. 5.2), can in
principle be superposed with complex amplitudes.
If we take space inversion (P) of those 64 states by using the same pro-
cedure as in Ref. [32], we obtain another 64 states of the 16× 16 matrix rep-
resenting Ψ, namely the states of mirror particles (P-particles). Under time
reversal (T) (see Ref. [32]), we obtain yet another 64 states corresponding
to time reversed particles (T-particles). And finally, under PT, we obtain
64 states of time reversed mirror particles (PT-particles). Altogether, we
have 4 × 64 = 256 states:
 
e ν e ν e ν e ν
 ē ν̄ ē ν̄ P ē ν̄ T ē ν̄ PT 
 
 u 
 d u d u d u d 
 
 ū d¯ r ū d¯ r,P ū d¯ r,T ū d¯ r,PT 
sαi =
 u
,
 (108)
 d u d u d u d 
 ū d¯ g d¯ g,P d¯ g,T d¯ g,PT 

ū

ū

ū

 
 u d u d u d u d 
ū d¯ b
ū d¯ b,P
ū d¯ b,T
ū d¯b,PT
c Unification based on Clifford algebras in phase space was considered by Castro [26].
638 M. Pavšič
where
 
eL ieL νL iνL

e ν  eR ieR νR iνR 
≡
 ēL
 (109)
ē ν̄ iēL ν̄L iν̄L 
ēR iēR ν̄R iν̄R
and similarly for u, d.

Those states interact with the corresponding gauge fields,d which
include the gauge fields of the Standard model, such as the photon, weak
bosons, and gluons. There exist also mirror versions, as well as T and PT
versions of the standard gauge bosons.e
Of the 256 particle states in Eq. (108), only 1/4 interact with our usual
photons, whereas the remaining 3/4 do not interact with our photons, but
they may interact with mirror photons, T-photons, or PT-photons. This
scheme thus predicts the existence of dark matter. If the matter in the
universe were evenly distributed over the ordinary particles, P-particles,
T-particles, and PT-particles, then 1/4 of the matter would be visible, and
3/4 dark. In reality, the distribution of matter in the Universe need not be
even over the four different version of the particles. It can deviate from even
distribution, but we expect that the deviation is not very big. According
to the current astronomical observations, about 81.7% of matter in the
Universe is dark, and only 18.3% is visible. This roughly corresponds to the
ratio 1/4 of the “visible states” in matrix (108).
d How this works in the case of Cl(1, 3) is shown in Ref. [32] (see also Refs. [27, 28]).
e Here we extend the concept of mirror particles and mirror gauge fields. The idea of
mirror particles was first put forward by Lee and Yang [35] who realized that “. . . there
could exist corresponding elementary particles exhibiting opposite asymmetry such that
in the broader sense there will still be over-all right–left symmetry.” Further they wrote:
“If this is the case, it should be pointed out that there must exist two kinds of protons pR
and pL , the right-handed one and the left-handed one.” Lee and Yang thus considered the
possibility of mirror particles, though they did not name them so, and as an example they
considered ordinary and mirror protons. Later, Kobzarev et al. [36], instead of P-partners,
considered CP-partners of ordinary particles and called them “mirror particles”. They
argued that a complete doubling of the known particles and forces, except gravity, was
necessary. Subsequently, the idea of mirror particles has been pursued in Refs. [37–42].
The connection between mirror particles and dark matter was suggested in Ref. [43], and
later explored in many works, e.g. in Refs. [44–50]. An explanation of mirror particles in
terms of algebraic spinors (elements of Clifford algebras) was exposed in Refs. [32, 33].
For a recent review see Ref. [51].
www.ebook3000.com
4.2. Quantized fields as generalized Clifford numbers

We can consider a field as an infinite dimensional vector. As an example,
let us take

Ψ = ψ i(x) hi(x) ≡ dn x ψ i (x)hi (x), (110)
where i = 1, 2, x ∈ R3 or x ∈ R1,3 are, respectively, a discrete index, and

(x) a continuous index, denoting, e.g. a point in 3D space, or an event
in 4D space–time. The infinite dimensional vector Ψ is decomposed with
respect to an infinite dimensional basis, consisting of vectors hi(x) ≡ hi (x),
satisfying [14]
1
hi(x) · hj(x ) ≡ (hi(x) hj(x ) + hj(x ) hi(x) ) = ρi(x)j(x ) , (111)
2
where ρi(x)j(x ) is the metric of the infinite dimensional space S. The latter
space may in general have non-vanishing curvature [9]. If, in particular, the
curvature of S is “flat”, then we may consider a parametrization of S such
that
ρi(x)j(x ) = δij δ(x − x ). (112)
In Eq. (111), we have a generalization of the Clifford algebra relations (4)

to infinite dimensions.
Instead of the basis in which the basis vectors satisfy Eq. (111), we can
introduce the Witt basis
1
h(x) = √ (h1(x) + ih2(x) ), (113)
2
1
h̄(x) = √ (h1(x) − ih2(x) ), (114)
2
in which we have
h(x) · h̄(x ) = δ(x)(x ) , (115)
h(x) · h(x ) = 0, h̄(x) · h̄(x ) = 0. (116)
The vector hi(x) and the corresponding components ψ i(x) may contain an
implicit discrete index µ = 0, 1, 2, . . . , n, so that Eq. (110) explicitly reads
Ψ = ψ iµ(x) hiµ(x) = ψ µ(x) hµ(x) + ψ̄ µ(x) h̄µ(x) . (117)

640 M. Pavšič
Then, Eqs. (115) and (116) become the anticommuting relations for fermion
fields:
hµ(x) · h̄ν(x ) = ηµν δ(x)(x ) , (118)
hµ(x) · hν(x ) = 0, h̄µ(x) · h̄ν(x ) = 0. (119)

The quantities hµ(x) ,h̄µ(x) are a generalization to infinite dimensions of the
Witt basis vectors θµ , θ̄µ , defined in Eq. (45).
Using h̄µ(x) , we can define a vacuum state as the product [14]

Ω= h̄µ(x) , h̄µ(x) Ω = 0. (120)

µ,x
Then, using the definition (117) of a vector Ψ, we have

ΨΩ = ψ µ(x) hµ(x) Ω. (121)
Because h̄µ(x) Ω = 0, the second part of Ψ disappears in the above equation.
The infinite-dimensional vector Ψ, defined in Eq. (117), consists of two
parts, ψ µ(x) hµ(x) and ψ̄ µ(x) h̄µ(x) , which both together span the phase space
of a field theory.
The vector ψ µ(x) hµ(x) can be generalized to an element of an infinite
dimensional Clifford algebra:

ψ0 1 + ψ µ(x) hµ(x) + ψ µ(x)ν(x ) hµ(x) hν(x ) + · · · (122)
Acting with the latter object on the vacuum (120), we obtain

ΨΩ = (ψ0 1 + ψ µ(x) hµ(x) + ψ µ(x)ν(x ) hµ(x) hν(x ) + · · · )Ω. (123)
This state is the infinite dimensional space analog of the spinor as an ele-
ment of a left ideal of a Clifford algebra. At a fixed point x ≡ xµ there is
no “sum” (i.e. integral) over x in expression (122), and we obtain a spinor
with 2n components. It is an element of a minimal left ideal of Cl(2n). In
4D space–time, n = 4, and we have Cl(8) at fixed x.
Besides the vacuum (120) there are other vacuums, such as

Ω= hµ(x) , hµ(x) Ω = 0 (124)

µ,x
and, in general,
  

Ω= h̄µ(x)   hµ(x) . (125)

µ∈R1 ,x µ∈R2 ,x
Here, R = R1 ∪ R2 is the set of indices µ = 0, 1, 2, . . . , n and R1 , R2 are

subsets of indices, e.g. R1 = {1, 3, 5, . . . , n}, R2 = {2, 4, . . . , n − 1}.
www.ebook3000.com
Expression (125) can be written as
  

Ω=  h̄µ(x)   hµ(x)  = Ω(x) , (126)

x µ∈R1 µ∈R2 x
where
  

Ω(x) =  h̄µ(x)   hµ(x)  (127)

µ∈R1 µ∈R2
is a vacuum at a fixed point x. At a fixed x, we have 2n different vacu-

ums, and thus 2n different spinors, defined analogously to the spinor (123),
belonging to different minimal ideal of Cl(2n).
The vacuum (125) can be even further generalized by taking different
domains R1 , R2 of space–time positions x:
  

Ω= h̄µ(x)   hµ(x) . (128)

µ∈R1 ,x∈R1 µ∈R2 ,x∈R2
In such a way, we obtain many other vacuums, depending on a partition of

Rn into two domains R1 and R2 so that Rn = R1 ∪ R2 .
Instead of the configuration space, we can take the momentum space,
and consider, e.g. positive and negative momenta. In Minkowski space–time
we can have a vacuum of the form
  

Ω= h̄µ(p0 ,p)   hµ(p0 ,p)  (129)

µ,p0 >0,p µ,p0 <0,p
which is annihilated according to
h̄µ(p0 >0,p) Ω = 0, hµ(p0 <0,p) Ω = 0. (130)
For the vacuum (129), h̄µ(p0 >0,p) and hµ(p0 <0,p) are annihilation oper-
ators, whereas h̄µ(p0 <0,p) and hµ(p0 >0,p) are creation operators from which
642 M. Pavšič
one can compose the states such as

0 0 0
ψ0 1 + ψ µ(p >0,p) hµ(p0 >0,p) + ψ µ(p >0,p) ψ ν(p >0,p ) hµ(p0 >0,p) hν(p0 >0,p )
0

+ · · · + ψ µ(p <0,p) h̄µ(p0 <0,p) + · · · Ω. (131)
The vacuum, satisfying (130), has the property of the bare Dirac vac-
uum. This can be seen if one changes the notation according to
hµ (p0 > 0, p) ≡ b†µ (p), h̄µ (p0 > 0, p) ≡ bµ (p), (132)

0 0
h̄µ (p < 0, p) ≡ dµ (p), hµ (p < 0, p) ≡ d†µ (p), (133)
Ω ≡ |0bare . (134)
A difference with the usual Dirac theory is that our operators have index µ
which takes four values, and not only two values, but otherwise the principle
is the same.
The operators b†µ and bµ , respectively, create and annihilate a positive
energy fermion, whereas the operators dµ , d†µ create and annihilate a nega-
tive energy fermion. This is precisely a property of the bare Dirac vacuum.
Instead of the bare vacuum, in quantum field theories we consider the phys-
ical vacuum

|0 = dµ (p)|0bare , (135)

µ,p
in which the negative energy states are filled, and which in our notation
reads

Ωphys = h̄µ(p0 ,p) Ω. (136)

µ,p0 <0,p
We see that in a field theory à la Clifford, a vacuum is defined as the

product of fermionic operators (generators in the Witt basis). The Dirac
(physical) vacuum is defined as a sea of negative energy states according to
(135) or (136). Today it is often stated that the Dirac vacuum as the sea
of negative energy states is an obsolete concept. But within a field theory
based infinite dimensional Clifford algebras, a vacuum is in fact a “sea” of
states defined by infinite (uncountable) product of operators.
With respect to the vacuum (129), one kind of particles are created
by the positive energy operators hµ (p0 > 0, p), whilst the other kind of
particles are created by the negative energy operator, h̄µ (p0 < 0, p). The
vacuum with reversed properties can also be defined, besides many other
possible vacuums. All those vacuums participate in a description of the
www.ebook3000.com
interactive processes of elementary particles. What we take into account

in our current quantum field theory (QFT) calculations seem to be only a
part of a larger theory that has been neglected. It could be that some of
the difficulties (e.g. infinities) that we have encountered in QFTs so far are
partly due to neglect of such a larger theory.
In an analogous way, we can also construct [14] bosonic states as ele-
ments of an infinite dimensional symplectic Clifford algebra. The generators
of the latter algebra are bosonic field operators. We will use them in the
next subsection, when constructing the action and field equations.
4.3. The action and field equations

A sympletic vector is (see Ref. [14])
Φ = φi(x) ki(x) = φ1(x) k1(x) + φ2(x) k2(x)

φ
≡ φ(x) k(x) Π
+ Π(x) k(x) ,
x ∈ R3 or x ∈ R1,3 . (137)
Here, φi(x) = (φ(x) , Π(x) ) are components and ki(x) , i = 1, 2, basis vectors,
satisfying
1
ki(x) ∧ kj(x ) = [ki(x) , kj(x ) ] = Ji(x)j(x ) , (138)
2
where

0 δ(x)(x )
Ji(x)j(x ) = . (139)
−δ(x)(x ) 0
The action is

1 i(x)
I= dτ φ̇ Ji(x)j(x ) φj(x ) − H , (140)
2
where
1 i(x)
H= φ Ki(x)j(x ) φj(x ) (141)
2
is the Hamiltonian, and
1 i(x) 1
φ̇ Ji(x)j(x ) φj(x ) = (Πφ̇ − φΠ̇) (142)
2 2
the symplectic form.
644 M. Pavšič
In particular [14], if x ≡ xr ∈ R3 , r = 1, 2, 3, and

2
(m + ∂ r ∂r )δ(x − x ) 0
Ki(x)j(x ) = , (143)
0 δ(x − x )
then we obtain the phase space action for a classical scalar field.
If x ≡ xr , r = 1, 2, 3, and

1 r 0 1
Ki(x)j(x ) = −
∂ ∂r + V (x) δ(x − x )gij , gij = , (144)
2m 1 0
then the action (140) describes the classical Schrödinger field.
If x ≡ xµ ∈ R1,3 , µ = 0, 1, 2, 3, and

1 µ 0 1
Ki(x)j(x ) = − ∂ ∂µ δ(x − x )gij , gij = , (145)
2Λ 1 0
then from (140) we obtain the action for the classical Stueckelberg field.
From the action (140) we obtain the following equations of motion
∂H
φ̇i(x) = J i(x)j(x ) , (146)
∂φj(x )

where ∂/∂φj(x ) ≡ δ/δφj(x ) is the functional derivative. By following the
analogous procedure as in Sec. 3.2, we obtain, [14] the equations of motion
for the operators:
k̇j(x ) = k i(x) Ki(x)j(x ) = [kj(x ) , Ĥ], (147)
where
1 i(x)
k Ki(x)j(x ) k j(x ) .
Ĥ = (148)
2
The Heisenberg equations of motion (147) can be derived from the action

1
I= dτ (k̇ i(x) Ji(x)j(x ) k j(x ) + k i(x) Ki(x)j(x ) k j(x ) ). (149)
2
The Poisson bracket between two functionals of the classical phase space
fields is
∂f ∂g
{f (φi(x) ), g(φj(x ) )}PB = i(x)
J i(x)j(x ) j(x ) . (150)
∂φ ∂φ

In particular, if f = φk(x ) , g = φ (x ) , Eq. (150) gives [14]

1

φk(x ) , φ (x ) = J k(x ) (x ) = k k(x ) ∧ k (x ) ≡ k k(x ) , k (x ) .
PB 2
(151)
On the one hand, the Poisson bracket of two classical fields is equal to the
symplectic metric. On the other hand, the symplectic metric is equal to
www.ebook3000.com
the wedge product of basis vectors. In fact, the basis vectors are quantum
mechanical operators, and they satisfy the quantum mechanical commuta-
tion relations
1
[kφ (x), kΠ (x )] = δ(x − x ) (152)
2
or
[φ̂(x), Π̂(x )] = iδ(x − x ), (153)
if we identify √12 kφ (x) ≡ φ̂(x), √i2 kΠ (x ) ≡ Π̂(x ).

A similar procedure can be repeated for fermionic vectors [14].
5. Towards Quantum Gravity

5.1. Gravitational field from Clifford algebra
The generators of a Clifford algebra, γµ , γ̄µ , are (i) tangent vectors to
a manifold which, in particular, can be space–time. On the other hand,
(ii) the γµ , γ̄µ are superpositions of fermionic creation and annihilation
operators, as shown in Eqs. (47) and (48). The two facts, (i) and (ii), must
have profound and far reaching consequences for quantum gravity. Here I
am going to expose some further ingredients that in the future, after having
been fully investigated, will illuminate the relation between quantum theory
and gravity.
As a first step, let us consider a generalized spinor field defined in
Sec. 4.1:
Ψ = ψ Ã sÃ = φA γA . (154)
We are interested in the expectation value of a a vector γµ with respect to

the state Ψ:
γµ 1 ≡ Ψ‡ γµ Ψ1 = ψ ∗Ã s‡Ã γµ sB̃ ψ B̃ 1 . (155)
The subscript 1 means vector part of the expression. Recall from Sec. 2
that ‡ means reversion. Taking
s‡Ã γµ sB̃ 1 = CÃcB̃µ γc , (156)
we have
γµ 1 = eµ c γc , (157)
646 M. Pavšič
where
eµ c = ψ ∗Ã CÃcB̃µ ψ B̃ (158)
is the fierbein. The vector γµ gives the flat space–time metric

γµ · γν = ηµν . (159)
The expectation value vector γµ 1 gives a curved space–time metric
gµν = γµ 1 · γν 1 = eµ c eν d ηcd (160)

which, in general, differs from ηµν . If ψ Ã depends on position x ≡ xµ in
space–time, then also eµ c depends on x, and so does gµν .
From Eq. (156), we obtain
eµ a = γµ · γ a . (161)
From the fierbein, we can calculate the spin connection

1 ρb a
ωµ ab = (e e[µ,ρ] − eρa e[µ,ρ]
b
+ eρb eaσ eµc e[σ,ρ]
c
). (162)
2
The curvature is
Rµν ab = ∂µ ων ab − ∂ν ωµ ab + ωµ ac ωνc b − ων ac ωµc b . (163)
ab
In order to see whether the curvature vanishes or not, let us calculate ω[µ,ν]
by using (158) in which we write
ψ ∗Ã ψ B̃ ≡ ψ ÃB̃ . (164)
We obtain
1
ab
ω[µ,ν] = [C bρ C a (ψ ÃB̃ ψ C̃ D̃ ,ρ ) ν − CÃbρ C a (ψ ÃB̃ ψ C̃ D̃ ,µ ),ν
2 ÃB̃ C̃ D̃µ B̃ C̃ D̃ρ
−CÃaρ C b (ψ ÃB̃ ψ C̃ D̃ ,ρ ),ν + CÃaρ

B̃ C̃ D̃µ
C b (ψ ÃB̃ ψ C̃ D̃ ,µ ),ν
B̃ C̃ D̃ρ
+ more terms − (µ → ν, ν → µ)]. (165)
The latter expression does not vanish identically. In general, it could be

different from zero, which would mean that also the curvature (163) is
different from zero, and that the generalized spinor field ψ Ã (x) induces
gravitation. This assertion should be checked by explicit calculations with
explicit structure constants CÃbρ
B̃
and/or their symmetry relations.
If ψ Ã (x) indeed induces gravitation, then we have essentially arrived
at the basis of quantum gravity. At the basic level, gravity is thus caused
by a space–time dependent (generalized) spinor field ψ Ã (x) entering the
www.ebook3000.com
µ
expression (158) for vierbein. If ψ Ã (x) is constant, or proportional to eipµ x
which, roughly speaking, means that there is no non-trivial matter, then
Rµν ab = 0. This has its counterpart in the (classical) Einstein’s equations
which say that matter curves space–time.
5.2. Action principle for the Clifford algebra valued

field
Let us assume that the field (154) satisfies the action principlef

1
I= d4 x ∂µ φA ∂ν φB ηAB η µν (166)
2
for a system of scalar fields φA that may contain an implicit index i = 1, 2,
denoting real and imaginary components. Here, ηAB is the metric of the
16D Clifford space, whereas ηµν is the metric of the 4D Minkowski space.
The action (166) is not invariant under reparametrizations of coordinates
xµ (i.e. of general coordinate transformations). A possible way to make the
action invariant is to replace η µν with g µν , and include a kinetic term for
g µν . Another possible way is to consider the action

1
I = d4 x det(∂µ φA ∂ν φB ηAB ) 2 . (167)
This is an action for a 4D surface V4 , embedded in the 16D space, the

embedding functions being φA (xµ ). The induced metric on V4 is
gµν = ∂µ φA ∂ν φB ηAB . (168)
The theory based on the nonlinear action (167) is complicated and dif-
ficult to quantize. Therefore, we will return to the action (166) and try
to explore how far can we arrive in inducing non-trivial space–time metric
according to the lines indicated in Sec. 5.1. The equations of motion derived
from (166) are
η µν ∂µ ∂ν φA = 0. (169)
A
A field φ that satisfied the latter equation satisfies also the Dirac-like
equation
γ µ ∂µ φA = 0, γ µ · γ ν = η µν . (170)
Ã
This is so because of the relation (154) and the fact that ψ are spinor
components belonging to all left minimal ideals of the considered Clifford
f If reduced to a subspace of the Clifford space, this action contains a mass term.
648 M. Pavšič
algebra. Equation (170) can be contracted by γA , and we obtain the Dirac–

Kähler equation
γ µ ∂µ φA γA = 0, (171)
where
γA = (1, γa1 , γa1 ∧ γa2 , . . . , γa1 ∧ · · · ∧ γa4 ). (172)
In (171), we have a geometric form of the equation. We can put it in a
sandwich between γ B and γA , or equivalently, between sB̃ and sÃ , according
to
γ B γ µ γA S ∂µ φA = 0 or sB̃ γ µ sB̃ S ∂µ φÃ = 0. (173)
Here, “S” denotes scalar part, 0 multiplied by the dimension of the spinor
space. Here, γ B γ µ γA S ≡ (γµ )B A and sB̃ γ µ sÃ S ≡ (γµ )B̃ Ã are 16 × 16
matrices, representing the vectors γµ . Those matrices are reducible to four
4 × 4 blocks
sβ γ µ sB̃ S ≡ (γµ )β α , α, β = 1, 2, 3, 4 spinor index (174)
which are just the (usual) Dirac matrices.
Equation (173) can be derived from the action

I = d4 x φA γA γ µ γB ∂µ φB S (175)
which can also be written in terms of the generalized spinors ψ Ã sÃ :

I = d4 x ψ Ã sÃ γ µ sB̃ ∂µ ψ B S . (176)
The action (175) or (176) is not invariant under general coordinate

transformations of xµ . For this aim one has to consider position dependent
Clifford numbers, giving the connection according to [28]
∂µ γA = Γµ B A γB , ∂µ sÃ = Γµ B̃ Ã sB̃ , (177)
from which we find [28] that ∂µ ψ B and ∂µ ψ B̃ must be replaced with the
covariant derivatives
Dµ φB = ∂µ φB + ΓBC
µ φC and Dµ φB̃ = ∂µ φB̃ + Γµ B̃ C̃ φC̃ . (178)
Then, in particular, the position dependent γ µ gives curved metric accord-
ing to γµ (x) · γν (x) = gµν . In addition, one also needs to include a kinetic
term for gµν or the connection Γµ BC (or for Γµ B̃ C̃ ).
www.ebook3000.com
Alternatively, one can find a solution φA (or, equivalently ψ Ã ) of the

flat space equation (171), with γµ · γν = ηµν , and calculate the expectation
value γµ according to Eq. (155), and then obtain the metric
gµν = γµ · γν , (179)
of a curved space–time, induced by the fields φA . No kinetic term for the
field gµν (x) or the corresponding connection is necessary in such a proce-
dure. A curved space–time metric comes directly from the fields φA (or ψ Ã )
which are solutions of the flat space equation (171).
In both procedures, the metric is given in terms of the fields φA (or ψ Ã ).
Equating the metrics (168) and (179), we have
∂µ φA ∂ν φB ηAB = φA CABµ
a
φB φC CCDν
b
φD γa γb . (180)
Here, we have used Eqs. (155)–(158) in which we replaced ψ Ã with φA , as

suggested by (154). Equation (180) is a condition that the fields φA must
satisfy. Such a condition can be satisfied if we start from the action

4 1 A B µν 1 A B C D
I= d ∂µ φ ∂ν φ ηAB η − λABCD φ φ φ φ (181)
2 4!
with a quartic self-interaction term. The equations of motion are then
1
∂µ ∂ µ φA + λABCD φB φC φD = 0, (182)
3!
from which we obtain

1
d4 x ∂µ φA ∂ µ φA − λABCD φB φC φD
3!

4 A µ 1 B C D
=− d x φ ∂µ ∂ φA + λABCD φ φ φ = 0. (183)
3!
The latter equation also comes from (180) after contracting with η µν and
integrating over x, provided that we identify
1 a b
λABCD = CABµ CCDν η µν ηab , (184)
3!
where ηab = γa · γb .
In the action (181), we have yet another possible generalization of the
non-interacting action (166) (the other generalization was the “minimal
surface” action (167)). We have thus arrived at a fascinating result that
the space–time metric gµν can be induced by Clifford algebra valued field
φA γA that satisfies the quartic action principle (181).
650 M. Pavšič
5.3. Fermion creation operators, branes as vacuums,

branes with holes, and induced gravity
The procedure described in Sec. 5.1 can be considered as a special case of
quantized fields (123) at a fixed space–time point x. We will now start from
a generic object of the form (123). It consists of the terms such as
ψ µ1 (x1 )µ2 (x2 )...µr (xr ) hµ1 (x1 ) hµ2 (x2 ) . . . hµr (xr ) Ω, (185)
where we assume that Ω is the vacuum given by Eq. (120). The oper-
ator hµi (xi ) creates a fermion at a point xi . The product of operators
hµi (xi ) hµj (xj ) creates a fermion at xi and another fermion at xj . By a generic
expression (185) we can form any structure of fermions, e.g. a spin network.
In the limit in which there are infinitely many densely-packed fermions, we
obtain arbitrary extended objects, such as strings, membranes, p-branes,
or even more general objects, including instantonic branes, considered in
Sec. 2.
Let us use the following compact notation for a state of many fermions
forming an extended object in space–time:
 

 hµ(x)  Ω. (186)
µ,x∈R
Here, the product runs over space–time points x ∈ R of a region R of

space–time MD . In particular, R can be a p-brane’s world sheet Vp+1 ,
whose parametric equation is xµ = X µ (σ a ), µ = 0, 1, 2, . . . , D − 1, a =
1, 2, . . . , p + 1, or it can be a brane-like instantonic object, also described
by some functions X µ (σ). Then the product of operators in Eq. (186) can
be written in the form

hµ(x) ≡ h[X µ (σ)], (187)

µ,x=X(σ)
where h[X µ (σ)] is the operator that creates a brane or an instantonic brane
(that we will also call “brane”). Here, a brane is an extended objects con-
sisting of infinitely many fermions, created according to
 

ψbrane = h[X µ (σ)]Ω =  hµ(x)  Ω. (188)

µ,x=X(σ)
To make contact with the usual notation, we identify

Ω ≡ |0, h[X µ (σ)] ≡ b† [X µ (σ)], ψbrane ≡ |X µ (σ) (189)
www.ebook3000.com
and write
|X µ (σ) = b† [X µ (σ)]|0. (190)
A generic single brane state is a superposition of the brane states:

|Ψ = |X µ (σ)DX(σ)X µ (σ)|Ψ. (191)
In the notation of Eqs. (185)–(188), the latter expression reads

Ψ = DX(σ)ψ[X µ (σ)]h[X µ (σ)]Ω, (192)
where
ψ[X µ (σ)] = lim ψ µ1 (x1 )...µr (xr ) . (193)

r→∞,∆xi →0
However, besides single-brane states, there are also two-brane, three-

brane, and in general, many-brane states. The brane Fock-space states are
thus
b† [X1µ (σ)]|0, b† [X1µ (σ)]b† [X2µ (σ)]|0,

b† [X1µ (σ)] . . . b† [Xrµ (σ)]|0, . . . .
(194)
A generic brane state is a superposition of those states.
If we act on the brane state (188) with the operator h̄µ (x ) , we have
 

h̄µ (x ) Ψbrane = h̄µ (x )  hµ(x)  Ω. (195)

µ,x=X(σ)
If x is outside the brane, then nothing happens. But is x is a position on

the brane, then (195) is a a state in which the particle at x with the spin
orientation µ is missing. In other words, (195) is a brane state with a hole
at x .
We may also form two-hole state, many-hole states, and the states with
a continuous set of holes,
    

 h̄µ(x)  Ψbrane =  h̄µ(x)   hµ(x)  Ω, (196)

µ,x∈R1 µ,x∈R1 µ,x=X(σ)
µ
where R1 ⊂ R = {X (σ)}. For instance, R1 can be a string or a brane of
a lower dimensionality than the brane X µ (σ a ).
If the space into which the brane is embedded has many dimensions,
e.g. D = 10 > p + 1, then the brane’s worldsheet Vp+1 can represent our
652 M. Pavšič
space–timeg which, if p + 1 > 4, has extra dimensions. The induced metric

on Vp+1 can be curved, and so we have curved space–time. We have thus
arrived at the brane world scenario. Holes in the brane are particles. More
precisely, the point-like holes in the worldsheet Vp+1 are instantonic point
particles, whereas the string-like holes are instantonic strings, which can be
either space-like or time-like (see Refs. [9, 52]).
Let me now outline how the induced metric on a brane Vp+1 could be
formally derived in terms of the operators hµ(x) , h̄µ(x) . The corresponding
operators in orthogonal basis are (see (113) and (114)),
1
h1µ(x) = √ (hµ(x) + h̄µ(x) ), (197)
2
1
h2µ(x) = √ (hµ(x) − h̄µ(x) ), (198)
i 2
satisfy the Clifford algebra relations
hiµ(x) · hjν(x ) = δij ηµν δ(x − x ). (199)
In particular,
h1µ(x) · h1ν(x) = ηµν δ(0). (200)
Comparing the latter result with
γµ · γν = ηµν , (201)
we find thath

h1µ(x) = γµ δ(0). (202)
This means that up to an infinite constant, h1µ(x) is proportional to γµ , a

basis vector of Minkowski space–time. Thus, a proper renormalization of
h1µ(x) gives γµ .
g Formore details on how an instantonic brane is related to our evolving space–time, see
Refs. [9, 52].
h Such notation could be set into a rigorous form if, e.g. in Eq. (199) we replace δ(x − x )
(x−x )2
with a√ 1
π
exp[− a2 ] and δ(0) with “δ(0)” ≡ a√ 1
π
. Then Eq. (200) is replaced by
h1µ(x) · h1ν(x) = ηµν “δ(0)”. By inserting into the latter equation the relation h1µ(x) =
p
γµ “δ(0)”, we obtain γµ · γν = ηµν , which also holds in the limit a → 0, because “δ(0)”
has disappeared from the equation.
www.ebook3000.com
In a given quantum state Ψ we can calculate the expectation value of

hiµ(x) according to
hiµ(x) = Ψ‡ hiµ(x) Ψ1 , (203)
where the subscript 1 means vector part of the expression in the bracket.
The inner product gives the expectation value of the metric:
ρiµ(x)jν(x ) = hiµ(x) · hjν(x ) . (204)
This is the metric of an infinite dimensional manifold that, in general, is

curved. In Ref. [9], a special case of such a manifold, for i = j = 1, called
membrane space M, was considered. It was shown how to define connection
and curvature of M.
Taking i = j = 1 and x = x in Eq. (204), we have
ρ1µ(x) 1ν(x) = h1µ(x) · h1ν(x) . (205)
Upon renormalization according to (202) (see footnote h), we obtain
gµν (x) = γµ (x) · γν (x), (206)
where
1
gµν (x) = ρ1µ(x)1ν(x) (207)
“δ(0)”
is a position dependent metric of space–time. We expect that the corre-
sponding Riemann tensor is in general different from zero.
As an example let us consider the expectation value of a basis vector
h1µ(x) in the brane state (188):

‡ ‡ 1
h1µ(x) = Ψbrane h1µ(x) Ψbrane 1 = Ψbrane √ (hµ(x) + h̄µ(x) )Ψbrane .
2 1
(208)
From Eq. (195) in which the vacuum Ω is defined according to (120), we
have

Ψbrane (x̌), x ∈ brane;
h̄µ(x) Ψbrane = (209)
0, x ∈ brane.
Here, Ψbrane (x̌), with the accent “ ˇ ” on x, denotes the brane with a hole at
x. The notation x ∈ brane means that x is on the brane, whereas x ∈ brane
means that x is outside the brane created according to (188).
654 M. Pavšič
Because (h̄µ(x) Ψbrane )‡ = Ψ‡brane hµ(x) , we also have


Ψbrane (x̌)‡ , x ∈ brane;
Ψ‡brane hµ(x) = (210)
0, x ∈ brane.
For the expectation value of h1µ(x) we then obtain

 √1 Ψbrane (x̌)‡ Ψbrane 1 + √1 Ψ‡brane Ψbrane (x̌)1 , x ∈ brane;
2 2
h1µ(x) =
0, x∈ brane.
(211)
A similar expression we obtain for h2µ(x) . The expectation value of the
metrici (204) is

ρiµ(x)jν(x ) |brane , on the brane;
ρiµ(x)jν(x ) = (212)
0, outside the brane.
An interesting result is that outside the brane the expectation value of the
metric is zero. Outside the brane, there is just the vacuum Ω. The expecta-
tion value of a vector hiµ(x) in the vacuum, given by (120), is zero, and so is
the expectation value ρiµ(x)jν(x ) . This makes sense, because the vacuum
Ω has no orientation that could be associated with a non-vanishing effective
vector. In Ω there also are no special points that could determine distances,
and thus a metric. This is in agreement with the concept of configuration
space, developed in Ref. [9], (see also Sec. 2), according to which outside
a configuration there is no space and thus no metric: a physical space is
associated with configurations, e.g. a system of particles, branes, etc.; with-
out a configuration there is no physical space. In other words, a concept
of a physical space unrelated to a configuration of physical objects has no
meaning. Our intuitive believing that there exists a three-(four) dimen-
sional space-(time) in which objects live is deceiving us. The three-(four)
dimensional space-(time) is merely a subspace of the multidimensional con-
figuration space of our universe, in which only position of a single particle
is allowed to vary, while positions of all remaining objects are considered
as fixed. Of course, this is only an idealization. In reality, other objects are
not fixed, and we have to take into account, when describing the universe,
their configuration subspaces as well. Special and general relativity in 4D
i Note that the expectation value of the metric is not defined as ρiµ(x)jν(x ) =
Ψ‡ ρiµ(x)jν(x ) Ψ, but as ρiµ(x)jν(x ) = hiµ(x) · hjν(x) .
www.ebook3000.com
space–time is thus a special case of a more general relativity in configu-

ration space. Quantization of general relativity has failed, because it has
not taken into account the concept of configuration space, and has not rec-
ognized that 4D space–time is a subspace of the huge configuration space
associated with our Universe. The approach with quantized fields presented
in this work has straightforwardly led us to the concept of many particle
configurations and effective curved spaces associated with them.
If in Eq. (212) we take i = j, x = x , and use Eqs. (205)–(207), then we
obtain

gµν (x)|brane = 0, on the brane;
gµν (x) = (213)
0, outside the brane.
It is reasonable to expect that detailed calculations will give the result that
gµν (x)|brane is the induced metric on the brane, i.e.
gµν (x)|brane = ∂a X µ ∂b X ν ηµν ≡ fab . (214)
Recall that the brane can be our space–time. We have thus pointed to a
possible derivation of a curved space–time metric from quantized fields in
higher dimensions.
6. Quantized Fields and Clifford space

In the previous section, we considered fermion states that are generated by
the action of creation operators on the vacuum Ω according to Eq. (185).
In particular, a many fermion state can be a brane, formed according to
Eq. (186). In Sec. 2, we showed that a brane can be approximately described
by a polyvector (5) (see also (6)), which is a superposition of the Clifford
algebra basis elements
1, γµ ∧ γν , γµ ∧ γν ∧ γρ , γµ ∧ γν ∧ γρ ∧ γσ . (215)
This means that a Fock-space element of the form (186) can be mapped
into a polyvector:
 

 hµ(x)  Ω −→ xM γM . (216)
µ,x∈R
As an example, let us consider the case in which the region R of space–

time is a closed line, i.e. a loop. The holographic projections of the area
enclosed by the loop are given in terms of the bivector coordinates X µν .
656 M. Pavšič
The loop itself is describedj by a bivectors X µν γµ ∧ γν . So we have the

mapping
 

 hµ(x)  Ω −→ xµν γµν . (217)

µ,x∈loop
With the definite quantum states, described by Eq. (186) or (188) (see
also (190)), which are the brane basis states, analogous to position states in
the usual quantum mechanics, we can form a superposition (192) (see also
(191)). To such an indefinite brane state, there corresponds a state with
indefinite polyvector coordinate X M :

DX(σ)Ψ[X(σ)]h[X(σ)]Ω −→ φ(xM ). (218)
In particular, if h[X(σ)]Ω is a loop, then we have the mapping

DX(σ)Ψ[X(σ)]h[X(σ)]Ω −→ φ(xµν ). (219)
The circle is thus closed. With the mapping (216), we have again
arrived at the polyvector xM γM introduced in Sec. 2. The polyvector
coordinates xM of a classical system satisfy the dynamics as formulated
in Refs. [10, 11, 28]. That dynamics can be generalized to super phase
space as discussed in Sec. 3, where besides the commuting coordinates xµ ,
µ = 0, 1, 2, 3, we introduced the Grassmann coordinates ξ µ . In the quantized
theory, the wave function ψ(xµ , ξ µ ) represents a 16-component field, φA ,
A = 1, 2, . . . , 16, that depends on position xµ in space–time, and satisfies
the Dirac equation (170) and the multicomponent Klein–Gordon equation
(169). In analogous way, besides commuting polyvector coordinates xM ,
M = 1, 2, . . . , 16, we have the corresponding Grassmann coordinates ξ M ,
and the wave function φ(xM ) is generalized to φ(xM , ξ M ). The expansion of
φ(xM , ξ M ) in terms of ξ M gives a 216 -component field, φA , A = 1, 2, . . . , 216 ,
that depends on position xM in Clifford space, and satisfies the generalized
Dirac equation, γ M ∂M φA (xM ).
As the evolution parameter, i.e. the time along which the wave function
evolves, we can take the time-like coordinate x0 , or the time-like coordi-
nate σ. Alternatively, we can take the light-like coordinate s, defined in
Eq. (15), as the evolution parameter. Then, as shown in Ref. [16], the
Cauchy problem can be well posed, in spite of the fact that in Clifford
j Of course, there is a class of loops, all having the same X µν .
www.ebook3000.com
space there are eight time-like dimensions, besides eight space-like dimen-
sions. Moreover, according to Refs. [53–55], there are no ghosts in such
spaces, if the theory is properly quantized, and in Refs. [56–60] it was
shown that the stability of solutions can be achieved even in the presence of
interactions.
We can now develop a theory of such quantized fields in Clifford space
along similar lines as we did in Secs. 4 and 5 for the quantized fields in
the ordinary space–time. So we can consider the analog of Eqs. (186)–
(214) and arrive at the induced metric on a 4D surface V4 embedded in
the 16D Clifford space. Whereas in Eqs. (186)–(214) we ad hoc postulated
the existence of extra dimensions, we now see that extra dimensions are
incorporated in the configuration space of brane-like objects created by the
fermionic field operators hµ(x) . Our space–time can thus be a curved surface
embedded in such a configuration space.
7. Conclusion
Clifford algebras are very useful to describe extended objects as points in
Clifford spaces, which are subspaces of configuration spaces. The Stueckel-
berg evolution parameter can be associated with the scalar and the pseu-
doscalar coordinate of the Clifford space.
The generators of orthogonal and symplectic Clifford algebras, i.e. the
orthogonal and symplectic basis vectors, behave, respectively, as fermions
and bosons. Quantization of a classical theory is the shift of description
from components to the (orthogonal or symplectic) basis vectors.
We have found that a natural space to start from is a phase space,
which can be either orthogonal or symplectic. We united both those phase
spaces into a super phase space, whose points are described by anticommut-
ing (Grassmann) and commuting coordinates, the basis vectors being the
generators of orthogonal and symplecting Clifford algebras. We have con-
sidered the Clifford algebra Cl(8) constructed over the 8D orthogonal part
of the super phase space. Remarkably, the 256 spinor states of Cl(8) can
be associated with all the particles of the Standard Model, as well as with
additional particles that do not interact with our photons and are therefore
invisible to us. This model thus predicts dark matter. Moreover, it appears
to be a promising step towards the unification of elementary particles and
interactions (see also Refs. [11, 27, 28]).
Both, orthogonal and symplectic Clifford algebras can be generalized
to infinite dimensions, in which case their generators (basis vectors) are
658 M. Pavšič
bosonic and fermionic field creation and annihilation operators. In the

Clifford algebra approach to field theories, a vacuum is the product of
infinite, uncountable number of Fermionic field creation operators. They
can form many sorts of possible vacuums as the seas composed of those
field operators. In particular, strings and branes can be envisaged as being
such seas. The field operators, acting on such brane states, can create holes
in the branes, that behave as particles. From the expectation values of vec-
tor operators in such a one, two, or many holes brane state, we can calculate
the metric on the brane. According to the brane world scenario, a brane
can be our world. We have found that holes in a fermionic brane behave as
particles, i.e. matter, in our world, and that the metric on the brane can
be quantum mechanically induced by means of the fermionic creation and
annihilation operators. We have thus found a road to quantum gravity that
seems to avoid the usual obstacles.
Acknowledgment
This work has been supported by the Slovenian Research Agency.
References
1. P.A.M. Dirac, The Principles of Quantum Mechanics, Oxford Univ. Press,
Oxford, 1958.
2. D. Hestess, Space-Time Algebra, Gordon and Breach, New York, 1966;
D. Hestenes and G. Sobcyk, Clifford Algebra to Geometric Calculus, D. Rei-
del, Dordrecht, 1984.
3. P. Lounesto, Clifford Algebras and Spinors, Cambridge University Press,
Cambridge, 2001.
4. B. Jancewicz, Multivectors and Clifford Algebra in Electrodynamics, World
Scientific, Singapore, 1988.
5. R. Porteous, Clifford Algebras and the Classical Groups, Cambridge Univ.
6. W. Baylis, Electrodynamics, A Modern Geometric Approach, Birkhauser,
Boston, 1999.
7. A. Lasenby and C. Doran, Geometric Algebra for Physicists, Cambridge Univ.
8. C. Castro, Chaos, Solitons and Fractals 10, 295 (1999); 11, 1663 (2000);12,
1585 (2001). C. Castro, Found. Phys. 30, 1301 (2000).
9. M. Pavšič, The Landscape of Theoretical Physics: A Global View; From Point
Particle to the Brane World and Beyond, in Search of Unifying Principle,
Kluwer, Berlin, 2001, [gr-qc/0610061].
10. M. Pavšič, Found. Phys. 33, 1277 (2003), [arXiv:gr-qc/0211085].
11. M. Pavšič, Found. Phys. 37, 1197 (2007), [hep-th/0605126].
www.ebook3000.com
12. C. Castro and M. Pavšič, Prog. Phys. 1, 31 (2005).

13. A. Crumeyrole, Orthogonal and Symplectic Clifford Algebras, Dordrecht,
Springer, 1990.
14. M. Pavšič, Adv. Appl. Clifford Algebras 22, 449 (2012), [arXiv:1104.2266
[math-ph]].
15. M. Pavšič, Phys. Conf. Ser. 437, 012006 (2013), [arXiv:1210.6820 [hep-th]].
16. M. Pavšič, Adv. Appl. Clifford Algebras 23, 469 (2013), [arXiv:1201.5755
[hep-th]].
17. M. Pavšič, J. Phys. Conf. Ser. 330, 012011 (2011), [arXiv:1104.2462
[math-ph]].
18. J. Barbour, Dynamics of Pure Shape, Relativity and the Problem of Times,
arXiv:gr-qc/0309089, 2003. J. Barbour, Class. Quant. Grav. 11, 2853 (1994).
J. Barbour, Class. Quant. Grav. 19, 3217 (2002). J. Barbour, The End of
Time, Oxford University Press, Oxford, 2000.
19. V. Fock, Phys. Z. Sowj. 12, 404 (1937). E.C.G. Stueckelberg, Helv. Phys. Acta
14, 322 (1941); E.C.G. Stueckelberg, Helv. Phys. Acta 14, 588 (1941). E.C.G.
Stueckelberg, Helv. Phys. Acta 15, 23 (1942). L.P. Horwitz and C. Piron,
Helv. Phys. Acta 46, 316 (1973). L.P. Horwitz and F. Rohrlich, Physical
Review D 24, 1528 (1981). L.P. Horwitz and R.I. Arshansky and A.C. Elitzur,
Found. Phys. 18, 1159 (1988). R.P. Feynman, Phys. Rev. 84, 108 (1951). J.R.
Fanchi, Found. Phys. 23, 287 (1993), and many references therein.
20. M. Pavšič, Found. Phys. 21, 1005 (1991).
21. J.R. Fanchi, Parametrized Relativistic Quantum Theory Kluwer, Berlin, 1993.
22. B. Libet, G.A. Gleason, E.W. Wright, and D.K. Pearl, Time of con-
scious intention to act in relation to onset of cerebral activity (readiness-
potential) — the unconscious initiation of a freely voluntary act, Brain 106,
623 (1983), http://dx.doi.org/10.1093.
23. H. Everett, Rev. Mod. Phys. 29, 454 (1957). H. Everett, The theory of the
universal wave function, in B.S. DeWitt and N. Graham (eds.), The Many
Worlds Interpretation of Quantum Mechanics, Princeton Univ. Press, Prince-
ton, 1973.
24. I. Bars, C. Deliduman, and O. Andreev, Phys. Rev. D 58, 066004 (1998).
I. Bars, Phys. Rev. D 58, 066006 (1998). I. Bars, Class. Quant. Grav. 18,
3113 (2001). I. Bars, Phys. Rev. D 74, 085019 (2006). I. Bars and J. Terning,
Extra Dimensions in Space and Time, Springer, Berlin, 2010.
25. P. Zenczykowski, J. Phys. A 42, 045204 (2009), [arXiv:0806.1823 [hep-th]];
P. Zenczykowski, Int. J. Theor. Phys. 49, 2246 (2010), [arXiv:0905.1207
[hep-th]].
26. C. Castro, Int. J. Geom. Meth. Mod. Phys. 6, 385 (2009); C. Castro, Int.
J. Geom. Meth. Mod. Phys. 4, 1239 (2007); C. Castro, J. Math. Phys. 47,
112301 (2006).
27. M. PavšičK, Phys. Lett. B 614, 85 (2005), [hep-th/0412255].
28. M. Pavšič, Int. J. Mod. Phys. A 21, 5905 (2006), [gr-qc/0507053].
29. M. Pavšič, Phys. A 41, 332001 (2008), [arXiv:0806.4365 [hep-th]].
30. J.O. Winnberg, J. Math. Phys. 18, 625 (1977).
660 M. Pavšič
31. P. Budinch, Phys. Rep. 137, 35 (1986). P. Budinich, and A. Trautman, Lett.
Math. Phys. 11, 315 (1986).
32. M. Pavšič, Phys. Lett. B 692, 212 (2010), [arXiv:1005.1500 [hep-th]].
33. M. Pavšič, Geometric spinors, generalized dirac equation and mirror particles,
in T. Kamalov (ed.), Theoretical Physics and its New Applications, Moscow
State Univ., Proc. of the 3rd International Conference on Theoretical Physics,
June 24–28, 2013, Moscow, Russia, arXiv:1310.6566 [hep-th].
34. M. Budinich, J. Math. Phys. 50, 053514 (2009). M. Budinich, J. Phys. A 47,
115201 (2014). M. Budinich, Adv. Appl. Clifford Algebras 25, 771 (2015).
35. T.D. Lee and C.N. Yang, Phys. Rev. 104, 254 (1956).
36. I. Yu. Kobzarev, L.B. Okun and I. Ya. Pomeranchuk, Soviet J. Nucl. Phys.
5, 837 (1966).
37. M. Pavšič, Int. J. Theor. Phys. 9, 229 (1974), [hep-ph/0105344].
38. E.W. Kolb, D. Seckel, and M.S. Turner, Nature 314, 415 (1985).
39. R. Foot, H. Lew, and R.R. Volkas, Phys. Lett. B 272, 67 (1991).
40. R. Foot, H. Lew, and R.R. Volkas, Mod. Phys. Lett. A 7, 2567 (1992).
41. R. Foot, Mod. Phys. Lett. 9, 169 (1994).
42. R. Foot and R.R. Volkas, Phys. Rev. D 52, 6595 (1995).
43. S.I. Blinnikov and M.Y. Khlopov, Sov. J. Nucl. Phys. 36, 472 (1982), [Yad.
Fiz. 36, 809 (1982)]; S.I. Blinnikov and M.Y. Khlopov, Sov.Astron. 27, 371
(1983), [Astron. Zh. 60, 632 (1983)].
44. H.M. Hodges, Phys. Rev. D 47, 456 (1993).
45. R. Foot, Phys. Lett. B 452, 83 (1999).
46. R. Foot, Phys. Lett. B 471, 191 (1999).
47. R.N. Mohapatra, Phys. Rev. D 62, 063506 (2000).
48. Z. Berezhiani, D. Comelli, and F. Villante, Phys. Lett. B 503, 362 (2001).
49. P. Ciarcelluti, Int. J. Mod. Phys. D 14, 187 (2005):
P. Ciarcelluti, Int. J. Mod. Phys. D 14, 223 (2005).
50. P. Ciarcelluti and R. Foot, Phys. Lett. B 679, 278 (2009).
51. R. Foot, Int. J. Mod. Phys. A 29, 1430013 (2014), [arXiv:1401.3965 [astro-
ph.CO]].
52. M. Pavšič, Found. Phys. 26, 159 (1996), [gr-qc/9506057].
53. R.P. Woodard, Lect. Lect. Notes Phys. 720, 403 (2007), [astro-ph/0601672].
54. M. Pavšič, Phys. Lett. A 254, 119 (1999), [hep-th/9812123].
55. M. Pavšič, Found. Phys. 35, 1617 (2005), [hep-th/0501222].
56. M. Pavšič, Mod. Phys. Lett. A 28, 1350165 (2013), [arXiv:1302.5257 [gr-qc]].
57. A.V. Smilga, Phys. Lett. B 632, 433 (2006), [hep-th/0503213].
58. A.V. Smilga, Nucl. Phys. B 706, 598 (2005), [hep-th/0407231].
59. A.V. Smilga, [SIGMA 5, 017 (2009), [arXiv:0808.0139 [quant-ph]].
60. M. Pavšič, Phys. Rev. D 87, 10 (2013), 107502 [arXiv:1304.1325 [gr-qc]].
www.ebook3000.com
Chapter 24
Non-commutative Einstein, almost

Kähler–Finsler and Quantum
Deformations
Sergiu I. Vacaru†
Theory Division, CERN, CH-1211, Geneva 23, Switzerland
and
Rector’s Office, Alexandru Ioan Cuza University
Alexandru Lapuşneanu street, nr. 14, UAIC — Corpus R, office 323
Iaşi, 700057, Romania
sergiu.vacaru@uaic.ro
We show how the geometric data for commutative and Non-commutative

NC models for Ricci solitons and Einstein spaces can be encoded equiva-
lently in terms of almost symplectic geometries and/or Lagrange–Finsler
spaces uniquely determined by the metric and/or almost Kähler struc-
tures. Such generic off-diagonal configurations are quantized following
methods of deformation quantization (DQ). Using almost Kähler vari-
ables, an associative star-product can be introduced for NC generaliza-
tions of Ricci solitonic equations. There are new classes of exact solutions
with NC variables describing NC black ellipsoid and solitonic waves gen-
erated as Ricci solitons. We conclude that the methods of non-holonomic
almost Kähler–Finsler geometry allows us to formulate a unified for-
malism for Ricci flows/solitons and DQ and NC generalizations of such
theories.
1. Introduction
Einstein’s gravity and various modifications can be equivalently formulated
in almost Kähler (symplectic) variables which allows us to perform defor-
mation quantization (DQ) [1] and A-brane quantization [2] of such theories.
† Associated visiting researcher.
661
662 S. I. Vacaru
The first approach is related to the Fedosov’s method [3–5] of constructing

a star-product in a covariant way for symplectic manifolds. The General
Relativity (GR) theory cannot be expressed in a self-consistent form as a
symplectic geometry which for a long time was considered as a problem
to perform DG. Nevertheless, there is a possibility to encode GR and var-
ious modifications/generalizations in terms of certain “canonical” almost
Kählera spaces with prescribed non-holonomic distributions. We can apply
certain constructions for DQ of metric-affine spaces with associated almost
symplectic connections due to Ref. [6].
It is possible to construct canonical almost symplectic forms and connec-
tions completely determined by a (pseudo) Riemannian metric g on a man-
ifold V if a 2 + 2 non-integrable splitting is correspondingly parameterized.
We can apply the DQ scheme to Einstein and Lagrange–Finsler theories
[1, 7] following certain methods from the geometry of non-holonomic mani-
folds with prescribed nonlinear connection, N-connection, structure.b Such
a quantization is not complete if it is not formulated as generally accepted
techniques for (non-) perturbative calculus and renormalization on Hilbert
spaces, etc. (see critics in Ref. [8]). There were two other geometrically
related approaches developed: the first one with two-connection renormal-
ization of gauge-like and Einstein gravity models [9] and, in the second
one, it was concluded that the A-brane quantization of the Einstein gravity
seems to be possible [2]. We proved also that the Hořava–Lifshitz theories
can be extended in a Finsler manner and/or with off-diagonal modifications
and covariant anisotropic quantization [10, 11]. Nevertheless, there are at
least two fundamental issues to be solved in any attempt to quantization
of gravity: (1) how beta-functions, Ricci flows and renormalization schemes
should be included in modern Quantum Gravity (QG) and (2) in which
manner such constructions would involve the NC geometry and modified
gravity models.
There are two approaches related to the NC and commutative Ricci
flow theory and spectral action/functional formalism [12, 13]. These con-
structions were performed using non-holonomic generalizations of Dirac
operators or two-tori NC variables. Certain models were related to NC
a We shall use also the term “almost symplectic”; for simplicity, we shall consider only
four-dimensional (4D) (pseudo) Riemannian manifolds and their geometric flows and
extensions with NC variables.
b This can be introduced as a Whitney sum in tangent bundle, N: T V = hV ⊕ vV.
www.ebook3000.com
Non-commutative Einstein 663
gauge gravity models and exact solutions with NC variables [24–26], see
also various theories and constructions on NC gravity in Refs. [14–23] (and
references therein). Perhaps, a formalism with NC generalizations of sym-
plectic geometry can be most directly related to DQ and gravity theories
encoded in almost symplectic variables Refs. [1, 7].
The goal of this work is to elaborate an unified almost Kähler geometric
formalism for DQ of Ricci solitons, and Einstein manifolds and modifica-
tions, with commutative and NC variables.c We shall define a N-adapted
covariant star-product (which can be associative) and consider its basic
properties via actions on non-holonomic distributions and tensor fields. In
certain sense, we shall generalize the constructions from [27–29] to the case
when the basic equations for Ricci solitons and Einstein manifolds with and
without NC variables can be integrated in very general forms. Here we note
that Ricci solitons play a fundamental role as stationary configurations in
Ricci flow theory [30–34], in Einstein and modified gravity (for reviews of
results, see Refs. [35–37]).
The chapter is organized as follows: In Sec. 2, we define almost Kähler
variables induced by non-holonomic distributions on (pseudo) Rieman-
nian manifolds and formulate the Ricci soliton equations in such variables.
There are discussed possible links of such constructions to modified grav-
ity theories. Section 3 is devoted to DQ of almost Kähler solitonic spaces.
We provide the main theorems of Fedosov–Ricci solitons. We study NC gen-
eralizations of almost symplectic Ricci solitonic structures in Sec. 4. Using
the canonical and Cartan canonical distinguished connections, adapted to
nonlinear connection structures, we define an associative star-product for
NC. There are formulated NC generalizations of the Ricci soliton and Ein-
stein equations. We prove the decoupling property of such equations and
show how we can construct very general classes of exact solutions. Certain
examples of generic off-diagonal solutions depending on NC parameters,
for NC Ricci solitonic black ellipsoids/holes and “non-Ricci” solitonic back-
grounds are provided. We conclude the work in Sec. 5. In the Appendix,
we present some necessary coefficient formulas.
c We shall not concern issues related to (non-) commutative Finsler geometry which are
considered in details in Refs. [1, 7, 10, 26]. Nevertheless, it is necessary to emphasize
that a number of ideas and methods in this chapter are taken from geometric models
with local anisotropy and re-defined, for instance, for Einstein manifolds with fibered
structure.
664 S. I. Vacaru
2. Ricci Solitons and Almost Kähler Geometry

Ricci solitons (introduced in Ref. [38], see a recent review in Ref. [39]) are spe-
cial solutions of the Ricci flow equations [30,31]. In some sense, Ricci flows are
dynamical systems on the space of Riemannian metrics modulo diffeomor-
physms and scaling. We can generalize the constructions considering (non)
commutative geometric flows for non-holonomic (equivalently, anholonomic,
i.e. non-integrable), and/or (almost) symplectic variables, etc. [12, 40].
Ricci solitons can be viewed as fixed points for such generalized non-
holonomic dynamical systems. They play important roles in study of singu-
larities of Ricci flows. They are also called as quasi-Einstein metrics (defin-
ing natural generalizations of Einstein spaces) and present a substantial
interest in modern gravity when renormalization group equations can be
related to geometric flows, exact solutions of modified field equations and
applications in cosmology [10, 35–37, 41, 42].
In this section, we introduce almost Kähler variables determined by
non-holonomic distributions on (pseudo) Riemannian manifolds V with
2 + 2 splitting and express the gradient Ricci soliton equations in canonical
almost symplectic variables.
2.1. Almost symplectic variables generated

by non-holonomic distributions
Let us consider a real 4D, (pseudo) Riemannian manifold V of neces-
sary smooth class endowed with symmetric metric structure g of signa-
ture (±, ±, ±, ±) (we can consider, for instance, constructions for Lorentz
manifolds in GR). For our purposes, we shall work with conventional 2 + 2
splitting when the local coordinates u = (x, y) are labeled in the form
uα = (xi , y a ), where i, j, k, . . . = 1, 2 and a, b, c, . . . = 3, 4; xi and y a are
respectively the horizontal (h) and vertical (v) coordinates for local fibred
structure to be defined below. A general (non-coordinate) local basis is
eα = (ei , ea ), with possible frame transforms eα = eαα (u)∂α , where
∂α = ∂/∂uα = (∂i = ∂/∂xi , ∂a = ∂/∂y a ), and corresponding dual trans-

forms with inverse matrices eαα (u), when eα = eαα (u)duα .d
d Priming indiceswill be omitted if that will not result in ambiguities. Here we note that in
order to construct exact solutions of partial differential equations (PDEs), corresponding
to Ricci solitons and generalized Einstein spaces, it is necessary to work with values
decomposed with respect to certain classes of non-holonomic frame. So, we shall follow
a geometric formalism using both abstract index and coordinate denotations and/or
www.ebook3000.com
2.1.1. Non-holonomic variables on Riemannian manifolds

We can introduce on V a non-holonomic distribution defined by a generating
function L(u), i.e. non-holonomic distribution, with non-degenerate Hessian
1 ∂2L
h̃ab = , det |h̃ab | = 0. (1)
2 ∂y a ∂y b
This allows us to construct the coefficients

∂ G̃a 1 a 2+i ∂2L ∂L
Ñia = , for G̃a
= h̃ y 2+k
− , (2)
∂y 2+i 4 ∂y 2+i ∂xk ∂xi
where h̃ab is inverse to h̃ab . We shall put “tilde” on some geometric objects
in order to emphasize that they are induced by L. In the above formulas,
the contractions of h- and v-indices, i, j, . . . and a, b, . . . , are performed
following the rule: for instance, an up v-index a is considered as a = 2 + i
which allows us to and contract it with a low index i = 1, 2. We can write y i
instead of y 2+i , or y a .e The local constructions with coefficients (2) can be
globalized in a form when Ñ = {Ñia (u)} determines a canonical nonlinear
connection (N-connection) structure. It can be defined as a Whitney sum
Ñ : T V = h̃V ⊕ ṽV, (3)
where T V is the tangent bundle with base V and h̃V and ṽV are, respec-
tively, h- and v-subspaces. We shall denote some spaces (geometric objects)
with boldface letters if they are endowed with (adapted to) a N-connection
structure. If such a h̃-ṽ-splitting is prescribed, we can define the so-called
N-elongated bases and co-bases,

ẽα = ẽi = ∂i − Ñia ∂a , ea = ∂a , (4)
ẽα = (ei = dxi , ẽa = dy a + Ñia dxi ). (5)
In general, such frames are non-holonomic.f Above variables model Finsler-
like geometries if L = F 2 for a Finsler generating function F which can
global constructions which will be more optimal for proofs of results and generating
exact solutions.
e There is a motivation [1] to introduce the values (2): the nonlinear geodesic semi-spray
configurations of L are equivalent to the Euler–Lagrange equations for L treated as a

regular Lagrange function in an analogous model of Lagrange mechanics. We omit such
details in this work, see Ref. [7].
f The term non-holonomic (equivalently, anholonomic, or non-integrable) manifold can
be related to the anholonomy relations

γ
[eα , eβ ] = eα eβ − eβ eα = wαβ eγ ,
666 S. I. Vacaru
be considered on tangent budles to some manifolds or on non-holonomic

manifolds with fibered structure, see details in Refs. [1, 7]. In this work, we
follow an approach when such generalized Finsler-like theories are encoded
into non-holonomic classical and quantum deformations of (pseudo) Rie-
manian manifolds and (modified) gravity.
A non-holonomic distribution L determines via (1), (2), and (5) a canon-
ical metric structure (distinguished metric, d-metric, which is adapted to
the N-connection splitting (3))
g̃ = g̃ij dxi ⊗ dxj + h̃ab ẽa ⊗ ẽb , (6)

g̃ij = h̃2+i 2+j .
It should be emphasized here that any metric g = {gα β } can be expressed

in a form g̃αβ = [ g̃ij , h̃ab ] (6), i.e. as a d-metric, via corresponding frame

transforms, ẽα = eαα eα and gα β eαα eββ = g̃αβ . For instance, we can pre-
scribe a non-holonomic distribution L, compute g̃αβ and then determine

any eαα as solutions of algebraic equations for a given set gα β (we have
to consider such atlases on V and a function L when the frame coefficients
preserve the local signature, certain physical conditions etc). We shall write
hereafter (g, N) without tilde if it will not result in ambiguities and consid-
ering that it is possible always to chose a necessary type L and introduce
geometric data (g̃, Ñ) induced by such a non-holonomic distribution.
For non-holonomic manifolds (V, N), linear connections can be defined
to be adapted to a N-connection splitting, or not. In the first case, a linear
connection D is called a distinguished connection, d-connection, if it pre-
serves under parallelism the Whitney sum (3). Such a d-connection is met-
ric compatible if Dg = 0. A d-connection D = (hD; vD) = {Γγαβ =
(Lijk , v Labk ; Cjc , Cbc )} can be expressed as 1-form Γγα := Γγαβ eβ , with
i v a
γ
coefficients Γ αβ computed with respect to N-adapted frames eα and eβ .
The torsion, T α = {Tαβγ }, and curvature, Rαβ = {Rα βγδ }, are respectively
T α := Deα = deα + Γαβ ∧ eβ = Tαβγ eβ ∧ eγ , (7)

Rαβ := DΓαβ = dΓαβ − Γγβ ∧ Γαγ = Rαβγδ eγ δ
∧e . (8)
The N-adapted coefficients are parameterized in the appendix, see formulas

(A.1) and (A.2).
with non-trivial anholonomy coefficients wia b = ∂ N b and w a = Ωa , where Ωa =

a i ji ij ij
ej (Nia ) − ei (Nja ) are used for the coefficients of N-connection curvature defined as the
γ
Neijenhuis tensor. If wαβ = 0, we get holonomic/integrable configurations.
www.ebook3000.com
The Ricci d-tensor Ric = {Rαβ := Rταβτ ; Rαβ = Rβα } is with

h–v-components
Rαβ = {Rij := Rkijk , Ria := −Rkika , Rai := Rbaib , Rab := Rcabc }. (9)
By definition, the scalar curvature of a d-connection D is

s
R := gαβ Rαβ = R + S = g ij Rij + hab Rab (10)
with R = g ij Rij and S = hab Rab .

For a given metric g and prescribed N on V, we can define three types
of preferred linear and d-connection structures uniquely defined in metric
compatible form by data (g, N):
∇: ∇g = 0; ∇ T α = 0, the Levi-Civita connection;
α α
g = g̃ → D : Dg = 0; hT = 0, v T = 0, the canonical d-connection;
D̃ : D̃g = 0; hT̃ α = 0, v T̃ α = 0, the Cartan d-connection.
The N-adapted coefficients for D and D̃ are given in the appendix (see,
respectively, formulas (A.3) and (A.5)). Here we note that the Levi-Civita
connection is not a d-connection because it does not preserve under par-
allelism the N-connection splitting. Nevertheless, we can always define the
distortion relations
=∇+Z
D and D̃ = ∇ + Z̃, (11)
where all connections and distortion tensors Z and Z̃ are completely defined
by g for a prescribed N, via certain algebraic combinations of coefficients
of respective torsion T and T̃ (see details in Refs. [25, 36, 40]).g
A (pseudo) Riemannian space can be described equivalently in terms
of any connections ∇, D, D̃ or other ones completely determined by data
(g, N). In general, we can consider metric non-compatible connections if
their distortions are uniquely determined by metric. Nevertheless, gravi-
tational field equations with Ricci tensors Ric, Ric, R̃ic (and correspond-
ing Einstein tensors) result, in general, in different classes of solutions.
and R̃ic can be introduced in GR. For instance, using Ric
Both Ric we
can decouple the Einstein equations with respect to N-adapted frames (4)
and (5) which allows to construct very general classes of exact solutions
[25, 35, 36]. At the end (after some integral varieties were defined in certain
g We have a similar situation in the Einstein–Cartan geometry but that torsion is not
defined by the metric structure; there are necessary additional, for instance, algebraic
field equations in order to define torsion not related to the metric tensor field.
668 S. I. Vacaru
general forms depending on different classes of generating and integra-

tion functions), we can impose additional constraints via additional frame
transforms, on coefficients Nia and choosing a subclass of generating and
integration functions when Z = 0. This way we can construct exact solu-
tions of the Einstein equations and for Ricci solitons and modified gravita-
tional equations (we shall provide examples in Sec. 4.2). The d-connection
D̃ is convenient for performing DQ of Ricci soliton geometries (and certain
modified/generalized gravity theories); for this, we have to prove that the
Cartan d-connection defines naturally an almost Kähler geometry.
2.1.2. Canonical almost symplectic variables

Let us consider a linear operator J acting on vectors on T V via actions
on N-adapted frames of type (4), J(ei ) = −e2+i and J(e2+i ) = ei , where
J ◦ J = −I, for I being the unity matrix. This defines an almost complex
structure completely determined by L(x, y) as a d-tensor field,
α ∂
J = Jαβ eα ⊗ eβ = J β ⊗ duβ
∂uα

= Jαβ eα ⊗ eβ = −e2+i ⊗ ei + ei ⊗ e2+i

∂ ∂ 2+j ∂
= − i ⊗ dxi + i
− Ñ i j
⊗ dy i + Ñk2+i dxk . (12)
∂y ∂x ∂y
Such a structure is canonical, we can write J̃ if N = Ñ.

The Neijenhuis tensor field can be computed for any almost complex
structure J being related to the N-connection curvature and anholonomic
coefficients,
J
Ω(X, Y) = −[X, Y] + [JX, JY] − J[JX, Y] − J[X, JY], (13)
where a vector X is represented as a d-vector (N-adapted) in the form

X = X α eα = X i ei + X a ea .
An almost symplectic structure can be introduced by a non-degenerate
2-form θ = 12 θαβ (u)eα ∧ eβ . It is defined by the d-metric structure if
θ(X, Y) := g(JX, Y) for any d-vectors X and Y. In N-adapted form,
1 1
θ= θij (u)ei ∧ ej + θab (u)ea ∧ eb , (14)
2 2
where eα = (ei , ea ) is given by (5).
www.ebook3000.com
An almost Hermitian model of a non-holonomic (pseudo) Riemannian

manifold (V, g, N) is defined by a triple H2+2 = (V, θ(·, ·) : = g (J·, ·) , J).
A space H2+2 is almost Kähler, denoted K2+2 , if and only if dθ = 0.
Let us prove that by prescribing a generating function L(x, y) on a
(pseudo) Riemannian space–time V, we can model equivalently this non-
holonomic manifold as a canonical almost Kähler space. Taking g = g̃,
N = Ñ and J = J̃ canonically defined by L, we can define θ̃(·, ·) := g̃(J̃·, ·).
In N-adapted form,
1 1
θ̃ = θ̃αβ (u)ẽα ∧ ẽβ = θ̃αβ (u)duα ∧ duβ
2 2
= g̃ij (x, y)ẽ2+i ∧ dxj = g̃ij (x, y)(dy 2+i + Ñk2+i dxk ) ∧ dxj , (15)
where θ̃ab = θ̃2+i 2+j are respectively the coefficients θ̃ij . One holds θ̃ = dω̃
∂L
for ω̃ := 12 ∂y i
i dx . This results in dθ̃ = ddω̃ = 0.
We emphasize that taking a general 2-form θ constructed almost com-

plex J structures on V we obtain dθ = 0. If θ is related to a canoni-

cal θ̃ via frame transforms, θα β eαα eββ = θ̃αβ , we positively construct

an almost Kähler geometry. Fixing a convenient type of L, we can con-

struct equivalent geometric models of non-holonomic manifolds given by
data (g, N) ≈ (g̃, Ñ) ≈ (θ, J) ≈ (θ̃, J̃).
Finally, we present this fundamental result: the Cartan d-connection
D̃ = θ D̃ is a unique almost symplectic d-connection which satisfies the
conditions θ D̃θ̃ = 0 and θ D̃J̃ = 0. These properties can be verified by
explicit computations using formulas (12), (15), and (A.5). Reformulating
the Einstein gravity in variables (θ̃, J̃, θ D̃), we can perform models of DQ
and A-brane quantization of GR, see Refs. [1, 2]. In this work, we shall
generalize those constructions for Ricci solitons and NC variables.
2.2. Non-holonomic Ricci solitons

We formulate the equations for Ricci solitons with respect to N-adapted
frames and discuss possible connections to field equations in GR and mod-
ified gravity theories.
2.2.1. Almost Kähler Ricci solitons

We can define a non-holonomic Einstein space for a metric compatible
d-connection D completely determined by some data (g, N) by solutions of
670 S. I. Vacaru
equations
Rαβ = λgαβ
for λ = const and coefficients determined with respect to N-adapted frames
or D = D̃/θ D̃, we get field equations for the canon-
(4) and (5). If D = D,
ical non-holonomic Einstein manifolds, or Cartan/almost Kähler–Einstein
manifolds. For D = ∇, we get usual Einstein manifolds. In this section and
Secs. 3 and 4, we shall work with D = D̃/θ D̃ which is necessary for DQ.
Exact solutions for modified Ricci solitons will be constructed in Sec. 4.2
Via non-holonomic deformations with distortions of linear con-
for D = D.
nections (11) all constructions can be mutually related.
A complete d-metric g = g̃ (6) on a smooth V is called an almost
Kähler–Ricci soliton if there exists a smooth d-vector field X for which
1
R̃αβ + (D̃α Xβ + D̃β Xα ) = λgαβ .
2
We can work with θ̃ (15) instead of g̃, for equivalent data (g̃, Ñ, D̃) ≈
(θ̃, J̃, θ D̃). It should be noted that L̃X = (D̃α Xβ + D̃β Xα ) generalizes for
D̃ the concept of the Lie derivative in the direction of X.
Considering a gradient d-vector Xβ = D̃β K(u) for some smooth func-
tion K(x, y) called the potential function, we get gradient almost Kähler–
Ricci solitons as solutions of
R̃αβ + D̃α D̃β K = λgαβ . (16)
By definition, a Ricci soliton is steady if λ = 0; shrinking, for λ > 0;
expanding, for λ < 0.
In some sense, the stationary points of geometric flows of almost Kähler
structures are characterized by “intermediary” properties between the Ein-
stein and Kähler–Einstein metrics. Taking D̃ instead of ∇ we can prove
similarly to Ref. [31] that:
(1) For solutions of (16), s R̃ + (D̃α K)(D̃α K)−2λK = C, where C = const
and the scalar curvature s R̃ is computed as in (10);
(2) Any compact steady or expanding almost Kähler–Ricci soliton must be
non-holonomic Einstein for D̃.
This allows us to conclude that Ricci solitons are natural generalizations
of Einstein metrics for various types of (non) holonomic and/or (almost)
Kähler structures even there are non-Einstein compact steady or expanding
Ricci solitons, see examples in Ref. [39]. We can relate such non-holonomic
configurations to modifications of GR.
www.ebook3000.com
2.2.2. Non-holonomic Ricci solitons and modified gravity

In order to formulate a gravity theory which would explain observational
data on accelerating Universe and dark energy/matter effects in cosmology,
there are considered modified gravity theories when the Lagrange density
R → f (R, T ) for a functional f on scalar curvature and trace of energy–
momentum tensor, see review of results in Ref. [37]. For simplicity, we con-
sider a gravity model with f (s R), when the effective vacuum gravitational
fields are
1
fR Rαβ − f gαβ + (gαβ Dγ Dγ − Dα Dβ ) fR = 0 (17)
2
for fR = ∂f /∂R. If D = ∇, we get a vacuum f (R) gravity.
We can associate Eq. (17) to a non-holonomic Ricci solitonic equation
Rαβ + Dα Dβ K = λgαβ , (18)
when K = fR and D̃ → D and g → g̃ in (16). Such non-integrable config-

urations are describe by effective gravitational field equations
Rαβ = Λ(xi , y a )gαβ , (19)
where the polarized cosmological “constant”

λ + Dγ Dγ fR − f /2
Λ= . (20)
1 − fR
Equation (19) defines non-holonomic generalizations of Einstein spaces.
A number of examples of exact solutions (in general, with NC variables)
were provided in Refs. [25,35,36], see also Sec. 4.2. Such generic off-diagonal
metrics can be generated in a “straightforward” form for configurations with
Killing symmetry, for instance, on ∂/∂y 4 when Λ ≈ Λ(xi ).
Finally, it should be concluded the Ricci solitons for non-holonomic Ricci
flows can be related to analogs of Einstein equations with polarized cosmo-
logical constants and modified locally anisotropic frame and (non)linear
connection structures. Considering quantum and/or NC versions of Ricci
solitons we study certain geometrized models of modified gravity theories.
3. DQ of Ricci Solitons
The equations for Ricci solitons can be written in terms of d-connections
The first representation defines an almost Kähler struc-
D = D̃, or = D.
ture which can be quantized following methods of DQ outlined in Ref. [6].
672 S. I. Vacaru
That program can be performed for such linear connections when torsion
is proportional to the Neijenhuis tensor with coefficient 1/4. In general, a
chosen D = {Γα βγ } may not satisfy such conditions but we can consider an

auxiliary d-connection Ď = {Γ̌α
β γ }, when
β γ α
Γ̌α α α α
β γ = ě α ěβ ěγ Γβγ + ě α eγ (ěβ ). (21)

The frame transforms ĕν = ěvv (u)eν and ĕν = ěvv (u)ev can be considered

for new sets of N-coefficients Ň = {Ňja } when Ťαβγ = (1/4)Ω̌αβγ . For this
type of transformations of connections, we have to solve some algebraic
equations for frame coefficients when for a prescribed 2 + 2 splitting the
necessary type d-connection are fixed and the torsion, Ťαβγ , and Neijenuis,
Ω̌αβγ , tensors are computed respectively following formula (19) and (13) for
re-defined N-connection and almost complex structure J̌ when J̌(ĕi ) = −ei
and J̌(ei ) = ĕi . The transforms (21) result in such torsion and curvature
coefficients:
β γ α
β γ τ α
Ťα α α α
β γ = ě α ěβ ěγ Tβγ and Ř β γ τ = ě α ěβ ěγ ěτ R βγτ , (22)
where Tα α
βγ and R βγτ are taken for D = D̃ or = D. Considering θ̌(X, Y)
:= g(J̌X, Y), we rewrite all data (θ̃, J̃, θ D̃) into equivalent ones (θ̌, J̌, Ď).
In particular, the approach can be elaborated for any (g, N, D) as solutions
of (18) or (19) when g can be represented in any necessary form (6) and/or
(A.4).
3.1. Fedosov operators and non-holonomic

Ricci solitons
Let us consider the space C ∞ (V)[[]] of “formal series” in variable with
coefficients from C ∞ (V) on a Poisson manifold (V, {·, ·}), where the bracket
{·, ·} will be defined below for non-holonomic (almost Kähler) structures.
We introduce the operator
∞

1
f ∗ 2f = 1 2
r C( f, f ) r , (23)
r=0
where r C, r ≥ 0, are bilinear operators on C ∞ (V) with 0 C(1 f, 2 f ) = 1 f 2 f

and 1 C(1 f, 2 f ) − 1 C(2 f, 1 f ) = i{1 f, 2 f }; i being the complex unity. This
defines an associative algebra structure on C ∞ (V)[[]] with a -linear and
-adically continuous star-product. Parameterizing local coordinates on T V
www.ebook3000.com
in the form (u, z) = (uα , z β ), where z β are the second-order fiber coordi-
nates, can use some elements as series of type

a(v, z) = ar,{α} (u)z {α} r , (24)
r≥0,|{α}|≥0
where {α} is a multi-index. Multiplication of such elements is considered to

define a formal Wick algebra W̌u , for u ∈ V associated with the tangent
space Tu V, with a formal Wick product

∂2
a ◦ b (z) = exp i Λ̌αβ a(z)b(z[1] ) |z=z[1]
2 ∂z α ∂z β [1]
for two elements a and b defined by formal series of type (24) and Λ̌αβ
θ̌αβ − i ǧαβ , see [1, 3, 4, 6] for necessary details on holonomic and non-
holonomic constructions which are necessary for DQ.
The d-connection (21) can extended for space to an operator on W̌ ⊗ Λ̌,
Ď (a ⊗ ξ) = (ěα (a) − uβ Γ̌γαβ z ěα (a)) ⊗ (ěα ∧ ξ) + a ⊗ dξ,
where z ěα is a similar to ěα but depend on z-variables. This operator is a

N-adapted dega -graded derivation of the distinguished algebra (W̌ ⊗ Λ̌, ◦)
(in brief, d-algebra).
Definition 3.1. The Fedosov N-adapted operators are


 i
α z −1
z α ěα (a), if p + q > 0,
δ̌(a) = ě ∧ ěα (a) and δ̌ (a) = p + q

0, if p = q = 0,
where a ∈ W̌ ⊗ Λ is homogeneous w.r.t. the grading degs and dega with

degs (a) = p and dega (a) = q.
We can consider the formula a = (δ̌ δ̌ −1 + δ̌ −1 δ̌ + σ)(a), where a −→

σ(a) is the projection on the (degs , dega )-bihomogeneous part of a of degree
zero, degs (a) = dega (a) = 0; the operator δ̌ is also a dega -graded derivation
of the d-algebra (W̌ ⊗ Λ̌, ◦).
674 S. I. Vacaru
Theorem 3.1. Any d-metric and equivalent symplectic structure, θ̌(·, ·) :=

g(J̌·, ·), defines a flat canonical Fedosov d-connection
i
Ď − δ̌ + Ď − adW ick (r).

It satisfies the condition Ď2 = 0, when a unique element r ∈ W̌ ⊗ Λ̌,
dega (r) = 1, δ̌ −1 r = 0, solves the equation
i
δ̌r = Ť + Ř + Ďr − r ◦ r.

This element is computed recursively,

i
r(0) = r(1) = 0, r(2) = δ̌ −1 Ť , r(3) = δ̌ −1 Ř + Ďr(2) − r(2) ◦ r(2) ,

k

i (l+2) (l+2)
r(k+3) = δ̌ −1 Ďr(k+2) − r ◦r , k ≥ 1.

l=0
where we denoted the Deg-homogeneous component of degree k of an ele-

ment a ∈ W̌ ⊗ Λ̌ by a(k) .
Proofs of theorems in this section are similar to those presented in non-

holonomic variables [1] for the Einstein spaces. We omit such constructions
in the work because they consist formal redefinitions of those for ∇ → D =

D̃ or = D.
3.2. Main theorems for Fedosov–Ricci solitons

For extension of Ď to W̌ ⊗ Λ̌, we can construct analogs of torsion and
curvature operators, respectively, given by 2-forms
zγ
Ť θ̌γτ Ťταβ (u) ěα ∧ ěβ , (25)
2
zγ zϕ
Ř θ̌γτ Řτϕαβ (u) ěα ∧ ěβ , (26)
4
where Ťταβ and Řτϕαβ are those from (22). By straightforward com-
putations, we can verify the properties [Ď, δ̌] =
i adW ick (Ť ) and Ď2 =
−
i adW ick (Ř). The bracket [·, ·] is the dega -graded commutator of endo-
morphisms of W̌ ⊗ Λ̌ and adW ick is defined via the dega -graded commutator
in (W̌ ⊗ Λ̌, ◦).
If Ď is taken for a solution of (non) holonomic Ricci solitonic equations
(18), the d-operators δ̌ and δ̌ −1 are for conventional Fedosov–Ricci solitonic
www.ebook3000.com
spaces. In particular, they can be considered for conventional Fedosov–

Einstein manifolds. The main purpose of DQ is to construct a star-product
which would encode the geometric structure and nonlinear dynamics of
certain fundamental field/evolution equations and constraints.
Theorem 3.2. A star-product on the almost Kähler model of a non-

holonomic Ricci soliton is defined on C ∞ (V)[[]] by formula
1
f ∗ 2 f σ(τ (1 f )) ◦ σ(τ (2 f )),
where the projection σ : W̌K D → C ∞ (V)[[]] onto the part of degs -degree
zero is a bijection and the inverse map τ : C ∞ (V)[[]] → W̌Ď can be calcu-
lated recursively w.r.t the total degree Deg,
τ (f )(0) = f,
k

(k+1) −1 (k) i (l+2) (k−l)
τ (f ) = δ̌ Ďτ (f ) − adW ick (r )(τ (f ) )
v
l=0
for k ≥ 0.
We denote by f ξ the Hamiltonian vector field corresponding to a func-

tion f ∈ C ∞ (V) on almost symplectic space (V, θ̌) and consider the anti-
symmetric part − C(1 f, 2 f ) = 12 (C(1 f, 2 f ) − C(2 f, 1 f )) of bilinear operator
C(1 f, 2 f ). A star-product (23) is normalized if 1 C(1 f, 2 f ) = 2i {1 f, 2 f },
where {·, ·} is the Poisson bracket which in our case is defined by θ̌. Sum-
marizing computations from Ref. [1] when the geometric space is a non-
holonomic Ricci soliton, we conclude that for a normalized ∗, the bilinear
operator − 2 C is a de Rham–Chevalley 2-cocycle. In such cases, there is a
unique closed 2-form κ̌ such that
1 2 1
2 C( f, f ) = κ̌(f1 ξ, f2 ξ) (27)
2
for all 1 f, 2 f ∈ C ∞ (V). We can consider the class c0 of a normalized star-
product ∗ as the equivalence class c0 (∗) = [κ̌]. This can computed as a
unique 2-form,
i
κ̌ = − J̌τα Řτα αβ ěα ∧ ěβ − i λ̌
8

for λ̌ = d µ̌, µ̌ = 16 J̌τα Ťτα β ěβ .
The canonical class ε̌ can be constructed for Ň T V = hV ⊕ vV with
the left label for a N-connection structure Ň which is related via frame
676 S. I. Vacaru
transforms, for instance, to the canonical Cartan N-connection. The com-

plexification of such non-holonomic tangent bundles can be performed in
the form TC (Ň T V) = TC (hV) ⊕ TC (vV). For such configurations, the class
ε̌ is the first Chern class of the distributions TC (N T V) = TC (hV)⊕TC (vV)
of couples of vectors of type (1, 0) both for the h- and v-parts.
The final result on DQ of Ricci solitons is to calculate the canonical
class ε̌, using Ď and the construction of ∗. It is possible to define the h-
and v-projections hΠ = 12 (Idh − iJh ) and vΠ = 12 (Idv − iJv ). The values
Idh and Idv are respective identity operators and Jh and Jv are h- and
v-components of almost complex operators, which are projection operators
onto corresponding (1, 0)-subspaces. At the next step, we construct the
matrix (hΠ, vΠ)Ř(hΠ, vΠ)T , where (. . .)T is for transposition. This is the
curvature matrix of the N-adapted restriction of Ď to TC (N T V). The final
step is to compute the closed Chern–Weyl form (see similar details in [6]
and, in N-adapted form, [1]),

T
γ̌ = −iTr (hΠ, vΠ) Ř (hΠ, vΠ) = −iTr (hΠ, vΠ) Ř
1
= − J̌τα Řτα αβ ěα ∧ ěβ .
4
So, we obtained that the canonical class is ε̌ = [γ̌] which presents a proof
of Theorem 3.3.
Theorem 3.3. The zero-degree cohomology coefficient c0 (∗) for the almost
Kähler model of a non-holonomic Ricci soliton is c0 (∗) = −(1/2i) ε̌.
Finally, we emphasize that star-products and cohomological properties

can be computed for various types of non-holonomic spaces which admit
almost Kähler variables. As a matter of principles, we do not need to solve
certain field equations in modified gravity or their Ricci soliton extensions.
The coefficient c0 (∗) encodes various quantum properties of the Ricci soli-
tons or gravitational field. Any metric defining a classical Einstein manifold
can be non-holonomically deformed into a corresponding quantum config-
uration for a Fedosov–Ricci solitonic space.
4. Non-Commutative Ricci Solitons

The data for (non-) holonomic Ricci solitons and Einstein spaces can be
encoded into almost Kähler data (θ̃, J̃, θ D̃) which can be non-holonomically
transformed and distorted to canonical ones (θ, J, D), when Dθ = 0 and
www.ebook3000.com
DJ = 0. The almost symplectic structure θ results in a non-degenerate Pois-

son structure. This allows us to define a N-adapted and covariant product
for NC.
4.1. Canonical and Cartan star-products

In this section, we shall follow the formalism elaborated in Refs. [27–29] but
modified for ∇ → D = D̃, or = D, and almost Kähler variables determined
naturally by data (g, N). The deformation parameter in formal power series
C ∞ (V)[[]] is taken = i, where i2 = −1 and where = h/2π is used for
respective Plank constants.
Definition 4.1. The canonical (Cartan) covariant star-product is defined

by formula
k
α˜
β := θµ1 ν1 . . . θµk νk (Dµ1 . . . Dµk ) · (Dν1 . . . Dνk ). (28)
k!
k
The product ˜ is adapted to N-connection structure (3) and maps

d-tensors into d-tensors. For D → ∇, it transforms into similar NC gener-
alizations of the (pseudo) Riemann geometry if θ is fixed for a symplectic
manifold, ˜ → . It is possible to define a h- and v-splitting, ˜ = (h ˜, v ˜) if
Dµ1 = (Di1 , Da1 ). For instance,
h k
α ˜
β= θi1 j1 . . . θik jk (Di1 · · · Dik ) · (Dj1 · · · Djk ).
k!
k
A similar formula for α (v ˜

)β can be written for abstract v-indices.
The star-product (28) can be expressed in the form
∞

α˜
β := αβ + k Ck (α, β), (29)
k
where the bilinear operators Ck are N-adapted, i.e. d-operators.

By straightforward computations, we can prove
Theorem 4.1. The product ˜ has such properties:
(1) Associativity, α˜
(β˜
γ) = (α˜β)˜γ;
(2) Poisson bracket,
C1 (α, β) = {α, β} = θµν Dµ α · Dν β,

678 S. I. Vacaru
with standard properties P1–P5 from Ref. [28] but distorted by (11),
here we note antisymmetry, {α, β} = −{β, α}, and the Jacoby identity,
{α, {β, γ}} + {γ, {α, β}} + {β, {α, γ}} = 0;
(3) N-adapted stability of type α˜β = α · β if Dα = 0 or Dβ = 0;
(4) The Moyal symmetry, Ck (α, β) = (−1)k Ck (β, α);
(5) N-adapted derivation with Leibniz rule,
D(α˜
β) = (Dα)˜β + α˜(Dβ)
= ((hD + vD)α) ˜β + α˜((hD + vD)β).
For applications in quantum physics, it is important the Hermitian prop-
erty, α˜
β = β˜ α. In this chapter, we work with (˜, D) in a similar man-
ner as with (, ∇) (the last variant is elaborated in details in Ref. [28])
and elaborate a d-tensor calculus, consider N-adapted gauge transforms of
star-products and restrict the gauge freedom, introduce a natural integra-
tion measure defined by θµν , perform integration, construct a “renormal-
ized” star-product and represent such star-products through a twist on a
suitable Hopf distinguished algebra. Nevertheless, there are certain differ-
ences resulting from the fact that an almost Kähler structure, in general,
is not a symplectic one. For instance, the metric compatibility condition
Dµ gαβ = 0, for some
1
gαβ = (eα ˜eβ + eβ ˜eα ) (30)
2
is not very restrictive as in the case of symplectic geometries.
Finally, we note that because θ D̃θ̃ = 0 we can write θµν ˜α = θµν · α.
Using canonical almost symplectic data (˜, θ D̃), it is possible to elaborate an
associative star-product calculus which is completely defined by the metric
structure in N-adapted form and keeps the covariant property.
4.2. Generating (non-) commutative Ricci solitons

The goal of this section is to study NC models of Ricci solitons defined by
N-adapted star-products ˜ (28). We shall prove that such PDE decouple
in a form similar to commutative cases and provided explicit examples of
exact solutions.
4.2.1. Non-commutative N-adapted frame deformations

Non-commutative structures can be introduced via “generalized uncer-
tainty” relations
α u
u β − u
β u
α = iθαβ (u), (31)
www.ebook3000.com
where uα are certain “quantum” operator analogs of coordinates on space–

time and θαβ is an anti-symmetric second-rank tensor (for some explicit
constructions θ ∼ ). In the simplest models with non-commutativity, one
considers θαβ (u) = const with respect to a local coordinate base. Such
conditions are very restrictive for various purposes in NC gravity and that
why θαβ (u) are associated to certain tensor fields, generalized symplec-
tic structures, etc. The relations (31) must be consistent with space–time
symmetries. In some theories, one considers that both gαβ and θαβ are
independent dynamical variables which should be determined from a (yet
unknown) NCly generalized version of gravity/brane/string theory and/or
Ricci flow evolutions. If we begin with a classical commutative model in
almost symplectic variables we anticipate that the NC space–time geome-
try is generated by quantum versions of gαβ and the values θαβ should be
completely determined by gαβ following certain geometric/physical princi-
ples.
Let us consider formal series non-holonomic deformations of a (pseudo)
Riemannian space, V → V using N-adapted frames with parametric depen-
dence on via certain NC values. We shall put a left bar “ ” in order
to emphasize that certain spaces and geometric objects are determined by
some NC structures generated by θαβ . We consider a constant valued matrix
h
0 θ=θ∼
θαβ = diag h ,
− θ = −θ ∼ − 0
v

0 θ=θ∼
−v θ = −θ ∼ − 0
for
uα uβ − uβ uα = iθαβ (32)
with uα ∼ uα . We begin with 4D space with conventional 2 + 2 splitting,

V, when NC non-holonomic deformation relations are stated with respect
to certain NC frames eα = ( ei , ea ). Fixing a local system of coordinates
uα = (xi , y a ) on V and ∂α ⊂ T V, we compute for the frame coefficients
α
eα = eα (u, θ)∂α such formal series
α α α
eα = eαα + iθα1 β1 eα α1 β1 + θα1 β1 θα2 β2 eα α1 β1 α2 β2 + O(θ3 ),
α
e α = eαα + iθα1 β1 eααα1 β1 + θα1 β1 θα2 β2 eααα1 β1 α2 β2 + O(θ3 ). (33)
680 S. I. Vacaru
α
The frame coefficients eα and eαα their duals are subjected to the con-
β β β
dition eαα
˜ eα = δα , for δα being the Kronecker tensor. The values
α α α
eαα1 β1 and eαα1 β1 α2 β2 can be written in terms of eα , θαβ and the spin dis-
tinguished connection corresponding to D or D̃, if such constructions are
preformed with generalized Dirac operators and fermions.
Via deformations (33) of a commutative metric g, we generated a NC
“target” metric of type (30),
h 1 β + β
gαβ = g, v g = ηαβ [ eαα ˜ eβ + eβ ˜ ( eαα )+ ], (34)
2
where (. . .)+ denotes the Hermitian conjugation and ηαβ is the flat Minkow-
ski space metric. We have developed [25, 26, 35, 36] a geometric method of
constructing exact solutions of gravitational and matter field equations in
different theories of gravity. The metric in such theories is parameterized
by ansatz of type (34) (commutative “prime” metric being of type (A.4)).
For certain assumptions, we can introduce into consideration any type of
parameters for commutative groups, supersymmetric generalizations, quan-
tum groups etc. in Einstein gravity and (non)commutative string, gauge,
Finsler and other generalizations.
The generalized N-adapted star-product encodes all the information of a
NC differential d-tensor calculus. We should follow a few simple rules that
the h-/v-adapted transformation of individual d-tensors is not deformed,
the geometric objects like d-tensors, d-vectors, etc. must be multiplied
via star-products and such products are transformed using the twisted N-
adapted Leibniz rule.
4.2.2. Non-commutative N-adapted connections

Considering any commutative linear connection, and d-connection, struc-
ture D = {Γβαγ } we can transform it into a NC one, D = { Γβαγ }, using
the star-product ˜
(28),
β ∂X β
Dα ˜
X = + X γ ˜ Γβαγ .
∂uα
We follow the geometric rule: take the partial derivatives as for commutative
spaces but twist the products via ˜ when the product results in series of
∞
type (29), for instance, X γ ˜ Γβαγ := X γ Γβαγ + k k Ck (X, Γ).
www.ebook3000.com
If V is endowed with N-connection structure, we can define D → D via

NC non-holonomic deformations (33) with respect to N-adapted frames.
For instance, for the canonical d-connection, we get
1 ταβ + Γ
γ ˜ gτ γ ),
Γ γαβ = ( gγτ ˜
Γ αβ
2

where Γ γ , L
= ( L i , C
a , C
i ), are computed
a
αβ jk bk jc bc
i 1
Ljk = g ir (h ˜) [ ek (v ˜) gjr + ej (v ˜) gkr − er (v ˜) gjk ],
2
a 1
Lbk = eb ( Nka ) + hac [ ek (v ˜) hbc
2
− hdc (v ˜ )eb ( Nkd ) − hdb (v ˜)ec Nkd ],
i 1 a = 1 had (v ˜)(ec hbd + ec hcd − ed hbc ).
Cjc = g ik (h ˜)ec gjk , C bc
2 2
(35)
In the zero approximation on powers on , we get the formulas for the
commutative analogs (A.3).
Applying N-adapted NC frames and star-products, we can define and
compute the NC expressions of curvature Rρσµν and torsion Tγαβ of a
general metric compatible d-connection D if its commutative prime ana-
log D is completely defined by the metric structure and N via distorting
relations of type (11),
[ Dµ , ˜ Yρ = Yρ ˜ Rρσµν − Tγµν ˜ ( Dµ ˜ Yρ )
Dν ] ˜
for an arbitrary d-vector Yρ = ( Y i , Y a ) ∈ T ( V), The non-trivial
h–v-components are computed: for the NC d-torsion,
i
T jk = Li jk − Li kj , T ija = − T iaj = C ija , T aji = Ωaji ,
a
T bi = − T aib = ∂b Nia − Labi , T abc = C abc − C acb
and for the NC curvature,
i
R hjk = ek Li hj − ej Li hk + Lmhj (h ) Li mk
− Lmhk (h ) Li mj − C iha Ωakj ,
a
R bjk = ek Labj − ej Labk + Lcbj (v ) Lack
− Lcbk (v ) Lacj − C abc Ωckj ,
i
R jka = ea Li jk − Dk (h ) C ija + C ijb (v ) T bka ,
c
R bka = ea Lcbk − Dk (v ) C cba + C cbd (v ) T cka ,
i
R jbc = ec C ijb − eb C ijc + C hjb (h ) C ihc − C hjc (h ) C ihb ,
a
R bcd = ed C abc − ec C abd + C ebc (v ) C aed − C ebd (v ) C aec .
682 S. I. Vacaru
These formulas transforms, respectively, into their commutative analogs

(A.1) and (A.2) if → 0.
4.2.3. Non-commutative Ricci soliton and Einstein

equations
The NC deformations of the Ricci tensor (9) and scalar (10) can be found
following a tedious NC calculus and contracting of indices and star-products
to inverse metric. In order to construct exact solutions, it is convenient to
work with the canonical d-connection D.h Using the principle of general
commutative covariance, we can consider that all NC geometric objects
possess certain general parametric dependence on θ, for instance, in the
form
∞

gαβ = gαβ (u, θ) = gαβ (u) + θk Ckαβ ( eγ ; g),
k
∞

Rαβ ≡ Rαβ (u, θ) = Rαβ (u) +
kαβ ( eγ ; g).
θk C
k
The Ricci solitonic/field equation (19) for NC metric compatible d-

connections can be written in the form
Rij = Λ(xi , y a ) gij , (36)

i a
Rab = Λ(x , y ) gab , (37)
Rai = 0, (38)
Ria = 0, (39)
where the NCly modified/polarized cosmological constant (20) has to be

computed recurrently in the form
λ + Dγ Dγ fR − f /2
Λ = .
1 − fR
h The concept of exact solutions should be discussed here, for instance, for gravity models
with ˜ (28). If an exact solution for certain generalized Ricci soliton/Einstein equations
was found for commutative configurations, we can use NC frame decompositions of type
(33). This way we generate analogs both for certain NC fundamental geometric/physical
equations and their formal solutions as series decompositions on powers of θ. There are
necessary additional considerations in order to conclude if the new classes of NC solutions
converge and may have certain physical importance. For simplicity, we can consider
that θ-corrections are very small in the vicinity of certain prescribed “commutative”
symmetries and well-defined classical solutions.
www.ebook3000.com
We can choose such non-holonomic distributions and NC deformations

g to g when
gij = gij (u) + g̊ij (u)θ2 + O(θ4 ), hab = hab (u) + h̊ab (u)θ2 + O(θ4 ),
3
Ni = wi (u) + ẘi (u)θ2 + O(θ4 ), Ni4 = ni (u) + n̊i (u)θ2 + O(θ4 ). (40)
In general, various commutative and NC modifications of Ricci soli-

ton and/or gravitational field equations depend on “sophisticate” rules
of adapted geometric objects, modified Lagrangians, definition of star-
products, etc. This results in less defined classes of PDE with complex
parameters, singularities, non-trivial topological structure, NC configura-
tions. Using decompositions of type (40), we can approximate such PDE to
well-defined nonlinear systems with dependence on certain real/imaginary
small parameters. Applying the principle of generalized covariance on such
complex/real non-holonomic manifolds, we “mix” all information on star-
products and non-holonomic distributions when certain polynomial approx-
imations on θ0 and θ2 can be considered for constructing solutions of certain
geometric/physically important systems of PDE.
4.3. Integrability of NC Ricci solitons

For 4D Ricci solitons and Einstein manifolds, the decoupling property can
be proven in a straightforward form for generic off-diagonal metrics with
one Killing symmetry, for instance, on ∂/∂y 4 . There are several important
results due to Geroch [43, 44] which show how using metrics with Killing
symmetries it is possible to generate new classes of solutions of vacuum
Einstein equations with, in general, a finite/infinite number of constant
parameters. In Ref. [36], we generalized the constructions following the
so-called anholonomic deformation method when the solutions depend on
generating and integration functions on 1–4 variables and, as a matter of
principle, may depend on an infinite number of commutative and NC con-
stant (real or complex) parameters. In particular, we can consider that a
NC structure with θ = const is such an example.
can be integrated in
In this section, we show how Eqs. (36)–(39) for D
very general off-diagonal forms for certain ansatz for (30) parameterized in
the form
k
k 3
gαβ (x , y , θ) = diag{ gi (xk , θ) = i e ψ(x ,θ)
= gi (xk ) + g̊i (xk )θ2 + O(θ4 ),
k 3
ha (x , y , θ) = ha (xk , y 3 ) + h̊a (xk , y 3 )θ2 + O(θ4 )},
684 S. I. Vacaru
3 k 3
Ni (x , y , θ) = wi (xk , y 3 , θ) = wi (xk , y 3 ) + ẘi (xk , y 3 )θ2 + O(θ4 ),
4 k 3
Ni (x , y , θ) = ni (xk , y 3 , θ) = ni (u) + n̊i (u)θ2 + O(θ4 ) (41)
and approximating Λ ≈ Λ(xk , θ) via a corresponding non-holonomic dis-

tribution. The values i = ±1 depend on chosen signature of metric for
θ → 0.
4.3.1. Decoupling with respect to N-adapted frames

A straightforward computation of the NC Ricci tensor R αβ (u, θ) in N-
adapted frame coordinates for ansatz (41) (see similar details in Refs. [25,
26, 36]), allows us to write (36)–(39) respectively as such a system of PDE
with parametric dependence on θ:
1 ψ •• + 2 ψ = Λ, (42)

∗ ∗
φ (ln | h4 |) = Λ h3 , (43)
β wi + αi = 0, (44)
∗∗ ∗
ni + γ ni = 0, (45)
when the partial derivatives are written in brief as a• = ∂a/∂x1 , a =

∂a/∂x2 , a∗ = ∂a/∂y 3 . The coefficients are determined in the form
γ = (ln | h4 |3/2 − ln | h3 |)∗ , αi = h∗4 ∂i φ, β = h∗4 φ∗ , (46)
when φ is given by h3 and h4 via

φ = ln |2(ln | h4 |)∗ | − ln | h3 |. (47)
All values in above equations can be decomposed in formal series with

coefficients before even powers on θ.
The system of Eqs. (42)–(45) reflects a very important decoupling prop-
erty of the Ricci solitonic and non-holonomic Einstein equations which holds
true both for commutative and NC models of gravity. Let us briefly explain
this construction. Equation (42) is a 2D D’Alembert, or Laplace, equation
with source Λ(xk , θ). We can approximate the solution as a polynom on θ0
and θ2 . The system of Eqs. (43) and (47) with partial derivatives ∗ = ∂/∂y 3
is for three unknown functions h3 (xk , y 3 , θ), h4 (xk , y 3 , θ) and φ(xk , y 3 , θ)
if a source Λ(xk , θ) is prescribed. Equation (44) algebraic for wi . Equa-
tion (45) contains only the first and second derivatives on ∂/∂y 3 of ni .
Such equations can be integrated for any found h3 and h4 when we can
compute the coefficients αi , β and γ, see formulas (46).
www.ebook3000.com
Constraints for the Levi-Civita connection ∇: For ansatz of

can be satisfied if
type (41), the zero-torsion constraints in (11), for D,
∗
wi ln | h4 |, ∂i wj = ∂j wi , n∗i = 0.
= ei ˜ (48)
The first condition introduces certain coupling for Eqs. (43) and (44). Nev-
ertheless, we can solve such conditions in explicit form via additional frame
and coordinate transforms and/or re-parametrization of generating func-
tions etc.
4.3.2. Constructing integral varieties

Let us perform a formal integration of the effective gravitational equations.
The horizontal metric is determined by ψ(xk , θ, Λ) as solutions of a 2D
linear Eq. (42) with source Λ(xk , θ).
For φ∗ = 0 and introducing variables P := (ln | h4 |)∗ and Q = | h3 |,
we can transform the systems (43) and (47) into an algebraic one,
∗
φ P = ΛQ2 , Qe φ = 2P. (49)
Excluding P, we express Q = (e φ )∗ /2Λ, i.e.

(xk , θ)
| h3 (xk , y 3 , θ)| = |0 h3 (xk , θ)| + Φ∗
2Λ(xk , θ)
for a redefined generating function Φ(xk , y 3 , θ) := e φ . The solution can be

written as
2
0 1 + Φ∗
h3 = h3 . (50)
2Λ |0 h3 |
Using the second equation in (49) and integrating on y 3 , we find

2
0 Φ
h4 = h4 exp ,
8Λ
where 0 h4 = 0 h4 (xk , θ) is an integration function.

Having defined the coefficients of d-metric we can compute the
N-connection coefficients. Equations (44) for (46) are solved
−∂i φ −∂i ( Φ)
wi = ∗
= ,
φ ( Φ)∗
686 S. I. Vacaru
Then, integrating two times on y 3 in (45), we find

1 2 h3
nk = nk + nk dy 3
( | h4 |)3
2
0 ∗
1 2 h3 3 Φ 3 Φ2
= nk + nk 0 3/2 dy 1 + exp − , (51)
| h4 | 2Λ |0 h3 | 16Λ
where 1 nk (xi , θ) and 2 nk (xi , θ) are integration functions.

The above solutions for coefficients determine a quadratic metric ele-
ment with parametric dependence on θ (for applications, we can consider
only terms θ0 and θ2 ),
2 2
φ ∗
2 ψ i 2 0 (e ) 3 ∂i φ i
ds = i e (dx ) + h3 1 + dy − ∗ dx
2Λ |0 h3 | φ
2 φ 2
e h 3
+0 h4 exp dy 4 + 1 nk + 2 nk dy 3 dxi . (52)
8Λ ( | h4 |)3
Such metrics depend on generating functions φ(xi , y 3 , θ) and

k 0 k 0 k 1 k 2 k
ψ[Λ(x , θ)] and integration functions h3 (x ), h4 (x ), nk (x ), nk (x ) as
we described in the above formulas. For small values of θ, we can approxi-
mate the solutions as power series on NC parameter.
4.4. Black ellipsoids and solitonic waves as

NC Ricci solitons
Metrics with spherical symmetries and solitonic configurations are typical
ones which are considered in (modified) gravity theories in order to study
possible observational/experimental objects. Deformations of symmetries
can be used for a comparative study of different types of theories or prop-
erties of new classes of solutions in the same theory. In Refs. [24–26, 36],
we studied off-diagonal deformations of black hole solutions into certain
ellipsoid and solitonic space–time metrics which can be derived as exact
solutions in GR and various commutative and NC Finsler-like theories.
Such small ellipsoid and solitonic configurations can be defined to be stable
and seem to have important physical implications. In this section, we shall
construct solutions of type (52) for Eqs. (42)–(45) for NC Ricci solitons.
www.ebook3000.com
α 1 2 3 4
Let us consider spherical
coordinates u = (x , x = ϑ, y = ϕ, y =
1

t), when x = ξ = dr/ |q(r)| for a usual radial coordinate r, on a
commutative (pseudo) Riemannian manifold V. A diagonal metric
◦
g = dξ ⊗ dξ + r2 (ξ) dϑ ⊗ dϑ + r2 (ξ) sin2 θdϕ ⊗ dϕ − q(ξ) dt ⊗ dt, (53)
2
defines an empty de Sitter space if q(r) = 1− 2 m(r) r − λ r3 , where λ is a
cosmological constant. The total mass energy within the radius r is defined
by a function m(r). For m(r) =0, we obtain an empty space with a cos-
mological horizon at r = rc = 3/λ. Possible modifications by NC Ricci
solitons are determined by an effective cosmological constant of type (20).
In this section, we analyze two classes of solutions related to possible
NC Ricci soliton modifications of GR when λ → Λ(xk , θ).
4.4.1. Non-commutative Ricci solitonic black ellipsoids

The generic off-diagonal ansatz (41) is chosen
= ηi ◦ gi , ha = ηa ◦ ha , Ni3 = ε wi (ξ, ϑ, ϕ, θ) , Ni4 = ε ni (ξ, ϑ, ϕ, θ),
gi
(54)
where the polarization functions
η1 = exp[ε ψ(xk , θ, Λ)], η2 = r−2 (ξ) exp[ε ψ(xk , θ, Λ)],

η3 = 1 + ε χ3 (xk , ϕ, θ, Λ), η4 = 1 + ε χ4 (xk , ϕ, θ, Λ) (55)
and the “prime” metric coefficients are determined by (53) when ◦ g1 = 1,
◦
g2 = r2 (ξ), ◦ h3 = r2 (ξ) sin2 ϑ, ◦ h4 = −q(ξ). The polarizations ηα (55)
and N -coefficients will be constructed in such a form that (54) will define
metrics of type (52). We consider a small parameter which for ε → 0, with
Λ ≈ λ + ε1 Λ(xk , θ), transforms a class of generic off-diagonal solutions
into the black hole metric (53). A subclass of solutions can be obtained for
ε = θ2 .
The data (54) generate solutions with parametric dependencies if

••
ψ + ψ = 1 Λ;
2 2 φ
◦ (e φ )∗ ◦ e
h3 = h3 1 + , h4 = h4 exp
◦
2Λ | h3 | 8Λ

−ε∂i ( φ) 1 2 h3
ε wi = ; ε ni = ε ni + ni dϕ
( φ)∗ ( |h4 |)3
(56)
688 S. I. Vacaru
for non-zero ha and h∗a , and (integrating) functions 1 ni (ξ, ϑ, θ), 2 ni (ξ, ϑ, θ)
and generating function φ(ξ, ϑ, ϕ, θ). The solutions (56) still define very
general off-diagonal deformations of the prime de Sitter black hole met-
ric with less clear physical implications. We can choose such generating
functions and parametrization when the target solutions may possess well-
defined symmetries and properties which are very similar to black hole
solutions.
Let us construct and analyze a class of rotoid de Sitter-like configu-
rations generated by small deformations with ε = θ2 by a NC Ricci soli-
ton. Applying the anholonomic frame method is convenient to transform h-
coordinates in a form (ξ, ϑ) → (ξ, ϑ̃(ξ, ϑ)) when dξ 2 + dϑ̃2 = dξ 2 +r2 (ξ) dϑ2 .
We parameterize data (56) for a solution (52) in the form
2
rot ψ(ξ,ϑ̃) 2 2 2 2 (e φ )∗
λ g = e (dξ + dϑ̃ ) + r (ξ) sin ϑ(ξ, ϑ̃) 1 +
2Λ |◦ h3 |

2
eϕ ⊗ eϕ − q(ξ) + θ ζ(ξ, ϑ̃, ϕ) et ⊗ et ,

2 ∂ξ φ ∂ϑ φ
e
ϕ = dϕ − θ dξ + dϑ ,
∂ϕ φ ∂ϕ φ

2
et = dt + θ n1 (ξ, ϑ̃, ϕ)dξ + n2 (ξ, ϑ̃, ϕ)dϑ , (57)
where ni are computed as ni ∼ θ2 . . . for corresponding coordinates and

values h3 and h4 . Prescribing
ζ = ζ(ξ, ϑ̃) sin(ω0 ϕ + ϕ0 ), (58)
as a generating function for some constant parameters ω0 and ϕ0 , we

generate rotoid configurations. For simplicity, we can consider ζ(ξ, ϑ̃)
ζ = const. The smaller “horizon” (when the term before et ⊗ et became
2
h4 = 0) is described by formula r+ 2 m0 /(1 + θ ζ sin(ω0 ϕ + ϕ0 )).
The generating function φ(ξ, ϑ̃, ϕ) contained in (56) is related to
ζ(ξ, ϑ̃, ϕ) via formula

1 − θ2 ζ
e2 φ = 8Λ ln
q(ξ)
for which a rotoid configuration (58) can be fixed. In general, such

off-diagonal deformations do not result in other classes of black hole solu-
tions but for small values θ2 we generate black ellipsoid solutions stud-
ied in Refs. [24–26, 36], see also references therein on stability conditions
www.ebook3000.com
and differences in various classes of gravity theories. Imposing additionally

the conditions (48) on the integral variety of such solutions, we generate
exact off-diagonal solutions for the NC Ricci solitons with the Levi-Civita
connection ∇.
4.4.2. NC Ricci solitonic black holes and “non-Ricci ”

solitonic backgrounds
We studied stationary ellipsoidal configurations for NC Ricci solitons. The
solutions can be generalized to non-stationary off-diagonal models with
induced torsion and depending on time solitonic waves (which are different
from Ricci solitons but included into a general Ricci solitonic deformation).
The local coordinates are paremetrized uα = (x1 , x2 = ϑ, y 3 = t, y 4 = ϕ)
in order to get dependence on time-like coordinate y 3 = t.
A nonlinear time propagating NC Ricci soliton moved by a solitonic
waves can be described by such a non-stationary ansatz
 2
φ
2
2 ψ 2 2  ∂t e  ∂ξ φ ∂ϑ φ
ds = e [dξ + dϑ̃ ] − q 1 + dt − dξ − dϑ̃
2Λ |q(ξ)| ∂t φ ∂t φ
2 φ
2 2 e 1 2 h3
+ r (ξ) sin ϑ̃ exp dϕ + n1 + n1 dt dξ
8Λ ( |h4 )3
2
1 2 h3
+ n2 + n2 dt dϑ̃ (59)
( |h4 |)3
for local coordinates x1 = ξ, x2 = ϑ̃, y 3 = t, y 4 = ϕ and q(ξ) = q(r(ξ))

computed as in (53) but the generating functions are of type ψ(ξ, ϑ̃, θ) and
1
φ(ξ, ϑ̃, t, θ) for ha (ξ, ϑ, t, θ); the integration functions are ni (ξ, ϑ̃, θ) and
2
ni (ξ, ϑ̃, θ), when y 3 = t and ( φ)∗ = ∂t φ.
Ricci solitons induced by solitonic backgrounds with radial

Burgers equation: We can prescribe 3D solitonic waves with φ =
η(ξ, ϑ̃, t, θ), when y 3 = t is a time-like coordinate. The function η depends
on a NC parameter θ introduced in a solution of Kadomtsev–Petviashvili,
KdP, equation [45, 46],

±η + (∂t η + ηη • + η ••• )• = 0 (60)
with dispersion and possible dependencies on a set of parameters θ. In the
dispersionless limit → 0 the solutions transforms in those for the Burgers’
equation ∂t η + ηη • = 0. Introducing generating functions φ determined by
690 S. I. Vacaru
solutions of such 3D solitonic equations in (59), we generate solitonic non-

holonomic deformations of the de Sitter black hole solutions. In general, the
new off-diagonal solutions do not have black hole properties. Such configura-
tions always define exact solutions of gravitational field equations (36)–(39)
for D. Constraining the solitonic integral varieties via conditions (48), we
generate solutions for the Levi-Civita connection ∇.
Generating Ricci solitons from solitonic backgrounds with

angular Burgers equation: Solitonic background variations are
angular if φ = η(ξ, ϑ̃, t, θ) is a solution of KdP equation
η •• + (∂t η + η η +
± η ) = 0. (61)
In the dispersionless limit → 0, the solutions are independent on x1 = ξ

and transform into those for Burgers’ equation ∂t η + η η = 0. Introducing
in (59), we generate solutions of (36)–(39) with angular anisotropy.
φ = η
For small values of η, we can models 3D solitonic and NC Ricci solitonic
deformations/polarizations of the de Sitter black holes.
5. Concluding Remarks
The almost Kähler (symplectic) framework presented in this chapter is
very broad. It was used to provide an unified geometric formalism to non-
holonomic Ricci solitons and generalized/modified Einstein spaces, to per-
form DQ of such models and consider NC extensions with associative star-
products, and develop a method of constructing generic off-diagonal solu-
tions of nonlinear PDE with NC parameters. Our general goal was to prove
that certain branches of modern geometry and physics can be treated in
the same commutative and NC geometric “language” with non-holonomic
parametric deformations, solutions with small parametric limits to phys-
ical important black hole and solitonic configurations, well-defined star-
products, etc.
Some aspects of our approach are also reminiscent to geometric
methods of quantization; problems of renormalization, Ricci flows and
their fixed stable configurations with associated non-holonomic struc-
ture and deformations of exact solutions in gravity; questions of gen-
eral integrability of PDE related to modern gravity and geometric anal-
ysis. We have not imposed the condition that the metric in field space
is positive definite but considered commutative and NC deformations
www.ebook3000.com
related to generic off-diagonal and non-holonomic configurations for Ricci

solitons and effective Einstein equations with “polarized” gravitational
constants.
Another aspect of the present work is the potential interest for mathe-
matical relativity and modified theories of gravity. The anholonomic frame
method of constructing generic off-diagonal solutions depending via gen-
erating/integration functions on all coordinates was generalized to include
NC parameters and almost symplectic variables. Such constructions also
model solutions of Ricci solitonic equations. The main physical problem
is to provide a realistic interpretation for the general classes of solutions
if they do not limit certain asymptotically flat space–times and/or well
known metrics in GR. The issue of Ricci solitons in connection to renor-
malized models of QG, for instance, with local anisotropies remains still
unexplored, even we can construct very sophisticate fixed/stable configura-
tions with NC variables.
Finally, we comment on possible use of our Ricci solitonic solutions
in order to encode geometric data for (non-) commutative spaces into
bi-Hamilton structures, curve flows and solitonic hierarchies [46]. This
program of transforming Ricci solitons into “non-Ricci” solitonic config-
urations is possible mathematically and seems to be important in gen-
erating exact solutions with various generalized commutative and NC
symmetries and methods of geometric quantization. Such issues will con-
sist a background for our “almost immediate” attention and further
research.
Acknowledgments
SV research is partially supported by the Program IDEI, PN-II-ID-PCE-
2011-3-0256 and visiting research fellowship at CERN.
Appendix A. N-adapted Coefficient Formulas

We provide some necessary N-adapted components formulas, see Refs. [25,
36, 40] for reviews of results and more details.
The non-trivial coefficients of torsion (7) and curvature (8) are
hT α = {T ijk = Li jk − Li kj }, vT α = {T abc = C abc − C acb },

T ija = C ija , T aji = Ωaji , T abi = ∂b Nia − Labi (A.1)
692 S. I. Vacaru
and

Ri hjk = ek Li hj − ej Li hk + Lmhj Li mk − Lmhk Li mj − C iha Ωakj ,

Rabjk = ek Labj − ej (Labk ) + Lcbj Lack − Lcbk Lacj − C abc Ωckj ,
Ri jka = ea Li jk − Dk C ija + C ijb T bka ,
Rcbka = ea Lcbk − Dk C cba + C cbd T cka ,
Ri jbc = ec C ijb − eb C ijc + C hjb C ihc − C hjc C ihb ,
Rabcd = ed C abc − ec C abd + C ebc C aed − C ebd C aec . (A.2)
Contracting indices R βγ := Rαβγα , one gets the N-adapted coefficients

for the Ricci tensor Ric := {Rβγ = (Rij , Ria , Rai , Rab )}.
With respect to N-adapted frames (4) and (5), the canonical d-
connection D = {Γ γ = (L i , L
a , Ci , C
a )} from (11) is determined by
αβ jk bk jc bc
coefficients

i = 1 g ir ek gjr + ej gkr − er gjk ,
L jk
2

L a = eb (N a ) + 1 hac ek hbc − hdc eb N d − hdb ec N d ,
bk k k k
2
Ci = 1 g ik ec gjk , C
a = 1 had (ec hbd + ec hcd − ed hbc ), (A.3)
jc bc
2 2
where
g = gij dxi ⊗ dxj + hab ea ⊗ eb ,

ea = dy a + Nia dxi (A.4)
which via frame transforms can be represented equivalently in the form

(6). Introducing values (A.3) in (A.1), we can check by straightforward
computations that for T γ = {Ti , Ti , Ta , Ta , Ta } the horizontal and
αβ jk ja ji bi bc
vertical coefficients, Ti = 0 and Ta = 0 and Dg
jk
= 0.
bc
Now we show how the coefficients of the Cartan d-connection can be
computed. There is a unique normal d-connection
n
D = {n Γα i v 2+i
βγ = (Ljk , L2+j
ijk ; C
=L jc
i 2+i , v C
= vC bc
a bc
=C a
)}
2+k 2+j c
which is metric compatible, n Dk gij = 0 and n Dc hij = 0, and

completely defined as n Dα = (n Dk , n Dc ), with N-adapted coefficients
www.ebook3000.com
Γ i , v C
α = (L a ), where
βγ jk bc
ijk = 1 g ih (ek gjh + ej ghk − eh gjk ),

L
2

i 1 ih ∂hjh ∂hhk ∂hjk
Cjk = h + − (A.5)
2 ∂y k ∂y j ∂y h
are computed for a d-metric (A.4). Introducing in (A.5) the values gjh =
hjh = h̃jh and Nia = Ñia as for (6), we get the coefficients Γ̃α i a
βγ = (L̃jk , C̃bc )
of the Cartan d-connection D̃ which is contained in the second distortion
relation in (11).
Finally, we note that introducing Γ γ = (L i , L
a , C
i , C
a ), or Γ̃α =
αβ jk bk jc bc βγ
(L̃ijk , C̃bc
a
), into, respectively, (A.1) and (A.2) we compute the N-adapted
coefficients of torsion and curvature of D or D̃.
References
1. S. Vacaru, Einstein gravity as a nonholonomic almost Kaehler geometry,
Lagrange–Finsler variables, and deformation quantization, J. Geom. Phys.
60, 1289–1305 (2010).
2. S. Vacaru, Branes and quantization for an A-model complexification of Ein-
stein gravity in almost Kaehler variables, Int. J. Geom. Meth. Mod. Phys. 6,
873–909 (2009).
3. B.V. Fedosov, Deformation quantization and asymptotic operator represen-
tation, Funktional Anal. i Prilozhen. 25, 1984–1994 (1990).
4. B.V. Fedosov, A simple geometric construction of deformation quantization,
J. Diff. Geom. 40, 213–228 (1994).
5. C. Castro, W-geometry from Fedosov’s deformation quantization, J. Geom.
Phys. 33, 173–190 (2000).
6. A.V. Karabegov and M. Schlichenmaier, Almost Kähler deformation quanti-
zation, Lett. Math. Phys. 57, 135–148 (2001).
7. S. Vacaru, Deformation quantization of almost Kaehler models and
Lagrange–Finsler spaces, J. Math. Phys. 48, 123509 (2007).
8. S. Gukov and E. Witten, Branes and Quantization, arXiv: 0809.0305, 2009.
9. S. Vacaru, Two-connection renormalization and nonholonomic gauge models
of Einstein gravity, Int. J. Geom. Meth. Mod. Phys. 7, 713–744 (2010).
10. S. Vacaru, Modified dispersion relations in Horava–Lifshitz gravity and
Finsler brane models, Gener. Relat. Grav. 44, 1015–1042 (2012).
11. S. Vacaru, Covariant renormalizable anisotropic theories and off-diagonal
Einstein–Yang–Mills–Higgs solutions, EPL 96, 50001 (2011).
12. S. Vacaru, Spectral functionals, nonholonomic Dirac operators, and NC Ricci
flows, J. Math. Phys. 50, 073503 (2009).
694 S. I. Vacaru
13. T.A. Bhuyain and M. Marcolli, The Ricci flow on NC two–tori, (2012), arXiv:
1107.4788.
14. E. Bianchi and C. Rovelli, A note on the geometrical interpretation of quan-
tum groups and NC spaces in gravity, Phys. Rev. D 84, 027502 (2011).
15. M. Kober, Canonical noncommutativity algebra for the tetrad field in general
relativity, Class. Quant. Grav. 28, 225021 (2011).
16. H. Seok Yang and M. Sivakumar, Emergent gravity from quantized space–
time, Phys. Rev. D 82, 045004 (2010).
17. V. Sahakian, Transcribing space–time data into matrices, JHEP 0106, 037
(2001).
18. F. Ardalan, H. Arfaei, M.R. Garousi, and A. Ghodsi, Gravity on NC D-
branes, Int. J. Mod. Phys. A 18, 1051–1066 (2003).
19. P. Aschieri, F. Lizzi, and P. Vitale, Twisting all the way: From classical
mechanics to quantum fields, Phys. Rev. D 77, 025037 (2008).
20. M.A. Cardella and D. Zanon, Noncommutative deformation of four-
dimensional Einstein gravity, Class. Quant. Grav. 20, L95–L104 (2003).
21. R.J. Szabo, Symmetry, gravity and noncommutativity, Class. Quant. Grav.
23, R199–R242 (2006).
22. P. Aschieri, C. Blohmann, M. Dimitrijevic, F. Meyer, P. Schupp, and J. Wess,
Class. Quant. Grav. 22, 3511–3532 (2005).
23. M. Chaichian, M. Oksanen, A. Tureanu, and G. Zet, Covariant star product
on symplectic and Poisson spactime manifolds, Int. J. Mod. Phys. A25, 3765–
3796 (2010).
24. S. Vacaru, Gauge and Einstein gravity from non-Abelian gauge models on
NC spaces, Phys. Lett. B 498, 74–82 (2001).
25. S. Vacaru, Exact solutions with NC symmetries in Einstein and gauge gravity,
J. Math. Phys. 46, 042503 (2005).
26. S. Vacaru, Finsler black holes induced by NC anholonomic distributions in
Einstein gravity, Class. Quant. Grav. 27, 105003 (2010).
27. D.V. Vassilevich, Diffeomorphism covariant star products and NC gravity,
26, 145010 (2009).
28. D.V. Vassilevich, Tensor calculus on noncommtuative spaces, Class. Quant.
Grav. 27, 095020 (2010).
29. D.V. Vassilevich, Is covariant star product unique? (2011), arXiv: 1101.4642.
30. R.S. Hamilton, Three manifolds of positive Ricci curvature, J. Diff. Geom.
17, 255–306 (1982).
31. R.S. Hamilton, Formation of singularities in the Ricci flow, Surveys in Dif-
ferential Geometry 2, 7–136 (1995).
32. G. Perelman, The entropy formula for the Ricci flow and its geometric appli-
cations, (2002), math.DG/ 0211159.
33. G. Perelman, Ricci flow with surgery on three-manifolds, (2003),
math.DG/0303109.
34. G. Perelman, Finite extinction time for the solutions to the Ricci flow on
certain three-manifolds, (2003), math.DG/0307245.
35. S. Vacaru, On general solutions in Einstein gravity, Int. J. Geom. Meth. Mod.
Phys. 8, 9–21 (2011).
www.ebook3000.com
36. S. Vacaru, Parametric nonholonomic frame transforms and exact solutions

in gravity, Int. J. Geom. Meth. Mod. Phys. 4, 1285–1334 (2007).
37. S. Nojiri and S.D. Odintsov, Unified cosmic history in modified gravity:
from F(R) theory to Lorentz non-invariant models, Phys. Rept. 505, 59–144
(2011).
38. R.S. Hamilton, The Ricci flow on surfaces, Contemp. Math. 71, 237–261
(1988).
39. H.-D. Cao, Recent progress on Ricci solitons, Adv. Lect. Math. (ALM) 11,
1–38 (2009), arXiv: 0908.2006.
40. S. Vacaru, Nonholonomic Ricci flows: II. Evolution equations and dynamics,
J. Math. Phys. 49, 043504 (2008).
41. I. Bakas, F. Bourliot, D. Lust, and M. Petropoulos, Geometric flows in
Horava–Lifshitz gravity, (2010), arXiv: 1002.0062.
42. I. Bakas, D. Orlando, and P.M. Petropoulos, Ricci flows and expansion in
axion–dilaton cosmology, JHEP 0701, 040 (2007).
43. R. Geroch, A method for generating solutions of Einstein’s equations, J.
Math. Phys. 12, 918–925 (1971).
44. R. Geroch, A method for generating new solutions of Einstein’s equations. II.
J. Math. Phys. 13, 394–404 (1972).
45. B.B. Kadomtsev and V.I. Petrviashvili, On the stability of solitary waves in
weakly dispersive media, Doklady Akademii Nauk SSS 192, 753–756 (1970)
[in Russian]; Sov. Phys. Dokl. 15, 539–541 (1970). [English translation].
46. S. Vacaru, Curve flows and solitonic hierarchies generated by Einstein met-
rics, Acta Appl. Math. 110, 73 (2010).
www.ebook3000.com
March 8, 2016 8:12 Beyond Peaceful Coexistence 9in x 6in b2362-index page 697
Index
A annihilation operators, 628, 635, 641,

645, 658
(angular) variables, 117 anomalous magnetic moment, 400
a quantum of time, 397 anti-periodicity, 124
Abraham–Lorentz theory, 396 anti-self-dual solutions, 419
Abraham–Lorentz–Dirac, 399 antimatter, 102, 109, 125, 130
abstract language, 230 antiparticles, 109, 637
accelerating Universe, 671 antipode, 194
action, 621, 630, 632–633, 644, 647, archaic (pre-)vacuum, 564
649 Archaic Holographic Principle, 559
action principle, 308, 315, 620, Archaic Holography, 561, 575
647 archaic phase, 564
adjacency matrix, 338 archaic precursor of time and space,
Anti-de Sitter/Conformal Field xvii
Theory (AdS/CFT), 95, 132, 172 archaic space
advanced formulation of the electron wave function in archaic space,
theory, 400 564
advanced Schrödinger equations, 401 Archaic Theory, 559
adynamical global constraint, 499, Archaic Universe, 409
504, 541 Archaic Vacuum, 560, 572
Agodi, A., 391, 413 archaic vacuum as a beable, 565
Alexandroff topology, 241, 245 theory of archaic vacuum, 565
algebra, 165, 170 Arcidiacono, G., 408–409
arrow of time, 101, 132, 466, 475, 483
algebraic deformation theory, 190
artificial life, 329
algebraic simplicity, 157
artificial neural networks, 208
algebraic stability, 173
ashtekar variables, 53
almost complex structure, 668
atomic orbitals, 103–104, 128
almost Hermitian model, 669 atomic variables, 229
almost Kähler, 669 attractors, 334, 347
almost Kähler–Ricci solitons, 669–670 autocorrelation, 129
almost symplectic structure, 668 Axiom A, 181–182
angular momentum, 128 Axiom A dynamical system, 182
angular variables, 106, 116 axioms of quantum mechanics,
anisotropy, 434, 439–441 112–113
697
698 Index
axiom of the measurement, 113 black-hole entropy-area law, 328

axiom of the motion, 112 block universe, 138, 502, 507–508,
axiom of the observables, 111 510–513, 621
axiom of the states, 108 block variable, 175
axioms of QM, 96, 98, 107, 118 blocking, 175
Boardman symbol, 185–186
B
Bohm approach, 12–13, 22
background, 548–551, 554 Bohm explicate order, xiv
background dependence, 70 Bohm hidden-reality, 592
background independence, 49, 52, 69,
Bohm momentum, 12–13
71–72
Bohm’s hidden variables, 579
background space–time, 47, 49, 51
Baker bracket, 10–11 Bohm’s quantum potential, 29
Baldo, M., 391, 413 Bohm’s theory of quantum motion, 29
bandlimit, 314 Bohm, D., xv, 15, 565
band-gap filters, 381 Bohm trajectories, 565
bandlimit, 315 Bohmian mechanics, 1, 18, 22
bandlimited, 322 Bohr, 16, 103
Barbour, J., 101, 624 Bohr atom, 130
Bars, 634 Bohr model, 545–548, 554
Bars action, 633 Bohr orbitals, 124
baryon number, 587
Bohr spectrum, 104
basis vectors, 109, 621, 626, 635, 643
Bohr–Sommerfeld quantization, 95,
beables, xiv
123–124, 126, 129–130
Bekenstein’s information, 289
Bekenstein, J., 288, 295, 562 Boltzmann constant, 129, 295, 562
Bekenstein limit, 570 Boltzmann, L., 573
Bekenstein–Hawking entropy, 64, 66, Bonifacio, R., 404, 407, 413
72, 74 Boolean algebra, 230
Bekenstein–Verlinde, 295 Born reciprocity, 266
Bekenstein–Verlinde entropy, 287 Born rule, 114–117, 488, 568
Bell Inequality, xii Born’s probability postulate, 2
Bell’s theorem, 95 Born, Max, 1, 447, 565
BH entropy area law, 366 Bose–Einstein condensate, 67
bialgebra, 173, 194–195
Bose–Einstein identical-photon
bifurcation, 173, 185, 188
statistics, 612
bifurcation set, 188
boson condensation, 467, 475
Big Bang, 561–562, 580
pre-Big Bang Universe, 562 bosonic creation, annihilation
Big Bang era, 336 operators, 635, 658
bilinear forms, 386, 412 bosons, 616–617, 657
binary question, 142 Boundary Conditions (BCs), 94, 105,
black hole space–times, 420 124
Black-Body, 125 boundary of space–time, 105, 133
Black-Body radiation, 104, 107–108 boundary term, 116
www.ebook3000.com
Index 699
brane, 620, 650–656, 658 causal set, 74–75

brane state with a hole, 651 causal set program, 490
instantonic, 650 causality, 76, 104–105, 132, 572
brane state, 651 causet, xv
brane world, 652, 658 Cayley, 361
branes, 650 cellular automata (CA), xiii, 72,
Breit–Wigner formula, 475 307–308, 347
Broglie–Planck period, 131 cellular automation action principle,
Broglie–Planck relations, 118 309
Brownian motion, 98 center of mass, 616, 618–619
central extension, 162
C changeables, xiv
chaotic evolutions, 132
1-cochain, 192
chaotic interactions, 98
2-coboundaries, 159, 192
characteristic function, 8
2-cocycles, 159, 192
charge conjugation, 472
CA Action Principle, 313, 318, 320
Chern–Simons-like Lagrangian, 429
CA conservation laws, 307, 316, 318 Chern–Weyl form, 676
CA observables, 313 Chew, G., 573
Caesium atomic clock, 113 Chiatti, L., xviii
Caldirola, 400 chirality, 587, 595
Caldirola, P., xvii, 393 chirality-carrying zero-mass
finite difference theory, 395 cosmological photons, 588
canonical almost symplectic variables, chirons, 425
668 chronon, xvii, 391–392, 406, 409, 412,
canonical and cartan star products, 561, 566
677 chronon associated with the
canonical class, 676 electron, 398
canonical d-connection, 667 chronon in quantum cosmology,
canonical formulation of QM, 107, 408
115, 118 internal solutions, 400
canonical QM, 116 quantum of time, 391
canonical quantization, 53, 62 chronon approach, 406
canonically conjugated variables, chronon formalism, 394
633–634 chronon procedure, 401
carbon nanotubes, 95, 130–131 chronon theory, 400
Carleman theorem, 378 chronon and its consequences for
Cartan d-connection, 667 classical and quantum physics, 395
Casagrande and Montaldi, 408 circuit, 114–115
Casagrande, F., 394 CL algebra, 579
Casimir effect, 124 CL symmetry group, 598
Casimir operator, 274 classes of equivalence, xiv
Castelnuovo chronotope, 560 classical logic, 230
category, 227 classical mechanics (CM), 137
category of classical snapshots, 235 classical particle, 102, 116, 123
Cauchy problem, 656 classical paths, 119
700 Index
classical physical quantities, 234 completeness relation, 110, 121

classical physics, 222 complex integer-valued, 312
classical propositions, 234 complex internal time, 552
classical snapshot, 257 complex mapping, 174
classical states, 234 complex network of events, 566
classical theory, 223 complex systems, xiii
classical variational principle, 123 complexity, 434
Clifford space, 616–617, 619–620, 622, composite systems, 325
625, 647, 655–657 Compton clock, 96, 111, 125
clique graph, 331, 353 Compton length, 125
clique number, 351 Compton periodicity (see also
cliques, 350 ‘internal clock’, ‘Compton clock’,
clock, 94, 99 ‘rest mass’), 97, 99–100, 103,
clock time, 334 105–106, 109, 113, 125–126, 128,
closed space–time orbits, 103, 124, 131
126–128 Compton periodicity of an electron,
closed spherical orbits, 128 113
closed string theory, 132 Computational Universe, 70–71, 74
Closed Timelike Curves theory, 105 concept of time, 132
CMB anisotropy, 461 condensate, 441, 460
coarse grained description, 392, 407 condensed matter model, 67–70
coarse-graining, 27, 355 condensed matter physics, 95
coboundary, 191–192 conditional expectation, 19
coboundary operator, 191, 338 conditional expectation value, 13, 15,
cocycle, 191–192 19
codimension-one, 176, 180, 187 configuration, 623–625
coherence, 484 configuration space, 623
coherent state wave function, 424 instantonic, 623–624
coherent states, 465–466, 476–477, configuration space, 224, 641,
483 654–655, 657
squeezed, 467, 475 connection, 171, 648–649, 653
coherent-state QFT, 612 connectivity dimension, 344
cohomology groups, 191 Connes distance metric, 342
collapse, 547, 550, 552 conscious experiences, 625
collapse postulate, xvii conscious observer, 625
collapsing matter, 420 conservation laws, 310, 320, 324
commutation, 127 constraint, 103, 128, 272
commutation relations, 101, 116–118 constraint “overdetermining”, 128
local commutation relations, 118 constraint of periodicity (see also
commutation relations: Dirac ‘periodic boundary conditions’,
quantization rule, 115 ‘intrinsic periodicity’), 108,
compact, 106, 182 context, 235
compact space–time dimensions, 105, context, relative external, 138–141,
124, 131 150
compactification length, 106 context, relative internal, 138–141,
complete coherence, 129 148, 150
www.ebook3000.com
Index 701
contextual completeness, 139 cyclic (compact) space–time

contextual incompleteness, 140 dimensions, 106
contextuality, xiii, 137–138, 141, 149 cyclic cosmology, 132
continuity equation, 382 cyclic dynamics, 114, 122, 130, 133
continuity relation, 374 cyclic evolution, 117–120
continuous time equation, 314 cyclic geometry, 122
continuum, 223–224, 257–258 cyclic nature of time, 132
continuum limit, 355, 360 cyclic proper time, 106
contraction, 162, 279 cyclic space–time, 95
coordinate representation, 383 cyclic space–time dimensions,
Copenhagen (S-matrix) quantum 105–106
physics, 592 cyclic space–time geometry, 119, 123
Copenhagen Interpretation, xii cyclic — relativistic mechanics, 94
Copenhagen statistical interpretation,
601 D
coproduct, 194
dark energy, 437, 500, 502, 507, 532,
corank, 189
536, 539–541, 579, 590
correlation length, 176–177
dark energy stars, 420
Cosmic Microwave Background
(CMB), 434, 456, 459 dark matter, 437, 579, 590, 592, 638,
657
cosmological constant, 423, 560, 570
cosmological horizon, 560 dark matter as non-particulate
cosmological photons, 580 (Bohm) hidden reality, 612
cosmological principle, 563 de Broglie, L., 2, 94, 96–97, 99, 133
Coulomb potential, 104, 124, 127, 129 de Broglie–Planck periodicity, 97–98
counit, 194 de Broglie–Planck relation, 96, 105
covariant derivative, 633, 648 de Gosson, 21
Covariant Loop QG, xii de Sitter, W., 172, 560
covariant quantization, 46, 50–52 Sitter geometry, 573
covariant theory of thermodynamics de Sitter cosmological horizon, 569
and statistical mechanics, 288 de Sitter groups, 165
creation and annihilation operators, de Sitter Universe, 560
127 decoherence, 392
creation operators, 641 decoherence from dissipation, 403
critical exponents, 177–179 decoherence through interaction with
Critical Phenomena, 173 the environment, 412
critical point, 173, 185, 187–188 Dedekind cuts, 259
critical properties, 187 defect gauge transformations, 301
critical surface, 176–179 deformation, 159–160, 164, 168, 185
current, 114 deformation parameters, 163
current density, 114 deformation theory, 154, 190
curvature, 617, 646, 653 deformation theory of algebras, 158
curved, 648–649 deformation-stable, 157
curved space, 617 deformations, 154, 164
curved space–time metric, 131 deformed algebra, 168
Cycles of Time, 94 degeneracy of classical paths, 119
702 Index
degrees of freedom (DOF), xiv, 309 direct action picture of fields, 488
fermion and boson, 419 direct action theory, 490
delocalization, 559, 566 Dirichlet boundary conditions, 124
Democritus, xv “discrete” Fourier transform, 108, 119
dense, 183 discrete (retarded)
density matrix, 20, 22 Liouville–von Neumann (LvN)
density matrix formalism, 396 equation, 403–404, 406
density of electrons, 114 discrete conservation law, 311
density operator, 403 discrete energy spectra, 384
derivations, 170–171 discrete formalism, 391
determinism, 114, 307 discrete frequency spectrum, 107
deterministic CA, 328 discrete geometries, 352
deterministically, 114 discrete net of interaction vertices,
diffeomorphism invariance, 47, 49, 565
53–54, 60, 69 discrete space–time, 64
differential forms, 170 discrete subgroups, 312, 318
differential structure of space–time, discrete theory, 406
105 discrete time theories, xvii
differentiation, 319 discreteness scale, 307–308, 313,
dimension, 331 323–324
Dirac, 1, 15 discretization of quc energy, 604
Dirac, P. A. M.; Dirac number, discretized QM, 396, 401
573 disordered locality, 506, 536, 538–539,
Dirac coordinates, 579 541
Dirac delta, 122 dispersion relation, 307, 315
Dirac dynamics, 104 dissipation, 392, 466, 468, 475–476
Dirac equation, 171, 635, 637, 647 dissipative behavior, 402
Dirac matrices, 648 dissipative quantum systems, 396
Dirac momenta, 579 distortion relations, 667
Dirac operator, 170–171, 330
domain theory, 258
Dirac quantization, 55–56, 107
“Double”, 466–467, 471
Dirac quantization for monopoles, 130
DQ of Ricci Solitons, 671
Dirac quantization rule, 116, 118, 127
dumping, 129
Dirac sea, 109, 617
dynamical system, 180, 183–184, 187
Dirac strings, 130
dynamics, 155
Dirac vacuum, 642
Dirac’s cosmology, 572
E
Dirac’s equation, 397
Dirac’s non-relativistic quantum -neighborhood, 181
theory, 579 -stable, 183
Dirac’s relativistic-electron −equivalent, 183
(quantum-physics) wave function, Eddington rubber-sheet, xii
583 effective gravitational field equations,
Dirac, Paul Adrien Maurice, 139, 150, 671
436, 573 effective Hamiltonian, 175–177
Dirac–Kähler equation, 648 effective speed, 169
www.ebook3000.com
Index 703
Ehrenfest, P., 376 emergent, 566

correspondence principle, 376 emergent gravity, 46, 63, 67–69, 71,
Einstein equations, 420, 647 73–74, 79
Einstein field equations, 66, 71, 75, 77 emergent phenomenon, 433, 436
Einstein special relativity, 438 emergent space–time, 63, 68, 71,
Einstein vs Bohr debate, xiii 73–74, 79, 491
Einstein, Albert, 95, 100, 103–105, energy (E-)representation, 374
133, 434, 439
energy gap opening, 130
Einstein–Bohr debate, 572
energy spectrum, 127
Einstein–Hilbert action, 289–290, 303
energy–momentum tensor, 466, 468
Ekeland–Hofer capacities, 35
electric current, 114 entanglement, 308, 466, 475
electro-gravitationally-stabilized entanglement structure, 328
double helix, 588 entire analytic functions, 477
electromagnetic field, 107 entropy S, 561, 570
electromagnetic interaction, 130 negative entropy, 562
electron internal clock, 113 entropy-action equivalence, 287–288,
element of the physical reality, xiii 290, 294–296
elementary clock, 94, 100, 108, 132 enveloping algebra, 170
elementary particle (see also ‘periodic EPR correlations, 421
phenomenon’, ‘elementary equations of motion, 126, 318–320
space–time cycle’, ‘elementary
equivalence, 184
clock’, ‘vibrating string’), 94, 98,
equivalence between Elementary
113, 564
cycles Theory and Quantum
anti-particles, 102
Mechanics, 115, 119
free relativistic particles, 98
interacting particles, 107 equivalence functor, 242
internal clock, 113 equivalence invariants, 187
massless particle (see also frozen equivalence with ordinary
internal clock), 100, 107 quantum-relativistic mechanics, 100
neutral particles, 114 equivalent, 183
non-homogeneous string, 102 equivalent maps, 184
non-relativistic particles, 103 ergodic system, 98, 132
rest particle, 106 étale bundle, 245
single particle, 114 etalé space, 252
vibrating string, 113 eternity, 101
elementary physical event, xvi ether, 434, 436, 438, 444, 459
elementary signal detector, 144–147,
ether-drift, 434, 438, 441–442, 449,
149
461–463
elementary space–time cycles, 94–95,
Euclidean Lie symmetry group, 579
106, 110, 122
elementary space–time vibrations, 111 Euclidean periodicity, 129
ellipsoidal localization-region, 389 evaluation map, 229
Elze, T., xviii “event”, 620, 624
emergence, 46, 63, 65, 70, 72, 564, 573 event horizon, 419
emergence of a time order, 568 Everett, 625
704 Index
Everett theory, 626 field theories, 103

evolution, 624–625 finite-difference (retarded) equation,
evolution parameter, 620, 398
622–624, 656–657 finite-difference equations, 310, 412
expectation value, 115, 646, 649, Finkelstein, D., xvii, xviii
653–654, 658 Fitzgerald, George Francis, 434
experiment, delayed choice, 143 flat canonical Fedosov d-connection,
experiment, double-slit, 146 674
experiment, monitored double-slit, flow, 181–182
138, 145, 147–148 Fock space, 111, 127, 129, 636, 651,
experiment, quantum erasure, 143 655
experiment, Stern–Gerlach, 143–145 Fock space basis, 636
explicate orders, 15, 22 Fock, V.A., 565
extended Dirac equation, 172 Fock–Bargmann representation, 477
extended object, 623 Fokker–Planck equation, 183
extended-like particle, 398 force, universal constant of, 267
extended-type (ellipsoidal) formal deformation of a Lie algebra,
localization, 411 190
extra-dimension, 132 four-momentum operator, 118
four-periodicity, 97
F Fourier coefficients, 108, 115
Fantappié, L., 408–409 Fourier transform, 101, 321
Farias, R.H.A., 372, 413 Fourier transformation, 322
Fedosov N-adapted operators, 673 Fourier, Jean Baptiste Joseph, 444,
Fedosov operators, 672 447, 451, 461
Fedosov–Einstein manifolds, 675 fractal dimension, 346, 481
Fedosov–Ricci solitons, 674 fractal dislocation, 484
feedback, 335 fractal operator, 478
Feigenbaum functional equation, 173 fractal self-similarity, 465–467, 476
Fermi, Enrico, 28 frame, 238–239, 259
Fermi blobs, 28, 37 frame homomorphism, 239–240
Fermi ellipsoid, 34 frames, 249
Fermi function, 32 free elementary particle, 108, 110
Fermi–Dirac statistics, 124 free energy, 176, 480
fermionic creation, annihilation free relativistic particle, 99, 101, 108,
operators, 628, 635, 645, 658 110, 117
fermionic field creation, 658 free will, 625
fermionic particles, 109 frequency operator, Hf , 111
fermions, 616–617, 650, 657 frequency spectrum, 104
Feynman, Richard P., 9, 115, 142, Friedmann, A., 563
150, 565 Friedmann model, 563
paths, 565 fringe shift, 442, 444, 446–447, 450,
Feynman path integral, 96, 107, 452
119–123, 132, 287, 632 “frozen” Compton clock, 132
fiber bundle, 241 function germs, 189
www.ebook3000.com
Index 705
fundamental constant, 154, 158–159, graph Laplacian, 330, 340

163, 166, 169 graph properties, 349
fundamental harmonic, 102 graphene, 129
fundamental length, 158, 166–170 graphene systems, 131
fundamental time, 169 Grassmann, 633–634
fundamental time scale, 308 coordinates, 633–634, 656
fuzzy cliques, 353 Grassmann variables, 635
fuzzy lumps, 353 gravistar, 420
gravitation, 646
G gravitational collapse, 419–420
Galilean, 156–158, 163–164 gravitational field, 45–48, 50
gauge fields, 171, 633, 638 gravitational interaction, 106, 130
gauge fixing, 504, 516–517, 522 graviton, 49, 51–52, 79, 132
gauge interaction, 94, 106, 130–132, gravity, 650, 671
171 gravity in the world crystal, 300
gauge invariance, 54, 95, 130, 334, gravity-wave, 437
504, 511–512, 516–517, 530 Gromov–Hausdorff-distance, 362
gauge transformations, 54, 60–61, 468 Gromov–Haussdorff Limit, 331
gauge/gravity duality, 132 Gromov-compactness theorem, 363
Gelfand–Naimark (GN) unitary group algebra, 337
Hilbert-space infinite-dimensional growth function, 344
representations, 580 growth type, 361
general covariance, 53
General Relativity (GR), xi, 105, 130, H
222, 257, 655 Haag’s Theorem, 488
generalized spinors, 648 Hall current, 424
generalized variational derivative, Hall effect equation, 422
318, 320, 322 Hamilton operator, 121, 311
generic, 187 Hamilton’s equations, 307, 324
geometric group theory, 361 Hamilton–Jacobi equation, 1, 12
geometric morphism, 242 Hamilton–Jacobi theory, 2–3
geometric renormalization group, 352 Hamiltonian, 629, 643
geometrodynamic theories, xvii Caldirola–Kanai, 393
geometrodynamical, 117 Dirac self-adjoint CL-invariant
geometrodynamics, 124, 131 Hamiltonian, 579
geometrogenesis, 75–77 equivalent Hamiltonian, 391
geometry of lumps, 353 non-Hermitian Hamiltonians,
ghosts, 657 391, 412
golden spiral, 480, 483 quasi-Hermitian Hamiltonians,
Goldstone, 130 412
grand unification, 617 Hamiltonian CA, 310, 313, 315–316,
graph characteristics, 347 318, 320
graph dimension, 346 Hamiltonian constraint, 56, 58, 60–62
graph Dirac operator, 340 Hamiltonian formalism, 53–54
graph isomorphism, 358 Hamiltonian operator, 111–112, 121
706 Index
Hamiltonian undulatory mechanics, Hilbert state, 110

2, 12 Hilbert-space Lorentz-group
Hardy paradox, 143 representation, 580
harmonic oscillator, 316 Hiley, 13, 15–16, 19–20
harmonic potential, 124, 127 Hiley, B., 573
harmonic system, 108 Hofer–Zehnder capacity, 35
Hartle–Hawking, 565 hole, 617
Hartman effect, 393 holographic conjecture, 570
generalized, 393 Holographic Entropy Bound, 64, 74
Hausdorff-metric, 362 holographic information, 575
Hawking, S., 572 holographic principle, xiii, 287–288,
Heisenberg (Lie) algebra, 5 296, 366, 437, 560
Heisenberg algebra, 5, 161–164, holographic relation, 570
167–169, 270, 272–274, 276
holographic screens, 72–74
Heisenberg cut, 139, 143
Holography, 131, 328, 365, 561
Heisenberg equation, 15
Holomovement, xii
Heisenberg equation-of-motion, 5, 13,
homogeneous string, 126
16
Heisenberg group, 5 Hopf algebra, 194, 467, 482
Heisenberg picture, 603 q-deformed Hopf algebra, 467
Heisenberg uncertainty principle, 6 non-commutativity, 468
Heisenberg uncertainty relation, 117, Hubble constant, 580
119 Hubble–Milne (τ ), 593
Heisenberg, Werner, 1, 139, 141, 572 Huygens construction, 2
commutation relations, 627 Hydrogen atom, 128
equations of motion, 631, 644 hyperbolic, 182
Heisenberg–Poincaré algebra, 158, hyperbolic motion, 399
164, 167 hyperbolicity, 181–183
Hermite function, 31
Hermitians operator, 111 I
self-adjoint operator, 111
Illingworth, K. K., 439, 455, 460
Hernández-Figueroa, H.E., 413
implicate order, 15, 22, 503
Heyting algebra, 228, 230, 257
Induced Gravity, 67
Hicks, W. M., 445, 447, 460
hidden variables, xiii, 95 infinite Compton periodicity, 108
Higgs bosons, 607 infinitesimal space–time evolutions,
Higgs, Peter, 440 121
geometric Higgs potential, 430 infinitesimal space–time paths, 122
Higgs potential, 430 infinitesimally close, 193
high-energy chaos, xiv information, 564, 568
high-temperature fixed point, 176 Information Retrieval, 201
high-temperature limits, 177 information theory, xvii, 288, 293
Hilbert basis, 111, 129 information void, 144–145
Hilbert space, 1, 4–7, 15–17, 22, 102, initial conditions, 103, 310, 315
109, 114–116, 126, 128 Initial-universe (planck-scale) wave
local Hilbert space, 118 function, 609
www.ebook3000.com
Index 707
inner product, 110, 626, 628–629, 653 It from bit, xii, 70, 73–74
local inner product, 118 It from qubit, 70, 73
instanton, 619
instantonic, 622 J
instantonic brane, 652 Jacobi identity, 192–193
instantonic extended object, 622–623 Janyes’ information, 290
instrumentalist interpretation, 222, Jaseja, T.S., 462
226 Jaynes’ non-equilibrium formulation,
integer valuedness, 311 292
integer-valued, 309 Joos’s fringe shifts, 454
integer-valued CA, 314 Joos, Georg, 439, 445, 452, 455–456,
integer-valued CA action, 309 459, 461
integer-valued symmetric matrices, Jordan, 1
315 Josephson effect, 130
integer-valued variations, 309, 313,
318 K
integer-valuedness, 315 Kahler potential, 421
interacting classical particle, 121 Kaluza’s miracle, 132
interacting system, 118 Kaluza–Klein theory, 132
interaction, 121–122 kinematical level, 155
interaction between defects, 304 kinetic term, 647–649
interactions, 106, 110, 117 Klein–Gordon, 411
interference, 120, 123, 169, 308 Klein–Gordon equation, 621–622,
internal algebra, 230 624, 656
internal clock, 94–96, 100–101 Koch curve, 476–477, 481
“frozen” internal clocks, 101 Kochen–Specker theorem, 143
internal complete Heyting, 262 Kolmogorov probability, 568
internal frame, 261 Kolmogorov, Andrej Nikolaevic, 438
internal Heyting algebra, 261 Kosterlitz–Thouless condensation,
internal language, 230 426
internal lattice, 260 Kosterlitz–Thouless transition, 305
internal locale, 240, 242, 248, 251 Kuramoto model, 366
internal motion of the electron, 398
internal relativity, 73 L
interval domain, 243, 258–259 laboratory, 143
intrinsic periodicity, 94, 97–98, 101, labstate, 145
116 ladder operators, 127
intuitionistic logic, 229 Lagrange multipliers, 291, 634
irreversibility, 406 Lagrangian, 121
irreversible processes, 392 Lagrangian formalism, 122
Isham–Doering Schema, 233 language, 229, 231
isochronism of the pendulum, 126 Large Hadron Collider, 143
isolated relativistic particle (see also lattice, 238, 259
free relativistic particle), 99, lattice parameter, 175
isomorphism, 466, 482–483 Laughlin, R., 420
708 Index
least action principle, 122 Lorentz forces, 469

Lee, T.D., 638 Lorentz group, 164
Leech lattice, 419, 421, 428 Lorentz invariance, 167
25 + 1 dimensional Lorentzian Lorentz transformations, 96, 105, 108,
Leech lattice, 419 125
left ideal, 16, 636, 640, 647 Lorentz, Hendrik, 434
Leibniz rule, 311 Lorentz–Minkowski forward
Libet experiment, 625 lightcone, 580
Lie algebra, 158, 160, 162, 164, 190 Lorentz-frame-independent, 581
Lie algebra-valued connections, 171 Lorentzian, 156, 164
Lienard–Wiechert (LW), 603 low-temperature fixed point, 176
like-Computer Universe, 560
linear potential, 126 M
linearity, 307–308, 324
macroscopic object, 496
Liouville equation, 1, 11, 17
macroscopic quantum systems, 467,
Liouville measure, 161
476, 483
Liouville operator, 403
macroscopic system, 132
Liouville-von Neumann’s (LvN), 392
macroscopic variables, 561
Lisi, G., 293
magnetic flux, 130
Little–Parks effect, 130
local CA-models, 328 Maldacena’s conjecture, 132
local inner product, 120 manifold, 617, 645
local modulations of periodicity, 112, infinite dimensional, 653
126 many worlds interpretation of
local modulations of the space–time quantum mechanics, 625
periods, 118 map, 158, 182, 184, 186
local nature of space–time, 105 map, validation, 140
local space–time periodicity, 105 mapping, 174, 181
local transformation, 118 Markopoulou-Kalamara, Fotini G.,
locale, 238–239, 242–243, 245–246, xvii, 575
249, 257–259 Markovian, 120
locality, 76, 324, 560 massive elementary particles, 586
localization, 559, 563, 566, 570, 572 massive particle, 125
spontaneous localization, 565 massless, 108
localization-region (an ellipsoid), 388 massless particle, 106, 108, 125
localization/delocalization, 568, 575 massless solution, 172
locally modulated periodicity, 122 Mathieu group, 423, 427
logarithmic spiral, 483 matter wave, 102
logical connectives, 229–230 maximal solvable algebra, 193
logical level, 154 maximum entropy production, 287,
long range order, 355 290, 292, 295–296
loop, 655–656 maxmal complete graph, 336
Loop Quantum Gravity, 46, 53–62, 64 Maydanyuk, S.P., 372, 413
Lorentz, 434, 436–440, 442, 444, 446, meanings, 484
460 measure (or weight), 374
Lorentz algebra, 159, 164 measurement, 112, 115, 226
www.ebook3000.com
Index 709
measurement problem in QM, 392 modulations of space–time

measurement process, 406 periodicity, 106
Meissner effect, 130 momentum operator, 111–112, 116,
Melucci metaphor, 213 121
membrane, 619–620, 650 momentum representation, 383
membrane space, 653 Montaldi, E., 394
Meroni, C., 413 Morley, Edward, 439, 445
metric, 620, 646–647, 649, 653–654, morphogenesis, 466, 482
658 morphogenetic paradigm, 468
curved, 648 Morse factor, 126–127
euclidean, 627 Morse function, 178
induced, 647, 652, 655, 657–658 Morse theory, 187
metric tensor, 615, 617 Moyal, 8, 12–14, 21–22
Minkowski, 627 Moyal algebra, 12–13, 161
symmetric, 629 Moyal bracket, 10–11, 159–160
symplectic, 616, 628 Moyal star product, 7
symplectic metric, 644 Moyal-Vey algebra, 163
metric spaces, 360 multiparticle states, 102
Michelson, 442–443, 445 muon
Michelson, Albert Abraham, 439 rest mass of the muon, 403
Michelson–Morley, 446, 449, 455–456,
460 N
microscopic degrees of freedom, xii n-cochain, 191
microscopic quantum dissipation, 392 n-cohomology group, 191
microstates, 561 n-tupling, 174
Miller, Dayton, 439, 445–446, 449, Naimark theorem, 372–373
460 natural cellular automata, 307
Milne, A., 579 Navier–Stokes equation, 435
Milne’s universe, 580 Nelson, Edward, 436
Milne’s negatively-curved network of events, 566
non-compact 3-space, 580 networks of transactions, 568
Milne–Lorentz cosmological Neumann boundary conditions, 124
symmetry, 584 neutral bosonic, 109
minimal substitution, 130 neutral bosons, 109
Minkowski space, 164–165 Newton’s law of inertia, 94, 98, 118,
Minkowskian periodicity, 129 133
Minkowskian space–time, 101 Newton’s laws, 97
Minkowskian time, 101 Newton–Maxwell (G − c), 593
mirror gauge fields, 638 Ng, Yee Jack, 437
mirror particles, 637–638 no-cycle condition, 182
modified lattice gauge theory, 500, Noether charges, 271
502 non-classical probabilities, 212
modified Regge calculus, 502 non-commutative, 22
modulated elementary cycles, 118 non-commutative algebra, 11, 19
modulated signals, 118 non-commutative algebraic, 18
710 Index
non-commutative dynamics, 2 nonlinear oscillator, xiv

non-commutative geometric algebras, nonlinear Schrödinger equation, 424
19 nonlinear Schrödinger equation for a
non-commutative geometry, 158, 170, superfluid, 420
342 nonlinear systems, xiv
non-commutative N -adapted frame nonlinearity, 320–321
deformations, 678 non-locality, 321, 324
non-commutative phase space, 9 normal ordering, 109, 125
non-commutative probability theory, normalization, 110, 115–116, 121, 325
22 nuclear optical model, 392
non-commutative product, 10 nucleation by 5-sphere vacuum, 564
non-commutative space–time, nucleus, 259–260
169–170 Nyquist rate, 314
non-commutative symplectic algebra,
1 O
non-commutative symplectic
geometry, 13, 19, 22 object, 618
non-commutative symplectic group extended object, 618, 620, 624,
algebra, 16 650
non-commutative symplectic space, instantonic, 619–620, 650
22 physical object, 617
non-commutative symplectic objective reality, 591
structure, 17 observable, xiv, 112, 115–116, 155,
non-commutativity, xvi, 12, 166 320, 324
non-degenerate, 187 observable properties, 203
non-equilibrium, 287–288, 290–292, observer, 137–138, 564, 625
294–295 observer, exophysical, 143
nonlinear Schrödinger equation, observer, primary, 138–140, 150
419–420 Occam’s principle, 581
non-local, 7, 9, 21–22, 143, 560 Olkhovsky, V.S., 372, 413
non-local star-product, 10 Onsager, Lars, 436
non-locality, xvii, 21, 560, 565 ontic structural realism, 502, 508
non-relativistic limit, 102–103, 125 ontic structural realists, 509
non-relativistic particle, 125 open in the locale, 240
non-renormalizability, 51 open string theory, 132
non-standard analysis, 193 operator
non-trivial deformations, 192 anti-Hermitian (or
non-unitary time-evolution, 405 skew-Hermitian) parts, 387
non-wandering set, 181–182 anti-Hermitian part, 373
non-commutative geometry, 482–483 bilinear Hermitian operator, 378,
non-holonomic (pseudo) Riemannian 386
manifold, 669 canonically conjugate operators,
non-holonomic Einstein space, 669 376
non-holonomic Ricci solitons, 669 damping of the non-diagonal
nonlinear CA processes, 325 terms of the density
nonlinear deformations, 308 operator, 385, 405
www.ebook3000.com
Index 711
extended-type position operator, overdetermination of relativistic

390 dynamics, 94, 95, 104, 108, 131
four-position operators, 373, 386, overdetermination of relativistic
390 mechanics, 100–101, 103, 133
Hermitian, 373
Hilbert-space self-adjoint P
operators, 579 Pais, A., 103
maximal hermitian, 374
Paleari, S., 413
(maximal) hermitian operators,
participatory principle, 137, 149
376
particle as a permanent object, 565
Newton–Wigner operator,
particle density, 115
388–389, 411
particle in a box, 94, 126, 128
non-Hermitian character, 391
particle on a circle, 95, 114
non-Hermitian operators, 373
particle physics, 500, 502, 532–533,
non-unitary evolution operators,
535–536, 541–542, 560
391
Pauli-matrix, 598
non-unitary time-evolution, 405
PBCs, 104, 111
not self-adjoint, 377
observable; in non-relativistic peaceful coexistence, xi
QM, 373 Pennisi di Floristella, A., 391
quasi-self-adjoint time operator, Pennisi, A., 413
386 Penrose, R., 94, 100–101, 560
retarded E-G field, 603 perfect recurrences, 129
self-adjoint Hilbert-space period-doubling bifurcations, 173
operators, 581 period-quadrupling, 174
self-adjoint single-quc period-tripling, 174
kinetic-energy operator, 605 Periodic Boundary Conditions (see
self-adjoint single-quc also intrinsic periodicity), 94,
potential-energy operator, periodic classical paths, 119–120, 123
605 periodic paths, 122
time as an observable, 373 periodic phenomenon, 94, 98, 113,
time in non-relativistic QM, 373 115, 117, 126
time operator, 373 isolated periodic phenomenon,
time operator also in a 112
hamiltonian form, 382 periodic time box, 125
Oppenheimer, R.J., 419 periodic variable (see also angular
orbit, 180, 192 variable), 116
order parameters, 354 persistent periodicity, 100
order-preserving functions, 237 persistent space–time periodicity, 99,
order reversing functions, 237 108
oriented area, 616, 618, 619 persistent time periodicity, 126
oriented volume, 619 phase fluctuations, 305
orthogonal Clifford algebra, 16, 616, phase harmony, 97, 99–100, 104, 108,
626, 628, 657 117
oscillator, 95 phase harmony condition, 105
overdetermination, 104, 126 phase locking, 366
712 Index
phase oscillators, 354, 366 Poisson summation, 119, 122

phase portrait, 180, 184 polyvector, 655–656
phase space, 6, 13, 15, 22, 627–630, Pontrijagin-like form in 13
635, 637, 640 dimensions, 429
phase space coordinates, 626 Popper, K.R., 586
phase space volume, 169 positive-operator-value-measure
phase spaces, 657 (POVM), 410
phase transition, 174, 176–177, 179, Possibilist transactional
185 interpretation, 488
phasors, 106 postulate, 401
photon postulate of intrinsic periodicity,
photon barriers, 381 96–98, 106–108, 111, 115, 117, 119,
physical observable, 111 126, 127
physical points, 355 potential well, 126
physical quantity, 224–225, 231–232, POVM approach, 378
238 pre-accelerations, 397
physical reality, 138 pre-space–time structures, xv
physical theory, xi, 231 preferred reference frame, 438
formal structure, xi presheaf, 237–238, 241, 243, 245, 247
Planck cells, xv primary observer, 141
Planck constant, 94, 96, 107
primeval quantum events, xvii
Planck frequency, 420
principle of equivalence, 97
Planck mass, 299
principle of the symplectic camel, 28,
Planck quantization, 107
38
Planck scale, xiii, 225, 299, 348, 570
probabilistic model, 202
Planck’s length, 169
probabilities, 224–225
Planck’s units, xvi
probability, 115, 123, 129, 561, 567
Planck, Max, 436, 570
probability amplitude, 622–623
Planck–Schrödinger–Dirac (), 593
planck-scale physics, 327 probability density, 623, 625
Poincaré cycle, 385–386 problem of time, 47, 60–62
Poincaré quasi-cycles, 386 product, 129
Poincaré, Henry, 440 Projective General Relativity (PGR),
Poincaré algebra, 165–166 569
Poincaré group, 165 projective relativity, 410
point in a locale, 239 proposition, 231–233
point of the locale, 253 proposition, generalized, 138,
point-like description, 125 140–142, 150
point-like localization, 389 propositions, classical, 141
point-like position, 411 propositions, generalized, 139
pointed metric spaces, 364 propositions, mathematical, 141
points of the locale, 255–256 propositions, metaphysical, 140
Poisson algebra, 163 propositions, physical, 140
Poisson bracket, 10, 116, 312–313, protogravity, 335
320, 324, 615–616, 627, 629, 644, pseudo-Euclidean groups, 169
675 pseudo-orthogonal group, 165–167
www.ebook3000.com
Index 713
pseudoscalar, 621 Quantum Gravity (QG), xii, 222, 226,

pseudoscalar coordinate, 625 287–288, 293–294, 296, 328, 617,
pure gauge, 130 645–646, 658
pure quantum phenomena, 129 quantum Hamilton–Jacobi equation,
pure quantum systems, 98, 129 13–14, 18
quantum harmonic oscillator, 101,
Q 114, 126–128
quantum information, 560
qbits, xvii
quantum leap, 545, 547–550, 552–554,
QFT identical-elementary-particle,
566
583
quantum mechanical information, 420
QM with friction, 396
quantum mechanical observables, 313
quadratic map, 173
Quantum Mechanics (QM), xi, 137,
quantity value object, 231–232, 237,
559
243, 245–247, 255, 257
foundational problems, 559
quantization, 615, 632, 634, 655, 657
transactional interpretation, 559
quantization condition, 98, 123, 128
quantum non-locality, xi
quantization method, 127 quantum number, 109, 128–129
quantization-by-deformation, 160 quantum potential, 15, 18, 29, 568
quantized energy spectrum, 107 quantum propositions, 141
quantized energy–momentum, 125 quantum recurrence, 129
quantized fields, 650, 655 quantum register, 138, 142–143, 149
quantropy, 293, 296 quantum space, 354
quantum (finite difference) equation quantum space–time, 47–50, 80
for dissipation and decoherence, quantum states, 103
394 quantum stochastic, xii
quantum algebra, 6 quantum substrate, 45, 62–63, 73–75,
quantum anharmonic oscillator, 126 80
quantum blob, 9, 21, 28, 30, 34, 40 quantum system, 112
quantum causal histories, 46, 73–79 quantum theory, 222–223, 257, 434,
quantum computers, 75 436, 442, 459
quantum cosmology, 141, 560 quantum theory of gravity, 420–421
quantum dissipation, 412, 467, 473 quantum vacuum, 327
quantum electrodynamics, 131 quantum Zeno effect, 149
quantum excitations, 102 quantum-cosmology, 581
Quantum Field Theory (QFT), xii, quantum-electrodynamics, 113
47, 49–52, 102, 109–110, 127, 146, quantum-gravity, 437
150, 617, 642, 643 quaplectic group, 268, 278
quantum fields, 171 Casimir operator, 273
quantum fluctuation, 590 Lie algebra, 273
quantum fluctuations of space–time, quasi-homogeneous function, 177, 179
49 quasi-isometric embedding, 360
quantum fluid, 428 quasi-isometries, 361
quantum friction, 412 quasi-linear Schrödinger-type
quantum geometry, 47–48 equation, 384
quantum graphity, 73 quasi-periodical evolution, 386
714 Index
quc in a five-dimensional fiber renormalization group, 174, 176–178,

bundle, 600 353
quc quantum-universe constituent, renormalization process, 330
585 renormalization transformation, 176
quc statistics, 587 resolution in time, 113
Quc-q’s 5-dimensional fiber-bundle rest mass, 96, 99–100
location, 601 retarded, 401
quc-time, 599 retarded “Schrödinger” equation, 392
quotient space, 191 retarded, symmetric, 395
reverses, 621
R reversion, 645
r-parameter unfolding, 185 Reynolds number, 437–438
R process, 560, 568–569 Riemann manifolds, 165
random collisions, 129 Riemann tensor, 653
random graph, 331, 347 right ideal, 16
random perturbation, 183 rigid, 154, 191, 193
randomness, 568 rigid algebra, 192
rank, 185 rigid bialgebras, 195
realist, 234 rigid Lie algebras, 193
realist theory, 234 rigidity theorem, 192
realistic quantum model for Ritz–Rydberg combination principle,
space–time, 421 2
reciprocal relativity, 267, 279 Riva, P., 413
recurrences, 103 Robertson–Mansouri–Sexl (RMS)
reduction, 401 scheme, 441
diagonal reduction process, 407 rolling die, 113–114
reduction to diagonal form, 412 Rota algebra, 210
Reduction, Vector State, 560 Rovelli, C., 95, 101
reference clock, 99–100, 131–132 relational interpretation, 296
regularization, 502, 532–534, 541 rules of inference, 229
Relational Blockworld, 500 runaway solutions, 397
relational time, 62
relativistic action, 131 S
relativistic clock, 100–105 SO(3), 162
relativistic covariance, 497 Σ2 -type, 174
relativistic cyclic dynamics, 104 S-matrix, 584
relativistic cyclic nature, 106 QFT S-matrix, 594
relativistic dispersion relation, 109 S-matrix asymptotic Hilbert space,
relativistic Doppler effect, 97 590
relativistic field, 103 Sacharov, A., xii
relativistic quantum field theory Sampling Theorem, 314, 321–323
(RQFT), 144 sampling theory, 308, 313–314, 324
relevant directions, 176 Santilli, R. M., 393
renormalization, 176, 533–534 scalar, 617, 657
renormalization fixed point, 177 scalar coordinate, 625
www.ebook3000.com
Index 715
scalar field, 644, 647 Shamir–Fox experiment, 440

scale free small world network, 365 Shannon information, 288, 569
scaling behavior, 173 Shannon’s Sampling Theorem, 307,
scaling dimension, 344 320
scaling hypothesis, 177 Shannon–Turing Information, 560
Schrödinger equation, 112, 311, sheaf, 241
314–315, 321, 323–324 Shimony, A., xvii
time dependent Schrödinger Sierpinski space, 253
equation, 112 signal operator, 145
Schrödinger field, 644 signature, 621, 625
Schrödinger problems, 126–127 simple algebra, 164
Schrödinger, 2–3, 12–13, 15–16, 94 simplicity criterion, 157
Schrödinger equation, 1–3, 15–16 single particle, 115
Schrödinger representation, 3, 6, 11, singular points, 185–186, 193
21 singularities, 186–187
Schrödinger, E., 567 singularities of the mappings, 185
Schrödinger–Milne Big Bang, 579 singularity, 174, 185–186, 419, 564
Schrödinger–Milne cosmology, 583 sinus cardinalis, 321
Schrödinger–Milne SL(2,c), 579
Planck-scale-originated, 580 slow motion camera, 113
Schrödinger–Milne universe, 579, 589 small world, 357
Schrödinger–Robertson inequality, small world networks, 354
276–277 smooth functions, 187
Schwinger, Julian, 149–150 smooth germ, 185
Scott interval domain, 254 Smrz, P., 372, 413
Scott topology, 243 SMU, 580, 583–585, 588, 591
second cohomology group, 159, SMU birth without any “particles” at
191–193 a Planck scale, 581
second quantization (see also Dirac SMU classical electromagnetic fields,
quantization rule), 101, 125–127, 607
self-adjoint quc-pair potential-energy SMU Dirac quc coordinate, 593
operators, 605 SMU elemental constituents, 585
self-dual, 419 SMU finiteness of photon, 612
self-interaction term, 649 SMU foundational numbers, 596
self-organization, 434 SMU gravity, 599
self-referential, 335, 625 SMU Hamiltonian, 606
self-similar geometry, 483 SMU Hamiltonian potential-energy
semi-classical limit, 275 operator, 605
semi-classical theories of gravity, 45, SMU Hamiltonian’s potential energy,
48, 62–63 586
semi-simple, 159 SMU Hilbert space, 600
semi-simple algebras, 157 SMU history, 613
Semi-simple Lie algebras, 193 SMU Lie symmetry group, 580
separability, 308 SMU local frame, 589
shadow phase, 15 SMU reality, 592
shadow phase spaces, 1, 14 SMU Schrödinger equation, 608, 612
716 Index
SMU wave function, 609 ‘de Broglie-Planck periodicity’),

SMU’s hidden reality, 592 108–109, 117, 125, 127, 129
SMU’s Lie symmetry-group center, local modulations of space–time
593 periodicity, 102
SMU’s multi-quc external-momentum space–time points, 251
csco, 589 space–time quantum recurrence, 94
SMU-foundational GUT scale, 594 space–timesource, 513
Snyder, H., 419 space–timesource element, 500–505,
SO(3), xvii 510, 514, 516, 522, 525–526,
SO(3,1), 584 532–536, 539
soliton, 435 spatial locations, 581
space Special Relativity, 437, 441–442, 462
flat space, 617 speed of light, 96
infinite dimensional, 640 spherical angles, 128
Minkowski, 647 spherical harmonics, 128–129
spherical membrane, 111, 128
subspace, 625
spherical periodicity, 111, 127–128
vector space, 615
spherical problem, 128
spinor space, 648
spin connection, 646
space–time (ST), 158, 222–223, 226,
spin foam, 74
236, 249–251, 256–258, 560,
spin foam models, 59
616–617, 620–622, 625, 646, 650,
spin network, 56–58, 650
652–653, 655, 657
spin-statistics, 117
contextuality of space and time,
spinor, 635, 640
149
algebraic spinors, 638
continuous space–time, 566
spinor space, 648
curved, 649, 655
spinor field, 645–646
de Sitter space–time, 560 spinors, 636, 641
emergent ST, xii splitting lemma, 189
local causality in space–time, xi squeezed states, 32, 277–278, 280
space–time coordinates, 566 stability, 157, 173, 184, 187
superfluid pictures of stability of physical theories principle
space–time, xvii (SPTP), 153, 157
superfluid model for space–time, stabilizing deformation, 166, 171
430 stable, 154, 159, 161–162, 164–165,
space–time boundary, 105–106, 131 182–184, 186–187, 191
space–time coordinates, 128 stable algebra, 165–166
space–time cylinder, 119 stable codimension-one, 179
space–time evolution, 119 stable dynamical family with
space–time foam, 437 degeneracies, 188
space–time geometrodynamics, 106, stable family of dynamical systems,
130–131 185
space–time periodicities, 100, 106, 128 stable functions, 187
space–time periodicity (see also stable interactions, 173
‘intrinsic periodicity’, stable map, 185
‘four-periodicity’, stable quantum algebra, 163
www.ebook3000.com
Index 717
stable sequences of higher order superposition principle, 308

bifurcations, 173 superstatistics, 437
stable theories, 154, 163 superstring models in 9 + 1
stage, 143 dimensions, 422
standard model, xiv, 48, 68, 79, 113, supersymmetric, 440
502, 525, 529–530, 532, 535, 541, supersymmetry, 342
637, 657 Swenson, Loyd S., 452
star-product, 7, 9, 21, 160, 675 symmetric, 401
state, 115, 637 symmetric formulation of the electron
state space, 180, 232–233 theory, 400
state space object, 231 symplectic, 32, 616
state vector, 311 symplectic capacities, 38
state-cycles, 334 symplectic capacity, 31, 37
states, 232 symplectic Clifford algebras, 16, 616,
stationary boundary conditions, 104 626, 643, 657
statistical behavior, xii symplectic form, 628, 643
statistical description, 113–114 symplectic geometry, 5
stochastic electrodynamics, 436 symplectic group, 5
Stokes theorem, 130 symplectic group Sp(2n), 5
Stone–von Neumann theorem, 3 symplectic non-squeezing theorem, 38
string, 618–619 symplectic space, 626
instantonic strings, 650, 652 symplectic vectors, 626
string theory, 46, 52–53 system, 115, 465
strings, 650, 658 closed, 465, 472, 481
strong rigidity, 194 dissipative, 465, 467, 472
strong transversality, 182 open, 465, 469
structural stability, 153, 158, 174,
systems under observation (SUOs),
176, 178, 180–183, 187
137
structurally dynamic cellular
network, 327, 330
T
structurally stable, 181, 183, 189
structurally stable codimension-one, ’t Hooft cellular automata, 95, 109,
178 114
structure constants, 192 ’t Hooft, G., xiii, 95, 114, 560
Stueckelberg equation, 622 temperature, 129
Stueckelberg theory, 625–626 temporal box, 123
SU(2), xvii tensor product, 111, 249
SU(3) × E6 gauge fields, 422 tensor product of Hilbert spaces, 110,
SU(N) Chern–Simons gauge field, 127
419, 421 Teresa Iaria, xviii
sublocale, 243, 260 theatre of coordinates, xvi
subobject classifier, 228, 242 theoretical physics, 131
supercausality, 103, 105 theory of nothing, 549
superconductivity, 95, 129–130, 420 thermal field theory, 475
superfluid helium, 305 thermal noise, 98, 129
superposition, 110, 112, 114–115 thermal time, 129
718 Index
Thomson, Joseph John, 434 transactional network, 575

three elementary-fermion generations, transactional ring, 567
613 transmission law, 398
threshold function, 349 transversality, 181–183
time, 564 Treder, H.J., xvi
advanced, 395 triangleland, 624
archaic precursor, 561 trivial, 191
cosmic time, 564 Troshkin, Oleg, 434
in non-relativistic QM, 373 truth object, 232
mean dwell time, 380 truth value, 137–138, 140, 229, 231
operator, 373
tunnel effect, 126
observable, 373
tunneling with dissipation, 393
particle traversal time, 380
tunneling time, 373, 410
thermal time, 288
tunnelings, 412
quantum theoretical observable,
373, 380 turbulence, 435–437, 451
time (t-)representation, 374 turbulent, 434, 436–438, 444, 451, 459
time capsule, 624 Turing, A.M., xv
time cycles, 132 twin-slit experiment, 517, 522, 532,
time evolution, 112, 120 535
time evolution operator, 120, 180, 567 twin-slit interference, 502, 523, 525
time flow, 95, 100–101, 105, 132 two-time correlation function, 325
time ordering, 132 two-time function, 317
time precursor, 575
time reversal invariant, 310 U
time–temperature relation, 561 U(1) × SL(2,c,D), 582
time-dependent Schrödinger equation
ultra-fast cyclic dynamics, 113
with dissipative terms, 392
ultra-fast cyclic universe, 132
time-reversal symmetry, 466, 472
ultraviolet catastrophe, 108
time-translation, 465
uncertainty correlations, 390
timekeeper, 113–114
uncertainty principle, 570, 572
timeless approaches, xvii
topological conjugacy, 181 uncertainty relation, 385
topological dimension, 342 undulatory mechanics (see also
topological equivalence, 181 ‘wave-particle duality’, ‘phase
topological space object, 252 harmony’), 98, 103
topological structure, 252 unfolding, 178, 189
topology, 109, 128 unification, 637
topos, 227–228 unification of all interactions, 52
topos quantum theory, 257 unification of quantum and
Topos theory, 227 relativistic mechanics, 95
TOTEM Collaboration, xvii unified description of physics, 133
trans-Planckian regime, 300 unified description of relativistic and
transactional interpretation of QM, quantum dynamics, 103
409 uniform rectilinear motion, 98–99
transactional loop, 568 unit, 194
www.ebook3000.com
Index 719
unitarily inequivalent representations, tangent, 645

465, 467, 474–475 trivector, 616
unitary evolution, 307 vector field, 181, 183
unitary operator, 120 vertex degree, 341
unitary symmetry, 320 vibrating string, 94, 99, 101, 103,
unitary transformations, 312 107–108, 111–112, 125
vibrational modes, 102
universal constant, 174
vierbein, 647
universal enveloping algebra, 194
virtual extra dimension, 132
universal unfolding, 188–189
Vitiello, G., xviii
universality, 173–174, 185 von Neumann, 6–8, 21–22, 411
universality class, 178 von Neumann algebra, 13, 19–20
universe, 654 von Neumann, J., 377
origin of Universe, 564 von Neumann–Moyal algebra, 16, 20
Universe as an information network, von Weizsäcker Ur Materie, xiv
xv vortex and anti-vortex, 426
unstable algebra, 163 vortex-like solitons, 425
unstable manifolds, 182
unstable states, 391, 412 W
wave function, xiii, 115, 294, 428,
V 622, 624–625, 634, 656
universe wave function, 572
vacuum, 434, 436, 438–440, 444, 451, wave function normalization, 317
459, 462, 560, 587, 591, 617, wave packet, 622, 625
635–636, 640–641, 653–655
wave particle duality phase harmony,
bare vacuum, 642 95
Dirac vacuum, 642 wave-function collapse, 404, 561
physical vacuum, 642 wave-packet instantaneous collapse,
Universal Action Reservoir, 560 401
vacuum state, 616, 640 wave-particle duality (see also,
vacuum energy, 124, 430 ‘undulatory mechanics’ and
validation, 140 ‘de Broglie, L.’, ‘phase harmony’),
variational derivative, 312, 320 94, 96–100, 105, 125
waves
variational principle, 104–106
anti-evanescent (increasing)
vector, 618, 640
waves, 381
basis vector, 616–617, 621, waves functions, 106
628–631, 643–645 wedge product, 627–628
bivector, 616, 618, 620, 656 Weisskopf, V., xviii
fermionic, 645 Weyl quantization prescription, 160
Hamilton equations of motion, Weyl–Heisenberg algebra, 477
630 Weyl–von Neumann algebra, 6
infinite dimensional, 630, Weyl–von Neumann algebra and the
639–640 Moyal algebra, 1
symplectic, 616, 627–628 Weyl–von Neumann approach, 7
720 Index
Weyl–von Neumann–Moyal algebraic world sheet, 650–652

approach, 9 worldline, 270
Wheeler’s participatory principle, world-line parameter, 131–132
137, 139 world-sheet parameter, 132
Wheeler, J.A., 573 worldline, 620–621, 632
Wheeler, John A., 139, 142 wormhole, 328
Wheeler, John Archibald, 137, 149, wormhole space, 331, 354, 357, 365
437
Wheeler–DeWitt equation, 56, 58 Y
Wick product, 673 Yang, T.D., 638
Wick rotation, 410, 560, 574 Yang–Mills gauge symmetries, 419
Wigner, 1, 9, 14
Wigner approach, 21 Z
Wigner function, 8–9, 21–22
Zachos, 11
Wilson/Kadanoff picture, 353
Zariski topology, 192
winding numbers, 119
Zeeman effect, 129
Witt basis, 616, 628–629, 639–640,
zero-degree cohomology, 676
642
zitterbewegung, 94
WKB method, 95, 123–124, 126
world crystal, 299, 304
floppy world crystal, 305
world crystal of lattice spacing lP
with defects, 299
www.ebook3000.com

Beyond Peaceful Coexistence The Emergence of Space, Time and Quantum

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Beyond Peaceful Coexistence The Emergence of Space, Time and Quantum

Caricato da

Copyright:

Formati disponibili

www.ebook3000.

This page intentionally left blank

P1045_9781783268313_TP.indd 2 7/3/16 10:18 AM

Library of Congress Cataloging-in-Publication Data

British Library Cataloguing-in-Publication Data

Copyright © 2016 by Imperial College Press

In-house Editors: Shilpa/Mary Simpson

Typeset by Stallion Press

Shilpa - Beyond Peaceful Coexistence.indd 1 23/12/2015 9:12:53 AM

From Peaceful Coexistence to Co-Emergence xi

The Algebraic Way 1

Fermi Blobs and the Symplectic Camel: A Geometric

Space–Time in Quantum Gravity: Does Space–Time

Introduction to the Quantum Theory of Elementary

Observers and Reality 137

The Stability of Physical Theories Principle 153

Factory of Realities: On the Emergence of Virtual

Space–Time from Topos Quantum Theory 221

From Born Reciprocity to Reciprocal Relativity: A

On Non-Equilibrium Thermodynamics of Space–

World Crystal Model of Gravity 299

Quantum Features of Natural Cellular Automata 307

Structurally Dynamic Cellular Networks as Models

On a Time–Space Operator (and other

Emergent Space–Time 419

The Idea of a Stochastic Space–Time:

. . .And Kronos Ate His Sons 465

The Emergence of Space–Time: Transactions

An Adynamical, Graphical Approach to Quantum

Is Bohr’s Challenge Still Relevant? 545

In and Out of the Screen. On Some New

Schrödinger–Milne Big Bang — Creating a “Universe

Quantized Fields à la Cliﬀord and Uniﬁcation 615

Non-commutative Einstein, almost Kähler–Finsler

indeed as strange as it looks? Or could it be that this is what we should

From Peaceful Coexistence

It may be that a real synthesis of quantum and relativity theories requires

A physical theory consists of a formal structure and one or more interpre-

for the foundational debate and the historical stratiﬁcations of Quantum

From Peaceful Coexistence to Co-Emergence xiii

oneness” or can we consider it “emergent” in some way? The quantum

relate to microscopic variables in the same way thermodynamic concepts

From Peaceful Coexistence to Co-Emergence xv

information. We should rather speak about manifestation theories as U.

The manifestation of the world consists in a transition from a condition

We will not dwell on the cultural and paradigmatic diﬀerences hidden

From Peaceful Coexistence to Co-Emergence xvii

From Peaceful Coexistence to Co-Emergence xix

11. D. Bohm, Wholeness and the Implicate Order, Routledge, 1980.

The Algebraic Way

In this chapter, we examine in detail the non-commutative symplectic

the Ritz–Rydberg combination principle, an empirical discovery in atomic

The Algebraic Way 3

for understanding individual experimental phenomena but also in deﬁning

so that in this case we are working in momentum space (p1 , p2 , . . . , pn ).

This representation enables us to work more easily with an arbitrary num-

3. Common Symmetries underlying both Classical

giving the special case {xi , pj } = δij .

The Algebraic Way 5

while in the classical case, it is commutative. Although these structures are

4. How do we Relate the Quantum Algebra

In the case of a translation in momentum space, we may similarly write

By recalling the Schrödinger representation, we can write these exponentials

Sψ (α, β) = ψ|Ŝ(α, β)|ψ.