Kotarbinski Logic PDF

KOTARBINSKI: LOGIC.
SEMANTICS AND ONTOLOGY

Nijhoff International Philosophy Series
VOLUME 40
General Editor: JAN T. J. SRZEDNICKI

Editor for volumes on Applying Philosophy: LYNNE M. BROUGHTON
Editor for volumes on Logic and Applying Logic: STANISLAW J. SURMA
Editor for volumes on Contributions to Philosophy: JAN T. J. SRZEDNICKI
Assistant to the General Editor: DAVID WOOD
Editorial Advisory Board:
R.M. Chisholm (Brown University. Rhode Island); Mats Furberg (Goteborg Univer-
sity); D.A.T. Gasking (University of Melbourne); H.L.A. Hart (University College.
Oxford); S. Komer (University of Bristol and Yale University); H.J. McCloskey (La
Trobe University. Bundoora. Melbourne); J. Passmore (Australian National Univer-
sity. Canberra); A. Quinton (Trinity College. Oxford); Nathan Rotenstreich (The
Hebrew University. Jerusalem); Franco Spisani (Centro Superiore di Logica e Scienze
Comparate. Bologna); R. Ziedins (Waikato University. New Zealand)
The titles published in this series are listed at the end o/this volume.
Kotarbinski:
Logic, Semantics and Ontology
edited by
Jan Wolenski
Jagiellonian University, KrakOw
and The Technical University, Wrodaw, Poland
..
KLUWER ACADEMIC PUBLISHERS
DORDRECHT / BOSTON / LONDON
Library of Congress Cataloging in Publication Data
Kotarblnskl logic. semantics. and ontology / edIted by Jan WOlenski.

p. cm. -- (Ni jhoff international phi losophy series; v. 40)
Includes bibliographical references and indexes.
ISBN ·13: 976-94·010-7442-1 e-ISBN-13: 976·94-009·2097·2
001: 10.10071976·94·009-2097·2
1. Kotarbinskl. Tadeuz. 2. LoglC. Symbolic and mathematical.

3. SemantIcs. 4. Ontology. I. WOlenskl. Jan. II. SerIes.
84691.K594K67 1990
199' . 438--dc20 90-40625
ISBN-13:978-94-010-7442-1
Published by Kluwer Academic Publishers,

P.O. Box 17, 3300 AA Dordrecht, The Netherlands.
Kluwer Academic Publishers incorporates

the publishing programmes of
D. Reidel, Martinus Nijhoff, Dr W. Junk and MTP Press.
Sold and distributed in the U.S.A. and Canada

by Kluwer Academic Publishers,
101 Philip Drive, Norwell, MA 02061, U.S.A.
In all other countries, sold and distributed

by Kluwer Academic Publishers Group,
P.O. Box 322, 3300 AH Dordrecht, The Netherlands.
Printed on acid-free paper
All Rights Reserved

e 1990 by Kluwer Academic Publishers
Softcover reprint of the hardcover 1st edition 1990
No part of the material protected by this copyright notice may be reproduced or

utilized in any form or by any means, electronic or mechanical,
including photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner.
TABLE OF CONTENTS
Preface vii
Notes on Contributors lX
Tadeusz Kotarbinski
Philosophical self-portrait
Kazimierz Ajdukiewicz
Review-article: T Kotarbinski's Elements of the Theory of
Knowledge, Formal Logic and Methodology of the Sciences 7
Alfred Gawro~ski
Psychologism and the principle of relevance in semantics 23
Peter Geach
Names in Kotarbinski's Elementy 31
Andrzej Grzegorczyk
Consistent reism 39
Henryk Hiz
A note about reism 47
Janina Kotarbinska
Puzzles of existence 53
Czesraw Lejewski
On the dramatic stage in the development of Kotarbinski's
pansomatism 69
Marian Przer~cki
Semantic reasons for ontological statements: the argumentation
of a reist 85
Tadeusz Pszczolowski
Philosophical and methodological foundations of Kotarbinski's
praxiology 97
vi
Vito F Sinisi
Kotarbinski's theory of genuine names 107
Vito F Sinisi
Kotarbinski's theory of pseudo-names 119
Barry Smith
On the phases of reism 137
Klemens Szaniawski
Philosophy of the concrete 185
Jan Wolenski
Kotarbinski, many-valued logic, and truth 191
BogusJaw Wolniewicz
Concerning reism 199
Ewa Zarneck"a-Bialy
The voice of the past in Kotarbinski's writings 205
References 213
Index of names 225
Index of subjects 229

PREFACE
Tadeusz Kotarbinski is one of towering figures in contemporary

Polish philosophy. He was a great thinker, a great teacher, a great
organizer of philosophical and scientific life (he was, among others,
the rector of the Uni versi ty of t6dz, the president of the Polish
Academy of Sciences, and the president of the International Institute of
Philosophy), and, last but not least, a great moral authority. He died
at the age of 96 on October 3, 1981.
Kotarbinski was active in almost all branches of philosophy. He
made many significant contributions to logic, semantics, ontology,
epistemology, history of philosophy, and ethics. He created a new field,
namely praxiology. Thus, using an ancient distinction, he contributed to
theoretical as well as practical philoso~hy.
Kotarbinski regarded praxiology as his major philosophical "child".
Doubtless, praxiology belongs to practical philosophy. This collection,
howewer, is mainly devoted to Kotarbinski' s theoretical philosophy.
Reism - Kotarbinski' s fundamental idea of ontology and semantics - is
the central topic of most papers included here; even Pszczolowski' s
essay on praxiology considers its ontological basis. ,Only two papers,
namely that of Zarnecka-Bialy and that of Wolenski, are not linked with
reism. However, both fall under the general label "Kotarbinski: logic,
semantics and ontology".
The collection partly consists of earlier published papers.
Kotarbinski's "Self-portrait" and Szaniawski' s "Philosophy of the
concrete" first appeared in Dialectics and Humanism 5 (1977), the
English translation Ajdukiewicz' s review of Kotarbinski' s Elements of
the Theory of Knowledge, Logic and the Methodology of Sciences - in
Kotarbinski, Gnosiology, Ossolineum-Pergamon Press, Wroclaw-Oxford 1966,
Kotarbinska's "Puzzles of existence" in Semiotics in Poland,
1894-1969, ed. by J. Pelc, Polish Scientific Publisher- D. Reidel,
Warszawa-Dordrecht 1978, Lejewski 's "The dramatic stage in the
development of Kotarbinski's pansomatism" - in Ontology and Logic, ed.
by P.Weingartner and E.Morscher, Duncker and Humblod, Berlin 1979,
Sinisi's "Kotarbinski's theory of genuine n.ames" and "Kotarbinski's
theory of pseudo-names" in Theoria 30 (1964) and 31 (1965),
respecti vely. I am very much grateful to editors and publishers for
their kind reprint permissions. Also many thanks are due to Lynn
Broughton for her checking of English in most papers and to Jaroslaw
Fall for his help in preparin~ the camera ready text.
Jan Wolenski
NOTES ON CONTRIBUTORS
Tadeusz Kotarbinski (1886-1981) was a Professor of Philosophy in

the University of Warsaw and the University of todz.
Kazimierz Ajdukiewicz (1890-1963) was a Professor of Philosophy in
the University of Lvov, the University of Poznan and the University of
Warsaw.
Alfred Gawronski is a Lecturer of Philosophy in the Pontifical
Theological Academy in Krakow.
Peter . Geach is an Emeritus Professor of Philosophy in the
University of Leeds.
Andrzej Grzegorczyk is a Professor of Philosophy in the Polish
Academy of Sciences.
Henryk Hiz is an Emeritus Professor of Linguistics and Philosophy
in the University of Pennsylvania in Philadelphia.
Janina Kotarbinska is an Emeritus Professor of Philosophy in the
University of Warsaw.
Czeslaw Lejewski is an Emeritus Professor of Philosophy in the
University of Manchester.
Marian Przel~cki is a Professor of Philosophy in the University of
Warsaw.
Tadeusz Pszczolowski is a Professor of Philosophy in the Polish
Academy of Sciences.
Vito F. Sinisi is a Professor of Philosophy in State University of
New York at Binghamton.
Barry Smith is a Professor of Philosophy in Internationale Akademie
der Philosophie in Schaan, Liechtenstein.
Klemens Szaniawski (1925-1990) was a Professor of Philosophy in the
University of Warsaw.
Jan Wolenski is a Professor of Philosophy in the Jagiellonian
University in Krakow.
Boguslaw Wolniewicz is a Professor of Philosophy in the University
of Warsaw.
Ewa Zarnecka-Bialy is a Professor of Philosoph~ in the Jagiellonian
University in Krakow.
ix
PHILOSOPHICAL SELF-PORTRAIT
Tadeusz Kotarbinski
The proposal to sketch an outline of my philosophical profile originates

wi th the editors of Ricerche Filosofiche who suggested that the essay
could be written either by one of my friends or by myself. However, I
realize how difficult it is to convey a realistic picture of a friend;
such a picture is prone to assume the guise of kindly eulogy. I would
rather then risk sketching my philosophical self-portrait in an attempt
to maintain a fair degree of objectivity.
I was born on March 31, 1886 in Warsaw, which was then occupied by
tsarist Russia. My father was a painter by profession and taught in
various art schools. His hobby was singing and composing music. My
mother played the piano and collaborated with my father in his musical
undertakings. Through the atmosphere in the home I imbibed the spirit of
painting, music, and poetry, and the spirit of romantic art - the style
which reigned in our home throughout my childhood and adolescence.
Although the background of both my parents was Catholic, they were far,
especially in their later years, from accepting religious dogmas.
In my intellectual development I was greatly influenced by the type
of schooling I received, a comprehensive secondary schooling of the
classical type with Greek, Latin, French, German, and Russian (the
official language at schools). It also provided rudimentary course and
training in mathematics though the main objectives were the languages
and, to a smaller extent, literatures. My interest in the philosophy of
language, unabated to this very day, was initially aroused by the
excellent way in which French was taught.
Having completed secondary education I hesitated as to my future
profession. Eventually I started with mathematics and physical scienCeS,
then attempted architecture, but finally decided on philosophy as my
principal subject with classical philology (Greek and Latin) as my com-
plementary course. I studied philosophy in Lwow (where a Polish uni¥er-
sity flourished within the Austrian monarchy) under the tutelage of
Kazimierz Twardowski, a disciple of Franz Brentano. Youth of my genera-
t ion owed a great deal especially to Twardowski, who was a master of
systematic work, quite indomitable in demanding exactitude and precision
of expression as much from himself as from everybody around hi ••
On returning to Warsaw I found a job as a form master and Latin
teacher in one of the private secondary schools where teaching was done
in Polish. Finally, in 1919, I was appointed professor extraordinary in
1. Woletfsld (ed.), Kotarbitfsld: Logic, Semantics and Ontology, 1-6.

© 1990 Kluwer Academic Publishers.
2 TADEUSZ KOTARBINSKI
philosophy at Warsaw University, reopened in independent, united Poland.

In 1929 I received the title of professor ordinary. During the period of
the Nazi invasion (1939-45) my activities were suspended, but after the
war I resumed working at the un i vers i ty where I taught philosophy,
mainly logic, until 1960 when I retired. Between 1945 and 1949, while
all the time holding my professorship at Warsaw University I was involv-
ed in organizing a new university in the city of todi, of which I was
rector. For fifty years I have been president of the Polish Philosophi-
cal Society, between 1960 and 1963 I held the chairmanship of the Inter-
national Philosophical Institute; while between 1957 and 1963 my duties
included those of the president of the Polish Academy of Sciences.
Included in this biographical sketch are circumstances that may
throw som~ light on my intellectual identity and, indirectly though,
help to understand the main lines of my activity together with 'individu-
al problems and personal convictions.
I began with ethics. The system and style of ethics in which I was
brought up and educated in childhood was evangelical ethics conceived as
the ethics of charity. At about fifteen years of age I parted - it was a
painful yet a complete parting - with the illusion of religious substan-
tiations of morality. However, since I was, and still am, attached to
its motivational content I launched into a quest to support it otherwise
and pondered the proposals of stoicism, epicureanism and utilitarianism.
This long search bore fruit after years of trying and erring in the out-
line of an ethical system that I evolved and which I propose to call in-
dependent ethics, the ethics of practical realism, the ethics of the
• reliable guardian'. I propose to refer to this ethics as independent
because it refuses to seek its substantiation in sources other than the
emotional evaluation developed through human interrelations. It does
without the doctrines of the hereafter or theories concerning the rela-
tion of spirit to matter. They seem as much superfluous in the domain of
ethics as they are spurious in the field of medicine, economy, or law.
Philosophy may be useful for ethics, but only insofar as it fashions a
critique of cognition, e.g. to prove the illusoriness 'of argumentation
derived from philosophical conceptions of the world when the latter are
grafted onto the domain of ethics. Practical realism, however, by no
means consists solely in liberating ethics from everything which is of
no concern to the matter; in formulating ethical postulates it imposes
coming to terms with limited possibilities as well as with the hierarchy
of values according to which an intelligent human being ought to decide
about his course of action. Do not destroy the existing, bearable system
of relations if the change is for the worse. Do not seek to attain the
maximum of joy both in your own life and in the lives of those depending
on you; instead, try to eliminate and prevent misfortune, for the great-
er the evil prevented, the more important the activity. This is precise-
ly the delland of consciense which assesses hUllan activity against the
scale ·commendable-disgraceful'. It is a demand of conscience to assist
those whose future life depends in sOlie measure on us in preventing mis-
fortune. The attitude to be illi tated in this respect is that of the
reliable guardian who can be counted upon in difficulties and in whose
conduct the following virtues are illplicit: courage, sympathy, right-
eou~ness, endurance, inner discipline. I had painfully arrived at these
PHILOSOPHICAL SELF-PORTRAIT 3
convictions, and this is the kind of ethics that I advance despite the
opposi tion of both the traditionalists and the revolutionaries. The
former accuse me of free-thinking which supposedly severs ethics from a
belief in justice ruling the world and in the redemption of guilt and
merit within eternity; the latter, on the other hand, reject any system
of universal ethics and consider moral judgements to be .intrinsically
relative: nothing can be permanent or universal as ethics is believed to
stem from the needs of social groups while these needs are viewed as
essentially changeable depending on circumstance and historical moment.
The kind of ethics described above, on the contrary, advances an intrin-
sic identity of human conscience in any society. For it seems that
assessment of human deeds against the scale 'commendable-disgraceful'
has evolved through recurrent situations when facing danger it was
necessary for the stronger to defend the weaker. Heroes of defence
action of this kind are worshipped everywhere while cowards and cravens
who cannot be relied upon in such circumstances are universally held in
bontempt. What differs from society to society is the emphasis placed on
various virtues of which the moral equipment of the human being is sup-
posed to be composed. Moreover, the degree to which vices as contrasted
to virtues are condemned is also variable.
Ethics has not'become the subject of my professional scholarly con-
centration even though its problems have always been of utmost signifi-
cance to me. I am too skeptical about the possibility of working out a
detailed system of principles underlying wisdom, a system which would
lend itself to intersubjective substantiation. What is more, teaching
ethics professionally, in my opinion, does not tally with the essence of
its problems unless one cultivates a historical, sociological, explicat-
ive knowledge of styles and conceptions of morality instead of tackling
the essential issues which always demand prospective, prescriptive
answers. Only then the subject taught would no longer be ethics but a
kind of science des moeurs, transcending the specific tasks of this
department of philosophy. I reached the chair of philosophy via logic.
Teaching logic became the field of my activity as a university professor
of philosophy, a member of other humanistic faculties. Emphasis is here
placed on the words 'teaching' and 'humanistic'. For my lectures and
classes were conceived as an organon in the classical sense of the term,
for philosophers as well as for those who, having completed their course
of study, would espouse the cause of disseminating humanistic knowledge
and thinking, particularly future secondary school teachers. Somewhat
later my activity embraced also law students. My linguistic equipment
proved to be very helpful in this respect. For it seems especially im-
portant when the problems of an organon of this kind are conceived his-
torically and is quite crucial when pondering the original Organon of
Aristotle (or to be more cautious, of the peripatetic school) and its
continuators. Conceived in this m~nner logic was by no means confined to
formal logic, but came to'comprise the problems of epistemology, seman-
tics and methodplogy. It is precisely the latter problems - not those of
formal logic - that were of particular interest to my mind. Neverthe-
less, I felt bound to contribute to the study of formal logic. The
feeling was encouraged both by my colleagues at Warsaw University and by
my awareness of the precise phase that logic had reached in its histori-
cal development. It was precisely the moment when mathematical logic was
triumphantly entering the scene. The names of Frege, Bertrand Russell,
Peano, Burali-Forti, Coutu rat and many others were on everybody's minds.
Mathematical logic was closely allied with the rapidly developing set
theory. The international periodical devoted to the latter, Fundaaenta
Mathematicae was, and still is, published in Warsaw. The distinguished
philosophic-mathematical logicians: Jan Lukasiewicz and his disciple and
my colleague, Stanislaw Lesniewski among many others, were active here.
I only mention the names of those persons to whom my studies in mathema-
tically oriented formal logic are particularly indebted. In this respect
lowe a lot to my close alliance with Professor Lesniewski. I simply
took over his original system of formal logic to suit my own purposes. 1
Relieved thus from the necessity to contribute to formal logic itself, I
could concentrate on the problems I faced as a teacher of logic to be
used by humanists. These centred around the problem of overcoming the
hypostases of linguistic origin, what Francis Bacon referred to as idola
for i. Both our everyday language and the language of the sciences as
well are teeming with nouns or noun-like forms. Hence the tendency to
perceive an object behind them even when the noun is an abstract one,
like for example, 'roundness', 'equality', etc. Once the existence of
the alleged objects of such names is admitted, once we agree to the
existence of such qualities or relations, human thought is made to wade
through a mire of apparent ontological problems. They in turn impose a
literal interpretation of the expressions like 'a quality inheres' in an
object in the same way as a nail is embedded in a wall: whereas, in
point of fact, their meaning is only metaphorical. Leibniz himself was
of the opinion (which he expressed in Nouveaux essais 2, XXII, § 1) that
problems bristling with difficulties can be dispelled as soon as we
stick only to the names of concretes in our discourse. Unaware both of
these words and equally ignorant of Franz Brentano' s similar idea 2 I
formulated in 1929 the principles of the so called reism. In its most
mature formula it declares war against the hypostases of linguistic ori-
gin on the following lines: inasmuch as it is possible try to formulate
statements in a way that would eliminate all names other than the names
of objects, that is, physical bodies or parts thereof. Persons ought to
be regarded as objects, i.e. sentient objects. Sentences may contain
words that are not names, e.g. verbs or conjunctions, etc. The point is,
however, to eliminate naaes other than the names of objects. Let me
hasten with an example of a reistic interpretation of sentences. 'Pru-
dence inheres in wisdom' simply menas: 'Every man who is possessed of
wisdom is prudent.' 'Bonds of brotherhood related Orestes to Electra'
simply lIeans: 'Orestes was Electra's' brother.' A reist by no means
demands that the use of sentences with abstract expressions like the
names of qualities or relations be completely abandoned. Quite the
contrary, the necessity of applying them is fully recognised just
i.rhose int.erest.ed in t.he .,st.em are referred t.o Lusch .. i (1962).
2It. waa he who oit. .. d L.ibniz'. word. in a foot.not.e t.o hi .. (1911) - a.e
'p', 163 in t.he 19:16 edit.ion, voL 193 of t.he Meiner Philoaophiache
8ibliot.hek, compare oleo Brent.ano'a dict-ot-ed works, •. 9. (1916).
PHILOSOPHICAL SELF-PORTRAIT 5
because their presence may often reduce the length of the statement. The
only thing he insists upon is to try to be able to do without nalles
which are not the names of things. I lIay add, by the way, that at
present I prefer to use the term 'concretism I instead of the term
'reism' as my readers were prone to identify 'reism' with 'realism'
while the meanings of the two are totally different.
Thus, reisll, that is concretism (or somatism - as I identify all
objects with bodies and in Greek 'solla' means 'body') proves to be a
certain innovation of my organon. It is however, . highly debatable as a
conception since a number of difficulties inhering in an attempt to
interpret reistically theorems of set theory have not yet been overcome.
If I were to be asked, however, about the main field of my scholar-
ly interest, I would disregard logic for the sake of a different discip-
line of science, namely praxiology. Until quite recently little has been
known about the existence of this branch of science, although as early
as 1890 Espinas delineated its .perspectives in the Revue Philosophique.
There is advised observation of the progress in various practical skills
with a view to improving the efficiency of any purposeful activity. I
approached praxiology via a different route as I was ignorant of Espinas
proposals. What struck me as odd was that concepts like 'activity',
'manner', 'method' or 'product' (which comprehend various elements or
phases of activity) are currently used together with those which convey
a technical, as it were, assessment of the activity itself: 'carefully',
'accurately', 'waste', 'bungler', 'a shoddy piece of work', while none
of the existing branches of science seemed to deal with them all. Yet a
separate science which would ponder the conditions of efficient action
in general, with reference to all spheres of human endeavour seelled
quite indispensable. For also in ethics, 'where the matter concerns among
others the pr9blem of living rationally, one invariably must have resort
to this or other recommendations of economy - for example when the
method of the golden mean is advised, or indolence and inaccuracy in
doing the job are censur~d. Similarly, an appropriate meting out of res-
ponsibility depends upon a rational general knowledge of the relations
between the activity and its intended and undesigned results.
On the other hand, even within the organon such matters as, for
example, economy in demonstrating the truth or falsity of a theorem or
in constructing the system of premises, are often taken into considera-
tion. And once we gain an insight into the wealth of possible modes of
acting, we are bound to encounter irritating paradoxes that call for
explanation, e. g. that one may act by simply doing nothing or that
greater forces can be overcome by means of smaller ones. It was also be-
cause of the~e reasons that I took an interest in praxiology. Praxiolo-
gical problems loomed especially large when the matters of organization
and management came to be reflected upon on a sufficiently general level
with respect to their application in solving the problems of industry
and trade. My interest in these matters dates back to my undergraduate
days at the university, and it was uninf1uenced either by my professors
or my fellow-students. Over sixty years ago I published in Polish my
first book on the subject, but I tackled the problems of praxiology
quite frequently afterwards. During the period when I held the presiden-
cy of the Polish Academy o~ Sciences the first team of scientists in ,the
world to work on praxiological problems was organized. In my later years

I published a comprehensive book, something of a first handbook of pra-
xiology - Traktat 0 dobrej robocie (1955a), which I consider to be the
work of my life (together with the Elementy, 1929).
REVIEW-ARTICLE: T. KOTARBINSKI'S ELEMENTS OF n£ THEORY OF
KNOWLEDGE. FORMAL LOGIC AND METHODOLOGY OF TI-E SCIENCES 1
Kazimierz Ajdukiewicz
1. A GENERAL CHARACTERIZATION OF THE WORK AS A TEXTBOOK AND AS A SCIEN-

TIFIC CONTRIBUTION.
In his preface to the book with which we shall be concerned here the
author describes his work as a textbook for those students of the human-
ities who are preparing for the examination in "the principles of philo-
sophy", which is obligatory for all those candidates for the degree of
Master of Arts in Polish uni versi ties who do not take philosophy as
their principal subject. He also thinks that it may serve as a.textbook
for those who are to sit for an examination in logic, methodology and
epistemology, obligatory for those who major in philosophy and pedagogy.
This presentation of the work as a textbook may give rise to a bias
against its scientific value, since textbooks usually are compilations,
expositions of things already accepted, and as a rule do not contribute
much new to science. In the present case, although the work is called a
textbook, it is not merely that. Professor Kotarbinski's book comprises
a wealth of tersely formulated original opinions which would suffice to
fill a number of monographs. To Polish culture it contributes new
values, not only didactic, but also scientific.
The author in many places presents his own views concerning episte-
Bology, methodology, ontology (in the Aristotelian sense) and semantics.
A.nd where he relates epistemological, metaphysical and methodological
trends in the history of human thought, he does not confine himself to
summarIzlng other people's oplnlons, but. undertakes original and
valuable analytical study in that he offers a thorough and penetrating
analysis of important philosophical issues and clearly describes the
various standpoints on those issues. In its analytical part the book
reaches a level which is approached by very' few works in the world
literature on the subject. Neither does the author confine himself to
interpretative analysis alone, but subjects the trends in question to a
cri tical and always profound scrutiny, in doing which he very often
formulates his own opinion.
1AII I>age-ref'erences are t.o t.h .. English t.ranslat.ion of' t.he Element.y.
t. •. t.o Kot.arbinski (1966).
7
I. Woletfsld (ed.), Kotarbitfsld: Logic. Semantics and Ontology, 7-21.
8 KAZIMIERZ AJDUKIEWICZ
The importance and variety of the issues in which the author

presents his own point of view, and the uniformity of the point of view,
revealed in all the author's own opinions, account for the fact that the
reader feels that the author has evolved a philosophical system.
In view of the double character of the book, which is an important
new scientific contribution to the treasure-house of Polish culture, and
also is conceived as a university textbook, we have to appraise it from
two points of view: scientific and didactic. In the present review we
shall be mainly concerned with the book as a scientific contribution. In
view of the wealth of the author's own opinions it would not be possible
to appraise all of them. We shall confine ourselves to such critical
remarks as seem to pertain to important points, and which we can at
present offer.
2. A REVIEW OF CONTENTS
The book consists of five parts. The first entitled Remarks on language,
is concerned with the principal semantic relations, such as expressing,
meaning and denoting, with the kinds, or categories, of expressions,
with defects of language and with definitions as a means of eliminating
such defects. That part also includes the exposition of the author's
basic standpoint, which he calls reism. We call that standpoint basic,
because it is from that point of view that the author settles many phi-
losophical issues which are considered classical. Here, too, the reader
will find an exposition of the problem of universals and the formulation
of the author's nominalist standpoint on that issue. All of the Part One
consists of an original contribution by the author, only broken in a few
places by the presentation of results known otherwise.
The second part deals with epistemology. We find there a profound
analysis of the various shades of idealism and realism, and the exposi-
tion of the author's own standpoint, which he calls radical realism. We
find there also an analysis of the concepts of truth, criterion of
truth, and an analysis of the terms: dogmatism, scepticism and criti-
cism. All these analyses are preceded by a study of the meanings of the
terms 'perceptual image', 'reproductive image', and 'productive image',
by a distinction as between the immanent and the transcedent object of
an image and the content of an image. The chapter concerned with con-
cepts naturally belongs here also.
The third part, the least original in the book, presents informa-
tion on formal logic, almost entirely known from other sources. It con-
sists of the theory of deduction (an axiomatic exposition of Russell's
system as simplified by Lukasiewicz), traditional logic, and some infor-
mation on what is called Lesniewski's ontology.
An original contribution in the last-named chapter seems to consist
in bringing out the fine relationship between the formulas of the sen-
tential calculus and the formulas of the calculus of terms.
The fourth part is concerned with a general methodology of sciences
and falls into three chapters. The first, on reasoning, works out the
classification of reasonings known from Lukasiewicz's article (1912).
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS •.• 9
Its new achievements consist apparently in that the author penetratingly

points to the ambiguity in the current usage of such methodological
terms as 'inference', 'demonstration', 'hypothesis', etc. The second, on
the deductive method, deals with the structure of axiomatic systems, but
brings no valuable novelties. The third, on the inductive method,
discusses the methods of Bacon and Mill and analyses them. It also
studies the methods of discovering functional relationships in empirical
ways (which gives an opportunity for penetrating remarks on relation-
ships). The fortunes of the concept of cause are described, and the
concept of hypothesis is.analysed. Finally, measurement, observation and
experiment are disoussed. This chapte'r offers a number of interesting
and very fine methodological analyses, rarely to be found in textbooks
of methodology.
The fifth part is entitled Analysis of the . specifying features of
the principal branches of sciences. Very sober and much needed remarks
on the concept of science and the classification of sciences are follow-
ed by further chapters. The first deals with the attellpts to define
mathematics by pointing to its distinct subject matter and by describing
the method used by that discipline. The author is not satisfied with
either solution, but declares himself in favour of the opinion that the
distinctive characteristic of mathematics consists in the purely formal
nature of its theses (which he explains in greater detail), combined
with its deductive method. The second chapter deals with the peculari-
ties of natural science. The author sees the distinctive feature of that
group of disciplines in that their problems pertain to those properties
of things which are their attributes as a result of the natural course
of events. But his principal analysis concentrates on the opposition
between physics and psychology, understood as the disciplines concerned
with body and soul, respectively, or with physical and psychic phenom-
ena, respectively. This gives the author an opportunity to discuss the
concepts of substance, soul and body, the issue of materialism (in fa-
vour of which, as the thesis stating that those objects which experience
something are bodies, the author declares himself), the concept of phe-
nomenon, and the difference between psychic and physical phenomena. This
chapter also brings the rejection of what are called ideal objects, and
the exposition of an exceptionally interesting doctrine, known from the
author's paper (Kotarbinski 1922) and concerned with the interpretation
of psychological statements. That chapter is completed by an analysis of
the concepts of time, space and motion, with an illusion to physical
relativism.
The next chapter characterizes the historical sciences as those
which are concerned with description and genetic and causal explanation
of past events. Reference is made here to the idiographic and nomothetic
aspects of the science of history, and to historical analysis -and syn-
thesis. The humanities are treated as a section of the historical
sciences and described as those historical sciences which investigate
the content, course, origin, importance and causes of human experiences
and actions, products of such actions, and dispositions to those experi-
ences and actions in human beings and in their environment. Reference is
made in this connection to the problems of determinism and vitalism.
The penultimate chapter, entitled On practical abilities, provides

original analyses of the concepts of valuation, nora and project.
The book is rounded off by a very short chapter on the philosophi-
cal disciplines.
The reader who has read this list of the issues raised in the book
under discussion immediately realizes that a review cannot examine all
these questions without assuming the size of a book. We shall therefore
now discuss the monograph part by part and present to the reader those
critical remarks which we consider to be of maj~r importance.
3. REISM
Let us begin with reisa. We are not in a position to formulate, on the

basis of the text, what it asserts. On p. 56 , where that term is intro-
duced for the first time, we read: "The stand taken by those in favour
of such a reduction Blight be called reisa." What reduction is meant
here? The reduction of the Aristotelian categories to the single ~atego
ry of things. "This is to mean that all sentences in which something is
seemingly said about some object which is not a thing, are treated as
substitutive formulations standing for other sentences, understood lit-
erally and predicating exclusively about things. In other words, any
'term' which is not a name of a thing is held to be an onoaatoid." (p.
51). This is a moderate formulation of the doctrine of reism, for in the
text we find grounds for a B1uch stricter formulation. That stricter for-
mulation is, as it were, a rule which authorizes us to make such state-
ments as "no object is an event", or "events do not exist"; "no object
is a relation", or "relations do not exist"; "properties do not exist" ..
"sets do not exist", "thoughts do not exist", "content of images does
not exist", because "only things exist". Thus, as it seems, reism has
two variations, one of which is milder and pertains to language (it may
be called semantic reism), while the other is stricter and pertains to
reality; the latter variation, which denies existence to relations, pro-
perties, etc., may be called real reism.
Conscious of the risk we incur, we shall try to explain the thesis
of semantic reism as we understand it.
In this interpretation we begin with an analysis of the author's
programlRatic declaration on p. 51: " any 'term' which is not a name of a
thing held to be an onomatoid". Thus we have to begin by opposing the
concepts: a 'proper term' and an 'onoaatoid' (the author does not use
the tera 'proper term', and speaks siaply of 'teras', which he places in
opposition to 'onollatoids', and such a terminology would suggest that
'onollatoids' are a kind of term).
On p.7 the author defines the proper term as a'phrase which aay be
used as the subjective coaplement in a sentence of the type 'A is B',
with the primary interpretation of the copula 'is'.
But how does the author explain when 'is' has been used in its
primary meaning? He gives examples of sentences in which it is used in
the ·primary meaning; all these sentences have this in common that they
are equisignificant with certain other lengthier sentences. Thus, for
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS ••. 11
instance, the sentence 'Seniority is a transitive relation' is equlslg-

nificant with the sentence: 'If an object is older than another object,
and that other object is older than a third object, then the first
object is older that the third'. In the sentence 'Seniority is a transi-
tive relation' the word 'is' does not stand between proper terms. "The
word 'is'," says the author, "does not play here its essential role, but
a secondary, substitutive one. For whenever we ask 'what is this?' with
reference to a [proper] term, for instance, with reference to the [pro-
per] term '1f, we have to answer by the phrase: '1f is the same as 'such
and such a thing (or person)'. And if we want to answer the question
'what is seniority?', this procedure fails." (p. 9) This quotation, and
in p.articular the word 'for' used after the statement that in the sen-
tence under consideration the word 'is' does not occur in its essential
role, permits us to guess such a definition of the term we are concerned
with: the word 'is' in its essential meaning is such a word as connects
two expressions, '1f and 'M', each of these having the same meaning as
the meaning of certain expressions of the type 'such and such a thing'.
For brevity's sake, let us call such expressions expressions of the
meaning of things.
All that system of definitions looks like a circle, though not a
vicious one. A proper term is defined as that which may be one of the
arguments of 'is' in its essential meaning, and that • is' as such an
'is' as may have certain specified arguments (namely expressions of the
meaning of things).
Anyhow, it follows from the system of these definitions that only a
term denoting a thing may be a proper term, and hence no term which does
not denote a thing is a proper term. By defining an apparent term as an
expression which may occur as the subjective complement in the sentences
of the type 'A is B' and is not a proper term, we obtain the reistic
declaration from p. 51: "any 'term' which is not a name of a thing is
held to be an onomatoid" as a definitional tautology. Reism so under-
stood is a thesis which follows from arbitrary definitions.
But probably reism as a semantic doctrine is not to be interpreted
in such a tautologous manner. Two other interpretations suggest them-
selves. In one of them, a reist would claim that an average English-
speaking man uses the word 'is' (and analogous forms) in a different
meaning when he connects with it terms denoting things, and in a differ-
ent one when he connects with it, so as to form meaningful sentences,
expressions which do not denote things (cf. p. 8, line 19 from. top). If
this were an empirical statement concerning the interpretation of the
word 'is' in different contexts, then we would say that both this claim
and its negation would require laborious researches and statistical
investigations, so that the most prudent thing would be to abstain from
advancing one's opinion on this issue.
The other interpretation seems to bring out better essence of
semantic reism. Its formulation is: terms denoting things form a closed
semantic category. This means approximately that if a term denoting a
thing and occuring in a meaningful sentence is replaced be an expression
which does not denote a thing (while the meanings of the other expres-
sions in that sentence remain the same), nonsense is obtained, whereas
if a term denoting a thing is replaced by another term denoting a thing,
12 KAZI~IERZ AJDUKIE~IrZ
the sentenre ~hich is thus ohtainerl may he true or false but is never-
theless meaningful.
Now jf we say that a class of expressions is a closerl semantic cat-
egory (cf. p. 56) we have to add in whirh language it is so. The author
sp<~ms to claim that. it must be so in every language. The antinomies of
'self-mpmbership', sur;h as Russell's antinom~-, the antinomy of rela-
t i nilS, the ant i nomy of prop<"rt i es, etc., haH' prov ioed arguments for the
assert ion that if ~e inclurle in the same semantir category expressions
which perform different funct.ions in the sentence, for inst:mc-" some-
t.hing which is a funr;tor and something which may be its argument, we run
the risk of contradictioll. The rule prohihiting, in any language,
inclusioll of such types of expressions in the same ~emantic r:ategory, is
t.hus to be observed under the penalty of contradiction. Now it does not
s(>em that tlwre are "'1ually strong arguments in favour of the prohibi-
t iOll to includic' in the same semllntic category, for inst:mce, the wor-d
'tllol,,' alld such words as 'death', 'paill', etc. In other words, I do not
think that ";e 1'110 the risk of an antinomy if we consider meaningful the
spntpllces ·'something is a pain' aorl 'sompthillg is Il tllble', if thp mean-
Irlgs of the other words occuring in such sentencPH are rptained.
The fact Ulat sPIlt.encec; including 'appal'ent terms' can he replacpd
hy e(lIlisigni ficant ones ~hich do not include Ilppllrent terms proyes only
that, Ilt IF-ast theorf'fically, ~"ca.n do without apparent terms. This
suggpsts ()ne more illtt"rpret.!1tiO)1 of semantic· reism, an interpretlltion
which is not a. thesis hilt a progrilmme. This programme might postulate
t.hllt ilpparent terms should not be lIsed unless the sentences in which
they oreur CRn, by mean~ of appropriate definitions, he reformulated as
e~lisignifica.nt sentences free from apparent terms. This would result in
the arhitrary Ildoption of the principle that in the language we use, the
h'rms which dpnote things form a. closed spmantie category, which means
that only tprms which denote things may meilningfully be substituted for
terms ..-hieh denotp things. I t must be admi tted that the advantage of
such a 1anguage cons i sts in the fad that it el i mi nates a number of
philosophical issues as fictitious.
Tn such a language the principal thesis of real reism, which we now
procepd to analyse, stating that eyery object is a thing (p. 56) turns
out to bf' a truism. 'All A is B' means 'for any x, if x is A, then x is
B, and some x is A' (p. 194).' Hence the thesis of real reism takes 0)1
the form: • for any x, if x is an ohject, then x is a thing, and some .Y
is an object'. It is a truism that some x is an object. And since in a
reist's language \ can take on as its values only names of things (the
variable x is limited to that semantic category hy the context in which
it occurs), hence any value of the variahle x which veri fies the antece-
dent also verifies the consequent of the implir;ation formulated above,
so that that implication must always be true. But real reism does not
confine itself to the positive assertion that every object is a thing,
hut pmphasizes a number of negative theses, such as 'relations do not
exist', 'properties do not exist', etc. Now these theses, if formulated
in -a reist's literal'language, entail an inr;onsistency on his part,
since they are meaningless in that language, and if formulated in a me-
taphorical sense, they are false.
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS ••• 13
In a reist's literal language it may meaningfully (though falsely)

be said that 'the earth does not exist', but the term "earth', as denot-
ing a thing, may not meaningfully be replaced by the term 'relation',
which does not denote a thing. Hence the statement 'relation does not
exist' is meaningless. On the other hand, in the metaphorical language
the formulation 'relation does not exist' means the same as 'for any R
it is not true that, for some x and y, x R y'. But this is a notorious
falsehood. Likewise the statement 'property does not exist' is meaning-
less in a reist's literal language, and in the metaphorical language,
being equisignificant with the statement 'no thing is of-some-sort', is
an obvious falsehood.
If it is so, then that which is tenable in real reism reduces, in a
reist's literal language, to the truistic tll.utology "every object is a
thing"; the annihilation of relations, properties, etc., cannot be for-
mulated as any meaningfuland true thesis. And yet we feel that reism
strives for such an annihilation. This must be formulated as the state-
ment of the rule: no sentence of the form 'relation exist' is true (in a
reist's literal language). This reistic theorem may appear too modest,
since it again ceases to be a thesis about what happens in the world,
and becomes a statement about a reist's language. As far as I comprehend
his intentions, I feel that he would not be satisfied if what he thinks
were rendered only by the statement, formulated above, about his own
language. Areist has some opinion of reality, which he cannot meaning-
fully state. If we hazard below a description of what a reist thinks, we
do so without a justified conviction that such a description complies
with the author's intensions. Thus, we think that the reist holds that,
should some person acquire an immediate contact with the world, some-
thing like Kant's intellektuelle Anschauung, he would not find anything
other than things. Should he really mean something like that, we could
not be at one with him, not because we think that in an immediate con-
tact with things he would find something more than things, but because
the very idea of such a contact and of such cognition taking place in
abstraction from any language and from any forms of thought is illusory.
In order not to disregard any such interpretation of real reism as
seems possible, let us add that on reist's part it would be correct to
say 'relations do not exist' if he heard, for instance, the definition
'a relation is such and such thing'and used the term 'relation' in
accordance with that definition. Such a sentence would not be meaning-
less, because in view of such a definition of a relation the term 'rela-
tion' would denote a thing and hence might occur in the sentence 'rela-
tions do not exist', in the literal interpretation of the words involv-
ed. Such a sentence would obviously be false, since by describing a
relation as such and such a thing it would probably ensure that no thing
would satisfy that description. But it seems dubious whether· anybody
would define a relation as such and such a thing.
In concluding this analysis of Kotarbinski's reism we want to state
that we consider as sound many elements inherent .in semantic reism,
understood as a programme concerned with the ways of using language. We
declare that we would be willing to speak so as not to consider all
nominal phrases as words belonging to the same semantic category, so
that, for instance, 'whiteness' and 'table' would not be included in the
14 ~ZIMIERZ AJDUKIEWICZ
salle semantic category; we would rather think that the words 'whiteness'
and 'looks white' are words of the same semantic category. But we do not
agree on the place in which the author draws the boundary, among the
nominal phrases, between the various semantic categories. We also do not
agree with the postulate of reducing all the sentences that include
nominal phrases to sentences containing nominal phrases of one and the
same semantic category. Neither do we clai .. that such a division of
expressions into semantic categories is found in the (Polish) language,
nor that any other division is faulty.
4. SEMANTICS
In a critical review of the author's opInlons on semantic issues we

shall concentrate on his definition of meaning. It is based on the defi-
nition of stating, which says: "a language phrase states directly a
given experience of a person ( ••• ) if and only if it is spoken or writ-
ten with the intention that a definite person, or any person whatever
who perceives that phrase, and understNmds the language in question,
should know froll that phrase that speaker experienced precisely ·that."
(p. 3-4) And "a given phrase states a certain experience indirectly
whenever, without stating it directly, such can, in view of linguistic
usage or adopted convention, be used in a given language as a direct
statement of such an experience." These definitions are followed by an
explanation of the term 'states as to content'. On p. 4 we read: "The
sentence 8 is a (direct) statement of John's thought as to content-
means: John thought that p ('p' stands for a whole sentence) and uttered
8 so that the listener might recognize that John thought that p". This
definition is not homogeneous, that is, on its right side it has .the
real variable p, which does not occur on its left side. Hence it is a
defini tion of the phrase: '8 is a statement of John's thought as to
content', and not a definition of the phrase '8 is a statement of John's
thought that q as to content'. Yet on p. 7 we read the following defini-
tion of meaning: "The sentence A means that q" is the same as: "The sen-
tence A states as to content, directly or indirectly, a person's thought
that q". This definition uses the ph.rase 'states as to c.ontent the
thought that' as a functor of two arguments, A and q, whereas what has
been defined previously is only the phrase 'states as to content' as a
functor of one argument, A. I do not doubt that in conforaity with the
author's intentions the definition of stating ought to be corrected so:
the sentence 8 states directly as to content John's thought that q, is
the sue as: John thought that q and spoke 8 so that the hearer should
know that John had thought that q. No definition of indirect stating as
to content is explicitly given by the author, and it should run: 8
states indirectly the thought q, is the saae as: 8 can be used, in view
of the usage of a given language, so that he who thinks that q should
speak 8 so that the hearer might thereby know that the speaker thinks
that q.
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS ••. 15
Now a number of objections might be raised against such a system of

definitions. First, in all those definitions .such technical terms as
'expresses', 'means', etc., are not related to language. This was, per-
haps, deliberately neglected by the author who wanted to simplify his
deUnitions for teaching purposes. Secondly, the definitions strike me
by their verbalism, though this is only a jarring impression, and I am
unable to supstantiate my point. Thirdly, a more important and quite
obvious objection. By the definitions of statement as to content, we may
say that, for instance, the sentence 'I think that 2x 2 = 4' states
(indirectly) as to content the thought that 2 x 2 = 4. -In fact, the
phrase 'I think that 2 x 2 = 4' can be used, in view of the linguistic
usage, so that John who thinks that 2 x 2 = 4 should utter 'I think that
2 x 2 = 4' so that the hearer might know thereby that John had thought
that 2 x 2 = 4.
Hence the phrase '2 x 2 = 4' states indirectly as to content the
thought that 2 x 2 = 4. If this is so, then in view of the definition of
the phrase 'means that' it follows that the sentence 'I think that 2 x 2
= 4' means that 2 x 2 = 4.
This objection can be avoided if an appropriate interpretation is
given to the phrase 'know thereby'. I suggest, for instance, the addi-
tional qaulification that the hearer knows by the statement uttered by
John what John thought it that hearer forms his opinion on what John
thought, regardless of his belief in any of John's statements.
I now turn to the fourth objection. Suppose that a certain X made a
lIistake in the choice of words (or, to put it colloquially, "his tongue
sliiped"), and, for instance, he said ',the horse is a mumal' although
he had in his' mind that the horse is a vertebrate, and he made his
statement so that the hearer who understands English should know thereby
that X had thought that the horse is a vertebrate. Now X, by the defini-
tion of direct statement as to content, by the sentence 'the horse is a
mamllal' direct'ty stated in English the thought that the horse is a
vertebrate. Since, by the definition of meaning, when the sentence A
states directly as to content a person's thought that q, then that sen-
tence A means that q, then should the case under discussion (which is
not at all fictitious) occur, we would have to conclude that the sen-
tence 'the horse is a l18IImal' means that the horse is a vertebrate. This
becomes embarrasing when, after relating all the semantic functions of
language, the conclusion has to be formulated as: 'the horse is a
mammal' means in English that the horse is a vertebrate.
This objection does not affect the-definition of meaning given on
p. 93, in which no reference is made to direct stating, and which runs:
"A given phrase means that: it is so and so - is the same as: a given
phrase can adequately be used to communicate to someone the idea that it
is so and so." By interpreting the se'nse of that 'communication of
thought' and taking into account both the definition of meaning from
p. 7 and the necessity of relating the technical terms 'means', 'states
indirectly', etc., to language, we shall make the following transforma-
tions of the definition of meaning in the conviction that we merely
explain the author's idea. We shall first be concerned with the'follow-
ing formulation:- "The sentence A means in language L that, q, is the same
as: in view of the usage in language L the sentence A can be used to
state directly as to content, in language L, a person's thought that q."

When we expand this definition by explaining the sense of the
phrase 'states directly as to content' we obtain: the sentence A means,
in langauge L, that q, is the same as: in view of the usage in language
L the sentence A can be used so that a person X who thinks that q should
utter the sentence A so that the hearer who understands language L
should know by that statement that the person X thinks that q. To say
that a sentence can be used for a certain purpose probably amounts to
saying that he who uses it will (at least in most cases) attain his
goal. Hence the definition of meaning can be simplified to the following
one: A means in language L that q, is the same as: the hearer who
understands language L having heard from the person X, who thinks that
q, the sentence A will (at least in most cases) know from that statement
that X thought that q.
We now pass to the last and the most important objection against
this definition. The definition of meaning is based on the concept of
statement as to content. Now a statement as to content is a certain
special form of statement. For a certain linguistic phrase may state a
certain thought in the various respects; for instance, the phrase 'cer-
tainly!' states my thought that war is not a factor of progress, in
respect of modality. It is because the phrase 'certainly!' can be used
so that a person who understands English should know thereby that I have
just experienced, in respect to modality, my conviction that war is not
a factor of progress. The phrase 'certainly!' states indirectly (in
respect to modali ty) any apodictic conviction. This follows from the
author's definition in which it is not said that a phrase states indi-
rectly this experience, but that it states such and such experience.
True, in the defintion of direct statement it is said that a given
phrase states a given experience, but on the condition that the inten-
tion is to indicate to the hearer that the speaker experienced not the
given, but such experience. Such experience is not this experience, but
is such experience in this and that respect.
Now to make a distinction between statement as to content and
statement in any other respect, the author uses the intensional function
of the sentential variable q, namely the function 'X thought that q'.
Sentences of this kind, that is, singular psychological statements,
are given by Professor Kotarbinski an original interpretation (p. 346).
Examples are given of such sentences as: 'A illagines B', 'A supposes
that B', A wishes that B'. The author's explanations concerning those
examples enable us to guess how he would interpret the sentence
'X thinks that q'. Any singular psychological statement can be squeezed
into the scheme which is always the saae and the consecutive parts of
which are: "a term (in the case of true sentences always a name of a
person); the word 'experiences' (or its equivalent); a constant element
which functions as a connecting word; and finally a nominal phrase, or a
sentential phrase or sOlie other phrase (for example, exclamatory). The
full scheme would, therefore, be as follows: 'A experiences this: B'. As
the constant element with the function of a connective we select
'this'." (p. 346) Let us try to explain the author's idea to the reader.
The point seems to be that a psychological statement, for example, 'John
experiences pain' be deprived of the appearance of a sentence consisting
REVIEW-ARTICLE: T.KDTARBINSKI'S ELEMENTS 17
of the two terms connected by the copula 'experiences'. For that purpose
such a sentence is interpreted so: 'John experiences so and so', where
that 'so and so' plays the role of an adverbial qualifier which deter-
mines more closely the kind of the experience involved. Now the author
so imagines that adverbial determination that he recommends the follow-
ing manner of speking: 'John experiences this: Dh! It hurts!' In this
way the sentence in question breaks into two parts: the announcement,
before the colon, and the information, after the colon.
By means of the connective 'this', which does not occur as a term,
but, as the reviewer thinks, is to be interpreted as an adverb, the
author continues to explain, "it is implied that the speaker will him-
self in some way experience something imitatively, in the same way as
John experienced something, and that the phrase which he utters will be
a formulation of that experience". (p. 346)
Now the following interpretation of the role of the connective
'this' suggests itself to us: that connective is an occasional adverb.
And the occasion determining the use of that adverb would be the imitat-
ive experience stated in the phrase that follows the colon. In a state-
ment like 'John experiences this: the sun shines!' only that which pre-
cedes the colon would be a complete sentence, while that which follows
the colon would play the sallie role as pointing one's finger in accompa-
niment of sentences in which the occasional demonstrative pronoun 'such'
is a (general) term standing for a physical object.
The second interpretation of the role of the connective 'this'
seems less satisfactory to render the author's intentions; it refers to
the second part of the explanation quoted above that "by means of that
connecti ve it is implied that the speaker will himself in some way
experience imitatively, taking as his model how John experienced, and
that the phrase which he will speak will be a statement of that experi-
ence." Now i f Peter by saying 'John experiences this: bats are probably
birds' implies by the use of the connective 'this' that the phrase which
he will speak after the colon will be a statement of John's experience,
then Peter's statement might be interpreted as: 'John experiences some-
thing which is (indirectly) stated by the phrase "bats are probably
birds'" (and not 'experiences the same as is now directly stated by that
phrase') •
Let us now consider both interpretations. Suppose that a person
says 'John has a suit like that' and points to a certain suit. Now in
the situation described above that sentence may be interpreted in
various ways. It may be that John's suit resembles the suit pointed to
in its cut, and perhaps also in its colour, and perhaps also in some
other respect. Likewise, when I say 'John experienced this: bats are
probably birds' and understand this statement in accordance with the
first interpretation as 'John experienced so', with that occasional 'so'
determined by the situation in which I say 'bats are probably birds',
then it is not known in what respect John experienced just so. Was it in
respect of modality, or in respect of content, or in respect of the
naivety of that belief?
Also in the second interpretation, by which the statement 'John
experienced this: bats are probably birds' turns out to be the same as
'John experienced something which is (indirectly') stated by the phrase
'bats are probably birds' the same ambiguity remains, for we may ask,
'what does it state in what respect?'
Let us now concentrate attention on the relationships between the
definitions we are concerned with. Now "states as to content" is defined
by the phrase 'John thinks that q'. The phrase 'John thiks that q' is
explained by 'John experiences this: q'. Now we see that the last phrase
is ambiguous for both interpretations described above, and it cannot be
made univocal by the phrase "as to content", for a vicious circle would
.result. Hence, for instance, the above explanation of 'John thinks that
q' by 'John experiences this: q' cannot be corrected by the replacement
of the second part by 'John experiences as to content this: q' without a
resulting vicious circle.
Let us now confront the definition of (simple) (indirect) statement
wi th the definition of statement as to content. The forRler, in an ab-
breviated and modified form, would be: the sentence A states indirectly
John's thought to be such and such (for instance, the thought that the
sun shines) if the hearer who understands the language in question knows
from that sentence that John had in some respect such a thought as the
thought described (for instance, the thought that the sun shines). The
latter would be: the sentence A states as to content John's thought that
q (for instance the thought that the sun shines), if the hearer who
understands the Janguage in question knows from that sentence that John
thought that q (that the sun shines).
What has been said above seems to indicate, when the phrase 'John
thought that q' is intepreted according to the scheme 'John experienced
this :q', that (simple) statement and statement as to content become one
and the same.
For in both intepretations 'John experienced this: the sun shines'
is more or less the same as 'John experienced in some respect so (as I
shall experience in a moment when saying): the sun shines', or 'John had
a thought which in some respect is the same as the thought that the sun
shines' •
In this way, statement as to content cannot be used in a defini-
tion of meaning. For if, for instance, the phrase 'certainly!' indirect-
ly states (in respect of modality, and hence in some respect, and hence
states simply) every firm conviction of mine, hence the conviction that
war is not a factor of progress, and the conviction that 2 x 2 = 4,
etc., and hence it would also state them as to content. Consequently
'certainly!' would mean in Polish both that war is not a factor of
progress and that 2 x 2 = 4. And not only would it mean this on one
occasion and that on another ocasion, but it would always mean both.
The situation might be saved if the phrase 'John experienced this:
bats are probably birds' were replaced by 'John experienced in respect
of content something like that: bats are probably birds'. But this is
impossible in view of a vicious circle.
Still another interpretation of the phrase 'John experienced this:
the sun shines' might suggest itself to us. It might be said that the
phrase used here is 'this', and not 'such'; and when a person says 'John
experienced this' and adds the sentence 'the sun shines', he does not
want to say that John experienced an identical (in the logical sense of
the term) thought with that which is directly stated by the phrase 'the
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS 19
sun shines' used in the context described above, but he wants to say
something more than that John experienced a thought which is the same in
some respect. He wants just to say that John experienced a thought which
in every respect is the same as that stated in the phrase in question,
though not identical with that thought. This will hereafter be called
the third interpretation.
Without raising the difficult problems connected with the differen-
tiation between 'the same in every respect' and 'identical','and inter-
preting intuitively the difference which it is difficult to bring out,
we shall point to other difficulties resulting from that third interpre-
tation of the definition of meaning now under discussion.
In such an interpretation of the scheme of psychological statements
as given by Professor Kotarbinski it would follow, it seems, that it
would be possible to say truly 'A means in language L that q' if and
only if q would have the same form as the sentence A. In other words, if
the sentence A runs: 'John is a teacher', then only the following state-
ment about the meaning of that sentence will be true: the sentence 'John
is a teacher' means in English that John is a teacher. Any Qther state-
ment about the meaning of that sentence will be false. For instance, it
will be false to say: the sentence 'John is a teacher' means in English
that John is a man whose profession is to teach.
According to our analysis of Professor Kotarbinski definitions:
'John is a teacher' means in English that 'John is a man whose pro-
fession is to teach' is the same as: "a hearer who understands English,
on having heard from the person X, who thinks that John is a man whose
profession is to teach, the sentence 'John is a techer', will know from
that statement that X experienced this: John is a person whose pro-
fession is to teach." Now a hearer who understands English, on having
heard from the person X the statement 'John is a teacher', will know
thereby that X experienced this: John is a teacher. He will not at all
know that X experienced this: John is a man whose profession is to
teach. To experience this: John is a teacher, is something quite other
than to experience this: John is a man whose profession is to teach, if
it is assumed that 'to experience this: John is a teacher' is the same
as to experience what is the same in every respect as what has been said
above in the sentence quoted after the colon.
An experience which is in every respect the same as the experience
stated directly by the sentence 'John is a teacher' is not in every
respect the same as the experience stated by the sentence 'John is a man
whose profession is to teach'. The characteristics of the experience of
the former kind includes this, that they have as an element an image of
the word 'teacher', which fact is not a property of the latter experi-
ence. Vice versa, the characteristics of the second experience include
the fact that when we have that experience we in someway think of a
profession, whereas in the case of the first experience no thought of a
profession comes into our head.
Hence whoever hears X say 'John is a teacher' will thereby know
only that X experienced this: John is a teacher. The same applies to
other examples. When X says 'this is a case of cancer' no one will
impute to X that X experienced this: this is a case of carino.a, since
everyone realizes the difference between the experience: this is aces
of cancer, and the experience: this is a case of carcinoma. The first

experience is popular in character, while the other is much more scien-
tific.
Therefore, only such statements will be true: 'the sentence "it is
thundering" means in English that is thundering', '''the sun is shining"
means in English that the sun is shining', etc. To put it in a general
way, the sentence 'A means in language L that q' may be asserted truly
only if the sentence A has the same form as q.
Hence it follows that in a given language we may speak only of
meanings of sentences belonging to that language, so that we may may not
state in English any true sentence about the meaning of a German sen-
tence in the German language. For suppose we say 'the sentence "Sonne
scheint" means in German that the sun is shining'. If the result
obtained just above are correct, that is, if the statement 'the sentence
A means in language L that q' is true only when the sentence A has the
form q, then the above statement about the meaning of the sentence
'Sonne scheint' is false. The sentence 'Sonne scheint' means in German
that the sun is shining; in the light of our analyses of the definitions
suggested by Professor Kotarbinski, this is tantamount to: a hearer who
understands German on hearing X say 'Sonne scheint' will thereby know
that X experienced this: the sun is shining. But, in the third interpre-
tation now under discussion, to experience this: 'Sonne scheint' is dif-
ferent from experiencing this: the sun is shining. The former experience
includes images of the German words 'Sonne' and 'scheint', while in the
latter experience those images are replaced by images of English words.
If we are not mistaken in the analysis carried out above, we may say
that for any q (where q may take as its values sentences belonging to
the English language) it is not true that 'Sonne scheint' means in
German that q. To put it simply, it would result from our analyses of
the definitions suggested by Professor Kotarbinski (iri their third
interpretation) that any sentences of languages other than English do
not mean anything in those languages. This result, which is paradoxical
enough, suffices to reject the assumptions on which it is based, unless
our reasoning can be refuted.
Thus, the definition of meaning offered by Professor Kotarbinski,
when brought together with his opinion on the interpretation of singular
psychological statements leads to undesirable results for any interpre-
tation of the latter statements. It is to be admitted that in our demon-
stration we have been using certain paraphrases of definitions given by
Professor Kotarbinski, but we consider these paraphrases faithfully to
render his ideas. Some of these paraphrases, for instance the paraphrase
of the definition of meaning (consisting in the elimination of the
phrase "can be used for"), given in the formulation of the last objec-
tion, but before making use of the scheme of singular psychological
statements, are given only in order to obtain a handier and shorter
formulation.
So much for meaning and expressing.
A few words more on denoting. On p. 7 we read: "For to denote a
given object in a given language, is as it were to supply a term for
that object in that language, in other words: a term usable in that
language as a subjective complement in a true sentence of the type
REVIEW-ARTICLE: T.KOTARBINSKI'S ELEMENTS ••• 21
'A is B' referring to that object (with the primary understanding of the
copula I is' )." Now it is not known whether this is a formal definition
of denoting, or a manoeuvre for teaching purposes, intended to secure a
fairly good understanding. From the point of view of teaching purposes
it would probably be advisable to precede this quasi-definition with a
number of examples showing the use of the word 'denotes'. And if this is
to be treated as a formal definition, the following objections arise.
First, I fear that the definition "The name N denotes in language L the
object 0, is the same as: the name N can be used in language L as a
subjective complement in a true sentence of the type tA is B' concerning
an object d', may lead to the antinomy of heterosemantic words.
Secondly, what does the formulation "a sentence of the type tA is B'
concerning an object d' mean? If the author's intention is to define it
as equisignificant with the formulation "a sentence of the type tA is B'
with the subject denoting the object d', then these two definitions
would form a vicious circle. And I do not know whether that formulation
could be defined in any other way. In any case, the formulation in
question is not defined by Professor Kotarbinski at all. Hence the
objection of a vicious circle is not to the point, but we have to raise
the objection of obscurity instead. The formulation "a sentence of the
type tA is B' concerning an object d' is not clear at all, for it is not
known whether, for instance, the sentence 'Galileo is Copernicus' is a
sentence concerning Galileo only, or concerning Copernicus as well.
Still worse, it is not known whether the sentence 'Kosciuszko's heart is
buried in Cracow' is a sentence concerning Kosciuszko, or Kosciuszko's
heart, or perhaps the body of King Casimir the Great buried in Cracow,
etc. In view of its obscurity that formulation requires a definition,
which, however, in my opinion, can be given in such a way as to lead to
a vicious circle.
So much for Professor Kotarbinski's semantics, although many other
remarks suggest themselves to the present writer. [ ••• ]
PSYCHOLOGISM AND TI£ PRINCIPLE OF RELEVANCE
IN SEMANTICS
Alfred Gawronski
To begin with let us pose a question: what do I aean by psychologism in

semantics, in the context of this paper? It seeas that we can distin-
guish two degrees of this theoretical standpoint, a moderate version of
psychologism and a strong version. After an atteapt to define thea
briefly, I am going to argue against both of these versions. The strong
version of psychologism - in accordance with the above differentiation -
is presented by Tadeusz Kotarbinski, in the first pages of his Ele.enty,
as follows: " ••• words and, in general, phrases of language, express the
experiences of speakers, their impressions, ideas, thoughts", and 'ex-
pression' is understood, thus conceived, as the essential constitutive
function of language. "Stating", Kotarbinski says further, "is therefore
a case of expressing: if a given sign states a given experience, then it
expresses that experience, but the converse does not necessarily hold
true. The elements of intentionality and co••unicative function are
joined here to in producing an expression." However, intentionality and
communicative function are treated in seaantic psychologism as
additional elements, which have no bearing on the structure of language
itself, and are thus disregarded in the description of this structure.
Moderate psychologisll, on the other hand, does not emphasize the
expressive function of language, i.e. the function of expressing experi-
ences, but, like antipsychologism, considers the communicative function
to be of prime importance. What, then, is communicated by language in
particular human utterances, according to the representatives of aoder-
ate psychologisa in semantics? Language coaaunicates the speaker's
thoughts in order that the hearer might know what the speaker thinks.
Moderate psychologisa basically does not differentiate between speaking
and thinking. Speaking equals thinking aloud, and thinking is silent
speech. In this approach there are no fundamental structural differences
between speaking and thinking. Proponents of this view claia that lan-
guage serves to express the thoughts of a speaker in such a way as to
cause similar thoughts in another person.
23
J. Wolenski (ed.), Kotarbinski: Logic, Semantics and Ontology, 23-29.
Ii:) 1990 Kluwer Academic Publishers.
24 ALFRED GAWRONSKI
Before embarking upon a critical analysis of the theses embodied in

these two versions of psychologism in semantics, it would be advisable,
for clarity of exposition, to say a few words about antipsychologism. We
do not often encounter the psychologistic approach in the works of ana-
lytical philosophers writing in English. The influence of Frege and
Wittgenstein is still so strong that it has left little rooll for psycho-
logistic interpretations of the nature of language. It is well known
that Frege ardently opposed connecting the theory of language with indi-
vidual internal experiences. He went far, perhaps too far, in forwarding
the claim that thoughts are not the creation of a subject; thoughts
exist objectively before anybody conceives them, and the hUllan subject,
by means of the psychological process that we call thinking, extracts
these or other elements from the set of existing thoughts.
In no way do I intend to defend Frege's opinions on this matter,
but we must keep them in mind if we want to understand antipsychologis-
tic tendencies in English language analytical philosophy, which has been
influenced by Frege to a greater extent than has German philosophy.
Another great opponent of psychologism in the philosophy of lan-
guage was Wittgensteinin both periods of his philosophical activity.
His famous theory concerning the logical illpossibility of a private lan-
guage undermines the basis of psychologisDl. World literature on this
theory is vast, and the controversy surrounding it continues; therefore
it cannot serve as the basis of my cri'tical reflections concerning psy-
chologism in some areas of Polish analytical philosophy. I have adduced
Frege's and Wittgenstein's theories only because they are representative
of the antipsychologistic attitude present in the Anglo-American philo-
sophy of language.
Let us return to the main topic of our present reflections, that is
to the semantic theses presented by Kotarbinski in the first chapter of
Elementy. After a brief analysis of his claims, I shall attempt to
present - equally briefly - my own solution to the long-standing
question of the relation of thought to language, taking the course set
by Wittgenstein and Austin, without, however, repeating their arguments,
but rather going, I dare say, one step further in a direction which may
eventually prove unrewarding.
Let us begin with an analysis of Kotarbinski's theory, which I have
labeled • strongly psychologistic'. In the sixth sentence of Elementy,
Kotarbinski writes: "When a person says with understanding 'Vesuvius de-
stroyed Pompeii', then the word 'Vesuvius' expresses his experience con-
sisting in imagining Vesuvius, and the whole phrase expresses his mental
experi~nce consisting in thinking that Vesuvius destroyed Pompeii. Such
expression is direct." There also exists, according to the author of
Elementy, indirect expression by means of language, but this is genetic-
ally and logically secondary to direct expression and we shall, there-
fore, disregard it here.
How can we reply to these claims made by Kotarbinski? First, it can
be said that although the words 'Vesuvius' and • Pompei i, can be associ-
ated by most persons with an inner representation that they can repro-
duce while uttering this sentence, in the case of many other elementary
sentences such a representation is impossible or unnecessary. If I say
'Sivi is a city in Southern India', it does not mean that in uttering
PSYCHOLOGISM AND THE PRINCIPLE OF RELEVANCE IN SEMANTICS 25
the word 'Si vi' I experience something particular or that I imagine

something. The hearer receives information about the city of Sivi, but
he/she has no data on the basis of which he/she could imagine Sivi. Yet
the sentence is fully comprehensible to everybody and has full informa-
tional value. If I say 'Aristotle lived before Kant', what do I experi-
ence, represent or imagine? The emphasis on experiences and represen-
tations obscures the fact that this sentence informs us that Aristotle
lived before Kant and says nothing about anybody's experiences.
The distinction between referential and predicative elements in
this sentence is classical, so there is no need to recall it here.
Let us further examine the semantic claims of the author of Elemen-
ty. We can analyze the next fragment by means of a simple manipulation.
We shall first present the fragment as worded by the author with the
symbols 'z' and 'p'. Subsequently, we shall render the same fragment,
substituting for these symbols a concrete sentence uttered in what is
called 'a speech act situation'.
Kotarbinski writes: "For instance, John has spoken the sentence:
'The River Bug flows into the Narew'. This sentence is the statement of
his mental experience as to content, since John, having experienced the
thought that the Bug flows into the Narew, uttered that sentence in
order that the listener, on having heard it and understanding English,
should recognize that John had experienced the thought that the Bug
flows into the Narew. To put it more generally and more briefly: Sen-
tence Z, as a (direct) statement of John's thought as to content, means
that John thought p (where 'p' stands for the whole sentence) and ut-
tered Z so that the listener might recognize that John thought p."
Let us imagine a real statement to substitute for the symbols 'z'
and 'p' in the latter part of this text. Let us imagine that John -
whose word one accepts - enters the room and says 'The house is on
fire'. According to Kotarbinski, the analysis of this statement as to
content is as follows: "The sentence 'The house is on fire', as a
(direct) statement of John's thought as to content, means that John
thought that the house was on fire and uttered the sentence tThe house
is on fire', so that the hearer would recognize that John was thinking
that the house was on fire."
It seems to me that there is no need to prove how remote this
analysis is from ordinary intuition and our natural reception of John's
statement tThe house is on fire' if such a statement were made. We would
simply think that John was informing us that the house was burning,
saying nothing about his mental experiences, and uttering the sentence
in order that we might learn something about an external fact taking
place in time and space.
One more of Kotarbinski.' s expressions requires an explanation, on
which I will end the part devoted to the trend in semantics here refer-
red to as strong psychologism. The expression in question is 'stating as
to content'. The author, besides discussing t stating as to content'
(direct and indirect), says nothing about other ways or modalities of
stating. Faced with this fact, we can assume that the insertion of the
phrase 'as to content' instead of being intended to distinguish one kind
of stating from other kinds has another function. We might ask: what
would stating which is not I as to content' be? There is no answer to
26 ALFRED GAWRONSKI
this question in Ele.enty, and so we can make the assumption, that the
author used the expression 'as to content' in order to mark the
difference between stating and other kinds of expressing, if we accept
that stating is a special case of expressing.
II
Unlike strong psychologism in semantics, moderate psychologism does not

explain stating in terlls of expression and draws a sharp distinction
between the expressive and informative function of speech acts. But what
kind of information is meant here? Again, it is information about the
speaker's thoughts. According to the proponents of moderate as well as
strong psychologisBl, speaking differs froll thinking only in that the
former informs a hearer about the speaker's thoughts. There is only one
substantial difference between thinking and speaking: speaking is
directly related to another person, whereas thinking is not.
We should recall, however, that psychologistic theories in seman-
tics predate transformational linguistics, or text linguistics. Thus
philosophers of language wanting to interpret and to make explicit the
semantic content of linguistic expressions and to explain what informa-
tion an uttered sentence conveys, in addition to what appears expressis
verbis on the surface, had to rely exclusively on their own intuition,
without having at their disposal any established scientific theory to
support their interpretations. Today the situation has changed substan-
tially: the placing of a sentence in a wide context of its possible
transformations and possible intersentential relations has become
a powerful instrument of clarifying and interpreting linguistic ex-
pressions. A similar role is played by the theory of presupposition
(e. g. when two sentences are proven to be contradictory though they
seemingly refer to different things), Austin's theory of illocution,
'etc.
On these grounds, some scholars point out latent elements or modal-
ities of speech acts, indicating a possible interpretation, different
from the one chosen by psychologists. Out of this arose the conjecture
(see Wierzbicka, 1969) that every sentence Z, when uttered in the in-
dicative lIood, means roughly: "I want you to know that 'Z'''. If John
states 'The River Bug flows into the Narew' or 'The house is on fire',
that means "I want you to know that 'the Bug flows into the Narew''', or
"I want you to know that 'the house is on fire'''. Since a prerequisite
to sensible speech is the requirement that neither John nor anybody else
utters true sentences randomly only because he happens to think them,
but because he has some reason to convey this particular information and
not any other.
The theory in question is going, 1 think, in the right direction,
but it cannot satisfy us fully. Other people's ignorance of a fact, e.g.
that 'three flies have just settled on one of the walls of the room', or
that 'the average number of hairs on the head of an inhabitant of Cracow
is 15,723' is not a sufficient reason for uttering such sentences.
Statements of this kind are objectionable in most contexts, irrespective
of their truth values, because their relevance is not cOllprehensible.

Such sentences do not invite the objection 'This is not true!', but the
objection 'So what?'. If the speaker answers 'So nothing! These sen-
tences happen to be true and express what. l've just been thinking', we
are justified in concluding that there is something wrong with him/her.
Is this a problem that concerns only the pragmatics of natural lan-
guages or does it have a bearing on their internal structure? In IIY
opinion, it does have such a bearing which is not difficult to prove by
examining the extensions of the terms utilized in any speech and the
functional necessity of their vagueness. However, there is no rooll here
to elaborate a theory of the informative content of speech acts in terms
of its relevance.
We have just said that uttering sentences cannot be accidental and
that a hearer's ignorance of what a sentence conveys cannot be a suffi-
cient reason for uttering it. We have the right to, and often do, utter
sentences in the indicative mood which do not inform a hearer about any-
thing in the strict sense of the word, because the hearer already knows
what we are informing him/her about. If I tell someone 'Tomorrow is
Sunday', I am not assuming that he/she is ignorant of this. I am drawing
his/her attention to a fact (or to a thought) to which - in my opinion -
he/she has not paid sufficient attention or which he/she has not taken
into account; and if I draw someone' s attention to something I assulle
not necessarily ignorance on his/her part but relevance of what I am
telling him/her, a relevance which he/she can acknowledge. If the
relevance of an uttered statement cannot be grasped by the hearer, some
clues have to be given in the continuation of the statement, or else the
whole message will border on nonsense.
We have come to the central point of our present reflections: a
statement, or a speech act is primarily an act of drawing the hearer's
attention to something that we shall for our purposes call 'a fact'. In
making a statement the speaker chooses one or several facts or thoughts
from a large set of those known to hill/her, and by drawing attention to
one or some of them he/she interferes in the hearer's awareness. In a
possible world in which everybody knew everything or knew exactly the
same, language would be as useful as it is in the actual world, because
it would still be necessary to point out certain things or facts. There
is no reason to think that a different structure ot language would be
more functional for this purpose.
Thus we have reached the conclusion that every uttered sentence is
primarily an act of drawing the hearer's attention to something, an act
which must have a justification going beyond the-affirmation of truth.
Thinking does not share this constraint, and - perhaps for this reason -
it is far 1I0re creative than speaking. In some respect, thinking can,
but does not have to, concern matters which are relevant. The essence of
thinking consists in the fact that a great number of thoughts and asso-
ciations appear in which we find nothing noteworthy, and yet they cOile
to our mind. We can let ourselves be occupied with thell for a while.
Stating, on the other hand, takes for granted the relevance of 'what is
stated. I do not mean the situational relevance, if situation is under-
stood as something transient. Neither do I mean the fact that in almost
every situation there are subjects "out of place" or having no connec-
28 ALFRED GAWRONSKI
tion wi th what is happening at that moment. I mean, in a relatively

neutral context, that in drawing someone's attention to something I
affirm the relevance of what I am stating, and I know how to answer the
question 'So what?', if only on the basis of the general situation of
man in the world. In the particular case of John's statement 'The house
is on fire' we can say that he is drawing the hearer's attention to
a relevant fact whose relevance is comprehensible, a fact which has no
logical connection with the proceedings in the room (but has a relevant
empirical connection). Such a statement does not draw our attention to
John's mental experiences. That is why it seems to me that the interpre-
tation of his statellent in terms of experiences, as Kotarbinski pro-
poses, is not correct.
III
The problems discussed above have so far been pushed aside by most phil-
osophers of language as belonging to the field of linguistic pragmatics.
However, if we conclude that the primary function of uttering sentences
and texts consists in selecting from a vast number of perceived or known
or possible facts, only (and necessarily only) those which enter through
mutual relations into a system of relevance(s), and that the second
function of language consists in drawing attention to those, and only
those facts at appropriate moments, with a presupposition substantiating
their relevance, then it is unlikely that these functions fail to have
a bearing on the structure of language.
Indeed, this is the case. If we observe the presently evolving
scientific discipline of systems of classification of objects in various
ethnic languages, we find that in the Sixties and Seventies a different
approach to verbal classification emerged. Until recently, including the
period of structuralism, the thesis proclaiming arbitrariness of the
linguistic sign was predominant; however, the new discipline, called
taxonomy (Robert Sokal, Eleanor Rosch and others), forced scholars to
reconsider the simplistic formulation of the problem. If we examine, for
example, the structures of the system of lexical terms in natural
languages, one can easily see that the fundamental factor delimiting
extensions and the degree of vagueness of sillple lexical teras is the
relevance of the type of object and of its delimitation in the
border-line cases in the total system of our knowledge about the world.
First of all, we do not naae all the objects that we perceive (and so,
in some sense, that we think about), but we extract from anonymity only
sOlie of them, to a different degree, by means of various types of naaes
or terms, ranging from singular to more and more general. Singularity of
a term and extensions of general terms are by no means accidental and do
not originate arbitrarily. As is well known, there is a close connec-
tion between the extension of a term and its intension or connotation.
If in aaking a statement I draw someone's attention to fact F, and I am
forced to eaploy a term whose extension is too narrow and intension too
rich for the purpose of my couunication (for there is no other, more
appropriate, terll in the language), then my message is fraught with
communicative disturbances, because I cannot narrow down someone's

attention to those aspects and only those of the fact F to which I
wanted to limit myself. Thus, I all not able to isolate the relevant
aspects in my message. It is not a matter of economy of speech. It is
rather a matter of distortion, of a deformation of the fundallental
principle stating that the condition of acceptability of every speech
act, of every statement, is the motivated choice of factual elements and
the justified interference in someone's awareness.
I have attempted to outline briefly the reasons which have led me
to reject any kind of psychologism in semantics, which treats speech
acts as an extension of thinking, and conversely, thinking as silent
speech. I have also attempted to present a rough outline of an alter-
native theory to psychologistic treatment of language, a theory based on
the principle of relevance.
The term 'relevance' does not exist in Polish, which uses the term
'significance' ('znaczenie'). This Polish term is, like its equivalent
in many other languages, aIIbiguous. 'To be significant' means either 'to
mean this or that', or 'to be important, relevant'. But even this
perplexing ambiguity may provide a useful insight. After all, the
relation between the two meanings of this term is probably not
fortuitous.
NAMES IN KOTARBINSKI'S "ELEMENTY"
Peter Geach
In the present essay I shall be largely concerned with the recognition

and elimination of sham names, 'nazwy pozorne', as reco-.ended and prac-
tised by Kotarbinski. I shall be much less concerned with his views on
genuine names; hardly at all with the view that only bodies are genuine-
ly namable. Let it pass, transeat, as aen would say in a aediaeval dis-
putation. It is likewise not beyond the reach of philosophical specula-
tion that somebody should restrict the range of the namable to what are
not bodies, e.g. to sense-data; but I shall not stop to consider that
position either. We have to start somewhere, and the thesis that bodies
are namable is well enough accepted and understood to start with.
I must in honesty put on record a disagreement with Kotarbinski as
to which expressions can serve to name even bodies. I do agree with hia
about the question of general names; I shall return to this in a moment.
But I am at least strongly inclined to reject from the category of names
all expressions of complex syntactical structure.
Following the lead of Plato's Sophist, Aristotle's De interpreta-
tione lays it down that the simplest sort of system consists of a name,
onoma, and a predicable expression, rheas; these are plainly aeant to
constitute two mutually exclusive classes, and in the simplest sort of
sentence each will be syntactically simple, having no parts that signify
separately. If Aristotle then had a program of turning non-complex sen-
tences into combinations of such simple two-element sentences, his ex-
tant works show no sign of his t.rying to carry out such a program, and
he must soon have convinced himself that it was unfeasible. But the
thought that a genuine name must be syntactically simple was to be re-
vived by Bertrand Russell and by Wittgenstein in the Tractatu8: a name
may per accidens be in some way complex, but as regards any given form
of complexity, one may show, from case to case, that it is inessential
and therefore irrelevant to the naaing role. This thought I am still
inclined to defend.
One must sharply distinguished the thesis that all names are simple
from the thesis that all names are singular. Maintaning the first
thesis, I have often been charged with maintaning the second; the
confusion is quite inexcusable, since I have on the contrary often
maintained that terms like 'lion', 'chair', 'book', are names, at least
in certain contexts of use. Here Kotarbinski and I are of one mind. The
31
J. Wolens1ci (ed.), Kotarbins1ci: Logic, Semantics and Ontology, 31-37.
32 PETER GEACH
naaing role of such common naaes was maintained continuously in

philosophical writing from Plato and Aristotle down to the 19th century;
even the young Frege who wrote Begriffsschrift counts names of chemical
substances as being naaes for the purposes of logic. (Quine indeed might
treat 'gold', say, as a proper name of a scattered individual consisting
of all the gold fragments throughout space and time; but it would be a
gross anachronism to ascribe such a view to Frege in 1878; no doubt
Frege, like his contemporaries, regarded 'gold' as a shared name of the
lIany gold fragments.)
By the time he was writing his Grundlagen, Frege had rejected the
idea of shared naaes; he now takes what others would read that way as
being expressions for Begriffe. I cannot account for this change of
mind; Frege more than once gives an argument for it that is clearly bad,
namely that if e.g. I make some predication about all men, I need not
have in mind some tribesman in Central Africa. As an argument the 'man'
does not, as a shared name, cover that tribesman among other men, this
is just as bad as would be an argument that 'Christopher Columbus' can-
not name the discoverer of America because someone might have used it
with no thought of discovery or of America. However this may be, White-
head and Russell use the ides that all naaes are singular as one plank
in the ground floor of Principia Mathematica; and the resounding success
of that work blinded people both to its many defects in rigour and to
the doubtfulness of these theses about names. Even somebody who is as
little of a Russellian as Strawson takes for granted that a naae must be
singular; with a lack of historical knowledge that is really astounding
Strawson describes the contrary view as a •proposal , by - Peter Geach!
Strawson like others believes that serious formal work cannot be done,
or at least has not been done, without a treatment of all naaes as sin-
gular; both Frege' s contrary treatment in the symbolisll of Begriffs-
schrift, and the work of contemporary followers of Lesniewski, is clear-
ly quite unknown to hill and to many others, although much of it is
available to non-readers of German or Polish.
I agree with Kotarbinski, then, on one point which I think is of
capital importance. Differences would emerge if I entered into a dis-
cussion of the theory of the copula ('A est B') as expanded in Elementy;
but I have gone into this matter elsewhere, in published papers both in
English and in Polish, and I think I need not repeat myself here.
As I said before, Kotarbinski accepts, whereas I reject, complex
structures of naaes. I aa not such a sworn upholder of the thesis of the
syntactical simplicity of naaes that I would not consider arguments
against it. But when Kotarbinski gives alleged instances of complex
names, he seems to me to be guided by nothing better than graamatical
considerations which a logician ought to regard as highly suspect - par-
ticularly if he is a thinker like Kotarbinski, who is ready to disqual-
ify noun phrases wholesale for the title of naaes and call thea pseudo-
naaes, naBWY poBorne. At this point, to my regret, I find a similarity
between Kotarbinski and the advocates (in Oxford and elsewhere) of
analysis in teras of 'referring expressions'; in both cases, what
appears to be taken as the voice of an oracle is the voice of the
writer's school teacher giving instruction about the way to analyse a
sentence into phrases.
NAMES IN KOTARBINSKI'S "ELEMENTY" 33
In many cases the unity of a phrase that seems to compel one's re-
cognition is demonstrably not a logical unity; and I think an alterna-
tive analysis, which would not recognize the binding together of words
into a complex name, is always possible. This holds even for the
simplest case, of a qual j fying adjective attached to a noun. 'Only a
shameless woman smokes' is not of the form 'Only an A !/Is', coming out
trne just in case '<j) s' is truly predicable of nothing not covered by the
shared flame '.-1'; rather it means 'A woman - i.e. an,Y woman - smokes only
if she is shameless', and here the unity of 'shameless woman' has been
broken lip. T will not di late upon thi.s, because as T argued in various
papers j n my co llect.ion Logic Matters (1972), many examples show how
brHtl~ is the intuitive foundation whim 'complex terms' arE' taken to be
]ogkal unities. r hasten to acid that I do not hereby doubt the exist-
ence of logieaJ ly eomplex predicates. As I indicated, Aristotle may at
Oil€' time have dreamt of analysing our discourse in terms of complex
hondings of simple one - onoma one - rhema. propositions; but there is 110
slIch symmetry as he might suggest between names and predicahle expres-
sions. A name c['tn be simpl€', in a sense, it must be simple, because any
inner complexity is alien to its naming role. But a predicable may need
1:0 be complex; having in mind some named objPct. or objects, no limit can
be drawn to t.h€' complexity of what we want to say about it or them.
The cat€'gory difference that I recognize between names and predi-
cahles j'3 not denied by the logic of Lesniewski, which Kotarbinski
follows in Elementy; for a name wi II still be of different category from
a functor taking one name as argument t.o produce a sentence; and as
Plato says in the Sophist, a pair of names, or of such functors (Plato's
'rhemat:3.', my 'predicables'), will produce no logos, no coherent dis-
course. In Ajdukiewicz's scheme of categories, two expressions of cat-
€'gories sin and n yield one of category s, that is a sentence, but two
expressions both of cat.egory n or both of category sin will form a syn-
tactically incoherent string. Of course I am not saying that the Les-
n.iew!'lkian system overrides t.his distinction, only that the distinction
is in practice obscllred. In using the form 'A est B' it is hard to sup-
press a feel i ng as if one t€'rm had subject ro Ie, t.he other predicate
role; now either 'A est B' and 'B est A' are both well formed, or
neither is; since a predicable may certainly be complex, as I have said,
it is then hard to see why a name in subject role may not be. But a cri-
tiquE' of the Lesniewskian apparat.us would take me too far.
I turn to Kotarbinski's valiant battle against onomatoids, na.zwy
pozorne. It is a worthy battle, but sometimes appears like the Irish
hero's fight with his sword against the sea waves; for Kotarbinski is
fighting against a strong tendency of natural languages, at least of the
Indo-European family. How readily we turn from 'When a man marries his
troubles begin' to something like • A man's marriage makes the beginning
of his troubles'! (Even in 'his troubles begin' there is of course an
onomatoid, but one that we can get rid of immediately by saying 'he
begins to be troubled'). And in many years of teaching logic I found a
strong resistance of my pupils' linguistic habits to the use of letters
for any category other than names. There was some resistance even to the
use of a letter (schematic letter or bindable variable) for verbs or
predicables; there was great resistance to this move for the category of
34 PETER GEACH
sentences. After one of ay lectures delivered to an adult.audience, one

of my hearers, a man of some education and sophistication, professed
himself quite unable to understand the construction 'If A knows that p,
then p': the idea that a single letter can stand in for a sentence (as
opposed to its naaing a sentence), so that the construction 'If A knows
that p, then p' is syntactically correct, was entirely alien to him. But
with training and good will the difficulty can be overcome.
How far can the elimination of onomatoids be carried out? In spite
of the intuitively clear and formally sustainable difference between
sentences and names, the temptation to treat what grammarians call
sentence nominalizations as if they were names is often found irressist-
ible. As is well known, Frege was eventually led to confound the
categories of sentence and name. He had been quite clear on the matter
in Begriffsschrift: 'Priam's wooden house' has an Inhalt that is not
possible matter of judgment (beurtheilbar) , in contrast to 'Priaa had a
wooden house'. An argument for the two categories' being distinct that
convinces me, though I have found others less easily convinced, is the
one based on duality. A sentence is dual to its contradictory; and it is
possible to set up rules for replacing each siaple or complex expression
in a sentence by its dual so that the result is the dual of the whole
sentence. By this set of rules a name will be dual to itself, but a
sentence occuring as a clause (or for that matter a predicable of
category sIn) will be dual not to itself but to its negation.
Denying a naming role to sentences does not mean that we must view
with an unfriendly eye such developments as Suszko's 'non-Fregean'
theory of propositional identity or Wolniewicz's theory of situations.
The prose explanations may use onomatoids, but the foraal development
does not blur out the distinction of names and sentences; on this I have
found remarkably clarifying a paper by Arthur Prior, "Intentionality and
Intensionality", in the posthumous collection (1976: Papers in Logic and
Ethics). Siailarly there is nothing in principle here against a foraal
theory of entailment; so long as entail.ent is expressed not by a term
for a relation between namable entities, but by a connective that forms
a sentence out of. two sentences; cf. .y own papers on entail.ent in
Logic Matters. Other difficulties (of detail) may arise, but we have not
here to do with enterprises misconceived in principle.
A wide class of sentence nominalizations purport to identify
certain events or changes or processes, and these next come in for our
consideration.
Kotarbinski, unless I have quite misunderstood hia, wished to
regard designations of events as only apparent designations; there are
changing objects, but not objects that are themselves the changes. He
certainly rejects vigorously the idea of reducing physical objects to
processes; quite fairly he co.pare& this to the idea of a clinic where
there are diseases, treatments and deaths, but no patients.
My own views on this matter have been shifting and undecided. In a
paper I once read to the Towarzystwo Filozoficzne in Warsaw, under the
title "0 porzllPku czasowym" (On Temporal Order), I tried to formulate
sharply a view of which I was not fully convinced - because truth is
likely to emerge better from a clearly formulated wrong view than froa a
confusion. The view was that all apparent names or designations of
NAMES IN KOTARBH~SKI 's "ELEMENTY" 35
events and processes shall be eliminated in favour of narrative proposi-

tions apt (as we say) to report such events or processes; thus, apparent
references to Queen Anne's death would be replaced by such a clause as
'Queen Anne died'. For example:
'The news o( Queen's Anne's death causes Lord Bolingbroke to swear.'

would become
'Lord Bolingbroke swore because he heard that Queen Anne died.'
What are called causal and temporal relations between events would
on this view be represented not by relative terms (functors apt to form
a sentence from a pair of names or designations) but by connectives
(functors forming sentences from sentences). A sign that this is on the
right track is the possibility of iteration; as Wittgenstein said in the
Tractatus, the result of an operation can again be the basis of an
operation. This is pretty obvious for temporal connectives, 'at the same
time' and 'before' and 'after' and so on, when used to join narrative
propositions which can fill the argument places of such functors. We
thus get narrative propositions of complex structure, say 'while p,
(q and then (r until s))', structures often found in works of history.
It is less easy to argue that the underlying structure of causal
propositions is given by the use of a, connective 'because', joining
clauses, rather than a verb like 'causes' joining designations of events
or processes or the like. I t is not so clear, because iterations are
less obviously available; though I think they are admissible; cf. the
example:
'Because Lord Bolingbroke cursed and swore because he heard that
Queen Anne was dead, Swift despised Lord Bolingbroke.'
(I have no idea whether this is~rue, but it is clearly well-formed);
or again:
'Because Smith did not unload the gun, 8mi th was wounded because
Smith's little boy played with the gun.'
What does speak strongly in favour of the reduction to a structure

of clauses joined by 'because' is that often the idea of designatable
entities related by a causal relation is very artificial or plainly
absurd. In English we may say either of the following:
'The car crashed because the driver was not careful and attentive.'
'The crash of the car was caused by the driver's lack of due care and
attention. '
To hypostatize individual car crashes may come natural to us, but
hardly: to hypostatize individual lacks of care and attention! It is
precisely here that theories treating causality as a proper relation
often go manifestly wrong - when they have to deal with negative
elements in causation. To mention one well-known example, John Mackie's
accounts of cause factors are open to this criticism.
36 PETER GEACH
think the 'demythologizing' of language about events can be

carried far;. but I am not wholly satisfied with the program. For one
thing, events seem to be countable, as in ordinary statistical tables.
Since a man cannot be murdered twice, the number of persons murdered co-
incides with the number of murders; but a similar reduction, for obvious
reasons, is not possible for other criminal offences - neither the num-
ber of criminals nor the number of victims would supply the correct
statistical entries ~ (It may of course be mere ineptitude or ignorance
on my part that I do not see how to avoid in such cases the attachment
of numbers to kinds of events.) I allow myself to add a further reason.
What I have called, for good historical reasons, the Cambridge concep-
tion of change appears to me inadequate. By the Cambridge conception a
thing changes just in case something is true of it at one time and not
at another; Theaetetus changes by coming to be taller than Socrates, but
also Socrates changes by coming to be shorter than Theaetetus. However,
if what has happened in this case is that Socrates has neither grown
taller nor shrunk with old age, whereas Theaetetus who was a boy is now
a full-grown man, we are inclined to say that only Theaetetus has really
changed. This could be justified if we recognized as an individual in
the world not only Theaetetus but a transient entity, his growth. I do
not pretend that this consideration is overwhelming; the intuition about
real changes may turn out to be thoroughly confused; or somebody might
explain the matter otherwise, so as to justify the intuition, without
hypostatizing changes.
I must add, however, that sounds appear to me rather resistant to a
reistic reduction. The counting of sounds is as primitive a matter as
the counting of fingers; so much is this so, that many thinkers, such as
Kant, have strongly associated the assignment of number (to my mind
wrongly) with temporal succession. If sounds are to be treated as ident-
ifiable individuals, there is a gap in the reistic fortress-wall. And I
do not see any other way of dealing wit.h them.
In considering whether we have or have not countable individuals,
am not forgetting the lesson of Frege: that a given number attaches not
to individuals or aggregates of individuals, but to what he calls Be-
griffe. People are mystified by the term, but his meaning is plain: a
number is a number of a kind of things; even if things of that kind are
such as to form an aggregate, the number does not attach to the aggre-
gate, we must bring the kind before our mind's eyes. In the diagram that
the slave is got to draw in the Menon of Plato,
all six areas in aggregate form one square, but the number of squares is
six (one big square, four small squares and one square formed by the
diagonals). When people jeer at Frege for rejecting the simple solution
that numerical predicates attach to aggregates, they simply stick to
cases when it is easy to think of an aggregate and forget cases when t.he
attachment of numbers cannot be thus explained. But although numbers
NAMES IN KOTARBnisKI' S "ELEMENTY" 37
thus attach not directly to things but to Begriffe, Le., to kinds of

things, the assignment of numbers (other than zero!) of course presup-
poses that there are individual things falling under these Begriffe or
kinds. Well, no doubt I should not say 'of course'; let me say more
modestly that I at least do not see how to avoid recognizing individuals
of the kind I have mentioned.
Kotarbinski ought not to have had any difficulty in applying his
reism to words themselves - which necessarily he must often mention - at
least so long as these were written words. He would have indeed less
difficulty than some other nominalists: for he need not think that only
names of token words were logically admissible as names. This takes us
back, in fact, to Kotarbinski's acceptance of shared names, with which I
have expressed agreement. Different token words, which by some criterion
(there are in fact several possible criteria) are 'the same word', may
share a name, not just a common predicate, in the language of syntax.
But once again, difficulties would arise for reism over spoken words, as
over other sounds. It seems to me that the philosophy of language has
suffered severely from the almost total loss of an insight constantly
found in mediaeval logic: the sense that spoken language is primary,
written language only an imperfect conventional proxy for it. (What
could music be if there were only scores?)
I conclude with a few words about numbers themselves. It is very
hard to resist regarding numbers as identifiable individuals; I do not
find that Kotarbinski has provided, or even convincingly sketched, any
alternative treatment. Numbers, like criminals, can even be identified
by their pictures; a set of cards consisting of a blank card and the
ace, 2, 3, •.• , 10 of hearts might very well serve to identify numbers 0
to 10. Of course we need a convention of picturing to do this identifi-
cation, and it might fail; but then, some tribes of men, we are told,
cannot understand how a man may be identified from a photograph. I may
add that to judge by recent writings the identity of a human being seems
to have become far more problematic, philosophically speaking, than the
identity of a number, for which we have a clear criterion. Rhetoric
about 'mysterious entities' moves me not at all. (Is not a man a IIYS-
tery?) And there is no threat of an antinomy such as Russell's, if we
confine ourselves to number theory as opposed to set theory; nobody has
shown that we are going to get a contradiction from such sentences as
'The number of prime numbers between 3 and 13 is 3'
or that the term '3' is here ambiguous.
Wittgenstein has put on record the conversational exchange between
him and Frege before they last parted company.
Don't you ever find any difficulty in your theory that nUllbers are
objects?
Sometimes I seem to see a difficulty, but then again I don't see it.
take Frege to have meant that he seemed to glimpse a difficulty, but

could never bring it into clear focus. I am certainly not able to do
better. Of course I am not offering them as a knock-down argument
against reism; rather, simply as an expression of honest perplexity.
CONSISTENT REISM
Andrzej Grzegorczyk
The philosophical programme of Tadeusz Kotarbinski has been called some-

where (by himself) a small one. He wanted to renounce dealing with
traditional ontological or epistemological problems and confine himself
to preparing the linguistic tools appropriate to philosophy. This goal
was not so small but rather maximalistic and demanding for his pupils.
He wanted to have a homogeneous rational linguistic structure univer-
sally appropriate for raising all scientific problems. As Kotarbinski
disciples we should be able to translating every not vanishing philo-
sophical problem into reistic language. Hence first I shall recall his:
1. Linguistic requirements and some general ideas
Exallining this experience I find it still useful. The argument for a

linguistic training in a sophisticated discipline seems to be epistemo-
logical and may follow in this way:
(1) In every style of utterance there are some hidden ontological
assumptions, and it is economical to assume as little ·as possible about
the structure of the reality described.
(2) The style of our scientific description should conform to our
conviction concerning the structure of the reality.
(3) Our capacity in this style becames the checking proof for sOlie
ontologico-epistemological assumptions.
The vantage-ground of reism is the Aristotelian distinction between
things and properties of things. Things do exist in the basic sense of
the word 'to be'. Properties do not exist in this same sense. They are
attributed to existing things. Then Kotarbinski launches at us the
following requirellent. If we believe that there is no redness as a thing
but only some things are red, then we should renounce the use of the
noun 'redness'. If we believe that people are sad or cheerful and
besides them there is neither sadness nor cheerfulness, then we should
not use the nouns 'sadness' and •cheerfulness' in our psychological
descriptions. Everything what we want to say about human states of mind
39
J. Wolerisld (ed.), Kotarbirisld: Logic, Semantics and Ontology, 39-45.
40 ANDRZEJ GRZEGORCZYK
should be formulated in utterances containing the nouns of human beings

and the adjectives 'sad' and ·cheerful'. Of course, there are not even
such things as states of mind, so that a consistent reistic speech is
really difficult and becomes rather artificial. The idea is not to
translate effectively all our knowledge into the reistic language (that
would be a very dull task), but only to feel the capacity of doing this
if required.
Using a noun always makes a suggestion of the existence of a thing-
like object. Then reistic ascetism is a consistent AristotelianislD com-
bined with Ockham's reduction principle, and also a materialistic empi-
ricism (assumed at least as a starting point) saying that we shall try
to describe reality as a multitude of material things.
When we admit that the world consists of a multitude of things what
we say about the world is likely to be formulated in terms of things
only. To put it in other words: adopting the reistic ontological thesis
inclines us to formulate the totality of our knowledge in terms of a
reistic language. Kotarbinski was only more radical than other philo-
sophers.
One can say that sOlie elellents of reistic language and reistic
atti tude are now almost generally accepted in science. But there are
many degrees of reism in science. One may even say that modern science
admits of no other language, since a form of the theory of types or of
set theory becomes a rather natural generalization of the Indo-European
linguistic structure. However, the reistic ontological attitude differs
from the pragmatical and methodological attitude of modern science. In
constructing a theory one has simply to choose some elements as the
fundamental constituents of reality; or, to say the least, we need sOlie
fundamental variables ranging over one type; the simplest solution is to
ascribe to those variables the lowest type of the theory of types or of
a set theory with objects that are not sets (uhrelements). The latter
may be identified with material objects, the particles, but also with
spatiotemporal points, states of matter or states of mental life, and
the like. One may say that the structure of our language compels us to
begin with something.
The attitude a philosopher-reist assumes is quite different. The
reason why he wants to describe the world as a lIultitude of things is
that he believes that such a description is adequate and true of the
reality we deal with; thereby it gives us a better understanding of
reali ty. Philosophical reism bas become one of the last ramparts of a
certain ontological style of thinking. Namely, it seems that reism plays
the same role in ontology as Newtonian mechanics played in physics. The
reistic theory of reality is a natural idealization of everyday experi-
ence. Owing to tbe reference to everyday experience reiBm provides us
with tbe feeling of philosophical understanding of the world. Unfortuna-
nately the attitude of scientists becomes pragmatic.
After a scholar-philospher, who strives mainly for ontological
understanding of the world, there comes a scholar-pragmatist who notices
that the familiar classical theory, perfect in itself, does not organize
the totality of experience. The scholar-pragmatist then says: let us put
aside niceties of philosophy and let us construct an ontologically
dubious, un-intuitive theory which, however, would offer us a uniform
CONSISTENT REISM 41
theoretical description of the totality of facts. Modern physics pro-

vides the best example of theoretical description which disregards the
tastes of philosophers fed upon everyday experience. It compels us to
talk about fields, duality of light, indeterminism of particles, and
about other problems we are not accustomed to; we must not dwell on the
problem of how something exists, what exists really, and what is merely
a make-believe. Science has ceased to strive for understanding; it is
content with behavioural testing of its consequences. We have become
accustomed to the idea that our knowledge may be doomed to this. Alien-
ation is unavoidable. It may seem doubtful whether the total view of the
world will ever become satisfactory. The structure of reality goes
beyond what our language can express.
In spite of arising doubts it seems worthwhile to have a closer
look at the advantages, or perhaps illusions, the reistic descriptions
bring about.
2. Formal advantages of the curtailaent of the problem of existence
The first philosophical advantage we get from the reistic attitude is

the curtailment of the problem of existence to the existence of one cat-
egory of being only, namely to the category of things. In the reistic
description of the world what we say in terms of properties, relations,
and other categories, appears a shorthand for what could be said in
terms of things. In fact there exist only things which are variously
interconnected.
That position may be regarded as a modern reformulation of what has
been already said by Aristotle: that there actually exist only sub-
stances whereas accidents exist only inasmuch as they are in substances.
If the world can be described in terms of things only, then the problem
of modes of being of real counterparts of the concepts like: 'nation',
'justice', 'will', 'majority', 'weight', etc., seems no longer enig-
matic. There actually exist only men: just, speaking this or that
language, wanting something; there exist also other things like trees,
houses, of which some are bigger, some are heavier than others.
The advantage of this mode of thinking may be formulated as
follows: There are no entities which would prove indispensable to the
scientific description of the world and whose mode of being would
present us with serious ontological difficulties.
The next consequent step for a reist philosopher is to choose some
formal system of logic as an ontology. Several formal theories may
compete for this position. Kotarbinski has chosen a theory of Lesniew-
ski, which was even called 'ontology' by its founder. I would like to
suggest that it was not the best (the most consistent) choice.
Lesniewski's ontology is a first order theory with identity having
one specific primitive notion 1 : 'is', two elementary axioms:
is .... L .. ~ni .. w .. ki <1930, SIup .. cki (1966),

x is a /\ y is x ---+ x = y
x is a ---+ x is x
and one axiom-schema:
3 an (x is an - - x is x A ••••• )
which is a version of the axiom of comprehension (ausonderungs-axiom).
Models for ontology are of course complete atomic Boolean algebras
(see Grzegorczyk, 1955).
Kotarbinski was greatly influenced by Lesniewski. They both
considered ontology as a theory of the word 'is' in its fundamental
sense, like e.g. in the sentence: 'Socrates is man' or 'Socrates is
Greek', but not in the sentences: 'Greek is man' nor 'Horse is animal'.
(I omit the prepositions which are of course necessary in normal English
speech patterns.)
Lesniewski's ontology is an elegant theory, but I doubt if it would
be suitable for a very reistic philosopher.
'Socrates' is of course the name of a thing, but 'man' and 'Greek'
are not. Hence in ontology the names for things and the names for sets
of things fall in the same type. This is a good formalization of the use
of the word 'is' in everyday speech. But for a philosopher there should
be a big difference between an individual thing which is the fundamental
enti ty of the universe, and a set of things which is an abstraction.
This distinction should be especially relevant for a reistic
philosopher.
Hence I claim that the appropriate ontology for reistic philosophy
is simply first order functional calculus with identity. The unique type
of variable may be interpreted as the names of things. Thus the quant-
ifier binding a variable: 3 x ( •.. ) can be read as: there is such a
thing x that •••• Then we can prescribe the meaning of real existence to
the existential quantifier, which fits well with the general ideas of
reisll. In Lesniewski's ontology we can not read the existential as:
there is such thing x that ••• , because x can be not a thing. It may be
a set like horses or animals.
One can say that there are several notions of existence or of
quasi-existence. Then the existential quantifier in other theories
(which are interpreted) may express the other notions of existence or
quasi-existence. Thus the simple type theory may of course be understood
as an ontology, in which only the existential quantifier binding the
first order variables express real existence while the existential
quantifiers binding variables of other types express only a kind of
quasi-existence. Such interpretation seems to be suitable for a reistic
philosopher. One can go farther and assume a set theory (and the whole
of mathematics) as a quasi-ontology or as a sci ence-fi ct ion-ontology. It
seems to me quite convenient to assume that mathematics is not a theory
about something really existing; but then perhaps we should not call
mathematics a science, if by science we mean a discipline providing
knowledge about reality. Mathematics may be comprehended as a useful
fantasy, which facilitates our manipulation of things. It becomes a
little mysterious why speculations about non-existent objects can enable
us to construct real existing things or to predict their functioning.
CONSISTENT REISM 43
This interpretation makes us aware of the essential insufficiency of our

understanding of reality. Doing mathematics we pretend that there are
some objects such as e.g. real numbers. We may realize that we do not
grasp the reality literally by our theories. We can only live as if.
Like Pascal's pledge.
3. The reistic semantics
The Elementy, the chief philosophical work of Tadeusz Kotarbinski begins

wi th philosophical semantics. In one of the first sentences of his
semantics Kotarbinski introduces the following expression: "A sign
expresses an experience of somebody", or itA sign expresses somebody as
experiencing so and so". These expressions lead to some confusions as
I~as shown by A. Gawronski. 2 It seems to me that it should be more
reistic to say that • a man expresses' than to say that •a sign ex-
D~esses'. Hence one can imagine that the philosophical semantics should
b~·gin rather in the following way:
First we should make some general observations about human beings
and their use of language. We should describe in our everyday language
how our human linguistic performances may be divided (and traditionally
are) into sentences and words. These may serve as separate tools in
human interactions. This description may be behavioural without semantic
notions. Then we introduce some semantic expressions as new primitive
notions. We should choose some formal shape of these expressions. E.g.
we choose the following phrases:
The human being X by using the expression A expresses
his/her feeling of C
understanding of D
joy over E
sorrow about F
fear of G
want of H.
Then by means of some examples we teach the reader about the correct
applications of these phrases.
At the end of the section it may be allowed in some non-formal
applications to use the "perversive" form of the above phrases:
feeling of C
understanding of D
The expression A expresses the joy over E of the person X.
sorrow about F
fear of G
want of H
assuming that the perversive form is only an ersatz of the primitive
one.
28 .... the present volume.

The other semantic notions should be introduced in a similar way.

The ambition to give an explicit definition for every theoretical notion
is a temptation of the devil (to become wiser than the experience of all
human-kind) and leads to contradictions.
4. Inspirations in geometry and topology
There are no doubt about some connections between the reistic philosophy
of Tadeusz Kotarbinski and some formal endeavours of Stanislaw Lesniew-
ski (hi s sJ'stems of ontology and mereology) and of Alfred Tarski
(geometry of bodies). 3 In this section I shall go farther in this
direction.
Interpreting geometry or· topology as descriptions of the relations
in real space a reistically educated topologist should of course dislike
speaking about points· as fundamental entities of real space. The real
components of the space should have the full dimension of the whole
space. Such components may be called geometrical bodies. (Topologically
they may be def ined as interiors of the i r own closures, i. e. open
domains in the topological terminology of Kuratowski (1948). E.g. ge-
ometrical spheres (in the sense of whole globes) are geometrical bodies.
Tarski (1956a) sketched a geometry having the notion of sphere and the
inclusion between bodies as the two unique primi ti ve notions Tarski' s
definitions of other spatial relations are very ingenious.
To take this further I would suggest founding a topoiogy without
points which may precede Tarski's geometry of bodies as the classical
topology precedes classical geometry. The topology of bodies may be
developed in two steps. The most general topology of bodies may be a
theory having two primitive notions: 1. the relation of inclusion
between bodies 'e' and 2. the predicate of connectedness 'cd'. I have
the candidates for two axioms: .
AI. A, B cd /'. A rI B '" 0 ----t A I) B cd
A2. X cd A C comp Y .', X - Y cd ----t X - C cd
where 'c comp Y' means: C cd /\ C c Y A 'II D (D cd ACe Dey ----t C = D).
Al and A2 are written in Boolean algebra notation. Using mereological
notation (which seems to be appropriate for this mode of thinking)
instead of An B'" 0 we shall write: 3X (X c A A X e B).
Al and A2 are known as very general properties of connectedness.
The notion of connectedness is very intuitive and in classical topology
may be considered as more general than the notions of closure and
interior, which are traditionally assumed as primitives of topology. One
can prove that the closure operation is not definable by means of the
notion of 'cd' on the ground of all sentences true in all topological
spaces. This observation suggests as the second step a theory having one
primitive notion more (Grzegorczyk, 1960).
3S "e Tarski (1956).

CONSISTENT REISM 45
5. Inspirations in the theory of definition
Even using higher types one can profit from some reistic intuitions in a
classification of definitions.
Suppose that we have a number of well founded empirical notions
which denote some properties of things or some relations between things.
Let us call them: empirical primitives. Starting from empirical primi-
tives, by means of logical operations one can define some other notions
concerning things.
These newly defined notions may be called more or less empirical
·according to the logical means applied in the definition. The classical
Jloolean operations of the calculus of propositions are most empirical.
The quantification limited to some empirically well defined sets is more
empirical than quantification which is not limited. If we pass from
quantifiers binding thing-variables to quantifiers binding higher types
then there is only one kind of definition which may be thought of as not
exceeding empirical definability. It is a definition with one existen-
tial quantifier at the beginning of the definiens:
aRb +-+ 38 (F(S,a,b»
S may be a variable of higher type, but the formula 'F(S,a,b)' should be
empirical. Some important empirical notions have such a shape. E.g. the
definition of numerical equivalence of two sets of things. It is
interesting to realize how many notions of science are very far from
empirical verifiability if we assume their speculative theoretical
definitions.
Our life runs in the world of things. We can imagine more. But we
should begin by the analysis of our thing-full environment. Then from
the methodological point of view the reistic philosophical programme of
Kotarbinski looks most attractive.
A NOTE ABOUT REISM
Henryk Hiz
It is a comllonplace by now that there are two theories of reisll: the

semantic and the material theory. This distinction was introduced by
Ajdukiewicz (1930) in his review of Kotarbinski's (1929) Elellenty.
(Kotarbinski's book appeared in English under the unrecognizable title
of Gnosiology; a Scientific Approach to the Theory of Knowledge, 1966;
this translation sometimes does not do justice to the original text).
Today. the difference can be formulated in familiar terms: semantic
reism states the conditions for sentences to be well formed and to say
something whereas material reism limits the models of such sentences.
A model is a structure composed of the primary objects, of functions
from one. two. three. etc., such objects to the logical values (truth
and falsity), of functions from such functions to the logical values.
and so on. At first. this description of a 1I0del, seems not to be stated
in a reistic language. It may be argued, however, that this definition
of a model is acceptable to a reist and that according to material reism
only such models are allowed which take things alone as the primary
objects. Semantic reism states that well formed primary sentences use
names of things. The rest of a primary sentence. those parts of it which
are not names of things. jointly constitute a functor. The functor,
together with names of things as its argullents, forlls a true or false
sentence. There may also be sentences, no longer primary, which have as
arguments not names of things but the functors just mentioned; as
functors these sentences have a new kind of expressions which forll
sentences with primary functors as their arguments. Generally:
(1) Nl,N2, ••• are names of things;
(2) If F(Nj.l.Nj.2, ••• ,Nj.x) is a sentence, then F is a functor of the
first order;
(3) If Fi.l,Fi.2, ••• ,Fi.v are functors of order II and if
G(Fi.l.Fi.2, •••• Fi.v) is a sentence, then G is a functor of the 11+1
order.
If there are x nalles which with F form a sentence, then F, when
used with them. is a functor of the first order froll x arguments, and if
there are y functors of order m which with G form a sentence. then G,
when used ·with them. is a functor of 1I+1st order from y arguments.
47
I. Wolenski (ed.). Kotarbinski: Logic. Semantics and Ontology. 47-52.
48 HENRYK HIi
Some authors accept the principle that the expressions that jointly
occur as arguments of a functor must all be of the same order. This
principle resembles the simple theory of types. Others allow arguments
of mixed order (in which case a changed is needed in (3), to the effect
that m is the highest order of an argument of G).
Kotarbinski mentions explicitly only the Nj,l where Nj,l is called
the subject. It is a relic of Aristotle's .notion that an elementary
sentence is composed of a name and a predicate. There is no good reason
for a reist to insist on the theory of singularity of subject, or singu-
larity of argument. In the sentence 'Johny and Jill got married' the
words 'Johny' and 'Jill' ~re names of things and can be considered argu-
ments of 'got married' which is a first order sentence-forming functor
from two arguments (in our culture). In the sentence 'The teacher gave
the book to a boy' the words "the teacher', 'the book', and 'a boy' are
names of things while 'gave' is a first order functor froll three argu-
ments. (The case indicators like 'to' show that these tlouns are argu-
ments.) However, according to Aristotle, and presumably according to
Kotarbinski, the functor of the first order in this sentence is 'gave
the book to a boy'. This functor can, in turn, be decomposed into a
functor-forming functor 'gave the book' and its argument 'a boy'. There
is no good reason to debate whether the grammatical theory of Aristotle
and Kotarbinski is reistically more correct than the theory shown in
points (1), (2) and (3). Each of them is compatible with reism.
Kotarbinski wants all sentences to be reducible to sentences which
have names of things as arguments. This reducibility can be understood
in two ways: either loosely, as in the theories just sketched, or more
rIgorously as the requirement that every sentence be translable into a
sentence in which the arguments are names of things exclusively. The
more rigorous understanding is suggested in Ele.enty and in Kotarbin-
ski's other writings. But it cannot be taken quite literally. In the
sentence 'Jill plays tennis when she has free time' the expression
'when' is a functor of two sentences, not of two names. The sentence
'For some F, F(the teacher, the book, a boy)', or lIore simply 'Something
took place between the teacher, the book and a boy', cannot be trans-
lated into a sentence in which there are only names of things and a
functor of them; at least not easily. The rigorous 'demand of reducibili-
ty leads to an artificial language difficult to study. Usually, it is no
longer English (or Italian, or Polish or the like). But sentences are
sentences of English (or Italian, or Polish). It is therefore worthwhile
to limit the discussion to the looser understanding of reducibility.
There are' functors of zero degree, which can for. a sentence with-
out requiring any arguments. In other words, they can constitute'senten-
ces by themselves. When one uses only functors of zero order, the hier-
archy given in (1), (2) and (3) is simplified. Zellig Harris (1982) pro-
poses to use only two kinds of functors in grammar: functors of argu-
me'nts that are nouns and functors of arguments at least sOlie of which
are sentences. Nouns are for him only those, expressions that can be
names of things. Harris is a reist among grumarians, and not without
Kotarbinski's influence.
A NOTE ABOUT REISM 49
Somebody may question whether, in fact, 'the teacher' is a noun. To

be sure, it refers to a man. But not exclusively. Not in the sentences
'History is a teacher of life' and 'Mistakes teach'. When I hear that
these are metaphors, I confess that I am not able to distinguish clearly
metaphors from non-metaphors. (A language continuously broadens its
supply of sentences by accepting commonly used metaphors as non-meta-
phors: 'I feel gloomy about the future', 'She has a sweet disposition',
'Military expenditure is a great drain on the national wealth', 'This
concept is sharp and clear', 'I have mixed feelings'.) Now, a teacher is
someone who teaches and the word 'teacher' is a nominalization of the
verb 'to teach'. Anytime I say 'A teacher got angry' I could say instead
'Somebody (a person) who teaches got angry'. The same will hold for
other sentences containing the word 'teacher'. 'Our physics teacher' is
the same as 'The person that teaches us physics'. Thus, 'the teacher' is
not a noun. In the same fashion other nouns formed from verbs, perhaps
all, can be considered verbs rather than nouns. Moreover, their elimina-
tion from among nouns can be carried out in accordance with transforma-
tional laws. Those are laws of a language (in the present examples they
are laws of English), and not necessarily directives of philosophy.
Rather, it is philosophy that should strive to be compatible with the
linguistic laws, for they, like the laws of physics, constitute general
conditions for philosophy. Kotarbinski, too, warned against taking such
nominalizations as 'teaching' as bona fide names. It is not clear what
would have been his advice concerning such words as 'teaching' (in
'teaching material', 'a teaching machine', or in 'the teaching of
Buddha'), 'taught' (as in 'a self-taught man'), 'unteachable', 'teacher-
less'. Kotarbinski (and Lesniewski, and Aristotle), for unclear reasons,
do not distinguish adjectives from nouns. A whole class of expressions
derive from 'teach': 'teacher', 'taught', 'self-taught', 'teaching',
etc. Any sentence which contains one of those-expressions can be treated
as containing the functor 'teach' together with an indication of whether
it is used as a functor or as the first argument in a sentence
('teacher'), the second ('the taught subject matter', 'teachings'), or
the third ('the taught person'), or whether 'teach' is an argument to a
still higher order functor ('Frank likes teaching', 'I doubt whether the
violin is teachable'). A reist, it seems to me, should not object to
such verbal nominalizations. But, when an argument' of the functor
'teach' is not the name of a thing, for· instance in 'Somebody's error
teaches us a lot', he requires an additional elucidation. One can re-
phrase the content of the last example by a sentence: 'When somebody
errs, it teaches us a lot'. In this sentence, the offensive onomatoid
'error' is replaced by the verb. But in this sentence, 'it' is 'not the
name of a thing. Rather, it is a referential to the preceding sentence.
So as to properly discern the order of functors, we must accept that the
word 'it' in 'When somebody errs, it teaches us a lot' is of a different
grammatical category than in my saying , I t is Johny'. Similarly, the
word 'teaches' in 'When somebody errs, it teaches us a lot' is of a
different category than in 'Johny teaches physics'. A reist may object
to the word 'it' in 'it teaches us a lot'. It is not the name of a
thing, perhaps not a noun at all, but short for the sentence 'Somebody
erred'. The whole can be restated as: 'That somebody made an error
50 HENRYK HIi
teaches us a lot'. Now we speak explicitly about a fact, not about a

thing. Languages cling to speaking about contents of sentences, about
facts, about events. This bothers a reist. His trouble lies in the dif-
ficulty of eliminating sentential referentials.
Kotarbinski ' s reism attempts to distinguish apparent names, the
onomatoids, from the veritable names. In the approach just presented,
the onomatoids becomes nouns used as functors. Thus, an alternative
theory - let us call it reism of a higher order - could do without the
distinction between veritable names and onomatoids, closely scrutinizing
functors instead. There are functors, for instance 'eat' and 'teach'
that take as argullents the expressions •something' or •somebody' • In
those contexts they are veritable functors of the first order. When such
a functor itself occurs as an argument, it often has a somewhat differ-
ent form, as mentioned before, for instance, 'teaching' or 'the teaching
of something to someone by someone'. Here are a few examples of 'teach'
used as a part of a functor from one argument, the subject: •.•• teaches
Frank physics', •••• teaches Frank how to play the violin', •••• teaches
Frank the violin', •.•• teaches Frank medicine', •••• teaches Frank how
to practice medicine'. Some of these expressions mean the same as some
others ('teaches the violin' = 'teaches violin playing' = 'teaches how
to play the violin'). An argument of • teach' may be a sentence, or its
shortened form. The first argument of 'teach', its subject, is either a
name of a thing, a sentence, or a disguised form of a sentence. In all
the above examples the subject could have been: 'Looking at people' or
'Frank's looking at people'. Similarly, we have 'Peter delayed Frank's
departure' and' A change of the time-table delayed Frank's departure'.
'Teach', 'delay' are, therefore, not only functors on nouns as argu-
ments. In the place of a noun sOlie form of a sentence can occur. In such
a case the functor is of a higher order. Also the second (or third)
argument of • teach' can be a sentence as in 'Experience teaches us that
trains do not run on schedule'.
To formulate the reisll of a higher order, one has to allow as argu-
ments in sentences not only nouns and functors, as prescribed by rules
( 1 ) - (3), but also sentences. Sentences, however, are functors of zero
arguments. The functors which accept sentences as their arguments will
be called functors of the sentential order. To the conditions (1)-(3) we
will now add:
(4) If G(Nj,l, ••. ,Nj,,,,Fi,l, ••• Fi,y) is a sentence, where x ~ 0 and at

least one of Fi,l, .•. ,Fi,y is of zero arguments, then G is a functor
of a sentential order.
In the rule (4) we have loosened the impositions of the theory of

types, which are somewhat artificial. The arguaents of functor G are not
necessarily all of the same type, or of the same grammatical category.
For instance, 'delay' in the sentence 'Peter delayed Frank's departure'
is a functor of a nominal argument 'Peter' and. a second, sentential,
argument 'Frank departs'. This is the analysis of the sentence 'Peter
delayed Frank's departure' when you transform it into: 'Frank departs;
Peter delays it'. Here 'it' is not a pronoun, for it is a referential
not to a proper name or a noun but to a sentence. One can call this 'it'
A NOTE ABOUT REISN 51
a "prosentence". Referentials are classified accord~ng to the gram-

matical category of their referend.
Perhaps it is worth noting that almost all of the functors illus-
trated in the present Note form sentences. One may remain with that, as
Harris does. One can also ramify the theory to the functor-forming
functors, as Lesniewski did.
There are functors which can sOlletimes be of sentential order
(almost all of them can). There are also functors which are always of a
sentential order, for instance, 'accelerate'. 'Coffee accelerates the
heart' is a transformational abbreviation of the sentence 'Drinking
coffee accelerates the heart beat', or of 'The heart beats faster in a
person after he drinks coffe'. Now, we can speak about rematoids, that
is, about functors some of whose arguments must be sentences. 'Acceler-
ate' is a rematoid. From what was said earlier, it follows also that
'teach' is a rematoid. The existence of rematoids is contrary to simple
reism as formulated in (1)-(3).
Sentences like 'Somebody teaches us physics' present a reist with
an additional difficulty. It is a short form of the sentence 'Somebody
teaches us how we should understand physics'. The difficulty is with the
word 'physics'. For, physics is not a thing, like a violin, and
'physics' is not a name of a thing. Nor is it evident of what sentence
can it be an abbreviation. It is not a functor, hence it must be a noun.
It seems that we are forced to accept nouns which are not nalles of
things. Consider a simple example. For a bookbinder, a book is a set of
sheets. For a printer, each sheet has two sides which count (and four
edges which do not count). The result is that a book has pages and the
first page of a book is not the same as the second page of the book even
though they are on the same sheet. If so, one must accept that there are
surfaces. Similarly, there are edges, ridges, angles, directions. Their
existence is no less well founded than the existence of an inch or a
street. It does not make sense to ask how much a street weights, even
though one may ask how broad it is. One can ask how long an edge is (but
not how broad), or how obtuse an angle. Such are the difficulties of
material reism. There are reasons to relax material reisll. If one
insists that streets, angles and edges are not things and that they do
not exist in the same sense in which houses and people do, it lIay turn
out that he will have to give up the inquiry about what is named by a
noun. But it suffices to speak about sentences and functors of senten-
tial order. Then we won't need rules (1) and (2), and in (4) we can
accept x = O. SOlie sentences are true and this is the link between the
language in which they are formed and reality. Reality is not necessari-
ly things, nor nouns necessarily names. It may be enough to have only
one anchor in reality: the truth. We can delete naming from a semantic
theory and rely solely on truth. Truth, however, is a property of sOlie
sentences and a property of no names. The sentence 'The point of the
knife aims at an apple' lIay be true. It does not follow, however, that
there is an aim, a point, or even a knife or an apple. A sentence may
say this or that without proclailling that something exists.
52 HENRYK HIZ
P.S. There is a difference between the use of the pair of English words
'noun' and 'name' and the pai r of roughly corresponding Polish words
'nazwa' and 'illlie:. The difference arises frolll the fact that English
'name' has the verbal 'to name' and 'noun' does not have a verbal form.
In Polish, on the other hand, 'nazwa' has the verbal fora 'nazywac' and
'imi~' does not have one. There are also disributional differences; e.g.
'the name of the city', *'the noun of/for the city', *' imi~ miasta',
'nazwa miasta'. Kotarbinski uses 'nazwa' for all the cases of nallles,
nouns and noun phrases.
PUZZLES OF EXISTENCE
Janina Kotarbinska
1. The controversy over universals has been revived, the parties being
the same that clashed in Antiquity and The Middle Ages: realism, con-
ceptualism, nominalism. The difference is that universals referred to at
present are neither "forms" nor "essences of things", nor are they ·Pla-
tonic ideas of concrete objects, but abstract entities of a special
kind: sets of individuals, sets of sets of individuals, sets of sets of
sets of individuals, etc., interpreted as objects which are essentially
non-perceivable, extra-temporal and extra-spatial. Just as centuries
ago, the realistic standpoint is that universals exist; the conceptual-
istic standpoint is that they exist in mente, but not extra mente.; and
the nominalistic standpoint, in opposition to the other two trends, is
that the existence of universals is firmly rejected. Conceptualism has
few adherents at present, but the other two movements are very strong.
Their fortunes are, however, variable. Until not so long ago we
might have been inclined to think that the nominalistic trend had final-
ly won the upper hand, and that realism, which was almost universally
declared to be metaphysical and unscientific, would never recover the
ground it had lost. Yet even a cursory knowledge of the current litera-
ture of the subject shows that the situation is otherwise. Recently, it
is nominalism which has been under fire. It is criticized for being at
variance, as far as its basic principles are concerned, with the funda-
mental assumptions. of contemporary mathematics, which is being pursued
in a realistic spirit, on the basis of set theory and its conceptual ap-
paratus. As we know, the important concepts of all mathematical disci-
plines have been defined by reference to the concept of set, and the
theorems accepted in those disciplines include existential theorems
which assume the existence of sets of specified kinds. The situation is
similar in other disciplines, particularly in logical semantics. Matters
have thus cOile to a head. It is obvious that mathematics cannot be the
loser in this unequal conflict •. It is only nominalism which can be the
loser, as it certainly will be unless it succeeds in demonstrating that
its inconformity with mathematics is merely apparent or can be elimin-
ated in a manner that will not require far-reaching compromise. There-
fore, it is not surprising that in these circumstances nominalists con-
centrate their efforts on finding a way out of the present crisis.
53
J. Wolerfsld (ed.), Kotarbirfsld: Logic, Semantics and Ontology, 53--67.
54 JANINA KOTARBINSKA
2. The most radical approach, initiated by Lesniewski in connection

with his analysis of Russell's antinomy, goes to the very roots of the
evil and calls for a fundamental revision of the foundations of mathema-
tics. His criticism is levelled mainly at the set-theoretical (distribu-
tive) concept of set. In Lesniewski's opinion, this concept is as a rule
introduced by means of highly unintuitive assumptions which, moreover,
impose such conditions on the sets that it would be difficult even to
guess what the objects to which the axioms of set theory would apply
would be like. As a result, set theory - and the whole of mathematics,
if the latter is reduced to set theory - becomes a sphere of specula-
tions concerned with mythical objects invented by logicians, and does
not in the least contribute to a scientific interpretation of the real
world. If we want to avoid these consequences, there is only one way
out: we have to do away with the set-theoretical concept of set - a
product of minds demoralized by speculative constructions dissociated
from the real world - and to revert to the concept of set in its ordi-
nary, and not denaturalized, meaning, according to which it would be
correct to say that Black Forest is the set of those trees which grow in
that area, and that Ursa Major is the set of those stars of which it
consists. In other words, the classical set theory, i.e., the theory of
sets interpreted in the distributive sense, should be replaced by mereo-
logy, i.e., the theory of sets interpreted in the collective sense. 1
The differences between these two interpretations are well known.
A set in the distributive sense is supposed to be an abstract object;
a set in the collective sense is supposed to be a concrete object whose
elements are its component parts, whether proper or improper. According
to the first interpretation we also have the following relationships:
(al) an object x is an element of a set of ~s if and only if x is an M:
(bI) a set of M's is identical with a set of N's if and only if, for
every x, if x is an element of the set of M's, then x is an element of
the set of N' s, and conversely: (Cl ) for every object x, which is
identical either with an individual or with a set, there is a set which
is not identical with x and of which x is its only element (hence every
object is a starting point for an infinite hierarchy of sets of one
element each, sets which differ both froll that object and froll one
another); (d!) there is an empty set which has no elements. For the
second interpretation we have the following relationships: (~) if x is
an M, then x is an element of the set of M's, but not conversely; (b.z) a
set of M's can be identical with a set of }fs, even if it is not true
that every M is an N and conversely; (C2) every set of one element is
identical with its only element (there are, accordingly, only as many
sets of one element as there are individuals; sets of one elellent thus
do not increase the number of existing individuals); (dz) there are no
empty sets; if there is a set of ~s, then there is at least one object
which is an M.
lL"~ni .. " .. ki
<1927 - .. specially pp. 166-9 and 186-20:3>, and (1928 - es-
pecially pp. 261-3). Similar ideos or .. t.o b .. ·found in Leonord • (foodmon
(1940); in Goodmon <1961>; ond in Slupecki <i96tO; olso Lusch .. i (1962).
PUZZLES OF EXISTENCE 55
It immediately follows that the collective interpretation of the

concept of sets differs essentially from the distributive one. The dif-
ferences between those two interpretations result however, not only in
intended consequences (the adjustment of the concept of set to the re-
quirements of nominalism), but in unintended and (what is worse) clearly
undesirable ones as well. It has become evident that the theory of sets
interpreted in the collective sense is much proper than the theory of
sets interpreted in the distributive sense, and does not sufice as a
foundation of the arithmetic of natural numbers. It is, therefore, too
poor to take over all the functions of set theory as the fundamental ma-
thematical discipline. Accordingly, it has become obvious that the ob-
stacles with which the nominalists have to cope have not been removed by
the measures described above.
3. Unlike Lesniewski's idea, the remaining proposals advanced by the
nominalists tend to "de-Platonize" mathematics while retaining set the-
ory as the fundamental discipline. The basic method they use is that of
semantic analysis. In this connection special attention ought to be paid
to the singling out, among utterances in the form of sentences, of three
kinds: (1) utterances which are meningful for the literal interpretation
of the expressions which occur in them; (2) utterances which are mean-
ingless for a literal interpretation, but meaningful for a metaphorical,
substitutive, interpretation; (3) utterances which are meaningless for a
literal interpretation and have no substitutive interpretation, and
hence are simply meaningless. Utterances of the first kind are those
which satisfy certain specified criteria of meaningfulness, criteria
which, let us add, are chosen so that the extensions of sentences coin-
cide with the extensions of those sentences which comply with the nomi-
nalistic requ i rements. Thus, in a li teral interpretat ion, only those
sentences are meaningful which are formulated in the language of indivi-
duals. Membership of the second category is determined by the secondary
criteria of meaningfulness, namely the reducibility of such sentences,
by semantic analysis, to sentences which satisfy the said primary crite-
ria. Finally, utterances of the third kind do not satisfy any criteria
of meaningfulness, whether primary or secondary, and are thus not in the
stock of expressions of a given language. It can easily be seen that
mathematical theorems, like all theorems in which the term 'set' or any
terms which are its derivatives. occur, are treated as having been formu-
lated in a substitutive language. The problem is to find a method which
will make it possible to translate any such theorem into a literal lan-
guage, and hence into a language in which we can speak solely about
individuals.
This is the main outline of the approach now under consideration.
As we can see, the problem of the criteria of meaningfulness comes to
the fore here. We shall examine the issue in greater detail, taking the
reistic (concretistic) doctrine developed and worked out in detail by
T. Kotarbinski, as our starting point and basis of analysis. We shall
note, too, that an analysis of that doctrine, which is usually believed
to be the most representative variation of contemporary nominalism, will
be the principal object of investigations in this paper.
In discussing reism, we must first of all Ilake a distinction be-

tween its ontological and semantic aspects. Both aspects are, after all,
closely connected. As an ontological doctrine reism claims that only
things exist, or, in other words, that every object is a thing. It is
assumed that every thing is a spatio-temporal object which is in prin-
ciple knowable by perception, and that the verb 'exist' is taken in its
fundamental sense. "In that fundaaental sense: A exists, is the same as:
a certain object is A, which [ ••• ] can either be shortened into: some-
thing is A, or expanded pedantically: for sOlie x, x is A. Now the reist
deems that only things and persons exist, since it is true only of
things and persons that certain objects are things or persons. Should a
person further ask about the definition of the term 'object', we should
have to refer to the meaning of the copula • is' in singular empirical
statements (such as 'this is green', with an indication of a leaf; or
'the Earth is spherical'; or 'I am gay'; or 'Peter is a carpenter') and
say that only that, and all that, is an object about which we may mean-
ingfully formulate a singular sentence (of the type 'A is B') with the
copula so understood."l Ontological reisll is a case of ontological nomi-
nalism which adopts the following more general thesis: only individuals
exist, in other words: every object is an individual. The reists also
assert that every individual is a thing, which the nominalistic approach
does not claim.
Semantic reisll has undergone certain transformations during its de-
velopment. Originally, its principal thesis was an inductive generaliza-
tion which stated that, in ordinary linguistic usage, sentences which
include nouns that are not names of things are interpreted as substitu-
tive abbreviations of sentences which do not include any such nouns.
That was, as we are told in Kotarbinski (1958), "the genetic and morpho-
logical nucleus of concretism". The ontological thesis present~d itself
as an explanatory guess: it is probably true because only things exist
and we can speak only about things. Hence if we appear to be speaking
about properties, relations and other alleged objects which are not
things, we are in fact speaking about things.
These two theses, the descriptively semantic and the ontological,
have resulted in turn in certain assumptions regarding the properties
and structure of language. These are intended, as may be supposed, to
bring about a reconstruction of ordinary speech according to certain
guidelines adopted in advance. The point is, on the other hand, to be in
agreement with the real world, which consists only of things (if only
things exist, then only names of things can occur in such a true, and
~ot.arbinski <194-9, p.4-:!4-). It. may be helpful t.o explain t.hat. reism is
based on Lesnievski's ont.ology and uses t.he concept.s charact.erist.ic of
t.hat. syst.em. In t.hat, ont,ology t,he only primit.ive t,erm is ,~, ('U'),
int,roduced by t,he axiom V A, B A ~ B '" V X X ~ A --+ X ~ B /\
3 X X ~ A /\ V X Y. X ~ A /\ Y. ~ A --+ X e·M y. . The hrm 'e/.X'
('~') i . defined t.hus: V A e/.X A '" 3 X X ~ A . This obviously
yields t.he t.heorem: V A B A ~ B --+ e/.X A , by which a st.at,ement, of'
t,he t,ype A ~ B has an exist,ent.ial consequence e/.X A vhich applies 1.0
t,he designat,um of' A.
hence also meaningful, utterance which states by implication that desig-

nata of such names exist), and on the other, to be in agreement with
ordinary speech, in which such statements as 'Seniority is a transitive
re lation' are accepted as both meaningful and true. Here assistance can
be found in the distinction between utterances interpreted in their
1 i teral sense and those which are interpreted in a metaphorical sense
closely c.onnected with the distinction between genuine names or onoma-
toids. Roughly speaking, semantic reism now becomes a viewpoint accord-
ing to which utterances which include names of properties, relations,
etc. (in general, names of alleged objects belonging to ontological cat-
egories other than that of things), are meaningful, and hence also true,
only in a non-literal interpretation of those names, and the said
supposed names of properties, relations, etc., are merely apparent
names, or onomatoids. The thesis of semantic reism includes a number of
variants. These are the most important, quoted verbatim or almost
verbatim.
(n) All sentences in which something is seemingly said about an object
which i.s not a thing, are treated as substitutive formulations standing
for other sentences, the latter understood literally and predicating
exclusively about things. In other word p , any 'term' which is not a name
of a thing, is held to be an onomatoid.:'!
(b) A n>ist is a person who interprets ordinary speech so that every-
thing which is a name by appearance, and denotes an alleged object of a
category other than that of things, is for him an onomatoid. 4
(c) Any sentence which formally implies that an alleged name of an
object which is of a category other than that of things, is a name of a
thing, can be true only on condition that it is interpreted so that it
cannot be proved by reference to it that the name in question is a name
of a thing. For the· sake of brevity, this can be reformulated as fol-
lows: any alleged name of an object which is not a thing ontologically
is an apparent name (an onomatoid).6
All these formulations are focused, above all, on the concept of
genuine name, or simply, name, on the one hand, and that of apparent
name, or onomatoid, on the other. What a name is, can be seen from the
following definition: "to be a term is to be usable as a predicate· (in a
later text: as a subject or a predicate - J.K.) in any sentence 'A is B'
with the primary understanding of the copula • is'. A given saying is
usable as a predicate not only if, when substituted for B, it makes the
whole sentence a true one, but even if it makes the sentence a meaning-
ful one." Since onomatoids can also be meaningfully substituted for B in
the sentences of the type 'A is B' (for instance, 'Seniority is a tran-
sitive relation'), the reservation with the primary understanding of the
3Kol.arbinski <1966, p.6D.

"'Kotarbinski (19680, p.76).
6'U~ -a-l./.fl/la, pp. 109-10. Bol.h (c) and (a) have two formulal.ions each: bol.h
cases formulations which form a given pair are I.reat.ed as synonymous.
These will be marked ·(01), (a2), (c1), (c2).
copula 'is' is intended to make sure that the words usable as subjective
complements are genuine names. 6
The interpretation of this reservation is, however, a source of
additional problems. The examples and comments given by Kotarbinski seem
to indicate that the copula 'is' is used in its primary sense if and
only if it stands between names and if, moreover, it satisfies the
formal comditions resulting from Lesniewski's axiom. Yet, as has been
noted by Ajdukiewicz, that combination of criteria appears to be a vi-
cious circle (Ajdukiewicz's original formulation was ~All that system of
definitions looks like a circle, though riot a vicious one.,,)7 and at any
rate it does not provide a method which would make it possible to dis-
tinguish, in concrete cases, expressions which are (genuine) names from
those which are not. This is why, in practice, when the reist has to
decide whether a given expression is, or is not, a (genuine) name, he
makes use of criteria which do not refer to the syntactic properties of
the expressions in question, -but to their semantic properties. In gen-
eral terms, he will class N as a name only (and probably_ always) if it
follows from the very meaning of that N that N is such and such a thing,
or simply that N is a thing. 8 It is perhaps worth mentioning here that
for such an interpretation the primary thesis of ontological reism is no
longer a hypothesis, but an analytic statement which is true on the
strength of the rules of language themselves, since in accordance with
these rules it is only about a thing that we can meaningfully say, given
a primary understanding of the copula 'is', that an object is precisely
that thing, or, that that thing does exist. '
As regards an apparent name we are told in turn that it is "any
word (or phrase) which may meaningfully stand for B in a structure of
the type 'A is B', but only if that structure plays not its prima~y role
but the role of a substitutive and an abbreviation.~ We are further told
that an apparent hame, when inserted in a singular sentence, yields a
meaningless whole if the fundamental, non-metaphorical sense of the cop-
ula 'is' is retained, and that those sentences in which apparent names
occur are substitutive expressions to which the structural criteria of
truth, used for sentences with genuine names, cannot be applied; in
particular, we cannot infer from them sentences of the type 'A exists'.
Finally, it is said that ~ apparent terms (onomatoids) [ ... ) are certain
nouns, adjectives, noun phrases and adjectival phrases which because of
their appearance are taken to be terms, but are not terms; they can be
used meaningfully only in substitutive formulations, and can always be
eliminated in favour of genuine terms alone.~9
.When all this is taken into consideration, we have the possibility
of the following interpretation of the terms in question, which allows
us to avoid the objection of a vicious circle (we offer these formula-
tjons with no claim to maximum precision).
6Ko t.arbinski <1966, p.7). S~~ also his -<1958a, pp.232-3).
7 From Ajduki~wicz's (1930) review of Element.y - see t.his volume p. 11.
8 Ko t.<l"rbinski (1958a, pp, 232-3),
9 Ko t.arbinskl <1966, pp,- 9 and 406); see also (1958a, pp, 74 and 232-3).'
(1) An expression N is a name if and only if it is assigned, by the os-

tensive method, to a special thing or things of a special kind (briefly:
if it is an ostensive term), or if it is definable, completely or par-
tially, by means of ostensive terms as the only ostensive terms occur-
ring in a given definition.l0
(2) The copula 'is' occurs in its primary role if and only if it occurs
in a sentence of the type 'A is B', if that sentence has structural cri-
teria of truth fixed by the axiom of ontology, and if both A and Bare
names.
(3) An expression of the type 'A is B' is meaningful in a Ii teral
interpretation if and only if the copula 'is' occurs in its primary
role.
(4) An expression of the type 'A is B' is meaningful in a metaphorical
interpretation if and only if A and B are not names and if it is trans-
latable into a sentence which is meaningful in a literal interpretation.
(5) If an expression 'W' occurs in a formulation Z of the type 'A is B'
as the subject or the predicate, then 'W' is an apparent name if and
only if it is not a name and if Z is meaningful in a metaphorical inter-
pretation. 11
(6) An expression which has the form of a sentence is meaningful if and
only if it is meaningful either in a literal or a metaphorical interpre-
tation.
In the light of these explanations the thesis advanced by semantic
reism, which states that any alleged names of objects which are not in
the ontological category of things are apparent names, affirms, first,
that utterances in the form of sentences in which alleged nues occur
are not meaningful in a literal interpretation, and, second, that they
are meaningful only on condition that they are translatable into formu-
lations that are meaningful in a literal interpretation. 11
10This interpretation is partly based on the formulation. found in El.-

m.nty, and in the paper <194-9>. It i . to b. not .. d that at th .. tim. when
th. EI ... enty fir.t appeared partial d .. finition. were not y .. t known. Yet
partial definition. should 01.0 b .. tak.n into con.id.ration; this i .
justifi.d by the pr .... nt-day approach and at th.. .ame ti... would not
app.ar to be at varianc. with the int.ntion. of the r . i . t •.
l~ormulation. (3), (4-> and (5) are, f'or the .ak. of' .implicity, given
only for .tatement. of' the type 'A is ~'. How.ver, sinc. in ontology all
.imple .tat.m.nt. or .. def'initionally r.ducibl. to .tat ..... nt. of' that
typ., and .inc. all compound stat •• ent. or. combination. of' .imple on •• ,
the .am .. f'ormulations can easily be mad .. to cover all other .tat ••• nt •.
11"ot .. that, if' this analy.i. i . corr.ct, th.n of th. two .tat.ment.,
(01) and (Cl>, which in the rei.tic th.ory or. con.ider.d to be .ynony-
moue with (02) or (c2> (cf. footnote 5 above>, (02) alon . . . voke. no
doubt. or objection •. In the ca ... of' .(C2>, it. ....m. to diff'.r in content
from (01) and (cl>; in particular, it do.s not . t a t . - not directly any-
vay - that tran.latability into the language of' thing. i . a necessary
condition. It point. rather to the motives which account f'or th. adop-
t.ion of that assu .. ption.
60 JANINA KOTARBUiSKA
In the reistic system, the above statement is accompanied by the

additional assumption that utterances which include apparent names and
which are in current use in ordinary speech and in the language of
science are nothing other than substitutive abbreviations of sentences
formulated so that they are meaningful in a literal interpretation. This
addi tional assumption, we are told, "is naively intui ti ve and based on
COllllon induction.,,13 It is therefore a hypothetical assumption, uncertain
and only partially substantiated. It must also be emphasized that the
range of observations on which it is based does not go beyond the
sillplest cases; all endeavours to find adequate translations into a
reistic language of more complicated sentences - such as sentences about
the properties of properties, about relations between properties, etc.,
in general and in another teminology: sentences about sets of higher
orders than sets of individuals - have so far ended in failure. Let us
add that in these circumstances the feasibility of this task with regard
to such 1I0re complicated sentences is highly problematic; hence the
meaningfulness of such sentences is problematic, too, since it has been
made dependent by definition on their translatability into sentences
susceptible to a literal interpretation, as both a necessary and a suf-
ficient condition.
At this point we have reached the main source of difficulties with
which reists have to cope. This becomes quite clear if we realize that
the aforementioned more complicated sentences include aathematical the-
orells as well as theorems of all sciences which make use of the concep-
tual apparatus of mathematics, and that the meaningfulness of these sen-
tences cannot be affirmed, even with a merely tolerably sufficient sub-
stantiation.The conflict with the exact sciences thus becomes quite
clear.
The question arises whether reists can avoid this conflict without
abandoning their most important goals. To answer this we must take a
further step in our analysis. To do this we shall go back to the assump-
tions of semantic reism quoted earlier, and try to comprehend as clearly
as possible the underlying intentions of that standpoint.
4. We can start from the statement (c) above, especially the formula-
tion that "any sentence which formally implies that an alleged nue of
an object which is of a category other than that of things, is a name of
a thing, can be true only on condition that it is interpreted so that it
cannot be proved by reference to it that the name in question is a name
of a thing."
A commentary on this formulation, expounding in greater detail the
ideas expressed here, would be to the point. We shall quote it there-
fore, almost in extenso, even though it is fairly long. It goes as
follows:
"A critical reader who sees that such words as 'property', 'rela-
tion', etc., are called pseudo-names or apparent names, will be right if
he asks which linguistic system is meant here: is it the terminology of
a certain scientist, ordinary literary language, or the language which
the present author has imposed upon himself?"
"But it is only apparently the case that what one is aiming at here
is to describe the usages and conventions of such and such a linguistic
system, and it is also only apparently the case that what one is aiming
at here is to suggest such and such linguistic conventions. In my inner-
most intentions, when I call the words 'relation', 'property', etc.,
apparent names I mean a certain condition of the truth of sentences in
any language in which, according to the intention of the speaker, those
words do not denote things. Now in any such language any utterance which
formally implies the existence of designata of such words can be true
only in so far as it is substitutive or non-literal in nature - if it
has a secondary interpretation in which such a proof of existence would
not be possible. Consider, for instance, the sentence 'Whiteness is an
attribute of snow'. Formally, the existence of whiteness follows froll
it, on the strength of the formula: 'A est B ~ ex A'. Now we say that
in this sentence the word 'whiteness' is an apparent name. By saying
this [ ••• J we claim that if this sentence is to be true in the speaker's
language, it must be interpreted in a substitutive, secondary sense
(e.g., as a substitutive of the sentence 'Snow is white'), for in its
literal interpretation we could prove the existence of whiteness on the
strength of this sentence. lfl 4-
When we consider this and similar statements we can hardly avoid
the impression that the motive by which the reist has been guided while
formulating his se.antic doctrine has been an endeavour to ensure to
sentences with abstract terms as subjects (briefly: sentences about ab-
stract entities) an interpretation for which those sentences would have
existential consequences incompatible with the main principle of ontolo-
gical eism. This motive seems to have been decisive for the imposition
on the concepts of name, the primary understanding of the copula 'is',
a meaningful sentence, etc., of definitional conditions which have
resulted in difficulties which are already familiar to us.
We may ask, however, whether - from the point of view of the said
intentions - all these conditions are really necessary, in particular,
whether it is necessary to refer to assumptions which are mainly respon-
sible for such untoward consequences.
The procedure adopted in the reistic system is basically as fol-
lows. Assumptions are adopted as a result of which:
(a) a sentence of the type 'A is B', with A standing for an abstract
term, is meaningful only on condition that the copula 'is' occurs in it
in its non-primary, metaphorical sense;
(b) a sentence of the type 'A exists' follows logically from a sentence
of the type 'A is B', with A standing for the same subject in both sen-
tences, if and only if in the latter sentence the copula 'is' is used in
its primary sense.
As a result:
(c) consequences of the type 'A exists', which would affirm the exist-
ence of abstract entities, cannot be deduced from the sentences de-
scribed under (a).
14-Kot.arbi~.ki <1930-31> - quot.ed from <1968 a, pp. 108-9).

It can be clearly seen now that, first, the meaninglessness of

sentences of the type in question, given a primary understanding of the
copula 'is', suffices to prevent existential consequences from being
deducible from such sentences, but this is by no means necessary: the
fact that the copula 'is' does not occur in these sentences in its
primary sense is by itself sufficient. Secondly, the sentences in
question are here deprived of all existential consequences, whereas the
ontological thesis of the reistic system requires merely that these
sentences have no consequences of the type 'A exists', given a
fundamental interpretation of the verb 'exist' (see pp. 55-56 above). In
both cases the requirements exceed what is really indispensable for the
attainment of the desired result.
It might therefore be suggested that this result could perhaps be
attained by reference to weaker means, which would not require such ex-
cessively severe restrictions, and that this would perhaps be the proper
method of avoiding difficulties which tax the reist so much.
The answer will be easier to find if we become aware in greater
detail how the verb 'exist' ought to be interpreted when used in its
fundamental meaning. Some light is shed upon this issue by a statement
which we have already quoted on another occasion (see pp. 55-56 above)
According to that explanation: "In that fundamental sense: 'A exists' is
the same as: 'a certain object is A'," and an object is defined as any-
thing, and only that thing, about which we can meaningfully formulate
singular sentences with the copula 'is' interpreted as in singular em-
pirical statements, while the latter can be meaningfully made only about
things. Such a standpoint has numerous consequences. First of all, it
follows that sentences of the type 'A exists', given a fundamental
interpretation of the verb 'exist', are empirical statementsj secondly,
sentences of the type 'A is B', if they contain existential sentences
interpreted in their fundamental sense, are also empirical' statementsj
thirdly, the fundamental sense of the verb 'is' and 'exists' is made to
depend not only on the syntactic and semantic criteria alone, as was
done previously (see p. 58-59, above), but also on the methodological
nature of the sentences in which those terms occur. 16 To distinguish
these two meanings, and for lack of better terms, we shall use the term
fundamental meaning to denote only that narrower, methodological sense,
and the term primary meaning, to denote only the broader sense, even
though in the reistic system the terms fundamental meaning and primary
meaning are used interchangeably.
16Th .. last. issu .. is c .. rt.ainly t.h .. most. cont.rov .. rsial of all. It. do .. s s .... m
how .. v .. r, t.hat. making t.h .. s .. mant.ic 1'01 .. of t.h .. t. .. rms 'is' and ' .. xist.s'
d .. p .. nd on t.h .. met.hodological prop .. rt.i .. s of t.he st.at.ement.s in which t.hose
I. .. rms are used is fairly nal.ural. Moreov .. r, I.his is not. specifically
linked 1.0 t.he reisl.ic st.andpoinl.. For insl.once, t.his ossumpl.ion is
adopt.ed by ·Ajdukiewicz (1961>. II. is also in harmony wit.h t.he I.endencies
discussed by Cornap (1960) and wil.h Reichenbach's views as fOl'mulat.ed in
<1936, Sec. 24. et. passilll).
Having made these explanations we can revert to the issue which has
become the starting point of the present considerations. The question
was to satisfy two requirements which appear to be discordant. One of
them calls for agreement with the guiding ideas of ontological reism,
and the other, for avoiding conflicts with statements which are accepted
in science. Now it seems that, contrary to appearances, a solution can
be found, at the price, however, of a number of radical changes in the
present form of the semantic doctrine of the reistic system. Such
changes would result in a far-reaching liberalization of the criteria of
meaningfulness so far adopted.
The basic method would be to extend the concept of name to cover
both those expressions which the reist accepts as genuine names, i.e.,
"names of things", and those which they class as apparent names (onoma-
toids); to make the primary understanding of the terms 'is' and 'exists'
depend, as a necessary condition, on their occurence together with
names, and the fundamental understanding of these terms, on their occur-
rence with names of things; to make the meaningfulness of statements
about abstract entities independent of their translatability into
reistic language.
This can be done in two ways. They differ from one another above
all by the fact that in one of them the same syntactic functions - us-
able both as the subject and as the predicative word in any sentence of
the type 'A is B' - are ascribed to all names, and hence all names are
classed in the same syntactic category: whereas in the other, names are
treated as expressions in the various syntactic categories, since it is
held that every name is usable as the subject or predicate only in such
a sentence of the type 'A is B' as satisfies certain additional condi-
tions, which are different for different kinds of names. The variety of
these conditions accounts for the variety of categories of names.
Without going into details, we shall briefly describe each of these
methods separately.
The multicategorial concept of names was developed by Ajdukiewicz
(by referring to the suggestions of Aristotle and Johnson) for use in
the controversy over the universals. 16 His point, however, was not con-
sistently to carry into effect the requirements of the reistic system,
but, on the contrary, to demonstrate that it is possible consistently to
defend the realistic standpoint against reistic cri ticisRl. This idea
came close to Russell's theory of logical types. In accordance with the
assumptions adopted here we can single out among names an infinite
number of separate syntactic categories which form a certain hierarchy:
the lowest category - that of the zero order - includes names which are
usable only as subjects in sentences of the type 'A is B' (names of in-
dividuals); names of the first order are those which are usable as pre-
dicates in those sentences of the type 'A is B' which have names of in-
dividuals as their subjects; names of the second order are those which
are usable as predicates in those sentences of the type 'A is B' which
have names of the first ord~'h as their subjects; etc. The general
principle is : names of the k order (where k is a positive integer)
16Cf , Ajdukiewicz (1935),

are those which are usable as prediq,aJes in those sentences of the type
'A is B' which have names of the k-l order as their subjects. The mul-
ticategoriality of names implies, obviously, the multicategoriality, and
thus the systematic polysemy, of all functors of name (term) arguments.
For instance, the copula 'is' may have the category of a sentence-
forming functor of two name arguments of zero and first order, respect-
ively; of first and second order, respectively; of second and third
order, respectively; etc. To use Ajdukiewicz's notation, its categories
could be symbolized thus:
z z z
no m m m m n3
. " ., etc •
A language which has such a syntax (let it be called language L) is

considered by Ajdukiewicz, apparently with justice, to be an admissible
interpretation of everyday language. It is, as we can see, a multilevel
language, consisting of a number of partial languages which differ from
one another by the category of names and the category of functors of
name arguments. Moreover,. each such language includes names of two
kinds: those of the k-l order,which are usable in that language as
s~Rjects in singular sentences (and only as subjects), and names of the
k order, which are usable as predicates (and only as predicates) in
such sentences. Each such language also includes name-argument functors
of the appropriate categories. The system of the calculus of names
(terms), on which such a language is based, is also multilevel. Under
additional assumptions, each of its partial systems is constructed
according to the pattern of Lesniewski's ontology with the salle shape of
functors and the salle structure of axioms, definitions, and theorems.
Hence sentences of the type 'A is B', 'A exists', 'A is an object',
etc., occur at all levels of language L; they have at all levels the
same structural conditions of truth, but at each level they have a dif-
ferent meaning. Let us suppose now, that the lowest-level language
satisfies the conditions which, in the reistic system, are imposed upon
the language about things, whereas the remaining partial languages are
not about things, but about abstract entities. The copula 'is' therefore
occurs at all levels in its primary meaning, but it occurs in its funda-
lIIental meaning only in the language about things, that is, in the
lowest-level language. 17
Now it is obvious that given such assuIIPtions neither the meaning-
fulness not the truth of statements which have the structure of elemen-
tary sentences with abstract names as their subjects depends on the
translatability of those statements into a language about things. It is
also obvious that the acceptance of 'statements about abstract entities
is now not at variance with the principal guidelines of the reistic sys-
17Cf . foot.not.e 16 above. To ovoid misunderst.andings it. should be said

t.hat. t.he primary meaning of t.he copula 'is' is int.erpret.ed here as in
Sec. 3, formulat.ion en, above, but. wit.h t.he underst.anding t.hat. t.he t.erm
'nome' is int.erpret.ed more broadlY.
tern, since none of these statements assumes the existence of abstract

entities in a fundamental interpretation of the verb 'exist'.
The situation is the same in this respect with regard to the second
of the concepts of name previously mentioned, that is, the unicategorial
approaeh. We are then dealing with a single-level language, in which
'is' and 'exists' (like all other functors of name arguments), when used
in t.he primary role, occur in the same meaning regardless of whether
their arguments a1'e names of concrete objects or names of abstract enti-
ties. In that language we can therefore truly assert, given a primary
understanding of the verb 'exists', both that such and such individuals
exist, and that the class of those individuals, the class of classes of
those individuals, etc., exist. On the other hand, given a fundamental
interpretation, as singled out in the by now familiar way (see pp. 61-62
above), we can truly predicate existence about things alone ..
The final result is, of course, the same as before. The reists can
accept the existence of abstract entities of the most diverse kinds
without running the risk of being at variance with their own doctrine,
regardless of whether given existential statements can be interpreted as
statements about things. This eliminates their conflict with mathe-
matics.
It could be argued, however, that this approach makes the reistic
doctrine so closely resemble the opposite standpoint that the difference
between the two is almost completely blurred. Now, this would in fact be
true if the opposition between those two standpoints were reduced to an
opposition of theses. 1S However, in addition to their theses, the reists
also advance a programme. They demand that despite partial failures we
should not stop searching for methods that will make it possible, on an
increasing scale, to reduce statements about abstract entities to those
about concrete objects, and to strive stubbornly for the elimination of
apparent names. 19 This programme has not lost its validity and continues
to be a specific trait of the reistic doctrine.
The point is, however, that to substantiate this programme one can
no longer refer to the argument that has been used so far as the princi-
palone. It is not poss ible to claim that the meaningfulness of state-
ments about abstract entities is ensured only if we are able to trans-
late them into a language about concrete objects. The other arguments,
however, retain their validity. They emphasize above all (a) the agree-
ment of the reistic interpretation with ordinary linguistic usage; when-
ever, in current speech, we use abstract terms we only seem to speak
18Kotar bi nski (1958) eve n s hi ft s hi s mai n emp ha si s f I'om t. heorem s t.o a

programme. He maintains ~hat the reistic doctrine in iLs earlier stages
vas hypothetical in naLure and concludes that. "in its mature form oon-
cret.ism absolut.ely insist.s on it.s programme only." <p. 4-35) This is cer-
tainly a much mOI'e cautious formulat.ion, and far less open to object.ions
and cr-it.icism t.han t.he previous one. It. do~s not., however, reject. t.he
previous concept. of meaningfulness (see t.he paper under considerat.ion ffl
(-ine) and t.hus does not. remove t.he essent.ial obstQcle~ I.e .. t.he con-
flict.. wit.h mat.hemat.ics and t.he exact. sciences.
19Ko t.arbi';ski (1958).

about abstract entities, and in fact our genuine intentions are to speak
about things; (b) the advantages of such an interpretation in philoso-
phical reflections: problems which have been the subject of endless
sterile discussions often vanish immediately as being ill-posed if we
stick to the principle that in our ultimate formulation we should use no
names which are not names of concrete objects; (c) the psychological
naturalness of the reistic guideline: we grasp the intuitive sense of
statements about abstract entities only when we are able to translate
them into statements about concrete objects. 20
The strongest emphasis is usually placed on this last argument.
This stress largely refers to the sometimes glaringly non-intuitive
nature of the concepts of set theory and af the disciplines derived from
it. There is nothing strange in this fact, for it would be difficult to
treat as intuitive such concepts as that of existence, for which in at
least some cases existence depends exclusively on the syntactic proper-
ties of language;21 or that of the distributive interpretation of sets,
since in this interpretation existence is ascribed to sets precisely
because of the somewhat peculiar criteria of applicability of that term;
or the concept of reality, which identifies reality with the totality of
not only all concrete objects, but all abstract objects as well (the
latter being sometimes termed ideal or "non-real"), Le., sets; or the
concept of semantics as that sphere of research which, while intended to
link language with extra-linguistic reality, links it, inter alia, with
that of "non-real reality" which consists of sets of all possible kinds,
including - let us note - sets of non-existent objects. The non-intui-
tive character of set theory is well known, being cited even by those
who cannot be classed as supporters of the reistic approach. 22
It would seem that the arguments to which reists refer are a suf-
ficient justification of their prograa.e. Even if it were to turn out
that this programme can be implemented only on a limited scale, it would
be reasonable to put it into effect wherever possible. To turn something
which is less comprehensible into something more comprehensible and to
eliminate at least some clashes with intuition is always an intellec-
tually prof!table undertaking.
In conclusion we might say that unlike Berkeley, who once recom-
mended that we should think like scholars and talk like the common
people,23 the reists strive to talk like scholars and to think like the
common people, whereas the realists want to think like scholars and to
talk like scholars. The remaining possibility is to think like the
com~n people and to talk like the common people. To complete the
lOCf. t.h .. t.vo pap .. r. on concr .. t.i •• in t.h .. Suppl ..... nt. t.o (1966).
211f a languag .. I? a •• u . . . . . . et. t.heory, and if an expre •• ion At i . a pr .. di-

cat.e in I? , t.h .. n a'. ve 'knov, ve .. ay .. ight.ly conclud .. in I? t.hat. t.h .. r .. i .
an infinit. .. hi .... a .. chy of ... t..: t.h . . . . . t. of At'., t.h . . . . . t. of t.h . . . . . t.. of'
t.he ... t.. of At'., et.c., cut -Ut(~.
22cr. Moet.oveki <1966, p. 7) and Mehlbe .. g <1963, pp. 79 and 8~).
23Cf . G. Be .. k .. l .. y, A T .. eat.i.e conc .... ning t.h .. Principle. of Human Knovl-

.. dge, Dublin 1710.
picture it is worth adding that the last formulation corresponds almost

exactly to the trends represented by Wittgenstein in the last period of
his activity and to the standpoint of the Oxford School.
ON THE DRAMATIC STAGE IN TI£ DEVELOPMENT OF KOTARBINSKI'S
PANSOMATISM
Czesiaw Lejewski
To Tadeusz Kotarbinski, the philosopher, on his ninetieth

birl.hday from a grateful, respectiful, and faithful but.
not. uncrit.ical follower.
Kotarbinski's philosophical doctrine, which originally he chose to call

reism, developed over a number of years beginning with a couple of
essays published in the early twenties. In a more mature form reism is
embedded in his opus magnum, by which I mean his Elementy published, for
the first time, in 1929. The principal tenets of reism at that stage in
its development have been stated by Kotarbinski himself with the aid of
the following three propositions:
(1) All objects are things
(2) No object is a property or relation or event or any other of the

alleged objects belonging allegedly to an ontological category other
than the category of things
(3) The terms • property', • relation', • event', and any other would-be
names of alleged objects belonging to an ontological category other
than the category of things are pseudo-names or onoDlatoids (see
Kotarbinski, 1930-31 reprinted in 1958a).
It is not claimed that this statement of reism is self-explanatory. In

order fully to understand its meaning one has to turn to the discussions
and arguments contained in the Elementy. And, indeed, for many of
Kotarbillski's early students this proved to be sufficient. They could
not find any inconsistencies in the doctrine. Nor were they left in any
doubt as to what Kotarbinski intended his doctrine to be. However, this
sympathetic attitude to reism was not universal. In 1930 Kazimierz
Ajdukiewicz, a colleague of Kotarbinski I published a rev iew of the
Elementy - see (1930) and also (1935). He welcomed the monograph as a
major contribution to Polish philosophical literature, but subjected
reism to a thorough and ruthless examination. This was the beginning of
69
I. Wolenski (ed.), Kotarbiriski: Logic, Semantics and Ontology, 69-83.
70 CZEstAW LEJEWSKI
the dramatic stage in the development of reism. Kotarbinski felt that

Ajdukiewicz's strictures were, for the most part; irrefutable. He found
it necessary to re-interpretthe reistic doctrine and reformulate some
of the arguments he had used in explaining and supporting it. However,
before we consider the details of Kotarbinski' s retrenchment, let IDe
state, briefly, the most salient points in Ajdukiewicz's criticism.
In the first place Ajdukiewicz complains that on the basis of the
Elementy one is not in a position to say with certainty what it is that
reislD asserts. On the one hand reism appears to be a semantical doctrine
concerned with language; on the other hand it appears to be an ontology
concerned with reality. As a milder, semantical doctrine, reism, in
Ajdukiewicz's view, is still open to various interpretations. It IDay be
understood as urging that (i) a noun-expression which does not purport
to designate a thing, should be held to be a pseudo-name or an
onomatoid. This, according to Ajdukiewicz, follows frOID arbitrary
definitions, and, being tautologous, can hardly be regarded as the
intended interpretation of reism. Alternatively, reism may be understood
as holding that (Ji) in ordinary language the copula 'is' (or 'is a') is
used in one sense whenever it occurs in concatenation with noun-expres-
sions which purport to designate things, and in another, quite different
sense whenever it occurs in concatenation with noun-expressions which do
not designate things. An empirical thesis of this sort, points out
Ajdukiewicz, would have to be supported by laboriously compiled
statistics. He goes on to suggest that a more appropriate interpretation
of semantical reism amounts to maintaining that (iii) noun-expressions
which designate things, form a closed semantical category. This means
that by replacing a noun-expression which designates a ;thing, by one
which purports to- designate something that is not a thing, we destroy
the syntactical cohesion,· and consequently the meaningfulness, of a
meaningful context. Now, we have learnt from a number of antinomies that
some sort of classification of expressions into different semantical
categories is essential under the penalty of contradiction, but no such
serious consequences seem to follow from assigning to one and the same
sellantical category such noun-expressions as I table' , I death.' , and
'pain'. Finally, Ajdukiewicz considers a fourth interpretation of
semantical reism. In accordance with it, (iv) semantical reism is ·taken
to be a programme which stipulates that onomatoids should not be used at
all unless the propositions in which they occur, can be adequately
paraphrased with the aid of propositions free of such pseudo-names. In
the language envisaged by this programme, noun-expressions which
designate things should form a closed semantical category. This is an
arbitrary stipulation, but ,within the framework of a language which
complies with it, a number of philosophical problems can,. concedes
Ajdukiewicz, be exposed as pseudo-problems.
The adoption of the reistic language, which appears to be enjoined
by semantical reism, entails, so Ajdukiewicz tells us, unwelcome conse-
quences for reism in its more severe version, that is to say for reism
conceived as ontology. The principal positive thesis of ontological
reism states that every object is a thing, which means that for some a,
a is an object, and that for all a, if a is an object then a IS a thing.
For Ajdukiewicz this is a truism. Clearly, he insists, the proposition
THE DRAMATIC STAGE IN THE DEVELOPMENT OF KOTARBINSKI'S PANSOMATISM 71
to the effect that for some a, a is an object, is a truism; and so is

the proposition which says that for all a, if a is an object then a is a
thing since, in the reistic language, only noun-expressions which
purport to designate things, can be substituted for the variable 'a'.
Whilst in Ajdukiewicz's opinion the positive thesis of ontological
reism turns out to be a truism, the fate of its negative theses such as
(4) no object is a property
(5) no object is a relation
is even worse. In the reistic language one can meaningfully and truly
say that
(6) no object is a centaur
One can also meaningfully but falsely Bay that

(7) no object is a star
But the noun-expressions 'centaur' and 'star', which purport to

designate things, cannot be replaced in (6) and (7) by the word
'property' or by the word 'relation' to yield (4) or (5) because neither
of these two words purports to designate things. They do not belong to
the semantical category to which the noun-expressions 'centaur' and
'star' belong. If (6) and (7) are syntactically coherent, as of course
they are, (4) and (5) must be declared to be syntactical misfits. It is
assumed in this argument that the expression 'no object is a' is used in
(4) to (7) in one and the same sense.
If, on the other hand, the words 'property' and 'relation' in (4)
and (5) are treated as onomatoids and if, consequently, (4) and (5) are
paraphrased to read
(8) it is not the case that for ~ome F and a, Fa

(9) it is not the case that for some B, a and b, B(a b),
and there does not seem to be any other plausible way of paraphrasing
them, then we must reject them as false because their respective para-
phrases are obviously so. We are thus left with a very strange result:
the positive thesis of ontological reism can be stated within the frame-
work of the reist's language, but it turns out to be trivially true
whereas the more interesting theses of the doctrine, namely the ones
which are meant to deny the existence of abstract entities, cannot even
be stated within the framework of the language to which the reist
appears to have committed himself. What the reist can still do, suggests
Ajdukiewicz, iB to replace, say (4), by the assertion which says that
(10) it is not the case that in the reistic language an utterance of the
form 'a is a property' is a true proposition
72 CZEStAW LEJEWSKI
In fact, on the strength of Ajdukiewicz' s argument, (4) should be

replaced by
(11) it is not the case that in the reistic language an utterance of the
form '8 is a property' is a meaningful sequence of words
Analogous assertions can be made in connection with utterances

which allegedly deny the existence of any other kind of entity. Re-
interpreted in this way the negative theses of ontological reism fail to
say anything about reali ty and turn out to be statements about the
language of the reist. Thus, ontological reism collapses into semantical
reism except for the thesis which urges that all objects are things. But
this latter thesis is, according to Ajdukiewicz, a truism.
In the semantical version of reism conceived as a programme for re-
forming ordinary language and making it more precise Ajdukiewicz finds a
number of sound ideas. He himself holds that noun-expressions of ordi-
nary language do not fall under one comprehensive semantical category,
but he does not accept the view implicit in Kotarbinski' s reisll that
there is a semantical category to which belong all and only noun-expres-
sions which designate or purport to designate things. Nor does he accept
the postulate which demands that one should distinguish a semantical
category formed by certain noun-expressions of ordinary language, and
reduce all propositions in which other noun-expression occur, to prop-
ositions with noun-expressions belonging to the sellantical category that
has been distinguished. He does not think that the classification of
noun-expressions into semantical categories as presupposed by Kotarbin-
ski's reism is foreshadowed in ordinary language or that any alternative
classification would necessarily be at fault.
Ajdukiewicz continued his critical examination of reism in his
(1935). In this important paper, which unfortunately is not yet avail-
able in a translation from Polish, he was concerned with Kotarbinski's
proof that there are no such objects as universals. The idea of the
proof is due to Lesniewski. It presupposes the following definition of
the term 'universal':
(12) for all 8 and b, a is a universal with respect to a noun-expression
b if and only if a is an object which has all those properties that
are shared by all the designata of b
In addition, the proof presupposes that there are universals with
respect to noun-expressions which designate more than one object. The
two presuppositions lead to a contradiction, and thus, from (12) we can
conclude that there are no universals with respect to a noun-expression
which designates more than one object. Ajdukiewicz found nothing wrong
with the proof. He pointed out, however, that the proof could be faulted
if one adopted a different definition of the term 'universal', but no
alternative definition which would comply with the 'gramllar' of Kotar-
binski's language and which could be put forward as equally appropriate
as (12), was known either to Kotarbinski or to hillself. This does not
lIean, as Ajdukiewicz was out to argue, that the proof could be repro-
duced, with the same cogency, in a language which exhibited a different
'gr8.lllJllar' •
Now, the characteristic feature of Kotarbinski's language, which is

modelled on Lesniewski's symbolic language, is that in it all noun-ex-
pressions form one semantical category. As is well known, the notion of
semantical category (Bedeutungskategorie) , which bears close relation-
ship to the notion of 'part.of speech' in ordinary grammar, was intro-
duced into logical enquiries by Edmund Husserl in connection with his
study of the structure of ordinary discourse. A systematic theory of
semantical categories, as worked out by Lesniewski, applies to a sym-
bolic language. In this language we distinguish the semantical category
of propositions, the semantical category of noun-expressions, and an
unlimited number of various semantical categories of functors. Proposi-
tions and noun-expressions have this property in comRlon: in compound
expressions they can be used only as arguments, never as functors. Now,
functors can be used in concatenation with other expressions, the latter
being used as arguments, or they themselves can be used as arguments in
concatenation with higher level functors. Propositions differ from noun-
expressions in that those of them which are constant expressions and not
just propositional functions, are deemed to be true or false whereas the
noun-expressions that are constant and not just nominal functions,
purport to designate objects outside the language. Some noun-expressions
do in fact designate either only one object or more objects than one;
others only purport to designate something but fail to do so. The
significant thing is that in Lesniewski's symbolic language constant
expressions which can be used only as arguments and which purport to
designate something directly" belong to one semantical category. To put
it in different words, in Lesniewski's language there is only one funda-
mental semantical category, nalllely the category of noun-expressions.
This category comprises also those functional expressions which satisfy
the criterion of substitutivity with respect to noun-expressions: for
expressions which satisfy this criterion are said to belong to the same
semantical category. In the case of a symbolic language the notion of
belonging to the same semantical category can be defined structurally.
With the application to ordinary discourse in view, Ajdukiewicz,
following Husserl, defines it with the aid of an informal equivalence,
which reads thus: .
(13) two expressions belong to the same semantical category if and only
if they can be substituted for each other in any syntactically
coherent context without destroying that context's syntactical
cohesion.
The criterion of substitutivity embedded in (13) is applicable to

expressions of any well constructed symbolic language. Its application
in the field of a natural language raises a few problems, and Ajdukie-
wicz is aware of this. The problem that is most relevant to our
discussion can be presented as follows. Consider the propositions which
say that
(14) Socrates is a philosopher

(15) the lion is a species
74 CZEStAW LEJEWSKI
(16) zero is a number

and try to answer the following questions:
(17) do the noun-expressions 'Socrates', 'the lion', and 'zero' belong

to one and the same semantical category?
that is to say, are they substitutable for one another in (14), (15),
and (16) salva congruitate?
(18) do the noun-expressions 'philosopher', 'species', and 'number'

belong to one and the same semantical category?
(19) do the noun-expressions 'Socrates' and 'philosopher' belong to one
and the same semantical category?
No ready answers to these questions come to one's mind, which means

that ordinary language is not categorially determined with sufficient
clarity. This, in turn, means that ordinary language can be idealized in
various ways, our preferred idealization depending in some cases on our
ontological presuppositions. In Kotarbinski's idealization of ordinary
language all noun-expressions occurring in propositions which are to be
taken in their literal sense and not treated as metaphors, belong to one
semantical category. They can be substituted for one another without
affecting the syntactical cohesion of the context. Thus, Kotarbinski's
answer to questions (17), (18), and (19) would be 'yes', always assuming
that propositions (14) to (16) are in need of no paraphrasing. However,
idealization of this sort is not the only one possible according to
Ajdukiewicz, and this, it seems to me, is the most important thesis of
his paper. It would be entirely in line with Ajdukiewicz's thought to
answer questions (17) and (18) in the negative. There may be some con-
vincing argument for having a language, call it unicategorial, in which
all noun-expressions of ordinary discourse, except those which occur in
a metaphorical context, are accommodated within one fundamental semanti-
cal category. But equally convincing arguments may be offered for accom-
modating noun-expressions of ordinary discourse within a number of dif-
ferent fundamental semantical categories. I shall refer to a language
idealized on these lines as multicategorial language.
Having distinguished various ways in which ordinary language can be
idealized, Ajdukiewicz turns against reism. He points out that in his
view the rules of Kotarbinski's language are not sufficient for the
purpose of deciding whether the proposition that all objects are things,
holds or not. Further idealization with a view to increasing the preci-
sion of expression would have to take place. For to say that a proposi-
tion holds in a certain language amounts to saying that the rules of
that language compel anyone who consistently observes and applies them,
and makes use, if need be, of the data of experience, to give his assent
to the proposition. Strengthening the rules related to asserting or re-
jecting propositions is, for Ajdukiewicz, a matter of convention. And
so, for him, is the reistic principle that all objects are things.
Equally well, however, one could adapt the rules of one's language,
again by convention and without violating the rules of ordinary lan-
guage, so as to make the contradictory of the reistic principle hold.
Thus, for instance, and here I follow Ajdukiewicz' s idea but not
his actual examples, if in one's language the words 'the lion' and 'spe-
cies' as used in (15) belong to a fundamental semantical category which
is not the same as the category to which the words •Socrates , and 'phil-
osopher' belong then by assuming that whatever is a species is a univer-
sal, one can prove the existence of universals from the truth of (15).
Ajdukiewicz concludes his discussion of reism in his (1935) by
saying that two philosophers, one asserting that only things exist and
the other maintaining that there are universals which are not things,
may appear to be using one and the same language and propounding theses
that are incompatible. In fact, having idealized ordinary discourse in
different ways, they are speaking two different languages, which are not
translatable into one another. The theses they propound need not be
viewed as contradictory. It may be more appropriate to describe them as
not comparable. Having committed ourselves to a possible idealization of
ordinary discourse, warns Ajdukiewicz, we should not think that those
who prefer a different idealization, are wrong even if what they say in
their language appears to contradict what we are prepared to assert in
terms of the language idealized in our way.
Ajdukiewicz, looking back after more than a quarter of a century,
was rather pleased with the results of his polemic with Kotarbinski' s
reism and with Lesniewski's nominalism. In the preface to his (1963) he
claims that under the impact of his criticism ontological reism was
weakened by its originator to become a programme for constrQcting one's
language in a certain way.
Kotarbinski's immediate reaction to Ajdukiewicz's analysis and eva-
luation of reism can be described as a retreat rather than counter-
attack. But it was a retreat not without a few rearguard battles. In the
early thirties Kotarbinski - see (1930), (1931a) and, in particular,
(1935) - tried to hold on to some of the positions taken in Eleaenty,
but on the question of the negative theses of ontological reism he
admitted defeat and withdrew to a second line of defence. In (1949) he
began emphasizing the semantical claims of reism at the expense of its
ontological aspirations. Retreat on this front continued during the
fifties and early sixties. In (1961) ontological claims of reism reached
their lowest, but in the late sixties, namely in (1968), one can detect
attempts at recovering at least some of the lost ground. Kotarbinski's
war against the use of onomatoids in what he calls final pronouncements,
has been relentless although at times he appears to have foresaken the
most convincing reason for waging it. But let us now turn to some
details of the controversy, and let us begin by considering the distinc-
tion between the ontological and the semantical versions of reism.
Ajdukiewicz's insistence on distinguishing between the two versions
is not surprising. His way of thinking was greatly influenced by the
philosophy of the Vienna Circle, and, for him, metaphysics including
ontology in the traditional sense had been eliminated through logical
analysis long before this was officially done by Carnap in his (1932)
paper. Thus, reism as ontology was bound to come to grief at the hands
of Ajdukiewicz. He could only accord qualified approval to a philosophi-
cal doctrine which concerned itself with problems of language. It is in-
teresting to note that Ajdukiewicz does not really try to clarify the
76 CZEStAW LEJEWSKI
relationship between the two versions he is so anxious to contrast with

one another. In his view semantical reism is a 'milder' doctrine whereas
ontological reism is 'more severe'. And while he has a few kind words to
say about the semantical version of reisBl, he takes the ontological
version to task for being in part tautologous and in part meaningless
unless reduced to semantics.
Judging by the general tenor of the Elementy it appears that before
the idea of the two versions of reism came to light in Ajdukiewicz' s
review, Kotarbinski regarded reism as a comprehensive doctrine contain-
ing both ontological and semantical theses. In characterizing the
reistic position he would emphasize either the ontological aspect of
reism or its semantical aspect, depending on the context which occasion-
ed such a characterization. In (1929) he describes the reist as
upholding a doctrine to the effect that among the categories of names,
and by names he understands noun-expressions, only the category of the
names of things, i.e. bodies, contains genuine names. All the remaining
categories of names contain, in fact, pseudo-names or onomatoids, and
these occur in expressions which are not to be understood in the literal
sense but are to be treated as metaphors and rephrased accordingly. In
(1931b), which seems to be later than the review, Kotarbinski defines
the reist as a person who so interprets ordinary language as to regard
any apparent name allegedly designating an object belonging to an onto-
logical category other than the category of things as a pseudo-name.
Here we have examples of Kotarbinski's referring to the semantical as-
pect of reisll, but in his (193{)-31), which is a sort of rejoinder to
Ajdukiewicz's review, and in (1931a) the reism of the Elementy is exemp-
lified with the aid of unmistakably ontological principles followed by a
principle of semantical nature. The corner stone of reism, writes Kotar-
binski (1930-31), is the thesis which says that all objects are things.
Similarly in (1935), which is the most outspoken post-review statement
of Kotarbinski' s doctrine, now called by him pansomatiSIl, ontological
theses take precedence over consequential semantical theses. However,
when fourteen years later he returned to the problem of reism, his ap-
proach was somewhat different. In (1949) the emphasis is put on seman-
tics, and reisBl is given the form of the following directive: try to
eliminate, from your final pronouncements, onomatoids in favour of
concrete names or noun-expressions. The new approach is marked by Kotar-
binski 's abandon ing the term 'pansollatism' and introducing the terms
'concretism' and 'concretist' as alternative to the .old 'reism' and
'reist'. Ontological considerations are removed to the background, and
serve as tentative justification of the directive. And the purpose of
concretism is not, so it seems, to obtain a correct description of
reality but rAther to enhance our capability for clear thinking. Thus,
concretism gives expression to Kotarbinski' s tendency to reduce the
claims of reism or pansomatisll to a minimum. In (1961) attempts at ven-
turing into the realm of ontology, although present, are explicitly
resisted, and the argument is confined to persuading the reader, mostly
by examples, that elimination of abstract noun-expressions from proposi-
tions in their final form is possible. Eventually, as Kotarbinski (1968)
puts it, reisBl has evolved from "an ontological doctrine with a semantic
spicing" into "a semantic doctrine with an ontological spicing".
Has this evolution been well advised? In my opInIon the answer is

'no'. A semantical doctrine in Kotarbinski's sense presupposes an onto-
logical doctrine of which it is an expansion. Semantics without ontology
is like a house without foundations. It cqllapses into a set of arbitra-
ry injunctions and prohibitions justified by ad hoc considerations.
Moreover, outside ontology reistic semantics loses its rationale. It is
in doing ontology, that is to say in making ontological statements, that
one should comply with the postulate of concretism. There is not much
point in avoiding abstract noun-expressions in disciplines of lesser ge-
nerali ty. Elimination of onomatoids from final pronouncements is of
paramount importance only if these final pronouncements are meant to be
used in ontological arguments. It is, for instance, difficult to see why
using abstract noun-expressions and, generally, hypostatizing that goes
with it, should be avoided in humanities or in social sciences as long
as pronouncements these disciplines offer, are not employed as evidence
in a debate concerning ontology. It seems to transpire from Kotarbinski
(1968) that he was not unaware of the need for some sort of limitation
as regards the applicability of the postulate of concretism. Reisll and
pansollatism began as an ontological doctrine with a semantical spicing,
to use Kotarbinski's own phraseology, and it should remain so.
Taking heed of the priority of ontology let us now examine Ajdukie-
wicz's contention that the positive principle of ontological reism is a
truism. The principle as Ajdukiewicz understood it says that every
object is a thing, which in more explicit terms can be stated as
follows: .
(20) for some a, a is an object and for all 8, if 8 is an object then 8

is a thing
I explained earlier how Ajdukiewicz had tried to justify his criticism

of the principle. Now, Kotarbinski seem to have tacitly agreed that the
first part of (20), namely the proposition which says that
(21) for some 8, 8 is an object
was a truism, that is to say an obviously or even trivially true asser-
tion. And, indeed, this is so. For the proposition simply amounts to
saying that there are objects. He argued, however, that contrary to
Ajdukiewicz's claim the second part of (20), which is
(22) for all 8, if a is an object then 8 is a thing
is not a tautology but a hypothetical credo, and he showed under what
conditions, which in his view were not fulfilled, it would be correct to
say that (22) followed from definitions.
The issue is a little obscured by the fact that in (20) the con-
stant terms 'object' and 'thing' are not primitive. Moreover, on Kotar-
binski's own admission, the notion of a thing as used in the Elementy is
ambiguous. Pointing out under what conditions (22) would be a tautology
does not show that it is not one, even if the conditions turn out not to
be fulfilled. As far as I can judge, if one is accused of having made a
78 CZEStAW LEJEWSKI
tautologous statement, the best thing to do is to go back to the founda-

tions of one's theory, present it in outline as a deductive system, and
show the place the incriminated statement is assigned within the system.
And this is what I propose to do in what follows.
A. Among the theories presupposed by reism we have

the logic of propositions
the logic of noun-expressions
The latter is based on the following axiom due to Lesniewski
(23) for all a and b, a is a b if and only if

(I) for some c, c is an a
(II) for all c and d, if c is an a and d is an a then c is a d
(III) for all c, if c is an a then c is a b
It is in the sense determined by (23) that the copula 'is' or 'is a' is
used in reistic theses.
B. Among the primitive undefined terms of reism we have

bulky (i.e., extended in space)
lasting (i.e., extended in time)
capable of experience (i.e., res cogitans)
C. Among the theses of reisa asserted as axioms we have

(24) for some a and b, a is a b
(25) for all a and b, if a is a b then a is bulky and lasting
(i.e., res extensa) or a is capable of experience
(26) for all a, if a is capable of experience then a is bulky and
lasting
D. Among the definitions within the field of reism we have

(27) for all a, a is an object if and only if for some b, a is a b
(28) for all a, a is a thingl if and only if for some b, a is a b
(29) for all a, a is a thing2 if and only if a is an object and
(a is bulky and lasting or a is capable of experience)
(30) for all a, a is a thing3, i.e., material thing, if and only if
a is bulky and lasting
E. Among the theses of reisa we have

(31) for some a, a is an object (follows fro. axiom (24) and defi-
nition (27»
(32) for all a, if a is an object then a is a thingl (follows fro.
definition (27) and (28»
(33) for all a, if a is an object then a is a thingz (follows from
axiom (25) and defintion (29»
(34) for all a and b, if a is a b then a is bulky and lasting
(follows from axiom (25) and (26»
(35) for all a, if a is an object then a is bulky and lasting {fol-
lows fro. axioms (25) and (26), and definition (27»
(36) for all a, if a is an object then a is a thing3, i.e.,
material thing (follows from thesis (35) and definition (30»
It is true that (32) follows from defintions but (32) does not mean
the same as (22), which is to be understood as equisignificant with
(36). And this last thesis could not be proved without reference to the
axioms. Even (21), trivial as it is, does not follow from definitions
alone.
Among further theses of the system we have
(37) for all a, a is an object if and only if (a is bulky and lasting or

a is capable of experience) (follows from axiom (25) and definition
(27»
(38) for all a, a is an object if and only if a is bulky and lasting
(follows from axioms (25) and (26), and definition (27»
Within the system outlined above, equivalences (37) and (38) cannot
be regarded as definitions let alone arbitrary definition~. They had to
be proved with the aid of axioms among other things.
It is easy to see that presuppositions (25) and (26) between them
are inferentially equivalent to (34). If we delete (26) from the axiom
system of reism we get a weaker ontological doctrine, which as Kotarbin-
ski was to learn soon after the publication of the Eleaenty, was in fact
propounded by Brentano in the later years of his life. By retaining (26)
or by using (34) as an axiom instead of the two axioms (25) and (26) we
get a stronger ontological doctrine, the one favoured by Kotarbinski. It
is to this stronger version of reism that Kotarbinski refers as somatic
reism or, simply, as pansomatism. And this is why this last term occurs
in the title of the present paper.
The problems raised by Ajdukiewicz in connection with the positive
ontological thesis of pansomatism are not as complex or difficult as the
problems surrounding the negative ones. According to Ajdukiewicz, to put
his criticism in a nutshell, the reist's utterance
(39) no object is a property
is syntactically incoherent if taken in its literal sense, and turns out

to be false if treated as a metaphor and re-interpreted as meaning that
(40) it is not the case that for some F and a, Fa
Kotarbinski (1930-31) agreed that the negative theses of reism were

ambiguous. He distinguished two cases in so far as the word 'property'
can be regarded as a genuine name or as an onoaatoid. In the first case
(39) should be taken in its literal sense, and in this sense it is a
true proposition. In the second case we have two further possibilities.
Taken in its literal sense (39) must be regarded as syntactically inco-
herent. Taken in its lIetaphorical sense it should be understood as
asserting that
(41) no utterance of the form 'a is a b' with a genuine name, whatever
it lIay designate, in the place of 'a' and the word 'property' in
the place of 'b' is a true proposition
80 CZEStAW LEJEWSKI
Now, if under certain conditions (39) is deemed to be syntactically

incoherent then under the same conditions the utterance
(42) some object is a property
is also syntactically incoherent, and (39), as a negative thesis of

reism, should be construed, so Kotarbinski would like us to agree, as a
rejection of (42) on the ground of the latter's lack of syntactical co-
hesion.
To be asked to agree with this interpretation of the negative
theses of reism is a tall order. The sequence of words
(43) every although if
is not syntactically coherent, but in order to say so we do not assert

that
(44) it is not the case that every although if

The idea that the notion of negation covers both negating false proposi-
tions and rejecting syntactically incoherent utterances, appears to have
little to recommend itself. And I do not think that Kotarbinski himself
was very happy with it. In (1968) he writes that originally, that is to
say in the early post-review period, reism claimed "that any declarati~e
sentence which cannot be freed from abstract terms, is a nonsense; now
it is less radical, it does not challenge those who think otherwise".
The first impression of the exchanges between Ajdukiewicz and Ko-
tarbinski is that there was in them a certain amount of talking at cross
purposes. In criticizing reisll Ajdukiewicz takes it for granted that
there is such a thing as reistic language peculiar to the theory. This
tacit assumption may have been suggested by Kotarbinski's talk about the
need of banning the use of onomatoids in final pronouncements. More
likely, however, it originated from Ajdukiewicz's own conception of lan-
guage. For in his view, as can be seen from his papers published in the
early thirties," a language is determined not by its vocabulary and the
rules of syntax alone but also by what he calls its deductive, axioma-
tic, and empirical rules. Now, if this is the case then one can hardly
talk about any difference between a theory and the language in which it
is stated. And since there is such a thing as reistic theory then, on
this view, there must be such a thing as reistic language. Moreover, on
this view, one cannot use one and the same language to propouhd two dif-
ferent theories. For differences of opinions do not consist in asserting
incompatible propositions. They consist in making clabs in terms of
different languages.
Kotarbinski has never challenged Ajdukiewicz's assumption concern-
ing reistic language. He seems to have tacitly accepted it.
As I have already mentioned, Ajdukiewicz's most significant contri-
bution to the discussion of rein is contained in his (1935). In it he
argued that ordinary discourse lends itself to what I call multicatego-
rial idealization. I do not think that he exploited this idea to the
0
full, and" the conclusions he drew from it, seem to be debatable. How-
ever, the notion of a language with more fundamental semantical catego-

ries than one appears to be sound and to the point. But we must reject
Ajdukiewicz's identification of the notion of language with the notion
of theory. If we take the view that vocabulary and syntax, between them,
determine a language, and that given a language the deductive, axioma-
tic, and empirical rules, in Ajdukiewicz' s sense, determine a theory,
then we can talk about different theories formulated in one and the same
language. Compared with a unicategorial language, a multicategorial one
is richer in forms of expression. It is a more sensitive tool, and
enables one to do things with it that cannot be done with the aid of a
unicategorial language. Its main advantage does not lie in the fact that
it is perhaps a more accurate idealization of ordinary language. The
most important thing, from our point of view, is that a multicategorial
language is sufficiently sensitive to be used as a means of expression
of certain ontological views that could not be expressed in terms of a
unicategorial language. Let me explain.
Ontologists who oppose reism are believers in so called abstract
entities. In a sense they can be said to follow Plato, and for this
reason they can conveniently be referred to as Platonists. Now, one can
distinguish two different breeds of Platonists. They may all of them try
to present their ontological views, in the first instance, in terms of
ordinary language, but given a chance of using languages constructed in
accordance with the theory of seman tical categories, some of them would
choose a unicategorial language as entirely sufficient for their purpose
whereas others would find such a language inadequate, and would insist
on using a multicategorial language. For the former, call them unicat-
egorial Platonists, believe that there is only one ontological category
or kind of entity, say objects. Some objects are material things, others
are non-things, properties for instance or relations or numbers. This,
however, is not the view of multicategorial Platonists, who hold that
there are several ontological categories or kinds of entity: there are
objects, in one sense of 'there are', and there are properties in a dif-
ferent sense of 'there are'. The noun-expressions of ordinary language
which purport to designate objects or properties or relations may appear
to belong to one and the same semantical category, but this, in the view
of multicategorial Platonists, is not the case. They belong to different
fundamental semantical categories. Some of them designate objects in one
sense of designating, others designate properties in a different sense
of designating and some others designate relations in yet another sense
of designating. All this, the multicategorial Platonist would claim, can
be expressed with sufficient precision in terms of an appropriate multi-
categorial language.
Now, if the mul ticategorial Platonist's assertions are made in
terms of a multicategorial language, the same language must be used to
negate those assertions, and I can see no reason why the reist should
not be allowed to use it to voice his ontological views.
The distinction between genuine names and onomatoids plays an im-
portant part in the statement of reism, and it may not be out of place
to make a few comments in this connection. For one thing, the distinc-
tion appears to be of little use unless it is related to the context in
which the name under consideration occurs. Consider, for instance, the
82 CZEStAW LEJEWSKI
noun 'pin'. It appears to be a genuine name as it purports to designate

certain objects that are material things. Yet, in the proposition which
says that
(45) John has pins and needles in his leg
the nouns 'pins' and 'needles' have to be treated as onomatoids. On the

other hand the noun 'property' appears to be an onomatoid since it pur-
ports to designate objects that are not material things. Nevertheless in
the proposition which says that
(46) something is a property
it can be, and has been, treated under certain conditions by both Kotar-
binski and Ajdukiewicz as a genuine name. The conditions amounted to
assuming that (46) was meant to be understood in its literal sense. And
there was a further assumption that the noun • property , belonged to the
semantical category to which nouns which purport to designate things,
belong. This latter assumption can be weakened. For it is enough to as-
sume that the noun 'property' belongs to one of the fundamental semanti-
cal categories.
We seem to have reached the stage when we can re-state the princip-
al theses of reism in the light of the preceding disquisitions. They
take the following form:
I. for all a and b, if a is a b then a is bulky and lasting
II. there are no properties, there are no relations, there are no

events, etc.
III. a proposition which appears to imply the existence of any kind of

abstract entity, is either false or is meant to be a metaphor; if
the latter is the case then the proposition can be rephrased, with-
out any loss of relevant context, so as to have no existential im-
plications or so as to imply the existence of material things only
Given the definitions which were discussed earlier, thesis I is in-

ferentially equivalent to the proposition which says that for all a, if
a is an object then a is a material thing. Thus, thesis I contradicts
the claim of the unicategorial Platonists. For he asserts that there are
objects which are not material things.
Thesis II is in fact an open-ended conjunction of propositions
which are meant to contradict possible claims of the lIulticategorial
Platonist. They are to be understood in the light of the aulticategorial
idealization of natural language. And on this assumption the nouns
'property', 'relation', 'event', etc. belong to different fundamental
semantical categories, which in turn determine the semantical category
of the exp~ession 'there are no' in each of the conjuncts.
Some statements are meant to be understood in their literal sense,
others are meant to be treated as metaphors. This sort of distinction
leaves ample room for disagreement and misunderstanding. And, indeed,
the saDIe statement may be taken by sOlie in its literal sense whereas
others may insist that it should be interpreted as a metaphor. In cogni-
zance of this, the reist tries to accomodate both possibilities and he
is particularly concerned with statements which appear to imply the
existence of objects that are not material things, in the case of unica-
tegorial Platonism, or, if one is allowed to make an illicit generaliza-
tion, the existence of properties and other abstract entities in the
case of multicategorial Platonism. Thesis III sets out the reist's views
in this matter.
The three theses embody the most important part of the reist's hy-
pothetical credo. The term 'hypothetical credo' is particularly apposite
to the third thesis. For there is no general proof that 'metaphors' im-
plying the existence of abstract entities can be rephrased in the manner
required by the reist. In fact it is the problem of possible counter-
examples to thesis III that keeps the reist's attention fully occupied.
But I feel I must leave this problem for another occasion.
As I see it, reism is an ontology with semantical ramifications.
Contrary to Ajdukiewicz' s v iew the reist is not committed to any par-
ticular categorially determined language; he has to use, and is at lib-
erty to do so, the language of whoever happens to be his opponent always
provided that it is a categorially determined langul!-ge. Contrary to
Ajdukiewicz's opinion, the positive ontological thesis of reism, i.e.,
thesis I, is not a tautology; it is the denial of a thesis held by
unicategorial Platonism. Contrary to Ajdukiewicz's insistence, the nega-
tive ontological thesis of reisll, i.e., thesis II, consists of proposi-
tions which, in the light of lIulticategorial idealization of ordinary
language, are meaningful and syntactically well constructed. They deny
equally meaningful and syntactically well constructed assertions of
lIulticategorial Platonism.
SEMANTIC REASONS FOR ONTOLOGICAL STATEMENTS:
THE ARGUMENTATION OF A REIST
Marian PrzeI~cki
There are different kinds of logical connection between semantics and

ontology - between semantic and ontological statements and theories. The
extreme case is that of reducing ontology to semantics by providing
semantic paraphrases for ontological theses. This may be Illustrated by
the analysis of the doctrine of transcendental idealism as caried out by
Ajdukiewicz. 1 The version of the idealistic thesis discussed by Ajdukie-
wicz amounts to the claim that "reality is nothing but a correlate of
the transcedental subject". This allegedly ontological thesis is trans-
lated by Ajdukiewicz into an expl ici tly semantic claim which asserts
that any statement of the language of science is true if and only if it
is "dictated by the meaning-rules of that language". So paraphrased, the
idealistic thesis is shown to be easily falsifiable.
In contradistinction to that case, the exposition of the doctrine
of reism as contained in Kotarbinski's work 2 may serve as an example of
another kind of relation between semantics and ontology. An ontological
statement is here given not a semantic paraphrase, but only a semantic
reason. Ascribing a genuinely ontological meaning to the thesis of
reism, the author adduces for it reasons of essentially semantic nature.
It is on this case that I shall concentrate in the present paper. The
problem is a general one. It pertains to the whole class of so-called
reductionistic ontological doctrines, which, as a rule, are being jus-
tified in a similar way. Generally speaking, it is a way that involves
an inference from the nature of language to the nature of the world.
That kind of inference is fairly plausible in the case of 'positive'
ontological statements - statements asserting the existence of some cat-
egories of entities. It is much less so in the case of 'negative' onto-
logical theses - claims denying the existence of certain kinds of
objects. And this is just what the reductionistic ontological doctrines
1S .... Ajdukie"icz (1937). I have di .. cu .... d t.hi. case in my (1982).
Arhe doct.rine "a" first. present..,d in Kot.arbinski (1929); it. "a .. t.hen de-
velop"d an comment..,d upon in some of hi .. lat.er papers, ... g. (1949) and
(1968).
85
J. Wolenski (ed.), Kotarbinski: Logic, Semantics and Ontology, 85-96.
86 MARIAN PRZEt~CKI
amount to. Reducing the whole realm of objects to a few chosen cat-
egories, they reject thereby the existence of any other kinds of enti-
ties. In doing so, they resort to certain facts about our language - to
its characteristic semantic features. Now, what exactly is the rationale
of that kind of argumentation? What does it consist in? We shall try to
analyze the problem, taking as our example the ontological doctrine of
reism.
The ontological thesi~ of reism amounts to saying that only things

exist, or - in an equivalent formulation - that:
(To) Every object is a thing.
Thing here is defined as an object located in tilRe and in space and
having some physical characteristics.
Though it is a statement about the world, the thesis (To) is jus-
tified by certain semantic reasons. The main premise on which it has
been founded is the so-called semantic thesis of reism - a statement
about the language, and not the world. In KotarbiIlski's writings, the
semantic thesis of reism occurs in different, logically non-equinalent,
versions. In what follows I shall concentrate on one of them. Being the
weakest of the existing versions, it is thereby the least controversial
one; but, at the same time, it is strong enough to embody what lRay be
identified as the fundamental reistic idea. The thesis I have in mind
may roughly be rendered as follows:
(Ts) Every statement about entities other than things is only true if it
is reducible to a statement about things.
So formulated, the thesis needs some explanations. A statement is said
to be about given entities if, literally understood, it entails the
existence of such entities. A statement which, in the sense just ex-
plained, is about things only will be called a reistic statement, for
short; that which is about entities other than things - a non-reistic
one. The concept of reducibility which occurs in (Ts) has a somewhat
specific meaning. A sentence S1 is said to be reducible to a sentence S2
if the proper sense of S1 is non-literal (e.g. metaphorical) and ident-
ical with the literal sense of S2. Reducibility so understood is based
on the concept of paraphrase rather than translation. Let us illustrate
the notion taking as an example a non-reistic statement of the simplest
kind:
(S) Whiteness is an attribute of snow.
Literally understood, it entails the existence of whiteness - a prop-
erty, not a thing. But, according to the reist, it is not to be taken
Ii terally. Its proper sense is a non-literal one; it is assumed to be
the same as the literal sense of the reistic statement:
(Sr) Snow is white.
SEMANTIC REASONS FOR ONTOLOGICAL STATEMENTS 87
The latter does not entail the existence of objects other than things;
it assumes the existence of snow and white things, but not that of
whiteness.
The formulation of the reistic thesis quoted is not what might be
called its 'official' version. This corresponds rather to the following
tenet.:
(Ts') Every statement about entities other than things is only meaning-
ful if it is reducible to a statement about things.
According to (Ts'), reducibility to a reistic statement is meant as a

necessary condition of the meaningfulness of the given non-reistic
statement, and not of its truth only. Formally speaking, it is thus a
formulation logically stronger than the former: (Ts') entails (Ts), but
not conversely. But which relation actually holds between these two
theses depends on some additional factors: on the logical structure of
the language in question and on the kind of rules which are to guarantee
the reducibility postulated by (Ts). The last point proves decisive,
because the rules might well be such that, guaranteeing the reducibility
of all true non-reistic statements, they guarantee thereby the reducibi-
lity of all false "Ones -' .and so, in effect, of all meaningful non-
reistic statements. This is what normally takes place, and the examples
of rules cited by the reist seem to co~firm this suggestion with respect
to reducibility postulated by his theses. The rules provide reistic
paraphrases not only for concretenon-reistic sentences, but also for
sentence-schemata. Rules of the. latter kind tell us how to paraphrase
any statement that falls under the given schema - independently of its
truth or falsity. We are thus given, besides the reistic paraphrase of
the statement: 'Whiteness is an attribute of snow', a reistic paraphrase
of its general schema:
X-ness is an attribute of Y,
which is claimed to be reducible to the reistic schema:
Yis X.
In view, however, of the fact that no complete list of such reduction-
rules has ever been supplied, we have to treat the formulations (Ts) and
(Ts') as two different versions of the semantic thesis. Our analysis
will be restricted to the weaker of them, for what it requires seems
quite sufficient as far as the reist's main objective is concerned.
The formulation of the thesis (Ts) given above is incomplete. Being
of a semantic nature, thesis (Ts) is to be understood as referring to a
definite language L (or to a definite class of languages). Now, what
language does the .thesis actually speak about? The language that Kotar-
binski explicitly refers to i.s, as a matter of course, the Polish lan-
guage. But the peculiarities of that ethnic language are not essential
to the problem in question. What seems'essential about the language re-
ferred to in thesis (Ts) is the fact that it is a language of our know-
ledge. This concept plays an important role in the reistic doctrine. The
body of our knowledge, identified with a given set of statements K, con-
stitutes one of the main data of the reistic construction. And it is to
this datum that the semantic thesis of reism refers. The language that
88 MARIAN PRZEt~CKI
thesis (Ts) speaks about may just be identified with the language of our
knowledge - in particular, of science. The reducibility which the thesis
postulates is meant to hold for any language L in which the body of our
knowledge K is expressible.
Now, what kind of justification is being provided for the semantic
thesis of reism so interpreted? In his later reflection on this point
Kotarbinski (1958, p. 434) states that "the principal foundation of
reism is naively intuitive and based on common induction". And he is
quite right in this qualification. The semantic thesis (Ts) has the form
of a general statement, which is being justified by resorting to its
particular instances. The premises of this argumentation are thus sen-
tences which assert the reducibility of particular non-reistic state-
ments to reistic ones. Given the explication of the concept of reducibi-
lity assumed by us above, what any such premise claims may be rendered
as follows: The proper sense of a given statement S, which apparently
refers to objects other than things, is non-literal and identical with
the literal sense of a statement Sr, which is about things only. Besides
premises referring to concrete non-reistic statements, use is made of
premises which refer to the schemata of such statements. It is clear
that those instances of the semantic thesis which the author actually
resorts to do not exhaust the class of all its instances. This is so in
spite of the fact that the premises referring to sentence-schemata deal
with the whole classes of statements, since the schemata actually taken
into account are far froll exhausting all possible types of non-reistic
statements. And so, the semantic thesis of reism is a kind of inductive
generalization, which goes far beyond the scope of its verified
instances.
The methodological character of this thesis depends ultimately upon
the methodological status of its premises. The last point deserves a few
words of explanation. The premises of the semantic thesis may generally
be characterized as statements about synonymy of certain sentences of
language L. They fall thus under the following schema:
Sentence Sl is synonymous with sentence 82 in language L.
Now, what is the methodological status of such statements? Within logi-

cal semantics they are usually regarded as analytic sentences. The argu-
mentation runs somewhat as follows. Sentence 81 and S2 are synonymous if
their biconditional: Sl if and only if S2, is analytic, that is, true in
virtue of the rules of language L alone. And a sentence of metalanguage
ML that ascribes analyticity to some sentence of language L is itself
regarded as analytic in metalanguage ML. This conclusion, however,
hinges on the characteristics of metalanguage ML - in particular, on the
way in which language L is defined in ML. Does L's definition include a
specification of its (syntactic and semantic) rules, or not? If it does,
an ascertainment of the synonymy of two sentences of language L bears
the character of an analytic metastatement. But if language L is identi-
fied in some different way, such an ascertainment need not be analytic.
And this is just what takes place in the case under consideration. As
pointed above, the language L referred to in the semantic thesis is
defined by means of descriptions such as: the Polish language (the lan-
guage spoken by the Polish people), or the language of our knowledge

(the language in which our knowledge has been formulated), and the like.
A reconstruction of the rules of a language so characterized is a matter
of experience; experience only can tell us what rules hold in language L
identified by means of such descriptions. In consequence, our premises,
which state synonymy of certain expressions of language L, acquire the
status of empirical propositions. Another situation'would obtain only if
the language L were to be a language which we do not find, but create -
by conventional acceptance of definite (syntactic and semantic) rules.
But this is not what the reist has in mind. He wants his thesis to hold
for the existing languages - in particular, for the actual language of
science. And with regard to such languages, the premises he employs
have, as we have seen, the status of empirical statements. The same
status, then, will characterize the semantic thesis of reism, which is
put forward as an inductive generalization of those premises.
Soon after its presentation the doctrine of reism was subjected to
penetrating cri ticism3 , which has shown the necessity of introducing
certain modifications into the original formulation of the semantic the-
sis. Identifying the language L referred to in the thesis as the Polish
language, or the language of knowledge, one does not determine it
uniquely. Ajdukiewicz (1935, pp. 104, 106) has called attention to the
fact that natural language - and, in consequence, the language of our
knowledge - is notoriously indefinite: syntactically and semantically
indeterminate. It admits thus of different interpretations which deter-
mine its syntactic and semantic features in different ways. "When they
make expressions more precise for their purposes, philosophers enter -
without violating ordinary language - one of the lanes left open by that
language". And so does, according to Ajdukiewicz, the reist. "To make
his thesis unambiguous it is necessary to specify the language in which
this thesis is to hold. If it is claimed that the thesis holds in every
language obtainable as a more precise reconstruction of ordinary lan-
guage, then this claim seems to me false." The truth of the semantic
thesis (Ts) is thus assumed to depend on which particular interpretation
has been chosen for the language L. If the thesis is claimed to be true,
it is claimed to be so with respect to some admissible interpretation of
language L. Its full formulation should thus contain such qualification.
Reference to language L should be replaced in it by reference to some
admissible interpretation of language L. The final version of the seman-
tic thesis of reism may thus be rendered as follows:
(T:) Every statement about entities other than things is only true if it
is reducible to a statement about things, on some admissible inter-
pretation of language L.
If the intuitive sense of the original version of the thesis amounts to

saying: "We speak, in fact, about things only", the sense of its modi-
fied version may be conveyed by the statement: "We can speak so as to
speak about things only."
3E . g . Ajdukieyicz (1930) & (1935).

90 MARIAN PRZEttCKI
Does the above modification of the semantic thesis affect its

methodological status? Ajdukiewicz's criticism suggests that it does. As
long as language L has not been precisely specified, the reistic thesis
is said to remain essentially undecidable. "Reisll then can, at best, be
a matter of convention." Such a conclusion, however,. does not apply to
the semantic thesis of reism, taken in its version (Ts) given above. The
thesis so formulated seems to retain the status of an empirical general-
ization, which characterized its former version. I t is true that the
premises which it is based upon differ from those quoted before. They
fall under a schema weaker than the former:
Sentence 81, in some of the meanings it has in language L, is syn-
onymous with sentenceS2.
But, so modi fied, the premises continue to be empiric'1l statements;
though weaker than the original premises, they remain empirically
decidable. The same character then must be ascribed to the reistic
thesis in its present version.
The concept of an admifsible interpretation of language L, which
occurs in the formulation (Ts), lIay be characterized more explicitly by
stating a condition which such an interpretation is bound to satisfy. In
any indeterminate language L the truth value of its sentence is, in gen-
eral, dependent on which particular meanings are assigned to L's ambigu-
ous expressions. At the same time, however, in any such language there
are sentences which are invariant in that respect: their truth value
turns out to be independent of how one disambiguates L's expressions.
Now, an interpretation of language L is admissible only if it preserves
the truth values of that class of sentences. A sentence which, whatever
its meaning in L, proves to be true (or, respectively, false) must re-
main such under any admissible interpretation of language L.
The kind of justification of the reistic thesis .outlined above is
not the only one that has been adduced by its author. In his writings
(Kotarbinski, 1949) we find attempts to justify the semantic thesis of
reisll by resorting to its ontological tenet. "The reist' will risk a
thesis which [ ••• ] is of an ontological nature. That thesis states that
every object is a thing [ ••• ], and that this justifies everything which
we have tried to explain above" - among them, the semantic thesis. "The
ontological statement serves to strengthen the foundation of the for-
mer." It is evident, however, that the argumentation might be said to
provide some justification of the semantic thesis only if the ontologi-
cal thesis had been justified independently of it. And this clearly is
not the case. As will be seen later, the actual justification procedure
goes in the opposite direction: it is the ontological thesis that is
being founded on the semantic one, and not conversely. This being so,
the inductive justification of the sellanticthesis discussed by us thus
far remains the only fundamental way of its validation. The question
then arises as to how its outcome is to be assessed. Has the reist suc-
ceeded in showing the truth of his thesis? I am not trying to give any
reliable answer to this question in the present paper. Its task is an
assessment of the connections between the two reistic theses and not of
the theses themselves. And so, I shall restrict myself to some loose
comments on the latter issue.
Being a general statement of unrestricted universality, the seman-

tic thesis cannot be proved conclusively by any finite number of its
particular instances. And this is how the reist tries to justify it, by
giving reistic paraphrases of particular non-reistic statements (or
their schemata). His paraphrases are, as far as I can judge, quite con-
vincing: the relevant statements may, indeed, be interpreted in the way
he propounds. It cannot be denied, however, that his achievements mean a
partial success only. This is especially true of that domain of knowl-
edge which has an openly abstract character: viz., mathematics, with its
set theoretical foundations and empirical applications. The attempts at
a reistic paraphrase of mathematical statements are very fragmentary and
somewhat problematic; they amount to nothing more than "a tiny hole made
in the ~uter wall of the mathematical fortress" - to use Kotarbinski' s
(1958, p.434) words. Therefore the actual justification of the semantic
thesis of reism is far from being conclusive; it confers on it, at best,
some plausibility. In fact, however, the situation of the thesis appears
to be much worse: there seem to be reasons that speak decidedly against
its truth. They point to certain counter-instances of the general claim
expressed by the semantic thesis. Let us realize the logical structure
of the the¥is and its counter-instances. Since the thesis in its final
version (Ta) is a statement containing three quantifiers (one universal
and two existential), its counter-instances involve two quantifiers
(both universal), in accord with the following schema:
S is a statement about entities other than things, true and irre-
ducible to any statement about things, on any admissible interpre-
tation of language L.
A claim like this cannot be conclusively verified by any finite number
of concrete examples. A proof of its truth can be obtained only on the
basis of some general theory. No wonder then that such proofs have only
been provided for certain mathematical statements, since in their case
we may resort to a relevant mathematical theory. Take, as an example, a
sentence which says that the number of objects M is finite. In contrast
to such non-reistic statements as those claiming that the number of M's
is equal to n, which are easily translatable into reistic statements,
the statement quoted above is shown to be irreducible to any statement
about things only. This is considered to be a consequence of the well
known mathematical theorem to the effect that the class of all finite
models is not elementarily definable.
There are thus reasons for thinking that the semantic the~is of
reism is false, if taken in its original form and interpretation. Raving
acknowledged this fact, one may contend that the thesis is true, if
taken in some restricted version, and try to formulate explicitly the
necessary restrictions. But there is also another way of guarateeing its
truth: by interpreting the thesis as a def ini tion of the concept of
meaningfulness, which is involved (explicitly or implicitly) in its for-
mulation. The thesis claims the reducibility of all meaningful (or mean-
ingful and true) non-reistic statements to reistic ones. In its original
interpretation it appeals to an existing concept of meaningfulness (a
common sense I or a scientific one). But the thesis may be interpreted
otherwise: as postulating a normative concept of meaningfulness. Any
92 MARIAN PRZEt~CKI
sentence not reducible to a reistic one would then, by the very defini-
tion, be qualified as a meaningless expression. A definition like this
might have been motivated by the ontological thesis of reism - provided
the latter has been accepted without, in turn, resorting to the. semantic
thesis (e.g., as a kind of "hypothetical credo", as the author sometimes
calls it). The meaningfulness referred to in this context is of seman-
tic, or philosophical, rather than purely syntactic nature. Those well
formed non-reistic sentences which have no reistic paraphrases will be
called meaningful in the syntactic, but meaningless in the semantic
sensej as such, they will be devoid of any truth value, taken in its
classical sense. According to this view, the abstract (in particular,
mathematical) part of our knowledge will playa purely instrumental, and
not cognitive role. Cognitive value will be attributed only to the empi-
rical basis of science - here identified with the reistic basis. On this
approach, the function of the abstract superstructure of science will
consist in an adequate forllulation and systematization of the reistic
basis - in accord with a contention that in order to say all there is to
be said about things, one has to speak not just about things.
II
The main object of my analysis is not the problem of truth of the seman-
tic thesis, but its bearing on the problem of truth of the ontological
thesis. Assuming thus, for the sake of argument, the truth of the seman-
tic thesis, I shall discuss the attempts to base the truth of the ontol-
ogical thesis upon the truth of the semantic one. As said before, this
is how the ontological thesis is actually justified by the reist, though
on th'] face of it, the situation might well be judged otherwise. Being
about the world, and not the language, the ontological thesis is clailled
to be justified by reference to "our global knowledge", to "the whole of
experience", etc. On a closer inspection, however, the knowledge the
reist has in mind turns out to be a kind of linguistic knowledge, and
the experiential data he resorts to - certain semantic facts. What they
justify directly is the semantic thesisj the ontological thesis is jus-
tified by them only indirectly. The main argumentation of the reist
amounts to a refutation of some counter-arguments advanced against the
ontological thesis. These counter-arguments point to certain statements
which are commonly regarded as true, and which, literaly understood, as-
sume the existence of objects other than things. The reist's argumenta-
tion aims at showing that the proper sense of these statements is non-
literal and reisticj so understood, they do not entail the existence of
entities other than things and, hence, do not contradict the ontological
thesis. But these are exactly the same arguments which have been em-
ployed in justifying the semantic thesis of reism: they claim the redu-
cibility of certain non-reistic statements to reistic ones. And so, what
they validate directly is the semantic thesis.
The problem then arises how to pass froll it to the ontological
thesis. In order to answer this question, we have to examine the logical
relations that obtain between those two statements. The result depends
on which version of the semantic thesis is taken into account. As is

easily seen, the ontological thesis (To) logically entails the weak
version of the semantic thesis, (T8) - on the assumption that (To) is a
sentence of the same language L which (Ts) referes to (or, in other
words, that L is a language spoken by the reist). If every object is a
thing, a non-reistic statement may only be true on the condition that it
is understood in a non-literal, reistic way; otherwise, it would assert
the existence of objects other than thillgs. The same holds for the
semantic thesis in its modified version (Ts); being weaker than (Ta), it
may be said to follow from (To) It fortiori. On the other hand, the
ontological thesis (To) does not entail logically the semantic thesis in
its strong version (Ts')' It does, of course, entail (T8') under the
additional (not implausible) assumption that any way of guaranteeing the
truth of (T8) guarantees thereby the truth ~f (Ts·).
In the case of the semantic thesis (Ts), which is the main object
of our discussi~n, we have thus to do with a one-way logical connection:
(To) entails (T8), but not conv~rsely. Though this connection does not
allow u~ to deduce {To} from (Ts), it makes it possible to infer (To)
from (Ta) by way of inductive (or so-called reductive) reasoning. In
doing so, we accept (To) a~ an ontological hypothesis that explains the
semantic fact stated by (Ts). This is a way in which we arrive at hypo-
theses in empirical science. Whether the hypotheses arrived at in that
way may be said to have been justified remains a controversial issue.
Even if we agree to call them justified, they are called so in a very
loose sense, indeed. The ontological thesis of reism justified in that
way only must remain, as Kotarbinski (1958, p. 434) himself admits, no
more than a "risky conjecture".
Is it then possible to infer (To) from (T:) in a more reliable way?
The question has, .by some critics, been answered in the negative; the
semantic thesis (T8) has been denied any ontological import. The thesis
assumes that, due to its indeterminacy, the language of our know ledge
admits of different interpretations; it is only some of them that are
claimed to possess a reistic paraphrase. This fact alone can hardly
entail any ontological consequences. This is how Ajdukiewicz (1930, p.
520) views the matter. "The fact that sentences which contain 'pseudo-
names' (i.e. names of objects other than things) can be replaced by sen-
tences which are synonymous with them, but which contain no pseudo-
names, proves nothing but that one can (theoretically at least) dispense
with pseudo-names." This view, however, has been questioned. The corre-
sponding argumentation runs as follows.
In order to deduce the ontological thesis of reism from the seman-
tic one, we have to resort to some additional premises. And there is
such a premise available: it is the principle known as Ockham's razor.
The principle has been regarded as an assum~tion that makes it possible
to pass from semantic statements such as (Ta) to ontological statements
such as (To). Let us analyze the kind of inference involved in that
step. Ockham's principle has the form of a methodological rule:
(Ro) One must not accept the existence of any entities unless it is
necessary.
94 MARIAN PRZEt~CKI
The concept of necessity which the rule (Ro) referes to may be expli-
cated in various ways. One of thea, directly relevant to the present
case, relates it to our knowledge, or - strictly speaking - to the class
of all adllissible ways in which our knowledge can be expressed. As
stated before, the language of our knowledge is susceptible of different
interpretations. These interpretations differ as to their ontological
cOlllli tments; there lIay be categories of objects which are assumed by
sOlie of them, but not by others. This suggests the following explication
of the concept of necessity:
The assullption that there are objects of a given ontological cat-
egory is necessary if and only if it is entailed by every admis-
sible way of expressing our knowledge.
Ockham's principle forbids us to accept any existential statements which
do not fulfil the above condition. But what is the rationale of Ockham's
rule so interpreted? Why are we not allowed to accept such statements?
There are two kinds of possible answer to that question: we must
not accept the statements because they are false, or - because they are
groundless. The answers correspond to two different kinds of postulates
that lIay be said to underlie Ockham's rule: an ontological and a method-
ological one. The ontological postulate may be rendered as follows:
(Po) There are objects of a given ontological category if and only if

the assumption that there are such objects is entailed by every
admissible way of expressing our knowledge.
The methodological postulate concerns not the existence of objects, but

the justification of existential statellents:
(R.) The assumption that there are objects of a given ontological cat-
egory is justified if and only if it is entailed by every admis-
sible way of expressing our knowledge.
So formulated, the postulate {Po} allows for ontological conclusions,

the postulate (P.) for methodological ones. But do the postulates them-
selves sound plausible enough? May they be said to be suficiently justi-
fied? In my opinion, they differ considerably in this respect. The
ontological postulate (Po) seems hfghly problematic. It claims that
there are only such objects as we happen to know of (or, rather, whose
existence is, so to say, enforced upon us by what we actually know).
But, as a matter of fact, why should it be so? Without some further
reasons, such a claim remains quite arbitrary. If, on the face of it,
the postulate (Po) appears plausible, this seems to be due to the fact
that it is not being clearly distinguished from the methodological
postulate (R.). And, in contrast to the former, the latter is fairly
convincing. It seems to be a consequence of how the very concepts of
knowledge and justification are being understood. To be justifiable a
statement must be semantically determinate; and so, its truth value must
be invariant over the class of all admissible ways of expressing our
knowledge. This pertains, among others, to existential statements, and
the postulate (P.) gives expression to this fact.
Now, what is the bearing of the above postulates in the case dis-
cussed in this paper? It is easily' seen that the postulates find
straightforwar~ application to the situation characterized by the seman-
tic thesis (Ts). As stated by it, the reistic way of expressing our
knowledge assumes only the existence of things. On the supposition that
the alternative ways of expressing our knowledge do not make any weaker
assumptions, it is only the assumption that there are things that is en-
tailed by every admissible way of expresing our knowledge and may, on
that account, be said to be necessary. This fact has important conse-
quences on the basis of either of the postulates mentioned. In virtue of
the ontological postulate (Po), it implies what amounts to the ontologi-
cal thesis of reism:
(To) Only things exist.
The postulate (Po) thus provides a necessary bridge which enables us to
pass from the semantic thesis of reism to the ontological one. The prob-
lem is that the bridge seems to rest on rather weak foundations. The
methodological postulate (P.) appears to be much more firmly grounded.
This postulate, however, cannot yield the ontological thesis of reism.
The only conclusion that one is entitled. to draw from the semantic
thesis of reism in virtue of the postulate (P.) is a statement which may
be called the methodological thesis of reism:
(T.) Only things may justifiably be said to exist.
Either of these statements - the ontological as well as the methodologi-
cal thesis of reism - provides a kind of justification for what might be
termed the methodological rule of reism:
(Rr) One must not accept the existence of objects other than things.
Referring to this formula, we can sum up the argument presented above as
follows. What Ockhall's razor, in fact, secures is a step from can to
ought: froll the fact that one can dispense with objects other than
things (as stated by the semantic thesis of reism) it allows us to infer
that one ought to do so (as required by' the methodological rule of
reism). By being feasible, the reistic paraphrase becomes obligatory.
Let me conclude this discussion with some remarks of a more general
nature. The case discussed above is not restricted to the doctrine of
reism. It may be generalized so as to comprise all ontological doctrines
of the reductionistic type. By these I mean doctrines whih may be char-
acterized as the weakest from a given class of doctrines comparable as
to the strength of their existential assumptions. The simplest case of
such doctrines can be schematically described as follows. Let two ontol-
ogical doctrines yield two different ways of expressing our knowledge K:
Kl and K2. Kl and K2 differ as to their ontological commitments: Kl
assumes the existence of objects of category Cl, K2 the existence of
objects of categories Cl and C2. In this-situation, Ockham's rule (Ro)
forbids us to accept the existence of obj~cts other than those of cat-
egory Cl. In view of the postulate (Po), this is to be so because- only
objects of category Cl exist; in view of the postulate (P.), because
only objects of category cl-may justif~ably be said to exist.
96 MARIAN PRZEt~CKI
There is also a way in which the above conception might be said to

apply to doctrines that are incomparable as to the strength of their on-
tological commitments. Let us assume, as before, th~t KI and Kz are two
interpretations of our knowledge K, corresponding to two different on-
tological frameworks. This time, however, KI is assumed to entail the
existence of entities of category CI, and Kz those of category cz, the
classes CI and cz being disjoint. So, neither of these categories sat-
isfies the requirements of Ockham's rule. The only existential assump-
tion that fulfils such requirements is a disjunctive sentence which
states that there are objects of category CI or C2. This is the only
ontological statement that can be said to be entailed by both inter-
pretations of knowledge K. And so, it is the only existential assumption
that escapes Ockham's razor. It has to be admitted that this conclusion
does not sound very convincing. This seems to confirm the view that the
proper application of Ockha~'s rule is restricted to ontological
doctrines of the reductionistic type, exemplified in our considerations
by the doctrine of reism.
PHILOSOPHICAL AND METHODOLOGICAL FOUNDATIONS
OF KOTARBINSKI'S PRAXIOLOGY
Tadeusz Pszczolowski
Tadeusz Kotarbinski, a representative of the famous Lvov-Warsaw School

of Logic was the founder of praxiology: the science of efficient action.
He began to form it in the period of his philosophical studies at the
University of Lvov, before the First World War. He continued to form
praxiology, also known as the logic of action, till the last years of
his life. The fullest exposition of it appeared in his book A Treatise
on Good Work (1955a) which was translated into many languages , and
whose English translation (1965) was entitled Praxiology: An Introduc-
tion to the Science of Efficient Action. Besides this, he wrote many
works on metapraxiology - projecting its contents and expanding the
problem connected with praxiology. He lectured on praxiology at the
University of Warsaw, and my opinions are based not only on my knowledge
of his written, but also on his lectures and on my memories of his sem-
inars and the problems discussed there. The subject of this discussion
is the philosophical and methodological assumptions of the theory of
efficient action.
Implications of reism
Kotarbinski declared himself as a supporter of the branch of materialism

which wa~ professed among others by Franz Brentano at the end of his
life, and which is called reism. Ontological views amount to accepting
the existence of material things and actions of individuals only. So the
agent is an acting individual. Only he acts and it cannot be said of a
human organization that it acts as an agent. 'The organization acts' is
an abbreviation, and means: the organization consists of separate indi-
viduals, each of whom acts. Here is no collective action, but a distri-
butive one.
The .concept of action in the determining definition limits its
scope to men. Maybe in another language we could assume that elephants,
dogs, computers etc. act; but in the language of praxiology, 'action' is
consciously confined to the individual subject. The term 'action' is
also used as an abbreviation, because in fact there is no action but an'
97
J, Wolerlski (ed.). Kotarbirlski: Logic. Semamics and Ontology. 97-105.
98 TADEUSZ PSZCZOtOWSKI
acting agent. Action belongs to the category of events, and they are
either changes or states of things. If we distinguish events considering
what happens to things,' considering variability or invariability of
certain features in a given interval, such distinction will also refer
to actions. Above the theory of action Kotarbinski constructs a theory
of events which is IBOre general than the theory of action, which, in
turn, is more general than praxiology, i.e. the theory of action evalu-
ated from the point of view of efficiency. .
The theory of events examines cause-effect relationships,.function-
al sequences and simultaneity of events; the theory of action is con-
cerned with particular cases of causal connection, that is causation
when the cause is an agent.
In his approach to praxiology Kotarbinski does not expose the as-
sumptions of ontological reason or of semantical reason, because he does
not state explicitly that genuine names are only those which have real
designations, and all the others are pseudo-names. Froll those we must
ei ther pass to genuine names or we must find out what the relation
between thell and concrete thing is. In the lectures on praxiology these
postulates are fulfilled implicitly. For instance, following Eugeniusz
Geblewicz (see: 1983), Kotarbinski adopts the definition of goal: John's
goal is a certain event (c) means, that John has made some effort in
order to bring about the event (c). John represents a concrete agent.
And in reality there is no goal without the agent. 'Goal' is. a pseudo-
name, and the same refers to 'event'.
Similarly in Plato's Dialogues the notion of courage is reduced -
it does not exist, there are only people whom we call courageous, i.e.
we ascribe certain features to them, for instance strength or character.
As a consequence of reisll one must accept that there is only one
ontological category - the category of things. However, we have several
semantic categories - events, features, relations, and many others; lan-
guage expressions belong to the same linguistic category if their ex-
change (replacement of one expression by another) does not turn a sen-
sible statement into a nonsensical one.
Precise ter.inology and adequacy of assertions
Kotarbinski's positivistic and scientific attitude was· shown in his

methodological formulation of praxiology; it was also methodology in
general, embracing all actions. Methodology of sciences including that
of deductive and inductive sciences, and on the lower level methodology
of particular sciences deal only with a certain sort of actions. As
methodology has not only theoretical' tasks - praxiology describes how
people act, and evaluates the efficiency of their actions, but it also
serves practical purposes - it formulates directions which inform one
how, in certain conditions and with assumed goals, to apply the sequence
of actions in order to achieve efficiency, that is, to realize the event
which was adopted as a goal. In this formulation the evaluation of effi-
ciency is the relation of agreement of the effect and the goal, and it
dan be verif.ied or falsified. Other estimations of efficiency also be-
PHILOSOPHICAL AND METHODOLOGICAL FOUNDATIONS OF PRAXIOLOGY 99
long to the category of relations and are defined precisely and with as
great accuracy as other terms in praxiology. It is assumed that the lan-
guag~ which praxiologists speak and write is already established and
there is no need of introductory explanations of what the terms which
are used mean or denote. It is also a realization of the postulate of
economic management. In connection with this, on Kotarbinski's initiat-
ive, work on the dictionary of praxiology started in 1959, and it lasted
for twenty five years (see: Pszczolowski, 1978).
In accordance with Aristotelian tradition and Leon Petrazycki's
postulate, his principle of adequacy of constructed assertions, of ad-
equacy of theory in regard to this fragment of reality which the asser-
tion or a set of assertions concerns, is taken into consideration in
praxiology.
A theory is adequate when its subject - predicate theses comprise
all classes of objects - and only them - not omitting those objects
which possess the feature designated by the predicate, and do not
ascribe such a feature to the objects which do not possess it. Mach's
thesis referring to so-called Denkoekonomie is not adequate, because
economical forms are demanded not only from thinking and, scientific
statements, but also from many other outcomes of human action. However,
the economy of human routine and repeated actions in the case of cre-
ation is sometimes replaced by the postulate of the greatest effort. The
search or adequacy of the theory of effective action brought about cre-
ation of more and more general theories - the theory of action, the
theory of events, and of theories which, in comparison with praxiology
are less and less general - the theory of struggle, the theory of organ-
ization of groups of people.
The role of the above-mentioned requirement of adequacy of science
is seldom emphasized. Striving to exhaust all the problems and to shape
a whole led the philosophers to propose a philosophical system composed
of various parts. An instance of this striving to overall formulation of
human action can be Kotarbinski' s proposal to set praxiology wi thin
broadly understood ethics to which felicitology (the principles of
achieving happiness), moral deontology (the principles of fair, that is
good, conduct) and praxiology (the efficient principles of fair achiev-
ing happiness) would be subordinated.
Only aesthetics was missing in the full range of aspects in which
action could be examined. In course of time, however, praxiologists'
became interested in it too, noticing the well-marked interdependence:
eff,icient actions are also beautiful and vice versa. This interdepend-
ence is prominent, for instance, in sport.
Two set theories
In the case of the formulation of ethics discussed above we have two

possibilities. Ethics is a science of human action consisting of three
components: praxiology; praxiology is included in this conception of
ethics. Or, otherwise, each of the above-mentioned fields of knowledge
is ethics, so praxiology is ethics as well. In his lectures Kotarbinski
specially emphasized the fact that some terms can be treated in two
different ways, according to set theory. Such terms besides ethics are,
for instance, 'philosophy', 'science', 'theory'. Philosophy is a whole
consisting of several branches - metaphysics, epistemology, ethics etc.
But metaphysics, epistemology, ethics are philosophy. Science is a
complex of particular sciences. On the other hand, physics is a science,
biology is a science, and every branch of biology - for instance, micro-
biology - is also a science. It is so because we have two set theories.
The first set theory operates on the notion of a class of M - elements,
a set of M - objects, and then a statement that an object belongs to the
class of M - elements is the statement that it is an M - object. If we
have defined what an instrument is, then the assertion that a given
object is an instrument means that this object is included in the class
of instruments. However, we can never say how many instruments there are
in the world, or imagine the set of instruments. We use 'set' in its
distributive sense - because 'set' is an abstract idea and we have to do
only with the elements of this set. In Stanislaw Lesniewski's mereology
we can find another notion of a set - in its collective sense. There
exist objects consisting of M - elements which constitute a whole M,
also called a set. However, it is a different kind of set. A human or-
ganization can serve here as an example. It consists of sections, which
are also organizations of men. Sections are further divided into depart-
ments, divisions, and these are organizations too. The organization con-
sidered here is then a whole consisting of organizations. To express it
quite clearly we must state that both the organization as a whole, and
its component organizations also belong to a set in a distributive
sense, namely to the set which in this case is human organization.
The reason why I use the predicate 'human' requires additional
explanation. The point is that Kotarbinski formulated a definition of
organization using "variables" like a logical formula. Organization is a
whole whose components contribute to prosperity of the whole. It is a
very general definition and if an agent is substituted for a "compo-
nent", we get a human organization. But these components may also be
bees or ants, and then the organization will be a hive or an ant-hill
respecti vely. Particular parts form a radio set or a car - these are
also certain kinds of organization in this general sense. In Poland
Kotarbinski was a precursor of the general theory of systems, which he
called theory of cOlllplex or theory of compound object. The notion of
organization, as proposed by Kotarbinski, is similar to the notion of
system on the understanding that according to reism an organization must
be a thing. It is so because only things (a cause-effect relationship
exists between material objects) contribute to the occurence of some
positively evaluated states of affairs, that is to prosperity. Since
Kotarbinski proposed a very general ·definition of organization, on the
basis of which concrete objects could be included in a set in its dis-
tributive sense, when discussing organization as a set in its collective
sense he should have added that it is human organization that is prefer-
red to. Nota bene, an organism is also embraced in the notion of organ-
ization, and Kotarbinski in his old age used to joke sadly that more and
more of his components had stopped contributing to the prosperity of his
whole.
Set theory in its distributive sense was created by mathematicians

and logicians who did not limit it by maintaining that the elements of a
set can be material things only. Nor did Lesniewski's set theory assume
the materiality of set, although it was based on ontology; but ontology
embraced among others the calculus of relations and the calculus of
classes.
Kotarbinski based his conceptions of efficiency evaluations refer-
ring to action on an idea supported .by two set theories - efficiency is
a set of evaluations in a collective sense, and efficiency is a set of
evaluations in a distributive sense. Hence two notions of evaluation of
efficiency; in the sense which is synthetic = general = collective, and
in a universal = distributive sense. What are the elements of the set of
evaluations then? I have already mentioned effectiveness. Except effect-
iveness, which does not take expenditure into account we have an alter-
native way of understanding economy, that is either chariness or produc-
tivity, together with economy understood non-alternatively and expressed
in the relation of positive effect to expenditure (input - output).
Kotarbinski formulated and defined several evaluations of efficiency,
e.g. adroitness, precision, neatness, accuracy, certitude, reliability,
boldness, etc. I am not going to enumerate all of them - this would be
sensible only if I were to quote the definitions and discuss them in
detail. Here it will be enough to explain that defining particular
evaluations of efficiency Kotarbinski took into consideration various
aspects - the occurrence of intended and unintended effects, effects
evaluated positively and negatively (such as agree with goals or dis-
agree with them), acquisitions and losses, the probability of achieving
the goal, cognitive justifications stating the rationality of action in
the case of evaluation of purposefulness. But the set of elements of
efficiency in its collective meaning is not closed yet. Kotarbinski is
cautious enough not to state that he has constructed and proposed all
possible evaluations of action, which constitute the synthesis of
efficiency. Efficiency is effectiveness and economy (alternatively
chariness or productivity) and adroitness and •.• When we are to evaluate
an action from the point of view of efficiency we choose an appropriate
set of constituent evaluations, and we leave out those evaluations which
are not significant in this case. In struggle, for instance, we do not
often take expences and costs into consideration. It is victory that is
important as a result. Then the evaluation of efficiency can be reduced
to efficacy. Victory was the goal and if it was achieved, nothing else
is important. That is how it happens when it is victory at all costs
that is our goal.
As if paralell to two sets of positive evaluations there exist two
sets of negative evaluations - inefficiency. Inefficiency may be regard-
ed as a "sum" - ineffectiveness, prodigality, maladroitness, etc., and
it may also be used as an equivalent for every particular form of
negative evaluation: ineffectiveness is inefficiency, prodigality is in-
efficiency, unproductivity is inefficiency etc. But this is not all.
Kotarbinski showed the need for a third pair of sets of negative
evaluations - counter-efficiency. He showed it on the example of effect-
iveness and ineffectiveness. In reality the following situations occur:
1) we perform an action leading to a goal, and we achieve the goal; the

action was effective,
2) we have a goal, but our action is not appropriate and we do not
achieve our goal; the action was ineffective,
3) we perform an action leading to a goal but, because of some reason
for which we are not to blame, we do not achieve our goal; we lose the
possibility of attaining it for ever or for a period of time; the action
was counter-effective.
A drastic example of counter-effectiveness may be a surgical operation
performed lege artis, after which the patient dies. It might seem that
counter-effectiveness can be reduced to ineffectiveness, as one of its
varieties. Logical analysis and an attempt at formalization of these
three situations, however, incline us rather to handle counter-effec-
tiveness separately. Let us try to write it down, adopting:
d - action leading to 'c', or rather performed in order to achieve
'c'
c goal
-, negation
--+ - causes I follows
v - exclusive disjunction, alternative denial
1. effectiveness:
2. ineffectiveness~
3. counter-effectiveness
The notation may provoke stipulations with regard to the meaning of
negation of a goal, negation of an event, negation of an action or with
regard to the use of the same symbol of negation for an event and for
the symbol of sequence, etc. 1 However, this conventional notation shows
that effectiveness is very close to counter-effectiveness, because:
d --+ c v -'c
We have a great class of actions which are either effective or
counter-effective. In this way we get to the theory of probability which
is included in praxiologieal evaluations. Kotarbinski, however, did not
solve the problem whether the notion of counter-effectiveness is suffi-
cient, or there is a need for enlarging the set of counter-efficiency.
lKotarbi~ski did not believe ,that adequate formalization of action was

possible. In the Department of Praxiology meetings of the Board of For-
mal Praxiology took place, but Kotarbi~ski never attended them, and he
neither expressed his opinion on the subject of the direction of re-
search, nor evaluated the effects. When he recorded his intuitions by
means of incidental variables, he made it clear that this record did not
render the ... fully. For instance, in the case of ineffectiveness only e!X
~~~ fac~ do we find out that the action did not lead to the intended
purpose. Besides, there are actions which can be either effective or in-
effective, or even more or less effective <effectiveness can be graded).
Theoretical basis of practical instructions
A famous Polish economist Oskar Lange, who introduced several praxiolo-

gical problems to political economy, disagreed with Kotarbinski. In
Lange's opinion praxiology should be a theory of methodologically ratio-
nal action. Rationality is based on cognitive justification -the better
justified the action and the more it is based on scientific background,
the greater is the probability that the action will be effective. Kotar-
binski's answer in this controversy was that praxiologists are interest-
ed not only in that which is positively evaluated with regard to effi-
ciency, but also in that which is evaluated negatively. They distinguish
not only kinds of efficient actions, but also kinds of inefficient ones,
for instance, practical errors.
By evaluation of purposefulness, that is by ascertaining that the
given action leads to the event desired as a goal, we get to practical
instructions. They indicate what must be done or what should be done in
order to achieve a goal, and performing on this account a certain action
we bring about conditions which are either necessary and sufficient, or
of assistance for the desired event.
Praxiology as methodology not only gives the scheme of practical
instructions (if you want to achieve goal C in conditions W you must
perform action D which, because of the natural sequence of events - from
D in conditions W to C - is justified by the theoretical basis T). This
means either that there exists a science which states that the event C
occurs after D in conditions W,· or that the same thing is maintained by
men on the basis of their personal experience. Such a statement then
does not necessarily have to be a scientific assertion. A large number
of Kotarbinski' s writings adduce theories outlined by himself - the
theory of events, the theory of action, and praxiology, which strives to
generalize and verify or falsify human experiences.
In the metodology of Claude Bernard's experimental medicine we find
the course used by research workers: a hypothesis - verification of the
hypothesis by means of experiments - probably making the hypothesis more
accurate - verification - etc. This method was presented by Le Chatelier
in the analysed philosophy of Taylor's system as the principle of
gradual improvement of all repeatable actions, of production, improve-
ment of inventions: we set a goal "for ourselves, then we choose the
appropriate means with regard to conditions in which we are going to
act, then we execute our planned actions, check up on the result, modify
the goal, and so on and so forth. What is the theoretical basis of the
course of experimental research? And what is the theoretical basis of
the principle of gradual improvement, which surely cannot be regarded as
a practical direction? Nowadays the former direction can be found in the
study of the progress of human knowledge. But where could it be found
before this st1,ldy was created? The latter direction is not a work of
~cientific management, but belongs to praxiology because of the adequacy
of assertions. Le Chatelier's direction which is called in Poland organ-
ization course does not refer only to production but also to a large
class of compound actions which are repeated and thanks to it improved -
made more efficient.
Kotarbinski tries not to reduce the theoretical basis to the theses

of accepted and honoured sciences. He introduces elements which seem to
come from the "outside", and tries to assign them to sciences which are
not necessarily mature, commonly known and accepted, e.g. to the theory
of events or the theory of action. He draws a typology of efficient
actions. He distinguishes e.g. anticipation (first come first served),
temporization (conscious delay and awaiting kairos - the best moment to
act and achieve the goal), potentialization (manipulating the possibi-
lity of action instead of acting - one of the forms of potentialization
is threat), instrumentalization (use of instruments). If the acting
subject knows about various kinds of efficient action, he is able to
choose in a given situation one of the distinguished forms. However,
Kotarbinski did not give any practical directions: if you have definite
circumstances and want to achieve your purpose, you must make the people
surrounding you face accomplished facts. It ought to be noticed that the
general quantifiers are carefully avoided in praxiology: we cannot find
there an assertion that something happens always and in all cases.
Relativism and relativization
I want to close this short survey of the philosophical and methodologi-

cal repercussions of Kotarbinski's praxiology with a discussion of rela-
tivism and relativization. Relativism is sometimes emphasized, and some-
times it is stated generally or neglected. Temporal relativism is as if
put in front of brackets and it holds for all elements covered by it.
However, less careful readers charge Kotarbinski with absolutization,
neglecting such concepts as 'phase' or 'period'. Man does not act all
the time of his existence. We can speak about action properly only if we
distinguish fragments of time - phases when man acts consciously, at
will and purposefully. In reism we would say that action is a man in the
phase of performing certain purposeful movements or in the phase of
purposeful immobility, because he acts even when he is apparently
motionless, but is ready to intervene. 2 Evaluations of economy in a
certain period may be positive, but over a longer period of time it may
turn out that the action has brought about heavy losses.
So, there is relativization with regard to time in praxiology.
Adequacy of declaration demands stating that there is more to it then
just reference to time and the trivial statement that sometimes things
go in one way, and some other times in another. Kotarbinski orders us to
"look round" and to perceive various possible objects of reference,
towards which we must try to make relativization, in this way avoiding
maximalism. According to his programme he was a minimalist. If we state
that a given event is a state of things - it is so from one particular
point of view, e.g. examined feature. From the point of view of another
loespit-e t-he exist-ing opposit-ion: oct-ion - speech, oct-ion - t-hinking,

){ot-orbinski includes bot-h speech and t-hinking in oct-ion. Thought. is
sometimes equal to transforming an unaware into an aware one.
feature the same event will be a change. Extracting an aching tooth we

cause a destructive event, but from the point of view of our health this
event is constructive. If two men compete in something, in some respects
they help each other, in other respects they disturb each other, and
disturbing is understood here in a very broad sense. Even striving to
win the first prize is disturbing - as the winner has prevented the
other man from getting the prize. Kotarbinski included struggle and
concurrence in the same class - for him concurrence is positive coopera-
tion, and struggle is negative cooperation.
Each science relativizes its interests to a particular fragment of
reality, and does not enter foreign lands. Still, familiarity with the
borderlands is advisable. It is so in theory. When practical problems
are being solved, they integrate various theoretical and practical
sciences. Theoretical sciences create a theoretical basis for the
schemes which are developed in practical directions, possibly being
appropriately outfitted. But the language of particular sciences are
different, and the same term may have different definitions - for
instance the term 'workllabour' (in Polish 'praca' - one word for the
various notions) in physics, physiology, psychology, sociology,
economics, praxiology and philosophy. Every correct - for a given
language - definition should contain in its definiens expressions which
belong to the class of terms of this science, and denote or designate
things, events, features, relations, etc. of this fragment of reality
which is the subject of this science. Genuine names designate things
which are nearly always compound, because .every particular science, if
it is not atomism, leaves atoms and elementary particles, and moves
towards more and more complex, animate and conscious things. Praxiology
has found its place among other sciences and is aware that the objects
it examines are among the branches of Porphirius's tree: thing - animate
thing - man - agent - ••• , thing - inanimate thing - instrument - outfit.
It uses colloquial language, but a precise one, and introducing new
concepts it relativizes them to the received sense - 'at least it tries
to do so. The development of praxiology is relativized to the number of
praxiologists and their productive abilities. And practical realism says
that one must desire only that which one can achieve.
KOTARBINSKI'S TI-EORY OF GENUINE NAMES
Vito F. Sinisi
While the work of the contemporary Polish logicians is fairly well known
by their colleagues in western Europe and America, little is known in
these countries about recent Polish philosophy. This is perhaps due to
the fact that the logicians generally published in universal languages
while the philosophers published their works mostly in Polish. This is
unfortunate because western philosophers, especially logical empiri-
cists, lIight have profited from a knowledge of the work of such men as
Tadeusz Kotarbinski and Kazimierz Ajdukiewicz, who are probably to be
eonsidered as Poland's outstanding contemporary philosophers. Both have
formulated philosophical views influenced by modern logic: Ajdukiewicz
has ereated radical conventionalism, and, Kotarbinski has' developed
rei sm.
Reism, a philosophical view akin to nominalism, countenances only
the existence of things, i.e. objects which are, as Kotarbinski says,
"extended and resistant". A reist Blight be described as one who system-
aticallyattempts to "reduce" all of the categories (e.g. the categories
of events, relations and properties) to the category of things. He con-
strues sentences purporting to be about events, relations or properties
as fa~ons de parle~, surrogates for sentences exclusively about .things.
For exa.ple, 'roundness is a property of this orange', see_inglyabout
the property roundness, is said to be a substitutive expression for
'this orange is round', which, it is held, is about things exclusively.
Expressions apparently referring to objects which are not things are not
genuine nues but "pseudo-n8.llles", e.g. such expressions as: 'concept',
'disposition', 'llotion', '.ethod',· 'operation', 'event', 'length',
'fact', 'velocity', 'number', and 'function'.
This view, like other ,.ore recent nominalistic views, is essential-
ly programmatic, which is to say that the reductive methods proposed are
i.nco.plete; for eXlUllple, there is no known way to reduce all sentences
prima facie about classes to sentences about things. And it is perhaps
101
J. Wole1lsld (ed.), Kotarbwki: Logic, Semantics and Omology, 101-118.
108 VITO F. SINISI
such difficulties which constrain Kotarbinski to speak of the basic

"postulate" of reism, one stating "Let us make ita point that no sen-
tence in its final form (that is, freed from metaphors, abbreviations
and substitute expressions) should include any pseudo-name (that is, an
expression that has the appearance of a name but is not a name of any
physicalobject).,,1
One of the fundamental constituents of reism is Kotarbinski's for-
mulation of the relation of denoting holding between genuine names and
things. In this paper I wish to sketch informally his views on denoting
as they are presented in his important work of 1929, Elementy teorii
poznania, logiki formalnej i metodologii nauk (Elements of Epistemology,
Formal Logic and Methodology of the Sciences). These views are interest-
ing because they are a development of nominalism and nominalistic seman-
tics generally unknown to philosophers and semanticists of the West.
They are also interesting because they reveal how some of the logical
concepts of Stanislaw Lesniewski (1886-1939), one of the leading members
of the Warsaw School of logicians which flourished between the two Wars,
influenced the development of a metaphysical system--reism. For example,
Kotarbinski's analysis of the denoting relation is given in terms of
'is', a primitive, undefined term characterized by the single axiom of
ontology, a logical system created by Lesniewski. Another interesting
point is that Kotarbinski uses the relation of multiple denotation
according to which a term truly predicable of a number of individual
things (in Kotarbinski's sense) is said to denote these things several-
ly; since the term 'man', for example, is truly predicable of Julius
Caesar and Bertrand Russell, it denotes Caesar and it also denotes
Russell. It does not denote the class of men but it denotes severally
what are sometimes spoken of as the members of the class of men;!. By
construing predicate expressions in this way Kotarbinski attempts to
free reism from commitment to classes and, more generally, to the exist~
ence of objects which are not things. Finally, these views are interest-
ing because they provide an adequate, formal analysis of non-designating
terms which does not have recourse to Russell's theory of definite de-
scriptions.
My purpose here is to state and explain Kotarbinski' s views as
clearly and briefly as possible. It is hoped that this sketch will help
to stimulate interest in philosophical developments in Poland between
the two Wars.
l F l'om an unpublished lecture, "The Basic Post.ulat.e of Concret.ism" (see

Kot.arbinski (1961) - Ed.l, delivered at. Pennsylvania St.at.e Universit.y on
April 9, 1969. I wish t.o t.hank Professor Kot.arbinski for a copy of t.his
IE>ct.urE>.
;!A similar view of common names is found in Ockham, Hobbes, and more
recE>nt.ly in J.H. Woodger (1961, p.196) and in R.M. Mart.in (1968).
KorARBINSKI'S THEORY OF GENUINE NAMES 109
II
In the Elementy Kotarbinski says: (1) " .•• to be a name is equivalent to

being suitable for the role of a predicate in every sentence of the type
'A is B' (with the fundamental sense of the copula 'is'). A given
expression is sui table for the role of a predicate when it is possible
to substitute it meaningfully for Bi it is not necessary that a true
sentence result from the substitution." However, (la) "To denote a given
object in a given language is equivalent to being a name of that object
in that language; in other words, to being !,uitable in that language as
Ii predicate in a true sentence (about that object) of the type 'A is B'
(with the fundamental sense of the copula 'is,).,,3
According to Kotarbinski there are three kinds of genuine names:
singular names (each denoting one and only one thing, in Kotarbinski's
sense of 'thing'), e.g. 'Rome', 'The discoverer of the circulation of
the blood'; common names (each denoting severally more than one thing),
e.g. 'Polish King'; and objectless or vacuous names (each of which de-
notes no thing), e.g. 'werewolf', 'Monarch of the Swiss Republic'. These
names are vacuous names because " ••• it is impossible to express a true
sentence in which either of them would be the predicate. For, there is
no one or nothing which would be a werewolf or Monarch of the Swiss
Hepubl ic. ,,4- '
It is readily apparent that a distinction is made between names and
denoting names; not all names are denoting names. Other question aside,
this distinction seems to preserve the one we often wish to and do make
i.n colloquial language between expressions or terms which purport to
name objects but fail, and expressions which both purport to and do name
objects. 'Cerberus', for example, is a term which purports to name an
object but fails, while 'Friedrich Ludwig Gottlob Frege', both purports
to and does name an object. For Kotarbinski 'Cerberus' is a vacuous name
while 'Friedrich Ludwig Gottlob Frege' is a singular name.
I think that the following examples might serve to illustrate (la)
above ('is' in each case functioning in the so-called fundamental
sense) :
(2 ) 'The man who wrote 29 Waverley novels' denotes Sir Walter Scott if
and only if Sir Walter Scott is the man who wrote 29 Waver lay
novels;
(3) 'Harvey' denotes the discoverer of the circulation of the blood if
and only if the discoverer of the circulation of the blood is
Harvey;
(4) 'A pupil of Socrates' denotes Aristippus if and only if Aristippus
is a pupil of Socrates;
3Element.y, pp. 6,7. Translat.ions of' Polish t.ext.s in t.his paper are by
V,P,S.
4- Ibid ., p.7.
110 VITO F. SINISI
(5) 'The winged horse captured by Bel1erophon' denotes Pegasus if and

only if Pegasus is the winged horse captured by Bellerophon;
(6) 'Zeus' denotes Zeus if and only if Zeus is Zeus.
It will have been noticed that the phrase "fundamental sense of the
copula 'is'" occurs in two crucial contexts: in the explanation of a
name, and in the explanation of denotation. In the Eleaenty Kotarbinski
did not explicate this so-called fundamental sense of 'is'; however, an
analysis of various passages in Kotarbinski's works reveals that he con-
strues this 'is' as the single primitive term of Lesniewski's ontology,
which is to say that a sentence of the type 'A is B' (with the funda-
mental sense of 'is') is (loosely) equivalent to: there is at least one
A, and there is at most one A, and whatever is A is B. In order to indi-
cate the meaning of the preceding sentence it will be necessary to say
something about Lesniewski's ontology.
Lesniewski created three systems of logic: protothetic (an extended
propositional calculus), ontology (a calculus of names), and Illereology
(a logical theory of the part-whole relation), which is based on the two
previous disciplines. Ontology is derived from protothetic by the intro-
duction of the primitive term 'E', which may be read as 'is'. The term
is introduced by means of a single axiom, and the theorems of ontology
are stated in terms of this primitive and the functors of protothetic
(material implication, conjunction, alternation, equivalence, negation).
'£' always appears as part of expressions of the form IX £ y'. Such
expressions may be called "singular expressions"; if appropriate con-
stants are substituted for the variables, we obtain a singular proposi-
tion. The meaning of 'E' must be gathered from the single axiom of onto-
logy, which may be expressed as follows5 :
'<:Ix'<:ly (xEy '" (3c ct:x 1\ '<:Ic'<:ld «ct:x 1\ dt:x) -+ ct:d) 1\ '<:Ic (ct:x -+ ct:y»)
The substituends of the variables in a singular expression may be

either singular, common or vacuous names, so that, for example, to say
that Socrates is the teacher of Plato is to say (colloquially) that
there is at least one Socrates, and there is at most one Socrates, and
whatever is Socrates is the teacher of Plato. It can be seen that a
necessary condition for the truth of a singular proposition is that its
subject term be a singular name; if the subject term of a singular prop-
osition is either a common name or a vacuous name, then it is false. 6
5 This axiom was first. formulat.ed in 1920, and subsequent.ly simplified.

Vid .. Le~ni .. wski (1930). This is t.h .. only work publish .. d by L .. sni .. wski
devot.ed t.o ont.ology .. xclusively. Some comment.s on ont.ology are t.o be
found in his <1931>. For Lesni"wski's at.t.empt. t.o explain t.he int.uit.ive
meaning of 't:' s .... my (1962), p.233. Perhaps it. should be point. .. d out.
t.hat. '£' is not. defined but. rat.her is "charact.erized" or "described" by
t.he axiom.
6Allowances must. be made for t.h .. fact. t.hat. t.h .. r .. are no d .. finit.e or in-
d .. finit. .. art.icle. in Polish, andt.hat. ont.ological synt.ax is based on t.he
synt.ax of Polish. cr. glupecki (1955), p.13. A suit.able inst.an" .. of t.h ..
KOTARBINSKI'S THEORY OF GENUINE NAMES 111
I have pointed out that while Kotarbinski did not explicitly ident-
ify the so-called fundamental sense of 'is' with Lesniewski"s 'c', there
is evidence in his works to justify the belief that sentences of the
type 'A is B' (with the fundamental sense of 'is') are to be construed
as Lesniewski's. singular propositions. 7 The necessary and sufficient
conditions for the truth of sentences of the type'A is B' are identical
with those for the truth of singular propositions; however,while these
conditions coincide, there is an additional factor present in Kotarbin-
ski's usage of 'is' which is absent in Lesniewski's use of 'c', namely,
a semantical rule, so to speak, to the effect that all the variables in
the axiom range over things (in Kotarbinski's sense) exclusively. Les-
niewski's ontology is not restricted in this way so that under some
other interpretation, for example, wherein the variables are restricted
to some set of non-things (again in Kotarbinski's sense), it is possible
that a singular proposition will be true but will not be true in Kotar-
binski's interpretation.
This explanation of what it means to say that a sentence of the
type 'A is B' (with the fundamental sense of 'is') is true will perhaps
make clear the purport of passages (1)-(6) and the significance of the
concepts of name and denoting name. (Henceforth I shall use "sentence(s)
of the type 'A is B'" to lIean "sentence(s) of the type 'A is B' (with
the fundamental sense of 'is')".) Knowing the truth conditions for a
sentence of the type 'A is B' we can easily distinguish between derioting
and non-denoting nalles. An· expression of a language is a name if and
only if it lIay be used "meaningfully" as the predicate in a sentence of
the type 'A·is B'. Denoting names form a subset of the set of names; an
expression of a language denotes an object if and only if it lIay be pre-
dicated truly of that object. According to this analysis, then, such ex-
pressions as • Zeus 'and 'Xanthippe's unmarried husband' are names if,
for example, such sentences as 'Plato is Zeus' and 'Socrates is
Xanthippe's unmarried husband' are meaningful. While 'Zeus' and
'Xanthippe's unmarried husband' are nalles, they are not denoting nalles
because no sentence of the type 'A is B' is true, wherein either func-
tions as a pre~licate. Furtherllore, this analysis entails that· some
(though not all) definite descriptions (or their analogues in Polish).
come to be classed as names. For exallple, 'The author of Waverley', 'The
author of Sla~kenburgius on Noses'. 'The relation of seniority', how-
ever, is not a genuine name but a pseudo-nalle, according to Kotarbinski,
because it cannot be predicated meaningfully of any thing in a sentence
of the type 'A is B'.
Some descriptive adjectives cOile to be classified as names and sOlie
are classified as denoting names. 'Massive' and 'bright' are, for
example, denoting names because sentences such as 'Gaurisanker is mas-
schema 'XCy' in Polish would b .. 'Jan III Sobieski j .. st. wyba.viciel ...
Wiednia', which would be t.ranslat.ed· int.o English (lit.erally> aa 'Jan So-
bieaki III ia lib.rat.or· of' Vi.nna'. It. might. b.helpf'ul t.o ke.p t.hia in
IDind in diacu~.ion of' t.h. schemat.a 'xC,' and ' . ia B'.
7F.'or example, ..... Element.y, pp. 7-8, 18, a. veli as Kot.arbin.ki'. more
r .. cent. <1955>, pp. 115, 117.
112 VITO F. SINISI
sive' and 'The Evening Star is bright' are true. While many substantives
are recognised as names, many are not. For example, 'wine', 'woman' and
'song' are genuine denoting names leach having multiple denotation) but
'fact', 'fiction' and 'forecast' are pseudo-names.
Let us now consider denoting names more closely. Having explicated
the so-called fundamental sense of 'is', the right side of the bicondi-
tional in (la) has perhaps been made clear. In (2) the left side of the
biconditional is equivalent to the right, and since the right side of
(2) is true, it follows that the left side is also true. Note that in
(2) the sentence of the type 'A is B' has a proper name as its subject
and a definite description as its predicate. In (3) the roles are re-
versed. These two examples indicate that in the analysis of 'A is B' the
uniqueness condition does not pertain to the definite description when
it is used as the predicate in such a sentence; it pertains to it only
when it is used as the subject. The displayed passage (4) differs from
(2) and (3) because the right side of its biconditional contains a
proper name as the subject and a common name as the predicate. (5) and
(6) present cases of names which do not denote any thing. 'The winged
horse captured by Bellerophon' does not denote any thing; specifically,
it does not denote Pegasus. This is so because no sentence of the type
'A is B' is true when 'The winged horse captured by Bellerophon' is sub-
stituted for 'B', and this is so because no individual thing is the
winged horse captured by Bellerophon. The falsity of the right side of
(5) is compounded because a vacuous name, 'Pegasus', has been substi-
tuted for 'A'. Neither 'Pegasus' nor 'The winged horse captured by
Bellerophon' is a name of a thing, though each is a genuine name.
The displayed passage (6) is perhaps the most interesting of the
examples. Previous discussions should make i t clear why the left side of
(6) is false. If the right side of (6) were true, it would follow that
'Zeus exists' would be true. This is made explicit by Kotarbinski' s
statement, late in the Elellenty,8 that "'M exists' for us", i.e. reists,
"is equivalent to 'something is M''', which may be expressed as "M exists
= 3y (y is M) Df.", wherein 'is' functions in the fundamental sense. 9 If
'Zeus is Zeus' were true, it would follow that something is Zeus, and at
most one thing is Zeus, and whatever is Zeus is Zeus. Consequently, 'If
Zeus is Zeus, then something is Zeus'. In virtue of the definition above
the consequent of 'If Zeus is Zeus, then something is Zeus' is equival-
ent to 'Zeus exists', and substituting 'Zeus exists' for this consequent
we obtain 'If Zeus is Zeus, then Zeus exists'. Since it is false that
Zeus exists, it follows that it is false that Zeus is Zeus. 10
9 Cf . t.he definit.ion of class exislence in Pl'incip/Q Malhemalico., 24.03. •

"We writ.e (3!o' ror 'Q exist.s'. The definit.ion is 24.03 3IQ'='<3x) xCQ
Df/'
iOCf . Principia Mat.hemat.ica, ·14.28, p.176. »·14.28. I-- : El (1 x><0x) . "'.

(1 x)(0 x)=( 1 x)(0 x). I.e. '( 1 x)(0 x)' only sat.i s f ies t. he re f lexi ve property
of identity if <1x)(0x) exists."
III
Russell's theory of definite descriptions provides for an analysis of

non-designating terms, e.g. 'Zeus', 'flamlet', 'Apollo'. According to
Russell, in a sentence such as 'Apollo is a god', " •.• 'Apollo' means
really 'the object having such-and-such properties', say 'the object
having the properties enumerated in the Classical Dictionary'. If these
make up the propositional function 0x, then 'Apollo' means '(1x)(0x)',
. . . ,,11 Hence, tApollo is a god' would be equivalent to t(1x)0x is a god',
and this in turn is equivalent to: (3b) (V'x (0x=x=b) /\ b is a god),
which says (colloquially) that at least one individual has the property
0, and at most one individual has the property 0, and whatever has the
property 0 is a god.
One of the virtues of Russell's analysis of definite descriptions
is supposed to be that it allows us to use such expressions as tz eus ',
tHamlet', and 'Apollo' without committing ourselves to the existence of
anything but the values of our bound variables; it does not commit us to
the existence of Greek gods or fictional entities. The reistic analysis
has the same virtue, since to say (reistically) tApollo is a god' is to
say '(3x) (x is Apollo) A V'xV'y «x is Apollo A y is Apollo) ~ x is y) A
V'x (x is Apollo ~ x is a god)', i.e. tat least one thing is Apollo, at
most one thing is Apollo, and whatever is Apollo is a god'. The reistic
analysis of such sentences as tApollo is a god' also permits us to use
expressions like Apollo without committing ourselves to the existence of
anything but the values of our bound variables, viz. "extended and
resistant" objects.
With the introduction of vacuous or empty names the reistic (and
ontological) analysis of 'Apollo is a god' does not require (as the
theory of definite descriptions does) the replacement of expressions
like tApollo' with other (supposedly) synonymous expressions. Russell's
analysis of definite descriptions seems to have been partly motivated by
the belief that a name must denote an object. Speaking of Russell's
views on names, Jaakko Hintikka has said "There is nothing so sacred
about the idea of a name that one could not conceivably drop the assump-
tion that it refers to something", and he adds "I would go so far as to
hold that a formal reconstruction of the logic of our language in which
this cannot be done without break inA the rules of the game just is not
comprehensive and flexible enough". It may be that ontology and reism
satisfy Hintikka's requirements for a formal reconstruction of the logic
of our language.
l1P l'incipia Mat.h8'lDat.ica, p.31.
1:.1uint.ikka <1969>, p.127.

114 VITO F. SINISI
IV
Denotation with respect to singular and vacuous names does not seem to
present any serious problems, while denotation with respect to common
names requires further explication.
It is sometimes said that predicate expressions denote (or have as
extensions) sets or classes. Does a common name, which denotes severally
the things to which it truly applies, ever denote the set or class of
its designata? If so, and if a set or class is construed as an "ab-
stract" entity, then reism is an ontology which seems to countenance
such entities. This, of course, is to be avoided if reism is to be con-
sistent; reism commits itself only to the existence of "extended" and
"resistant" objects. An analysis of the terms "set" and "class" allows
Kotarbinski to say that some common names not only have multiple denota-
tion but denote the class of their own designata as well. This does not,
however, commit him to belief in the existence of classes as abstract
entities. I shall now try to explain this apparent paradox.
In the Ele.enty (pp. 13-15) Kotarbinski distinguishes the distribu-
tive meaning of "set" from its collective meaning. In the first sense
"set" is said to be a pseudo-name, while in the second it is said to be
a genuine name. Consider "X is an element of the set of M's". In this
expression 'set' is used in its so-called distributive sense, and so
used is a pseudo-name. This expression is said to be equivalent to 'X is
an M'. For example, 'Mars is an element of the set of solar planets' is
equivalent to 'Mars is a solar planet'. In spite of the fact that 'Mars
is an element of the set of solar planets' is true, no object is the set
of solar planets. In the distributive sense 'the set of solar planets'
does not denote any thing because to use the expression 'the set of so-
lar planets' (in the distributive sense of "set") as a predicate in some
sentence of the type 'A is B' would be an absurdity, according to Kotar-
binski; something on the order of 'John is hold on!,13
I think that at least part of what Kotarbinski means may be restat-
ed as follows. "Set", in the distributive sense, can be eliminated from
a sentence without changing the meaning of the sentence. For example,
'Mars is an element of the set of solar planets' would mean the same
thing as 'Mars is a solar planet'. Thus in the distributive sense of
"set" the expression 'is an element of the set of' would be "reducible"
to 'is' (in the so-called fundamental sense).14 'Mars is a solar planet'
would be an instance of the schema 'A is B'; 'Mars' being a singular
name and 'a solar planet' a common name.
In the collective sense of "set" the term is said to be a genuine
denoting name. In this sense "the set of M's" denotes a certain whole
composed of M's - just as a chain consists of links, or a swarm of bees
13Loc . cit,.
14cf . Soboei';ski <1964-6>, p.2. I "ish to thank Prof •• sor Soboei';ski for
an offprint of this artiel •.
of particular bees, or a heap of sand of grains of sand. In this sense a

chessboard is a set of squares because whatever dissection of the chess-
board we consider (whether it be a particular square, or a piece of a
square, or a whole square plus a piece of an adjacent square, or the
amalgamation of two squares lying at opposite corners of the board) it
turns out that it contains a piece of some square which enters into the
composition of the chessboard. Briefly, in the collective sense a set of
M's is construed as an object P each of whose ingredients contains an
ingredient of some M which is an ingredient of this object. 15
The last sentence above is Kotarbinski's paraphrase of a definition
in Lesniewski's mereology.16 In Lesniewski's system this definition is
preceded by previous definitions and some axioms. Rather than cite the
defini tion out of context it might be helpful to state the axiolls and
definitions preceding this definition (in each case 'is' is to be taken
as the single primitive term of ontology).
Axiom I. If P is part of the object Q, then Q is not part of the object

P.
(Lesniewski points out that he uses sentences of the type 'A is not b',
e.g. 'Q is not part of the object P', as being equivalent to sentences
of the type 'A is an object, and not (A is b)'. They are not equivalent
to sentences of the type 'not (A is b)').
Axiom II. If P is part of the object Q, and Q is part of the object R,

then P is part of the object R.
Definition I. P is an ingredient of the object Q if and only if P is the

same object as Q, or P is part of the object Q.
Definition II. P is the class of objects a if and only if the following
conditions are satisfied:
a) P is an object,
b) every a is an ingredient of the object P,
c) for all Q - if Q is an ingredient of the object P, then
some ingredient of the object Q is an ingredient of some
8.
(Lesniewski points out that he uses sentences of the type 'every a is

b', e.g. 'every a is an ingredient of the object P', as being equivalent
to sentences of the type 'some object is a, and for all X - if X is a,
then X is b'. They are not equivalent to sentences of the type 'for all
X - if X is a, then X is b'. He says that he uses sentences of the type
'no a is b' as being equivalent to sentences of the type 'some object is
a, and for all X - if X is a, then X is not b'. They are not equivalent
to sentences of the type 'for all X - if X is a, then X is not b'.)
l5cr . El .. ment.y, p.13.
16Vide L .. ~nie"8ki <1928>, pp.27orr.

116 VITO F. SINISI
The following examples (taken from Lesniewski) might help to illus-

trate these notions. The segment AB of Figure 1
Alr-__~C~I____~D+I____-=E~I----~B~I
Figure 1
is the class of parts of the segment AB because all three conditions of

Definition II are satisfied. The segment AB is not the class of parts of
the segment AD because while the first two conditions of the definition
are satisfied, the third is not; the segment EB is an ingredient of the
segment AB, but it is false that some ingredient of the segment EB is an
ingredient of some part of the segment AD. The segment AC is not the
class of ingredients of the segment AB because while the first and third
conditions of Definition II are satisfied, the second is not; the
segment AB is an ingredient of the segment AB, but it is not an ingredi-
ent of the segment AC.
Axiom III. If P is the class of objects a, and Q is the class of objects

0, then P is Q.
Axiom IV. If some object is a, then some object is the class of objects
a.
Definition III. P is the set of objects 0 if and only if the following
conditions are satisfied:
a) P is an object,
b) for all Q - if Q is an ingredient of the object P, then
some ingredient of the object Q is an ingredient of some
a, which is an ingredient of the object P.
This may be illustrated by the following examples (taken again from Les-
niewski). The segment AC of Figure 1 is a set of parts of the segment AB
because both conditions of Definition III are satisfied. The segment AC
is not a set of ingredients of the segment DB because while the first
condition of the definition is satisfied, the second is not; the segment
AC is an ingredient of the segment AC, but it is false that some ingred-
ient of the segment AC is an ingredient of some ingredient' of the
segment DB, which is an ingredient of the segment AC.
Perhaps these axioms, definitions and examples will help to expli-
cate Kotarbinski's use of "set" and "class". In the collective sense of
"set", then, the term is a genuine name. In this sense a set of M's is
an object P. In the distributive sense a set of M's is not an object.
According to Kotarbinski, 'x is an object' implies 'x is a thing', i.e.
• x is extended and resistant'. This stipulation is not to be found in
the axioms and definitions of mereology, and hence. Kotarbinski's use of
"set" may be thought of as a specialization of Lesniewski's use of the
term.
Let us now consider the term "class". According to Kotarbinski,
"the class of M's" is equivalent to "the set of all M's". In the dis-
tributive sense of "class" the term is a pseudo-name, and it behaves in
the same manner as "set" (in its distributive sense). For example, it is
said that "X is an element of the class of M's" is equivalent to "X is

an M". Consequently, 'Mars is an element of the class of solar planets'
is equivalent to 'Mars is a solar planet'. Thus in the distributive
sense of "class" the expression 'is an element of the class of' (like
the expression 'is an element of the set of') is "reducible" to 'is' (in
the so-called fundamental sense).
"Class" as a genuine name deviates from "set" (when the latter is
construed as a genuine name). In the collective sense the class of M's
is the set of all M's, that is, an object P each of whose ingredients
contains some ingredient of some M which enters into the composition of
this object: and furthermore, whatever is an M is also an ingredient of
the object P. This is Kotarbinski's version of Lesniewski's second defi-
nition above. For example, in the collective sense a particular tree is
not the class of cells, nor is it the class of fibres, nor is the class
of particles. For, while the first condition of the definition is satis-
fied, the second is not; it is not true that whatever is a plant cell is
an ingredient of this particular tree (and similarly with regard to
fibres and particles). However, a particular tree is the class of its
own cells, it is the class of its own particles. For, not only does
everyone of its ingredients contain at least sOBle ingredient of some
cell of the tree, but, moreover, whatever is a cell of the tree is also
an ingredient of the tree. Similarly, the fauna of Poland is the class
(in the collective sense) of animals living in Poland (past, present and
future) because each or its ingredients (whether a particular animal, or
the head of some animal, or one anima-I plus the leg of another) contains
some ingredient of some animal living in Poland; and furthermore7 every
animal living in Poland is an ingredient of the fauna in Poland. 1
Let us now return to our initial question: does a common name,
which denotes severally the objects to which it truly applies, ever de-
note the set or class of its own designata? In the distributive sense of
"set" and "class" these terms are pseudo-names; there is no object which
would be either the set or class of designata of a given common name.
Consequently, in this sense a common name does not designate either the
set or class of its own designata. In the collective sense, however,
there are cases when common names designate the set or class of their
own designata, but there will also be cases when common names do not
designate the set or class of their own designata. For example, consider
the common name 'something located wi thin the confines of the solar
system', which designates severally e.g. the moon, the earth, a mountain
on the earth, a particular man. This common name also designates the
class of its own designata because it (the set of all objects of the
solar system) is an object located within the confines of the solar
system. However, the common name 'chair', for example, whose designatum
is each chair, does not designate the class of its designata because
this class is not a chair.
17Element.y, p.16. Perhaps it. should be point.ed out. t.hat. i1' a part.icular
tree is t.he set. 01' it.s cells and also t.he set. 01' it.s,part.icles, t.hese
sets do not. have t.he same number 01' element.s, and hence in t.he set. -
theoret.ic usage 01' "set." are di1'1'erent..
118 VITO F. SINISI
Kotarbinski points out, finally, that a singular name always de-

notes every set (in the collective sense) of its own designata, and con-
sequently it denotes the class of its own designata because such sets
and classes are identical with the only designatum of such a name. For
example, the class of first emperors of the French is identical with
Napoleon Bonaparte, the only designatum of the singular name 'first
emperor of the French'. On the other hand, a vacuous name, e.g. 'Zeus',
never denotes any set or class of its own designata because it has no
designatum at all. 18
Perhaps this sketch of Kotarbinski's views on genuine names has revealed

that interesting and significant results were obtained in the develop-
ment of nominalism and nominalistic semantics in Poland between the two
Wars. The search for an alternative to construing predicate expressions
as names of classes or properties seems to have been a common character-
istic of nominalistic views. Kotarbinski's view that many predicate ex-
pressions can be construed as terms having multiple denotation seems to
provide such an alternative. However, as stated earlier, this view is
incomplete; serious difficulties arise for the reist when he is con-
fronted, for example, with sentences which quantify on relation or class
variables, e.g. 'Some relation is the converse of the relation greater
than', and the definition of ancestor in terms of parent: x is ancestor
of y if x belongs to every class which contains y and all parents of its
own members. Two courses seem open to the reist: to ban such sentences
from his language, thereby limiting its expressive power, or to show how
such sentences can be "reduced" to sentences about things. While Kotar-
binski is aware of such difficulties, he has not, to my knowledge, indi-
cated how these problems are to be treated.
18EleIDe nty, p.16.

KOTARBINSKI'S THEORY OF PSEUDO-NAMES
Vito F. Sinisi
In the paper "Kotarbinski's Theory of Genuine Names", which was

published recently in this journal 1 , I attempted to sketch and explain a
theory of genuine names formulated by Tadeusz Kotarbinski, who is
perhaps Poland's outstanding 20th century philosopher. The theory is a
fundamental ingredient of Kotarbinski's nominalistic views, and it marks
a development of nominalism and nominalistic semantics generally unknown
to philosophers and semanticists of the West. The theory also reveals
how some of the logical concepts of Stanislaw Lesniewski, one of the
leading members of the Warsaw School of logicians which flourished
between the two Wars, influenced the development of a nominalistic
metaphysics - Kotarbinski's reism.
In the present paper I shall present and analyze Kotarbinski' s
theory of pseudo-names, i. e., expressions purporting to refer to non-
things. The discussion will be informal and will focus on the theory as
it is presented in Kotarbinski's fundamental and important work of 1929
(Elellenty). I shall assume the reader's acquaintance with tlKotarbinski's
Theory of Genuine Names".
Perhaps this sketch and analysis of Kotarbinski's views on pseudo-
names will help to stimulate an interest in his work and in philosophi-
cal developments in Poland between the two Wars.
According to Kotarbinski, an expression is a genuine name if it may be

meaningfully predicated of a thing (an extended and resistant object) in
a sentence of the form 'A is B' (wherein the copula has its so-called
fundamental sense); if the result of so using the expression is a true
sentence, then it is said that the expression denotes the subject of
that sentence. According to this analysis, there are three types of
names: singular names (expressions denoting one and only one thing),
1Theoria, A Swedish Journal of' Philosophy and Psychology, vol. XXX, 2;

1964-, pp. 80-96; reprinted also in t.he present volume, pp. 107-118 [Ed.]
119
I. Woletfski (ed.), Kotarbitfski: Logic. Semantics and Ontology, 119-135.
120 VITO F. SINISI
e.g., 'Rome', 'The discoverer of the circulation of the blood', common

nlJllles (expressions having multiple denotation), e.g. 'red', 'Polish
King', 'man', 'mammal'; and vacuous or objectless nlJllles (expressions
which may be meaningfully though not truly predicated of a thing), e.g.,
'Apollo', 'The author of Slawkenburgius on Noses', 'Monarch of the Swiss
Republic' •
It does not follow from this analysis that every noun and adjective
is a genuine name, for not all names are genuine names; some names are
pseudo-names. Kotarbinski has described a pseudo-nalle as being any "word
(or set of words) which may indeed figure coherently as 'B' in a struc-
ture of the type 'A is B', but only if this structure ·fulfils a substi-
tuting-abbreviating role, not a fundamentai role"". The sentence
(1) Uranus is a planet
exhibits the fundamental or principal use of 'is', i.e., (1) is an in-
stance of the schema 'A is B' (wherein 'is' functions in the fundamental
sense). However, some sentences have the form 'A is B' without exhibit-
ing this fundamental or principal use of the copula; they display a so-
called derivative or secondary us~ of 'is', e.g.,
(2) Seniority is a transitive relation,

(3) A whale is a mammal.
Kotarbinski maintains that ·(2) is both meaningful and true; it has the
form 'A is B', and it would seem that "transitive relation" is a name of
an object, viz. seniority. He asserts that this is not so because (2) is
merely an abbreviation for
(2') If some object is older that another object, and the latter
object is older than some other object, then the first object
is older than the third object.
In the abbrev iating sentence (2) , is' does not stand between genuine
names although 'seniority' and 'relation' are substantives. In (2) we
have a sentence of the form 'A is B' "but in a secondary use, wherein
'is' sounds like or look~ like the copula in the sentence" (1) " but it
fulfils another role, it means something different, and the set of words
in the place of 'A' and 'B 'are not n8.lles3 ". These expressions are
called "pseudo-names". Sentence (3) is described in the same way: it is
said to be a substituting-abbreviation for
(3') Whatever is a whale is a mammal.
It may be recalled that ontologically reism countenances only the
existence of things. Semantically the reist countenances among genuine
names only singular, common, or vacuous names. Expressions apparently
"Element." p.9. Tl'on.lot.ion or Poli.h "ext.. in t.hi. popel' are b, V.F.S.

3 Ibid .
KOTARBINSKI'S THEORY OF PSEUDO-NAMES 121
referring to classes, events, relations, or properties are not genuine

names but pseudo-names; sentences purporting to be about such entities
are constructed as fapons de parler, surrogates for sentences about
things exclusively. (2) and (3) are examples of such surrogates. Addi-
tional examples are provided in Kotarbinski's discussion of the "reduc-
tion" of the ontological categories (events, relations, and properties)
to the category of things.
For the reist there is only one ontological category - the category
of things. Kotarbinski maintains that every other category reduces to
the category of things. Every meaningful sentence purporting to be about
a non-thing is construed as a substitutive expression for another
sentence which is understood literally and which is about a thing4.
KotarbiIlski takes Wundt's (1906) list of categories as the basis
for his discussion. Wundt's four categories are: things, properties,
states, and relations. Unfortunately, 'Kotarbinski does not characterize
these categories precisely, saying, "We shall not be disturbed by
Wundt's interpretation of 'things', 'properties', 'states', and 'rela-
tions' •.. We shall take these terms in their prevailing meaning"s.
Instead of 'state' Kotarbinski uses 'event' which, he says, is synony-
mous with 'state' in its more general sense. If one considers a watch,
for example, one might be inclined to, say that the movement of the
second-hand and the movement of the minute-hand are examples of events,
and one might be tempted to say
(4) The movement of the second-hand was faster than the movement
of the minute-hand
is true, and since (4) is about events it is likely that some object is
an event. The following also seems to be a true sentence about an event:
(5) The eruption of Vesuvius destroyed Pompeii and Herculaneum.
Since (5) is true one might be induced to believe that there are events,
e.g., the eruption of Vesuvius. Kotarbinski insists that (4) and (5) are
only prima- facie about events; they are not about events because they
are substitutive expressions for
(4') The second-hand moved faster than the minute-hand.

and
(5') Vesuvius erupted in such a way that it destroyed Pompeii and

Herculaneum.
In both (4') and (5') ther.e are no names of events; in (4') and (5')
each nalle denotes a thing. Acording to Kotarbinski, "No object is an
event because only spuriously (by construing a sentence in a substitu-
4 Ibid , p.61.
sElemeni.v, p.61.
122 VITO F. SINISI
tive. non-literal. non-fundamental way) and never genuinely is it

possible to express a true sentence about an event".6
Kotarbinski's claim is that the category of events is "reducible"
to the category of things because sentences which seem to be about
events are surrogates for sentences about things. He maintains that for
the sake of colorful expression. brevity. and out of habit as well. we
speak as if we were referring to such objects. He does not want to re-
nounce such locutions. but he cautions that one should remember that
these are abbreviative. substitutive expressions. and that their occur-
rence in a language does not force us to accept the existence of
'events' •
The category of relations is· the next to be "reduced" to the cat-
egory of things.
It is true that one mountain. Gaduch. is higher than another.
Giewont. and it is equally true that John is older than his son. How-
ever. from the reistic standpoint the relation higher than between the
two mountains (and. more generally. the relation higher than) does not
exist. nor does the relation older than exist. One can indeed say
(6) The relation higher than holds between Garluch and Giewont.
and this is correct. One may also say
(7) The relation older than joins John with his son.
According to Kotarbinski. (6) and (7) are also substitutive expressions

which need not be taken literally; they mean nothing more than
(6') Garluch is higher than Giewont,

(7') John is older than his son.
It is said that in (6') and (7') there are no names of relations; all of
the names in (6') and (7') denote things. The expressions 'older than
•••• and 'higher than ••• ' are not names of relations; these expressions
are, respectively, names of John and Garluch because the first can be
predicated of John in a true singular sentence of the form 'A is B'
(with the fundamental sense of 'is') 'and the second can be predicated of
Garluch in a similar sentence.
The claim. again, is that the category of relations is "reducible"
to the category of things because sentences which seem-to be about rela-
tions are mere substitutive-abbreviations for sentences which are about
things exclusively.
The category of properties is the last to be "reduced" to the cat-
egory.of things.
Consider the following sentences:
(8) Whiteness is a property of snow.

(9) The property of generosity characterizes noble people,
6 Ibid , pp.62-63.
(10) A magnet has the property of attracting iron,

(11) Every body has the attribute (or property) of extension,
(12) To every plane triangle applies the property that the sum of
its interior angles is equal to the sum of two right angles.
Now, (8)-(12) are apparently sentences about properties, and it would
seem that something is whiteness, generosity, etc. Kotarbinski claims
that these are also substitutive sentences for sentences which do not
contain so-called names of properties. (8)-(11) are equivalent, respect-
ively, to:
(8' ) Snow is white,
(9' ) Whoever is noble is generous,
(10' ) A magnet attracts iron,
(11') Every body is extended.
(Unfortunately, Kotarbinski did not complete this list with a sentence
equi valent to (12). He maintains that no object is a property, i. e. ,
properties do not exist, there are no properties. It is impossible, he
adds, to express any true sentence in which words (including 'is' or its
synonym) are used literally, and in which the subject would be some sin-
gular name of a property. No name designates a property because there
are no properties. There are, of course, pseudo-names of properties. It
does not follo~ from the denial that there are properties that snow is
not white. One may very well use such pseudo-names for brevity or pic-
turesqueness of speech. A pseudo-name of a property may in fact be
called an "abstract term" in the sense of Mill, Kotarbinski adds.
The reduction of the categories has shown, Kotarbinski claims, that
only the category of things remains, i.e., there are no other objects
besides things; every object is a thing, whatever is is a thing. When
metaphorical, abbreviative, figurative or (more generally) substitutive
assertions are supplanted with the aid of "definitive" sentences (which
are understood literally), there will not be any expressions which will
be names of anything other than things. These definitive sentences will
be about things exclusively. According to Kotarbinski, the doctrine of
the advocates of the reduction sketched above lIay be called "reism".
II
In the preciding I have attempted to give a complete summary of Kotar-

binski's views on pseudo-names as they are presented in the Eleaenty7.
It will have been noted that in some cases, e.g., in (2) and (3), Kotar-
binski's description of a pseudo-name depends upon the notion of a sub-
stituting-abbreviating sentence of the type 'A is B'. What precisely is.
a substituting-abbreviating sentence of the type 'A is B'? Unfortunate-
7Pseudo-naDles are also discussed (wIt.hout. significant. changes) in SODle

of Kot.arblnski's lat.er works. (1931>, (1936), (1966), (1968).
124 VITO F. SINISI
ly, Kotarbinski does not give an adequate explication of such sentences

ei ther in the Eleaenty or in his later works. I shall try to give an
explication of the notions of pseudo-name and substituting-abbreviating
sentence of the type 'A is B' which is consistent with his discussion of
the subject. (Hereafter I shall use 'SAS' as an abbreviation for 'sub-
stituting-abbreviating sentence'.)
It would be of relatively minor interest if Kotarbinski intended to
limi t his discussion of pseudo-names and SAS' s only to the instances
cited above. I think that his discussion is intended to cover all sen-
tences of a certain type, so that, e.g., to claim that (2) is "reduc-
ible" to (2') is to imply that every sentence of the form 'A is B' (in
which the subject term purportedly refers to a binary relation and the
predicate apparently characterizes that relation) is "reducible" to
a sentence which makes no reference whatsoever to relations but refers
exclusively to extended and resistant objects. Similarly, I believe that
when Kotarbinski claims that (3) is "reducible" to (3') he is also
claiming that every sentence of the form 'A is B' (in which both terms
seem to refer to classes) is "reducible" to a sentence referring
exclusively to things. I think that what might be called the reistic
general thesis of reducibility is that every meaningful sentence about
non-things is "reducible" to a sentence exclusively about things. This
thesis is of major philosophical interest since it marks an important
and relatively unknown stage in the development of nominalism.
We notice that the SAS' s of the type 'A is B' which Kotarbinski
cites are accepted as being true, and that each stands in some relation
to some other sentence which apparently is not a SAS and which is a sen-
tence about things. Furthermore, these latter sentences, e.g., (2') and
(3') are built up out of 'is' (in the fundamental sense), sentential
connectives, variables (apparently ranging over things exclusively), and
genuine names.
Let us consider (3) and (3'). In logical works containing a theory
of classes it is customary to define class inclusion in terms of class
membership:
df
(13) A s;: B === 'Vx (xEA .... xEB)
's;:' expresses 'are' and 'is' in contexts like 'Cats are animals' and
'A whale is a mammal'. The copu1a in (3) could be construed as
expressing class inclusion, and its terms interpreted as referring to
classes; we could say that (3) and (3') are, respectively, instances of
the definiendum and definiens of the definition schema (13), and that
(in an appropriate language) (3) and (3') are definitionally equivalent.
In such cases one usually speaks of the definiendum as being an abbrevi-
ation of the definiens, and one also says that the definiendum may be
anywhere substituted for the definiens. It would not be inappropriate,
then, to speak of the definiendum of the definition as being (in an
appropriate language) a substituting-abbreviation of the definiens.
Let us consider (2) and (2'). Suppose that in some language L
"transitive relation" is defined in the familiar way:
(14) R is a transitive relation.:!.!. 'Vx 'Vy 'Vz (xRy A yRz .... xRz).
If for 'R' we take 'older than', then we could say that in L (2) is
definitionally equivalent to (2'). Hence, (2) could be thought of as an
instance of the definiendum and (2') as an instance of the definiens of
(14); and it would be appropriate to say that in L the instances of the
definiendum of (14) are substituting-abbreviations of the instances of
the definiendum of (14).
I think that the relation between (2) and (2'), and (3) and (3') is
similar (though not identical) to that expressed in (14) and (13), re-
spectively, I shall try to show that a SAS is an instance of a definien-
dum of a definition schema in a reistic language (hereafter abbreviated.
as 'RL').
Consider (3). In some contexts 'whale' is a genuine name, e.g., in
(15) The largest animal in the aquarium is a whale.
The subject term of (15) may be construed as a singular name of a thing,

the copula as functioning in the so-called fundamental sense, and
'whale' taken to be a common name. If 'whale' is construed as a common
name, then to use it as the subject term of a singular proposition would
result in a false proposition since a necessary condition for the truth
of a singular proposition (according to Kotarbinski and Lesniewski) is
that the subject term be a singular name. If (3) is assumed to be in
some sense true, then the copula cannot function in its so-called funda-
mental sense in (3). Thus it seems necesary to introduce another copula
into RL. How shall this be done? I suggest that this be introduced via a
definition in much the same way as IS' is introduced into the, usual
theory of classes. Let X and Y be two expressions which could be con-
strued (in the proper context) as genuine common names. We introduce 'a'
into RL by means of the following definition:
(16) XaY~Vx(xisX-+xisY).
This is offered as a contextual definition of 'a', which may be read as
'is' or 'are'. The expression 'a' is defined in terms of what might be
called the "primitive" terms of RL, Le., the definiens contains vari-
ables ranging over things, occurences of the undamental sense of 'is',
and the usual logical connectives.
This explication, however, goes only part way in explaining the re-
lation between (2) and (2') since neither can be construed as an in-
stance of the definiendum or definiens of (16). If the relation between
(2) and (2') is that of definitional equivalence, it would seem that
what is being defined is the expression 'is a transitive relation'
rather than a use of the expression 'is'. If so, then (2) cannot be con-
strued as an instance of the definiendum of (16), as' (3) was, since (16)
offers a definition of the ,binary predicate 'a'. It would seem that a
different definitional schema is necessary to account for the defini-
tional equivalence of (2) and (2').
Consider the singular sentence
(17) John is older than Peter.

126 VITO F. SINISI
We may construe the copula as functioning in the fundamental sense of

'is', 'John' as a singular name, and 'older than Peter' as a common name
denoting, among others, John. Kotarbinski has suggested that 'older than
•.• ' is a name. The expression 'older than Peter' may be analysed into
two main parts: 'older than' and 'Peter'. The latter may be taken to be
a singular nl1llle, and the former may be construed as a nalle-creating
functor with one nominal argument, i.e., the expression 'older than ___ '
forlls a name when a name, e.g., 'Peter' is substituted for the blank. Of
course, 'older than' is not a name (according to Kotarbinski) although
the concatenation of 'older than' with a name forms a name.
Very often one can use a name-creating functor as the apparent sub-
ject of a sentence, e.g.,
(18) Older than is a transitive relation.

Generalizing on (18), if we let R be a nl1llle-creating functor with one
nominal argument (e.g., 'taller than', 'smaller than', 'a brother of'),
we obtain
(19) R is a transitive relation.

We can express (2')
(20) Vx Vy Vz (if x is older than y and y is older than z, then x

is older than z),
and generalizing on (20) we obtain
(21) Vx Vy Vz (if x is Ry and y is Rz, then x is Rz).

If we construe the variables in (21) as ranging over things exclusively,
and the copula as functioning in the fundamental sense, then (21) is
taken to be in what might be called the "primitive" nota:tion of RL. I
think it plausible to suggest, then, that in RL (19) is definitionally
equivalent to (21), and that Kotarbinski would accept the following as a
definition in RL:
(22) R is a transitive relation ~ Vx Vy Vz (if x is Ry and y is

Rz, then x is Rz),
If I am correct, then to say that (2) is a SAS for (2') is to say that
in virtue of (22), (2) is in RL definitionally equivalent to (2'), the
former being an instance of the definiendull of (22), and the latter
being an instance of the definiens of (22).
The discussion thus far suggests that it would be plausible to ex-
plicate the notion of substituting-abbreviation in the following way. In
RL an expression R is a SAS (of the type 'A is B') for an expression T
if and only if R and Tare definitionally equivalent (R being the defi-
niendull and T the definiens), where T is in the "primitive" notation of
RL, i.e., T is an expression built up out of 'is' (in the fundamental
sense), variables having things only as values, quantifiers, the usual
logical connectives of the propositional calculus, and (possibly) genu-

ine names. If this explication of SAS of the type 'A is B' is correct,
then an express ion in RL is a pseudo-name if it may be "meaningfully"
substituted for 'A is B' in a SAS of the type 'A is B'. Thus, for ex-
ample, we could construe (in RL) "proximity" and "symmetrical relation"
as pseudo-names if 'Proximity is a symmetrical relation' is a SAS of the
type 'A is B' for "if one object is near another, then the latter is
near the former". I believe that this explanation is consistent with
Kotarbinski's discussion of the relation between (Z) and (2'), and (.3)
and (3').
If the explanation proposed above is correct, then it seems clear
that there are at least two distinct kinds of SAS's of the form 'A is
B'. It would seem that only a complete formalization of R1 would reveal
precisely how many distinct kinds of SAS's of the type 'A is B' there
are.
Let us recall that the purpose of the discussion which has led us
to this point was to give an account of "SAS of the type 'A is B'" and
of "pseudo-name". It should be remembered that the discussion was limit-
ed to sentences like (2) and (2'), and (3) and (3'), hence if any light
has been shed on the notion of pseudo-name, it has been relative to
sentences like (2) and (3)·only.
Let us consider now two sentences analogous to (2):
(23) Divisibility is a transitive relation,

(24) Relative primeness is a symmetrical relation.
There are many sentences containing pseudo-names of relations which may

be reduced a
la Kotarbinski, e.g., 'parenthood is an asymmetrical rela-
tion', 'ancestor of is a transitive relation', 'motherhood is an intran-
sitive relation', 'brotherhood is an irreflexive relation'. The first
example could be "reduced" to 'if one object is parent of a second, then
the second is not parent of the first', and the others may be "reduced"
in a similar fashion to sentences about things. Yet (23) cannot be so
"reduced"; for it will not do to say that (23) is a SAS for
(23' ) If one object is divisible by a second object and the second
is divisible by a third, then the first object is divisible by
the third.
Nor will it do to say that (24) is a SAS for

(24' ) If one object is relatively prime to another object, then the
latter is relatively prime to the former.
(23') and (24') will not do because Kotarbinski stipulates that "every'
object is a thing, whatever is is a thing,,8, and a thing is extended and
resistant. No object (in Kotarbinski's sense) is divisible by another
object, and no object is relatively prime to another object; divisibili-
8EleDlent.y, p.67.
128 VITO F. SINISI
ty and relative primeness are not relations holding between objects,

they are relations holding between numbers, and numbers are not objects
(in Kotarbinski's sense).
This, however, is no objection to Kotarbinski's concept of SAS of
the type 'A is B', but it does indicate that (23) and (24) are not "re-
ducible to (23') and (24'). The problem, then, is to find two meaningful
sentences K and L such that K is equivalent to (23), and L to (24), and
both K and L are about things.
Another problem would be to find a sentence which is about things
and which is equivalent to the apparently true sentence
(25) The sperm whale is a species of whale.
Kotarbinski has said that (3) is equivalent to and is a SAS of the type
'A is B' for (3'). Both (25) and (3) are of the form 'A is B', and each
has a subject and predicate which are "class-names", and both (25) and
(3), according to Kotarbinski's criterion, are SAS's. (25) differs from
(3) in that 'is' in (3) takes as its arguments expressions which could
(in a different context) function as genuine common names, while 'is' in
(25) does not. (25) seems to affirm a relation of membership holding
between two classes; the class of sperm whales is a member of the class
Species of whale. The class Species of whale has as its members classes
of whales (not individual whales), e.g., the class of sperm whales, the
class of right whales, the class of hump-back whales, etc.
Kotarbinski has said that to say 'x is an element of the class of
M's' is equivalent to saying 'x is an M,.9 He nowhere places any
restriction on the type of name which may be substituted for 'x',
although in examples of 'x is an element of the class of M's' he uses
what are ordinarily called "individual" names as substituends, e.g.,
'Mars'. When we "reduce" 'the sperm whale is an element of the class
Species of whale' we obtain (25). Now, (25) contains as its subject term
a classname (or pseudo-name), and (25) is of the form 'A is B'. Since
sentences containing pseudo-names are SAS's, then (25) wo~ld seem to be
a SAS, and it calls for further reduction. We cannot "reduce" it in the
same manner as we "reduced" 'a whale is a mammal', for if we did, we
would obtain the evident falsehood
(26) Whatever is a sperm whale is a species of whale.
The problem, then, would be to find a meaningful sentence K which is
equivalent to (25), and which is about things.
In the preceding I have attempted to clarify Kotarbinski's concept
of SAS of the type 'A is B', and on the basis of this clarification I
have attempted to explain his concept of pseudo-name. I have also
attempted to show that there are a number of unresolved problems arising
from his concept of SAS of the type 'A is B'. I shall now consider other
types of SAS and pseudo-name which result from Kotarbinski's discussion
of the reduction of the ontological categories to the category of
things.
9Elementy, p. 14..
III
Kotarbinski 's general claim is that there is but one ontological cat-
egory, and that the putative categories of events, relations, and prop-
erties are "reducible" to the category of things. The basis for this
claim is that certain sentences, viz. (4), (5), (6), (7), (8)-(12), are
SAS's, surrogates for sentences which are about things exclusively.
A more general claim would be that not only these but all sentences like
these are surrogates for sentences about things. And a still more gen-
eral claim would be that liny meaningful sentence which purports to be
about events, relations, or properties is "reducible" to a sentence
about things exclusively.
Although Kotarbinski does not give or even suggest a general method
for translating all sentences purporting to be about events, relations,
and properties into sentences about things, it would seem highly desir-
able to prov ide a systematic method whereby any meaningful sentence
about these bogus entities can be shown to be "reducible" to a sentence
about things exclusively. I shall try to show that such a procedure (at
least for some sentences about relations and properties) -is forthcoming
by showing first that it is possible to formalize (using some elementary
devices of logic) a small fragment of a reistic language. I shall formu-
late a part of this language, and show that it gives the reist a general
procedure for systematically "reducing" at least certain types of sen-
tences, SAS's, about relations, and certain types of sentences, SAS's,
about properties to sentences about things. And I shall try to show that
this allows the reist to say that in certain cases the relation of being
a SAS is the relation 9f definitional equivalence within RL. I shall
suggest, for example, that in RL (6) and (6') are instances, respective-
ly, of the definiendum and definiens of a contextual definition of the
expression 'the relation higher than', and that (7) and (7') are in-
stances, respectively, of the definiendum and definiens of a contextual
definition of 'the relation older than'.
Just as Russell in his theory of definite descriptions showed how
one might use a seeming name without supposing the existence of any en-
tjty named, so too one can use, in an appropriate language, 'the rela-
tion higher than' and 'the relation older than' without assuming the ex-
istence of two relations. This will be possible provided that such ex-
pressions can be so defined in context that all reference to these rela-
tions are eliminated in the definiens.
In the following discussion "relation" is construed extensionally;
two relations are identical if they have the same extension.
We begin by assuming that the schema 'Aa,b' represents sentences in
RL which consist of a dyadic predicate A (e.g., 'is higher than', 'is
older than') and two singular names (in Kotarbinski's sense). (6') and
(7') would be instances of this schema. It may be said that a dyadic
predicate is true of ordered pairs of objects, and that the ordered
pairs of which a dyadic predicate A is true constitute a relation, the
relation of anything x to anything y such that Ax,y. With each dyadic
130 VITO F. SINISI
predicate A, then, we associate a relation R. Clearly, the relation R

will hold between an object a and another object b if and only if the
dyadic predicate associated with R is true of the ordered pair (a,b). We
shall abbreviate' holds between a and b' as '(.;.) a,b', and
abbreviate 'the relation of anything x to anything y such that Ax,y' as
'(Ax,y) (Ax,y)'. Hence, we shall consider '«Ax,y)(Ax,y» a,b' as a
paraphrase of 'Aa,b', and we introduce into RL the followirig defini-
tion schema:
df
(27) «Ax,y)(Ax,y» a,b =- Aa,b.
In light of (27) let us consider the relation between (6) and (6'),
and (7) and (7'). Be making appropriate substitutions into (27) we
obtain a contextual definition in RL, of 'the relation higher than'.
Thus, to say that (6) is a SAS for (6') would be to say that (6) is, in
RL, definitionally equivalent to .(6'): and the same may be said of (7)
and (7'), Le., by making appropriate substitutions we obtain a contex-
tual definition of 'the relation older than'. As Russell might say, we
have defined a use of 'the relation older than' and a use of 'the rela-
tion higher than'. Of course, there may be other uses of relation sym-
bols in RL: for these additional uses further definitions are requisite.
If this analysis is correct, then to say that 'the relation higher
than' and 'the relation older than' are pseudo-names is to say that in
RL these expressions are instances of '(Ax,y)(Ax,y)' appearing in (27).
Generalizing, we can say that in RL any instance of '(Ax,y)(Ax,y)'
appearing in (27) will be a pseudo-name, e.g., 'the relation brother
of', 'the relation larger than'. (27) shows how, in at least one ,con-
text, Kotarbinski may meaningfully use what appears to be the name of a
relation without assuming the existence of that relation. Expressions
apparentty referring to relations are pseudo-names, and are eliminable
by contextual definition.
If we examine (6) and (7), we note that each asserts that a binary
relation holds between a -couple of things (in Kotarbinski' s sense) • .It
should be pointed out that in his discussion of the reduction of the
category of relations to the category of things Kotarbinski does not
consider sentences which assert that a binary relation holds between a
couple of non-things, e.g., 'the relation greater than holds between 9
and 7'. The relations he considers in (6) and (7) are such that their
respective domains and counter-domains are sets of things. It would seem
that he means to confine his discussion to relations whose relata are
things. The definition schema (27) was suggested with this restriction
in mind; it is based on the assumption that the dyadic predicate takes
two singular naaes as arguments. A singular name, it will be recalled,
denotes one and only one thing. If this restriction is lifted, if, for
example, 'a' and 'b' in (27) took vacuous names as substituends as well,
then it would be possible to say that there is a sentence x which is a
SAS for a sentencey, and y is not a sentence about things. It would be
possible to say, for example, that 'the relation taller thBll holds
between Zeus and Apollo' is a SAS for 'Zeus is taller than Apollo".
Let us note that the definition schema (27) is idle for such con-
texts as
(28) The relation father of is identical with the relation male

parent of,
and
(29) The relation father of is distinct from the relation mother
of.
In short, (27) cannot be used to eliminate all occurrences of a pseudo-

nalle purporting to designate a binary relation. However, if the reist
wanted to use (28) and (29) without committing himself to the existence
of these relations, he could do so by introducing further definitions,
e.g., by introducing the schemata
df
(30) «h,y)(Ax,y) = (h,y)(Bx,y» -
Vw Vz «(h,y)(Ax,y» W,z :; «h,y)(Bx,y» w,z),
and
df
(31) «h,y)(Ax,y) J4 (Ax,y)(Bx,y» =-
-«Ax,y)(Ax,y) = (lx,y)(Bx,y».
He could, then, use (28) and (29) because he could always replace them,
via (30) and (31), by expressions whose variables range over things ex-
clusively.
While many sentences ostensibly referring to relations can be
handled in the manner indicated above, serious (if not insuperable) dif-
ficulties arise for the reist when he is confronted with sentences in-
volving quantification of relation variables, e.g., 'Some relation is
the converse of the relation greater than'. Two courses seem open to the
reist: to ban such locutions from his language, and thus to make it a
language incapable of expressing important logical truths, or to show
that such sentences can be "reduced" in some way to sentences about
things. Kotarbinski has never to my knowledge indicated how the reist
would "reduce" such sentences to sentences about things, so that in this
respect at least reism (like other nominalistic views) cannot hope to be
more than a program.
IV
Let us now consider Kotarbinski' s discussion of the reduction of the

categoryof properties to the category of things.
It is unsatisfactory merely to say, for example, that (8) is a SAS
for. (8'). What is desired is a fuller explanation of the relation be-
tween (8) and (8'), (9) and (9'), etc. I shall propose such an explana-
tion, examine it in the light of (8)-(11), and (8')-(11'), and consider
its limitations. Again, which only accounts for the displayed sentences
and no others would be of slight interest. What is wanted is an explana-
tion which systematically accounts for the relation among sentences of
certain types.
132 VITO F. SINISI
In the following discussion we shall assume that "property" is con-

strued extensionally. We regard two properties as being identical when
their instances are the same". Hence, a property may be thought of as a
class.
We shall also assume that the schema 'Aa' represents sentences .in
RL consisting of a monadic predicate A and a denoting name (in Kotarbin-
ski's sense) a. It may be said that the various things of which a monad-
ic predicate A is true have a -common property: the property of all x's
such that Ax. The things of which the predicate 'is white' is true, for
example, may be said to have the property of white or whiteness. Now it
is clear that a thing a can be said to have a given property, the
property of all x's such that Ax, if and only if the monadic predicate
A, which is associated with the property expression , is true of a. Let
us abbreviate 'the property of all x's such that Ax' as '(Ax)(Ax)', and
let us indicate that a property is ascribed to an individual thing a in
the following way: '( ... )a'. Since a has a given property if and only if
the monadic predicate A (associated with the property expression) is
true of a, we may considel' '( (Ax) (Ax) )a' -as a paraphrase or substitute
for 'Aa'. Generalizing, we introduce the following definition schema:
df
(32) ( (Ax) (Ax) ) a - Aa
We must emphasize a restriction mentioned earlier, viz. that the

individual variable 'a' range over things exclusively. This restriction
is dictated by Kotarbinski's stipulation that a SAS which contains what
appears to be the name of property is a surrogate for a sentence which
is about things. If we ignored this restriction, it would be possible to
say, for example, that 'Having a beard is a property of Zeus' is a SAS
for 'Zeus has a beard', which clearly is not a sentence about a thing
since 'Zeus' is not a denoting name of a thing.
The schema above allows Kotarbinski to use a seeming name of a
property without assuming the existence of any property. Clearly, the
application of the above definition schema is limited; it does not, for
example, apply to contexts in which two properties are said to be ident-
ical or to be different, although it would be easy to provide further
definition schemata to accomodate these cases.
Let us now consider the schema (32) in the light of (8)-(11), and
(8')-(11'). If we substitute the denoting name 'snow' for 'a' and
'white' for 'A' in the schema, we obtain (8) as an instance of the defi-
nienduIB. The same substitutions into the definiens of (32) gives us
(8'). Hence, we may assert, on the basis of (32), that to say (8) is a
SAS for (8') is to say that (8) is, in RL, definitionally equivalent to
(8'). Making the proper substitutions into our schema we show that (10)
is definitionally equivalent to (10°). If the analysis is correct, then
to say that 'whiteness' and 'the property of attracting iron' are, in
(8) and (10), pseudo-nues is to say that in RL these expressions are
instances of '(Ax)(Ax)', appearing in (32). Generalizing, we can say
that in RL any instance of '(Ax) (Ax)', appearing in (32), will be a
pseudo-name, e.g., 'the property man', 'the property animal'. (32) shows
how, in at least one context, Kotarbinski may meaningfully use what
appears to be the name of -a property without assuming the- existence of
that property. Expressions (in these contexts) apparently referring to

properties are pseudo-names, and are eliminable by contextual defini-
tion.
We cannot say that (9') and (11') are instances of our definiens.
For clearly (9') .and (11') are not sentences of the same form as (8')
and (10'), nor are (9) and (11) of the same form as (8) and (10). If we
agree that the form of (8') and (10') is exemplified by the schemata
'Ax', then it seems plausible to say that the form of (9') and (11') is
exemplified by 'Vx (Nx ~ Gx)'. If the form of (8) and (10) is exemplifi-
ed by the schema '«\x)(Ax»a', then it seems plausible to say that the
form of (9) and (11) is exemplified by 'Vx (Nx ~ «\y)(Gy»x)'. If this
is so, then (9) and (11) are not instances of the definiendulB, and (9' )
and (II') are not instances of the definiens of (32). However, in view
of (32), 'Vx (Nx ~ «\y)(Gy»x)' could be reduced to 'Vx (Nx ~ Gx)', and
we can plausibly claim that (9) and (11) are, in RL, instances of the
schema 'Vx (Nx ..... «\y)(Gy) )x)', while (9') and (11') are instances of
'Vx (Nx ..... Gx)'. Thus, to say that 'the property of generosity' and 'the
attribute of extension' are, in (9) and (11), pseudo-names is to say
that in RL these expressions are instances of '(\y)(Gy)', appearing in
'Vx (Nx-~ «ly)(Oy»x)'. We can say that (9) and (Ii), and all sentences
of similar form, are reducible to sentences about things if we assume
that the variable in 'Vx (Nx ~ Ox)' ranges over things exclusively.
Let us consider the SAS (12). We have noted that Kotarbinski does
not give the corresponding "thing" sentence. (12), like (9) and (11),
does not have the form of the definiendum of (32). Furthermore, 'plane
triangle' is not the name of a thing. For what sentence could (12) be a
surrogate? Since (11) is a SAS for (11') it might be suggested that (12)
is a SAS for 'every plane triangle has the sum of its interior angles
equal to the sum of two right angles'. If this is so, then we have not
produced a sentence about a thing which is equivalent to (12) since
'every plane triangle .•• ' is not a sentence about a thing. Once, again,
the reist is obliged either to ban such sentences as (12) from RL or to
show how sentences l-ike (12) are to be reduced to sentences about
things.
Just as Kotarbinski did not consider sentences asserting that a
binary relation holds between a couple of non-things, so too, in his
discussion of sentences purporting to be about properties he does not
indicate how sentences asserting that some non-thing has some property,
e.g., 'primeness is a property of the number three i , can be reduced to
sentences about things exclusively. For such sentences the definition~
offered above are clearly inoperative. ,
There is even more crucial deficiency. While a number of sentences
which seemingly refer to properties or classes may be accomodated by
using the techniques developed above, the reist meets serious difficul-
ties when he encounters sentences about properties or classes which
involve quantification on property or class variables, e.g., the defini-
tion of ancestor in terms of parent - x is ancestor of y if x belongs to
every class which contains y and all parents of its own members. Another
example would be the statement of the principle of mathematical induc-
tion - for any property P, if zero has the property P, etc. Once again,
then, the reist (like the' nominalist) Bust either forswear s.entences
134 VITO F. SINISI
involving essential quantification of class or property variables, or

show that such sentences can in some way be reduced to sentences about
things. Kotarbinski has given no indication in his published works as to
how such cases are to be treated.
I have presented Kotarbinski' s views on pseudo-names and substituting-

abbreviating sentences, and I have attempted to clarify and (for some
cases) systematically develop these concepts. I have also indicated some
of the more obvious limitations inherent in such a systematic develop-
ment. While KotarbiIlski's views are now perhaps only of historical
interest, nevertheless they do reveal, I believe, that during the 1920's
he was making interesting and significant contributions to nominalism
and nominalistic semantics lO . It seems clear that Kotarbinski thought one
of the main tasks of reism (and nominalism) to be the reduction of sen-
tences containing pseudo-names to sentences about things. There are some
similarities between this view and more recent appraisals of the task of
nominalism. In one of his early articles Quine, for example, says "As a
thesis in the philosophy of science, nominalism can be formulated thus:
it is possible to set up a nominalistic language in which all of natural
science can be expressed. ,,11 Such a language would be a formalized lan-
guage, and the nominalist claims that a language of this kind can be
10Th .. El .. m .. nt.y v .. r .. publish .. d in 1929, but. as .. arly as 1920 Kot.arbinski,

in his art.icl .. (1920>, vas arguing against. t.h . . . . xist.ence of fict.ional
charact. .. rs, po .. t.ic .. v .. nt.s, imaginary obj .. ct.s, mat.h .. mat.ical obj .. ct.s,
classes, set-s, universals, propert.ies, and relat..ions. He reject.ed prop-
.. rt.ies and r .. lat.ions in much t.h .. sam .. vay as h .. r .. j .. ct. .. d t.h .. m in t.h ..
El .. m .. nt.y. H.. says (pp. 164--166) "Wh .. n v .. say 't.h .. prop .. rt.y vhit. .. ness
inh .. res in snoy' or 't.he propert.y vhit.eness applies t.o snoy' or 'snow
has t.h .. propert.y vhit.eness' et.c., y .. say neit.her more nor l .. ss t.han t.hat.
snov i .. whit. ... And t.h .. propert.y melt.., just. os snow m .. lt.s. There are
simply no prop .. rt.i .... in t.h .. vorld; but. t.h .. r .. or .. expr .. ssions in vhich
t.he words ·propert.y', ·"hit.enes.' or 'just.ice' .t.e. occur, circumlocut.o-
ry, subst.it.ut.iv • • xpr .. ssions which or .... quival.nt. t.o s .. nt..nc .. s fr ... of
such .. l .. m .. nt.s. And,. only t.h.se s .. nt. .. nces hav .. a lit.eral m .. aning .... There
ar .. simply no relat.ions in t.h .. yorld; but. t.h .. re or .. expr .. ssions in vhich
t.he words (relation', It.-he JDajorit.y', 'brot.herhood' et.c. ocour, circum-
locut.ory, sub .. t.it.ut.iv . . . xpr.ssions vh·ich or . . . quival .. nt. t.o s.nt. .. nc .. s
fr .... of such .. l .. m .. nt.s. Th .. r .. is no r .. lat.ion of s.niorit.y b .. t.w .. en a fa-
t.her and his son, alt.hough a fat.h .. r is old .. r t.han his son. Hoy .. ".r, v ..
or .. fr .... t.o say t.hat. t.h .. r .. lat.ion of s .. niorit.y holds b .. t.veen fat.her and
son provided t.hot. y .. r .. m.mber precis .. ly vhat. we are .,aying in t.his vay;
ve are saying no mol''' and no less t.hat. t.h. fat.her is older t.han his son.
That.'s all."
U Quin & (1939). All quot..d pa .... ag ... in t.his paragraph are from Quine's
aJ-t.icle.
construed in such a way that "its variables admit only concrete objects,
individuals, as values - hence only proper names of concrete objects as
substituends. Abstract terms will retain the status of syncategorematic
expressions, designating nothing, so long as no corresponding variables
are used." Yet, a nominalist, in Quine's sense, is free to use variables
taking abstract terms as substituends provided "that he can explain this
usage away as a mere manner of speaking". Quantification of this kind
"can often be introduced by a contextual definition - a mere convention
of notational abbreviation". If the nominalist can devise contextual
defini tions explaining quantification with respect to entities of an
abstract kind as a mere abridged manner of speaking,
he becomes justified in speaking as i f there were such entities with-
out really forsaking his nominalism. The entities remain fictions for
him; his reference to such entities remains a mere manner of speak-
ing, in the sense that he can expand this sort of quantification at
will into an official idiom which uses only variables having proper
names of individuals as substituends. But if the nominalist can not
supply the relevant contextual definitions, then his nominalism for-
bids his use of variables having abstract entities as values. He will
perhaps still plead that his apparent abstract entities are merely
convenient fictions; but his plea is no more than an incantation, a
crossing of the fingers, so long as the required contextual defini-
tions are not forthcoming.
The similarities between Kotarbinski and Quine are readily

apparent, but so are the differences; the main difference is that Quine
states the nominalistic thesis with reference to a formalized language,
while Kotarbinski seems to be concerned with the reistic (and nominalis-
tic) analysis of sentences from colloquial language. In recent discus-
sions Kotarbinski explicitly acknowledges the programmatic character of
reism. He now speaks of the basic "postulate" of reism: "Let us make it
a point that no sentences in its final form (that is, freed from meta-
phors, abbreviations and substitute expressions) should include any
pseudo-name (that is, an expression that has the appearance of a name
but is not a name of any physical object) • .,1:.J
It is unfortunate, I think, that the work of such a major philoso-
pher as Kotarbinski is so .little known outside of Poland. I hope that
this sketch and analysis of his theory of pseudo-names will help stimu-
late an interest in his work.
1~rom an unpubliahed lect.ul'e, "The Basic Poat.ulat.e of Concret.iam" [aee

1C0t.al'bin8ki, 1961 - Ed.l, delivel'ed at. Penn8ylvania St.at.e Univerait.y on
Apl'il 9, 1969. I wiah t.o t.honk Profesaor ICot.arbinski for a copy or t.hia
lect.ure. See also Kot.arbin8ki (1966).
ON THE PHASES OF REISM 1
Barry Smith
iii 1. Introduction
Along with almost all the more important Polish philosophers of the
twentieth century, Kotarbinski, too, was a student of Kazimierz
Twardowski, and it is Twardowski who is more than anyone else
responsible for the rigorous thinking and simplicity of expression that
is so characteristic of Kotarbinski' s work. Twardowski was of course
himself a member of the Brentanist movement, and the influence of
Brentanism on Kotarbinski's writings reveals itself clearly in the fact
that the ontological theories which Kotarbinski felt called upon to
attack in his writings were in many cases just those theories defended
either by Twardowski or by other thinkers within the Brentano tradition.
Lesniewski, too, inheri ted through Twardowski an interest in Brentano
and his school, and as a young man he had conceived the project of
translating into Polish the Investigations on General Grammar and
Philosophy of Language of Anton Marty, one of Brentano's most intimate
disciples. Lesniewski, as he himself expressed it, grew up "tuned to
general grammar and logico-semantic problems la Edmund Husserl and the a
representatives of the so-called Austrian School" (1927/31, p.9).
The influence of Brentanism on Polish analytic philosophers such as
Kotarbinski and Lesniewski has, however, been largely over looked
principally as a result of the fact that the writings of the Polish
analytic school have been perceived almost exclusively against the
background of Viennese positivism or of Anglo-Saxon analytic philosophY.
It is hoped that the present paper, following in the footsteps of Jan
Wolenski's recent work 2 , might do something to help rectify this
imbalance. The paper will consist of a critical survey of Kotarbinski's
development from his early nominalism and 'pansomatistic reism' to the
later doctrine of • temporal phases '. It will be shown that the surface
clarity and simplicity of Kotarbinski's writings mask a number of
11 am grateful to Audoenus Leblanc, Czeslav Lejevski, Dieter Munch, Karl

Schuhmann, Peter Simons and the editor for helpful comments on an
earlier version of this paper.
:lSee WOlenski (1989), and compare also Schnelle (1982) and the papers by
Schnelle in Cohen and Schnelle (1986).
137
I. Wolelfski (ed.). Kotarbmski: Logic. Semantics and Ontology. 137-183.
@ 1990 Kluwer Academic Publishers.
138 BARRY SMITH
profound philosophical difficulties, connected above all with the

problem of giving an adequate account of the truth of contingent
(tensed) predications. The paper will examine in particular the attempts
to resolve these difficulties on the part of Lesniewski. It will
continue with an account of the relations of Kotarbinskian reism to the
ontology of things or entia realia defended by the later Brentano.
Kotarbinski's identification of Brentano as a precursor of reism is, it
will be suggested, at least questionable, and the paper will conclude
with a more careful attempt to situate the Brentanian and Kotarbinskian
ontologies within the spectrum of competing ontological views.
§2. Stages in the Development of Reism
We shall be concerned, in the first place, with Kotarbinski's magnum

opus, the Elements of the Theory of Knowledge, Formal Logic and Method-
ology of the Sciences, first published in 1929 and hereafter referred to
as Elementy. The principal doctrine expounded and defended by Kotarbin-
ski in this work is that of 'reism', a doctrine according to which all
existence is made up entirely of individual things, realia or concreta.
A more specialised version of the doctrine is referred to by Kotarbinski
as the doctrine of 'somatism' (or sometimes also 'pansomatism'), which
results when one adds the thesis that individual things are to be
identified in every case as physical bodies - a thesis which Kotarbinski
also accepts. In an essay of 1958 appended to the second edition of his
Elementy, Kotarbinski speaks retrospectively of seven stages in the
development of the reistic theory, from his own early acceptance of
nominalism - which he himself preferred to call • concretism' - to the
working out of a full-blown pansomatist ontology in the 1930s3 •
It is especially in relation to the early stages that Twardowski's
influence is most strongly felt. Stage 1 consists in the rejection of
universals, properties, or general objects. All entities are individ-
uals, on this ('concretist') view, though it does not thereby follow
that they must all be things. Kotarbinski' s adoption of nominalism in
this sense may be attributed on the one hand to the effects of his early
exposure to the thinking of the British empiricists at the hands of
Twardowski. On the other hand however it can be seen as a reaction
against Twardowski's own thesis that there are general objects, objects
which result when the features common to the particular objects 'falling
under a given concept are "unified into a whole,,4. A discipline like
geometry, in Twardowski's view, is concerned precisely with general
objects in this sense (Triangle, Circle, Square, and so forth), and a
similar thesis may be extended to the other sciences. As Husserl points
out in his Logical Investigations, general- objects as conceived by
Twardow$ki are subject to all the disadvantages of the Lockean general
3J(ot.Ql'bi~ski distinguished also an eight.h st.age, in which I'ei$m is no

long .... a t.h .. o .. y but. .. at.h .... a 's .. ..;ant.ic p .. obl .. m'.
4Twa .. dowski <1894" p.l06, Eng. p.100)

triangle5 • This suggestion is taken still further by Lesniewski,

Kotarbinski's colleague in Warsaw, who offers a proof that Twardowski's
theory (together with a range of similar theories, including Husserl's
own) is contradictorl, and Kotarbinski would later claim that it was
only concerning universals or general objects that nominalism "succeeded
in convincingly proving their non-existence by a reductio ad absurdum".
(1966, p.55)
Stage 2 consists in the rejection of events, processes, states of
affairs, and other putative individuals falling outside the category
thing. This, too, may be interpreted as a reaction on Kotarbinski's part
to the Brentanist views of his teacher Twardowski, given that the
ontology of states of affairs or Sachverhalte was, in the first decades
of this century, a quite peculiar preserve of the Brentanist movement7 •
Stage 3, the rejection of sets or classes, reflects the influence
of Lesniewski, and above all of Lesniewski's criticisms of the theory of
sets8 • The set-theoretical antinomies had resulted, in Lesniewski's
view, not from any inherent contradict ion in the notion of set as
originally conceived by Cantor, but from a departure from this notion in
the direction of a conception of sets as abstract entities. As Cantor
puts it: "Every set of well-differentiated things can be conceived as a
uni tary thing in which these things are constituent parts or
constitutive elements" (Cantor 1887/88, p.379). Thus for example a
musical composition is a set consisting of the sounds which are its
consti tuents, a painting is a set consisting of various patches of
colour. A set, therefore, on Lesniewski's interpretation of Cantor's
views, is a concrete whole made up of concrete parts, not an abstract
enti ty sealed off from changes in the real world of material things9 •
According to Lesniewski, it would be correct to say, for example, that
the Black Forest is just the set of trees now growing in a certain area,
and that this set becomes smaller as trees within it die. Clearly, on
this view, there can be no empty set, and a set consisting of just one
object as member will be identical with that object. Further, there can
be no sets of higher type, which means also that there is no way in
which the more usual antinomies may be generated.
Frege, too, as Lesniewski points out, attacks those mathematicians
who introduce into their theories such arbitrary t inventions' as the
5<1900/01, voL II Ii 1t)

6 S .... L .. sni .. wski <1913, p.319), and also th.. summary of L .. sniewski's
argum .. nt in Kotarbinski (1920), and Lej .. wski <1979, pp.200f). As
Wol .. nski (forthcoming) shows Lesniewski was influ .. nc .. d here by Marty's
criticism of Huss .. rl's Platonism in his Inv .. stigations of Gen .. ral
Grammar (1908, §71).
7Smith (1989, 1990).
8 See L .. sni .. wski (1914-, 1927/31). It may b.. also that L .. sniewski's
criticism of prop .. rti .. s in his (1913) help .. d to provok .. Kotarbinski's
initiol nominalism.
9L .. sni .. wski <1927/31, p.17), citing Cantor <1887/88, pp.4-21f).

140 BARRY SMITH
empty set, merely because they prove expedient for certain purposes 10
Lesniewski's own strictures in this respect are directed in particular
against axiomatic theories of sets such as were developed by Zermelo.
These do not merely lack the sort of naturalness that would dispose one
to accept them; they lack also that intrinsic intelligibility which
would make their meaning clear, so that Lesniewski can in all honesty
assert that he does not understand what is meant by 'set' as this term
is supposed to be 'implicitly defined' by theories like Zermelo's11.
Lesniewski, himself, in contrast, starts not from 'inventions' or
from axioms or hypotheses selected for pragmatic reasons, but from what
he calls intuitions, commonly accepted and meaningful to all, relating
to such concepts as whole, part, totality, object, identity, and so on 12 •
The language of Lesniewski's theories is therefore an extrapolation of
natural language, a making precise of what, in natural language, is left
inart.iculate or indistinct. His work forms part of that strand in the
development of logic, represented also by Frege and by the early
Russell, which sees logic as a descriptive enterprise, part and parcel
of the attempt to produce formal theories adequate to and true of the
actual world 3. Hence he is mistrustful, too, of the model-theoretic
semantics that has been built up on an abstract set-theoretical basis,
and he is opposed also to the work of those formalist logicians who
embrace an essentially abstract-algebraic approach to logic, or see
logic as having to deal essentially with un interpreted formal systems 14 •
St.age 4 in the development of reism, which consists in the
rejection by Kotarbinski of mental images and other 'immanent contents',
again reflects the influence of Twardowski. The status of mental
entities was an issue of particular importance to Kotarbinski, since it
marked one of the very few areas of disagreement between himself and
Lesniewski. For Lesniewski admitted contents and images into his
ontology, remaining in this respect faithful to the heritage of Brentano
and Twardowski 15. Thus Lesniewski, in this sense and perhaps also in
others, is not a reist. Since, however, like Brentano, Marty and
Twardowski, he held that contents are concrete items, tied, in effect,
10Lesniewski <1927/31, p.18), cit.ing Frege <1983, pp.U·., Eng. p.3t>.
11Le s niewski <1927/31, p.22).
1~esniewski <1927/31, p.24.).

13Lesniewski <1929, pp.6, 78), Lejewski <1958, pp.123f.).
14In t.erest.ingly, Tar ski, at. least. in his early years, up t.o and including
his paper on '''The Semant.ic Concept.ion of Trut.h",. agreed wit.h Lesnievski
in t.his (see esp. pp.34.2f. of Tarski 194.4.). Tar ski, be it. not.ed, was
never a formalist.: Tar ski and Lesniewski part.ed company rat.her because
Tar ski came gradually t.o accept. t.he Use of set. t.heory and infinit.ist.ic
met.hods in his work.
15See Twardowski <1894., §§ 1-2>. Lesniewski did not. himself develop a

~~ of cont.ent.s, since he held t.hat. t.he problems involved would be
t.oo difficult. t.o allow him t.o achieve t.he appropriat.e degree of
t.heoret.ical rig our.
to specific mental episodes, his acceptance of contents does not imply a

departure from nominalism or 'concretism'.
Stage 5, which consists in the awakening of Kotarbinski's interest
in certain precursors of his own way of thinking, was provoked by the
discovery of what· he took to be reist tendencies in Brentano's later
work - a matter which will be dealt with in more detail below.
Stage 6 consists in an amendment to the reist doctrine, provoked by
criticisms put forward by Ajdukiewicz in his 1930 review of the
Elementy16. These criticisms concern in particular the question as to how
the negative theses of reism ('properties do not exist', 'events do not
exist', and so on) are properly to be treated. Are such formulations to
be accepted as they stand as literal renderings? Certainly not,
Ajdukiewicz claims, if 'exists' is taken in the literal sense - the
sense it has in sentences like 'rabbits exist', 'dinosaurs no longer
exist' and so on. For the subjects of such sentences are in every case
the names of things, which is ex hypothesi not the case where we have to
deal with expressions like 'property', 'event', and so on. Yet the reist
allows no other sense of 'exists'.
Kotarbinski himself initially responded to this criticism by taking
up Ajdukiewicz' s suggestion that the negative theses of ontology be
reformulated on the level of semantics, as theses to the effect that
certain kinds of apparent statement are nonsensical 17 • This solution is
unsatisfactory for a number of reasons. It implies, first of all, a view
of 'words with negative meanings' which would seemingly allow them to be
employed as a means of constructing well-formed sentences more or less
at random from nonsensical strings of words. Further, as Lejewski points
out in his paper "On the Dramatic Stage in the Development of Kotarbill-
ski's Pansomatism", it implies that "the negative theses of ontological
reism fail to say anything about reality because they are merely
statements about the language of the reist" (1979, p.200). A semantical
doctrine in this sense must however, as Lejewski argues, presuppose a
prior ontological doctrine: "Semantics without ontology is like a house
without foundations. It collapses into a set of arbitrary injunctions
and prohibitions justified by ad hoc considerations" (1979, pp. 205f.).
Moreover, how, in the absence of some more deep-seated ontological
theory, could the reist be assured of the truth of his semantic claim
that all onomatoids will vanish in ultimate formulations? And how could
he account for the fact that, as he will want to insist, translation
into the language of things is both natural and clarificatory?
There is, however, an alternative response to Ajdukiewicz's criti-
cism. This is to accept that the reist's negative theses make good sense
(are in good grammatical order) as they stand, not, however, in the lan-
guage of the reist but in the language of his opponent, i.e. of someone
who accepts both a multicategorial ontology (accepts categories other
than that of thing) and the concomitant multicategorial language. For
if, as Lejewski puts it, the multicategorial ontologist's assertions
16See t.he d .. t.ail .. d account. in L .. j<>wcki (1979). Kot.al'binski's init.ial

I'''oct.ion t.o Ajduki .. wicz's cl'it.icism was in POI't. incpil' .. d by COl'nop.
17Ct . Kot.al'binaki (1966, p.433).

142 BARRY SMITH
"are made in terms of a multicategorial language, the same language must

be used to negate those assertions". Propositions such as 'there are no
properties', 'there are no relations', 'there are no events', etc., are
properly
to be understood in the light of the multicategorial idealization
of natural language. And on this assumption the nouns 'property',
'relation', 'event', etc. belong to different fundamental semantic-
al categories, which in turn determine the semantical category of
the expression 'there are no' in each of the conjunts.
(Lejewski 1979, pp.211f.)
The reist, we might say, can accept his opponent's multicategorial

language as a ladder, to be thrown away when once it has served its
polemical purpose.
Stage 7 sees the re-insti tution of reism as an ontological doc-
trine, founded on a recognition of the need to supply non-tautological
definitions of notions such as 'thing', 'object', 'body', etc. This
development, too, was provoked by a criticism of Ajdukiewicz, a criti-
cism to the effect that, if 'exists' has a literal sense only wheri used
in conjunct ion with names for things, then the posi ti ve statement of
reism, to the effect that only things exist, is tautological. It is
equivalent to the truism: 'only things are things'.
Here, also, Kotarbi~ski's initial reaction was one of retreat to
semantics. Later, however, he responded to Ajdukiewicz's objection in a
more ontologically-minded fashion, by seeking definitions of concepts
such as 'thing', 'object', 'body', etc., in a way which he hoped would
render non-tautological the fundamental theses of reism and somatism.
Since a formal statement of such definitions has been provided by
Lejewski in his just-mentioned paper, it will be sufficient here if we
examine briefly (and critically) the concepts Kotarbinski here employs.
All reality, according to Kotarbinski, is composed exclusively of
things, and things are in every case bodies. Kotarbi~ski initially
sought to define body as that which is extended in space and time, as
that which is 'bulky and lasting'. Then, however, he saw reason to add
the further condition that bod ies are 'inert'. Certainly it would be
stifficient, Kotarbinski holds, to define 'body' as "that which is
extensive":
But in order to avoid misunderstandings which might lead someone to

suppose, on the strength of that definition, that physics is also
concerned with 'fragments of empty space' (which also seems to be a
hypostasis), we prefer to narrow the definition as to intension -
without thereby, as we think, narrowing its extension - by adopting
the formula stating that 'a body is what is extensive and inert.'
(1966, p.330)
In other contexts Kotarbinski preferred to define body as: that

which is extensive and such as to offer resistance. All is not quite
clear, however, about the application of either formula. Thus Kotar-
binski is on the one hand keen to insist that the term 'body', as he
understands it, embraces not only planets, rocks, etc., but also objects
investigated by physics "such as electrons, protons, magnetic fields"
(1966, p.331). On the other hand, however, he stresses that it excludes
for example "immanent coloured patches". Consider, however, the example
of a glass cube that is uniformly red in colour. Is the transcendent
redness of this cube (an individual three-dimensionally extended moment
of colour), a body, on Kotarbinski's view? Certainly it is bulky and
lasting and, perforce, such as to offer resistance. Kotarbinski, it
would seem, was able to ignore such cases in framing his account of
'body' only because his attentions were concentrated on instances of
surface colour, entities which fall short of three-dimensionality and
can be excluded on this count. In order to rule out examples like the
cube of colour, Kotarbinski would have to add the condition that a body
is that which exists (is extended and inert) in its own right (has need
of no other thing in order to eXist)18. As we shall see, a condition of
this sort is very much in the spirit of Aristotle. Certainly such a
condition would capture the sense in which the given example gives
grounds for suspicion - that the cube of colour exists merely as a
dependent moment of the cube of glass, and enjoys no separate existence.
Yet how are we to formulate the condition in question in such a way that
it would not rule out other examples which we would wish to count as
bona fida bodies? Does a human being, for example, exist 'in his own
right', given that he has need, for example, of nourishment, and
processes of breathing and metabolising (to say nothing of parents), in
order to exist? How, moreover, are we to make precise the sense of
'other' in 'has need of no other thing'? Simple non-identity will not
do, since everything may in this sense stand in need of its own proper
parts in order to exist. On the other hand spatiotemporal discreteness
or disjointness will not serve, either, since the cube and its colour
would seem to coincide in space and time. All that can be said here is
that considerations such as this have exercised Kotarbinski (and
Lesniewski, et. a1.) too little, so that the project of a somatist
ontology still leaves much to be desired in terms of a clear statement
of what is meant by 'body'.
18some t.hing .imilar would be r.quir.d t.o exclude from t.he r.alm of t.hing.
01.0 cert.ain .ort.. of' event.s. Con.ider, f'o.. example, a .. ot.at.ion of a
met.al .phe .... Thi . . . ot.at.ion i . ext.end.d in .pace and t.ime and, again, it.
is such as t.o off' ... re.ist.ance. Not.e t.hat. what. i . est.ablished by t.he
a .. gument. in t.he t..xt. is me .. ely t.hat. t.he canonical .. ei.t.ic not.ion of
~ is indet. ... minat.. in it.s applicat.ion. The a .. gument. is not. designed
t.o .how t.hat. on. could not. deal .at.isfact.orily wit.h colou .. s (0 .. t.hree-
dimensional shapes or mas.es of sound or heat.) wit.hin t.he Le8niew.kian
framewo .. k. As Lejewski has .uggest.ed, Ju.t. a. Chronology and St.e .. eology
(t.heories of' t.ime and space) can in p .. inciple be obt.ained f'rom Me .. eology
by t.he addit.ion of cert.ain ext.ra-Iogical con.t.ant.s, .0 it. would be
po.sibl. t.o conceive a di.cipline of" Colourology, obt.ain.d by adding
const.ant. t..rm. such 'r.d', 'blu.', et.c., and a relat.ional pr.dicat.e .uch
ae 4is t.he 8ame colour as".
144 BARRY SMITH
§3. Reisll and Truth
Kotarbinski's reism is, as we have seen, a doctrine according to which

all existence is made up entirely of individual things. At the same time
he defended in the Elellenty a form of the correspondence theory of truth
derived from his teacher Twardowski. Twardowski himself had at least at
one stage in his career come to the conclusion that a conception of
truth as correspondence requi res special 'states of affai rs', unitary
entities which would stand to sentences or acts of judgment in something
like the way in which things or objects in the narrow sense would stand
to names or acts of presentation 19 • Not only Twardowski, but also
Husserl, Meinong and Marty, as well as the early Moore, Russell and
Wittgenstein were drawn to similar views. How, then, was it possible for
Kotarbinski to maintain a correspondence theory of truth and at the same
time embrace the view that there are no entities other than things?
Kotarbinski himself holds that this dilemma may be resolved by re-
jecting that ontological interpretation of correspondence (derived from
the scholastics) which would interpret truth in terms of 'copies' of re-
ality existing somehow in the mind of the judging subject. He advances,
rather, what might nowadays be called an adverbial theory of truth:
The point is not that a true thought should be a good copy or sim-
ile of the thing of which we are thinking, as a painted copy or a
photograph is. A brief reflection suffices to recognize the meta-
phorical nature of such a comparison. A different interpretation of
'accordance with reality' is reQ,uired. We shall confine ourselves
to the following: 'John thinks truly if and only if John thinks
that things are so and so, and things in fact are so and so'.
(1966, pp.106f.)
He came, in other words, to iflterpret the correspondence theory in
the superficially neutral terms of the Aristotelian "to say of what is
that it is not, or of what is not that it is, is false, while to say of
what is that it is, or of what is not that it is not, is true,,20. Con-
ceptions of truth in terms of the copy theory are to be avoided, from
Kotarbinski 's' point of view, not merely because they involve a hypo-
statisation of states of affairs or other special entities on the side
of the object; they comlli t us also, on the side of the subject, to
19S .... my (1989) for an account. of Twardowski's t.h .. ory of st.at. .. s of

affairs. Lat.er, however, Twardowski crit.icised t.heories of t.rut.h, such
as t.hat. of Russ .. ll, which employed cat. .. gories such as st.at. .. of affairs
or fact..
10Met.., 1011 b 16ff. Less neut.ral is Arist.ot.le's r·emark, somewhat. lat.er

in t.he Met.aphysics, t.o t.he effect. t.hat. t.rut.h and falsehood depend ·on
t.he side of t.he object.s on t.heir being combined or separat.ed, so t.hat. he
who t.hinks t.he separat.ed t.o be separat.ed and t.he combined t.o be combined
has t.he t.rut.h, while he whose ~ U m a~.taJ.e c~ ~ .uu:u
~{
~ ~J!J-j4Bcu is in error". (1061 b 3, emphasis supplied.)
'immanent contents', 'thoughts', 'judgments', 'propositions' or 'mean-

ings' - and all of these terms are mere fal]ons de parler, to be
eliminated from any language adequate to the purposes of philosophy.
When I judge truly, then I judge in accordance with the things, and that
is all that need be said.
Can matters really be so simple, however? Certainly in the case of
judgments expressed by posi ti ve existential sentences such as 'John
exists' or 'cheetahs exist', it is plausible to account for their truth
or falsehood exclusively on the basis of an appeal to things or bodies
as commonly understood. 'John exists', ,on a view of this sort, is made
true by John himself; 'cheetahs exist' by some one or more cheetahs 21 •
But how, on this basis, are we to deal with negative existential judg-
ments like 'Ba'al does not exist' or 'there are no unicorns'. It was
precisely difficulties in the treatment of judgments such as this which
led some Brentanists to the view that what makes a judgment true are
special sui generis entities designated by expressions of the form: 'the
existence of A', 'the non-existence of B', 'the existence of an A which
is B', and so on, where 'A' and 'B' stand in for expressions like 'a
horse', 'the redness over there', 'unicorns', 'God Ba' a1', and so on.
The consideration of sentences like 'John is suntanned', 'John is
eating', 'John is a heavy eater',. 'John's eating is on the increase',
'John has a bad case of dispesia', 'there is a ridge of high pressure
over the Atlantic', suggests moreover that the domain of such special,
non-thingly truth-makers must be extended even more widely, to embrace
events, processes and conditions, as well as complex states of affairs
involving these as parts.
How, then, can Kotarbinski cope with cases such as this in a way
that will not stretch ontologically beyond the realll of things? Two
answers to this question lIay suggest themselves:
(1) that it would be possible to effect a logical or linguistic
analysis of the sentences in question, of a sort that would reveal their
underlying form as involving a relation only to things;
(2) that it would be possible to embrace special sorts of things as
truth-makers for the given sentences, so that reism would be saved,
though only at the e,xpense of our embracing a notion of 'thing' which
would depart in some degree from common sense.
As we shall see, elements of both solutions are present in
Kotarbinski'g work.
210n t.h ...... rminology of' 'making t.ru .. ' s .... Mulligan, Simons and Smit.h
(19S~>. The t.erminology has a numb .. r of' advant.ag .. s over t.h .. mol''' usual
t.alk of' correspond .. nce. It. is dis .. mbarrassed, f'irst. of' all, of' all
connot.at.ions of' 'copying'. It. does not. s.uggest. t.ha't. t.h .. r, .. lat.ion bet.ween
a sent.ence and t.hat. in virt.ue of which it. is t.ru .. would be a sYmmet.rical
relat.ion. And it. can cop .. wit.h t.h .. fact. t.hat. t.h .. re may b .. mar .. t.han on ..
ent.it.y which makes or h .. lps t.o mak .. a given sen"enc .. t.ru ... Thus, in t.he
simplest. possible case, 'I have a headache' may be made t.rue by my
present. headache ('f'rom .. he beginning t.o t.he end of' it.s exis"ence'), or
by ~ny phase of' t.hi~ h .. adache ov .. rlapping wit.h my pr .. sent. ut.t.erance, or
by relevant. st.a"es of' nervous "issue upon which my headache supervenes.
146 BARRY SMITH
§4. Problems of Semantics
Let us look, first, at the semantic side of Kotarbinski's doctrine.

Consider the sentence 'John's jump cleared the hurdle'. This seems to
refer to a certain concrete individual event or process - John's jump -
which occurred at a certain time. And it must therefore surely
correspond, if true, to a segment of real ity containing this jump as
part. We have a strong intuitive disposition to suppose that any account
of what makes the given sentence true will be inadequate if it takes no
account of this specific jump. Acording to Kotarbinski, however, this
intuition cannot even be properly expressed. For all apparent references
to jumps and other events or processes are "merely substitutive".
A literal rendering of the intentions of one who utters the sentence in
question would be: 'John jumped clear of the hurdle', a sentence in
which the only names that occur are names for things 22 • It is renderings
of this sort, Kotarbinski insists, that reproduce "the intention of any
statement that says something about an event or events". For, "it is
only seemingly (and never in fact) that we can make a true statement
about an event, namely if we take that statement in its substitutive,
and not literal and fundamental role". References to events are mere
'onomatoids' or 'apparent names'. They are terms which merely sound like
names. When the attempt is made to establish a literal interpretation,
then it becomes clear that the expressions in question belong to
a cate10ry quite different from that of names in the strict and proper
sense l .
Physics too, of course, along with many other sciences such as
'phonology, military history and meteorology, seems to trade largely in
sentences of greater or lesser generality about events. Such disciplines
are, accordingly, in need of radical linguistic reform, so that, as
Kotarbinski points out, "one of the most topical but unperformed tasks
of concretism is to work out a dictionary of mathematics and physics
[and of other sciences] in the reistic interpretation,,24.. It is not
difficult to appreciate the obstacles confronting such a project in
relation, say, to the physicist's talk of energy-fields characterising
points or regions of spacetime. Kotarbinski in fact copes with the
latter not by semantic means, but ontologically. As we have seen, he
accepts into his ontology fields and other creatures of physics. These,
too, are extended in space and time and are "such as to offer resis-
tance" (or, at least, they are presumably such as to be involved in
causal relations of certain sorts). Reality, as the reist conceives it,
22Similarly when a person states the fact of London's lying somewhere on

the Thomes, "he merely sLat.es, in a devious ",oy, t.hot. London lies .on t.he
Thames" and here - as Kot.arbinski conceives it. - no reference is made t.o
facts or st.at.es of affairs or anyt.hing ot.her t.han t.hings. <1966, p.4.29)
23Kot.arbi~ski (1966, pp.52, 4.01,n.4.)
24.(1966, p.4.26). Compare Szaniawski (1977).
is not therefore 'a "static conglomerate" ("a mere sum") of "rigid and
changeless solids'" j it is a "fabric composed of changing things". in a
new and extended sense of 'thing,25. No explicit criterion is provided,
however, as to what is 'thing' and what 'event' on this more liberal
dispensation. so that one does not know, for example, whether quarks,
neutrinos, or flashes of lightning are to be admitted as (short-lived)
things or rejected as events.
Moreover, even where we are dealing with non-scientific sentences
of the everyday world, the reist's literal renderings are not in every
case so easy to come by. What, for example, is to count as a 'literal'
rendering of a judgment like: 'John's jump impressed the spectators'?
Perhaps: 'John jumped and impressed the spectators'. Yet it is far from
clear that this rendering is even roughly adequate. John's jump, after
all, may have impressed the spectators, but not John himself. Or John
may have jumped, and impressed the spectators, without it being the case
that it was his jump by which they were impressed~.
Kotarbinski's problem here results from the fact that there is as
it were a selectivity of intentional verbs like 'see' or 'think about'
or 'be impressed by'. It seems that such verbs may relate their subjects
to entities such as events, processes, images, contents, meanings, sur-
faces, boundaries, states of affairs, absences, and so on, in ways not
accountable for exclusively in terms of any mere di~ectedness to things.
Such selectivity is characteristic especially of memory, which may as it
were conceal from our present consciousness the things which serve as
supports for events or circumstances remembered. Thus Harry may remember
the intonation of Mary's voice, yet he may have forgotten both Mary
herself and the voice that had this certain quite specific intonation 27 •
Further problems arise for an approach of the sort sketched by
Kotarbinski when we cosider sentences apparently involving quantifica-
tion over events or types of event (John" danced the same jig twice), or
when we consider relational or comparative sentences like Mary's blush
was redder than Susan's, The beginning of John's jump was more elegant
than the end of Jack's, and so on28.
26(1966, pp.330f., 4,26)
26Simila!" difficulties a!"ise fo!" the p!"oposed !"eistic t!"anslation of

'Justice is a virtue of honest people' by 'Any honest man is just'
(Wole';ski 1987, p.168). For it may be that all honest men are as a
matter of fact just, though not in virtue of being honest.
27See §3 of Mulligan, Simons and Smith (1984,), fa!" a discussion of this

example. The!"e is, noto!"iously, a pa!"allel selectivity of 'cause'. Con-
sider fa!" example a sentence such as' 'The fact that ag!"eement WaS
!"eached cau"sed unive!"sal Joy', which Kotarbi';ski (somewhat counterintu-
itively> wants to" !"ende!" as, 'All we!"e ove!"joyed when they agreed'.
(1935, p.49t>.
28See Tegtmeie!" (1981) fo!" an extended discussion of such cases and of

the !"ea80n8 why they 8eem to dictate an ontology !"iche!" than that of the
!"eist.
148 BARRY SMITH
Kotarbinski himself, of course, since he believes that somatistic

reism is true, can countenance neither a selectivity of mind to non-
things, nor the possibility of relations involving apparent non-things
in ways which could not be cashed out satisfactorily in terms of
corresponding things. Hence he is constrained to hold, for example, that
when Harry remembers the intonation of Mary's voice, then there is of
necessity a sense in which he remembers Mary also, and that the precise
content of his memory can be accounted for without loss of content in
termR of his relation to this and other t.hings.
Lat.er reists in the Pol ish t.radition t.ake a more relaxed view of
such translation prohlems, conceiving reistic semantic analysis as of
value only when confined to t.heses purely philosophical in character.
Outside ontology, as Lejewski puts it, reistic semantics "loses it.s
rationale":
There is not much point avoiding abstract noun-expressions in
disciplines of lesser generality. Elimination of onomatoids from
final pronouncements is of paramount importance only if these final
pronouncements are meant to be used in ontological arguments 29 •
Now there is, cert.ainly, some justice to this, if it means t.hat the
reist is restrained from emharking on gratuitous attempts to reform the
language of his fellows, language which must surely be in order as it
is. What Lejewski has to say should not, however, be interpreted as
implying that we may properly ignore those forms of everyday and
scienti fie language which pose prima facie problems for the would-be
reist translator.
§5. Kotarbinskian Psychology
As we have seen, Kotarbinski rejects the doctrine of mental contents

propounded by his teacher Twardowski. Contents and images are, as Kotar-
binski points out, commonly held to come into being when someone recalls
something or dreams of something. The subject who dreams or remembers is
then "ready to formulate various true judgments, allegedly pertaining to
those images" (l966, pp.430f.). Brentanists such as Twardowski had de-
fended the thesis that such contents or mental images endoy an immanent
existence "in the subject" or "in a person's head,,3. Kotarbinski,
29L .. j .. wski <1979, p.206). A compromis .. position is put forward by Wolni .. -

wicz in his pap .. r in this volum ... Wolni .. wicz maintains that suffici .. nt
support for r .. ism is provid .. d by a demonstration that it is possibl .. to
r .. duc .. appar .. ntly non-r .. istic th .. ories to th .. ori .. s having a reistic
axiomatisation. R .. ism, on this basis, would th.er .. by concern whole th .. -
ories and not separate sentences.
3OS .... e.g. Br .. ntano <187~, pp.12~-32, Eng. pp.88-9~), Twardowski (189~,
p.3, Eng. p.t>, Hofl .. r (1890, §6).
ON THE PHASES OF REISN 149
however, could not see how this 'in' is properly to be interpreted.

Surely not spatially, "as though it referred to the nervous tissue in
the brain". (And where, for example, would we locate such mental
phenomena as the pain in a phantom limb?) Yet it seems equally inappro-
priate to regard mental images as located outside the brain, for
instance where imagined external objects seem to be located. Lesniewski
had been prepared to conclude from these difficulties that contents and
images exist "nowhere", a conclusion perhaps in the spi.rit of Descartes,
with his view of res cogitans as unextended, and accepted also among the
Brentanists 31 • Kotarbinski, however, could permit himself no such radical
departure from somatistic realism and concluded that immanent contents
and images are not to be accepted as bona fide things at all. This
conclusion he saw as being supported further by the fact that such
putative entities are not three-dimensional. Thus they cannot count as
bodies as ordinarily understood, and this, for Kotarbinski, rules out
their counting as things in any sense32
But how, then, are we to cope semantically with our apparent refer-
ences to images and other like phenomena? Here, again, Kotarbinski' s
attack is two-pronged, both semantic and ontological. On the one hand he
hopes, with Lesniewski, to 'deintensionalise' psychological statements,
to find means of converting such statements to extensional forms. On the
other hand, however, he hopes to develop a reistic conception of the
discipline of psychology i tse If, a conception according to which psy-
chology would deal not with mental acts of hearing or thinki ng or
desiring and with the contents of such acts, but rather with things of
certain sorts - with the sentient person, the hearer, thinker, or
desirer.
That which sees and hears and desires i.s, Kotarbinski holds,
identical with a certain organism (or at least with some fart of the
organism such as the brain or the system of nerve receptors)3 • To think,
then, is to be a thinking brain or body, a brain or body which, in non-
reistic language, enjoys certain special states or processes of
thinking. As Kotarbinski is himself careful to stress, this is not a
materialist or behaviourist doctrine. For while he certainly holds that
physics investigates all that there is, Kotarbinski does not suppose
that all scientific statements about what there is will turn out to be
statements of physics. As for Spinoza, so also for Kotarbinski, it is as
if, in the case of sentient beings, one single substance is able to
support two different systems of mutually incommensurable modifications.
Physics describes how sentient organisms (and other bodies) move and how
their particles are located. Psychology describes how sentient organisms
think and fee13~.
3iSee e.g. Mart.y <1908, p.~01).

3aCf . Kot.arbinski <1966, p.342). From t.his it. follows, t.oo, t.hat. t.here
cannot. be psychic subject.s if t.he lat.t.er are conceived as syst.ems 01'
sequences of cont.ent.s 01' images: see Kot.arbinski <1936, p.493).
33(1966, p.344).
34Cf . Kot.arbinski <1966, p.346; 1936, pp.496f'.).

150 BARRY SMITH
Suppose, however, that during some given period of time one and the
same sentient organism is both thinking and jumping. The same thing, in
such circumstances, is both a thinker and a ,jumper. Is not the reist
left in such circumstances with no means in his ontology to distinguish
between what are, surely, activities of different sorts? Clearly, he
cannot solve this problem by appealing to the fact that different
(mental and physical) predicates are applied to the thing in question,
for the issue here is precisely that of establishing in virtue of what
such different predications count as true, and to this end the reist has
only things to which he can appeal. The problem cannot be solved,
either, by appealing to any special understanding of the m~terial 'thing
that thinks' (which had been left indeterminate by Kotarbinski himself).
For whichever concrete thing' is fixed upon by the reist as that which
thinks, be it the brain, the central nervous system, or some other
proper or improper part of the organism as a whole, there will 'always be
physical truths about the thing selected in relation to which the given
problem will recur. Moreover, whatever the nature of the material thing
that the reist puts forward as his candidate 'thing that thinks', it
seems not logically excluded that two parallel consciousnesses should be
realised simultaneously within it, after the manner of Siamese twins. We
might then have occasion to assert that consciousnessl is thinking this,
while consciousness2 is thinking that, and then it seems that the reist
- short of assuming special immaterial things - would have ,no way of
doing justice to the presence of parallel thinking processes in the
given case, for there are ex hypothesi no separate bodies which might
here serve as subjects of the respective predications. What is not
logically excluded seells thereby to be ruled out by the fiat of the
reist's linguistic predilections.
Reism has consequences not only for the subjects of mental experi-
ence, however, but for the objects of such experiences also. As already
stated, the reist insists that that to which our experiences are related
is in every case a thing. In everyday perception, as also in hallucina-
tions, dreams and memories, we are typically presented with external
things which seem to us to be coloured and shaped in 'this or that par-
ticular way. And in dreams and memorie.s, as Kotarbinski puts it, we as
it were "observe, though somehow in a secondary manner, things from our
past enviroment, which seem to us to be such or another" (1966, p.431).
This account will clearly face problems in connection with iterated
reference to what is mental - dreams about dreams, for example ~ as also
. in connection with that peculiar selectivity of memory {lnd other acts
discussed in §4 above. Kotarbinski's view, nevertheless, is that our
mental experiences are in every case a matter of our being related in
special ways to things. From t.his it follows that all (third person)
psychological statements must have literal r~adings of one or other of
the forms:
A feels this: B,
A experiences this: B,
A thinks this: B,
and so on - where 'A' stands in for the name of some sentient body and
'B' for words or phrases which answer the question 'what?': 'What does
John imagine'?, 'What does John think?', 'What does John want?', and so
on. 'B' will stand, typically, for a "summary description. of [A's]

surroundings made in extrospective terms", and is supposed in every case
to involve reference exclusively to things 35 •
A slightly different analysis may be required for statements like
'my tooth aches' or 'I feel sick'. These may on the one hand be compared
to statements like 'my shoe is pinching'. Taken in this sense, 'I feel
sick' would be formulated as 'This is sickening', "where the indicative
pronoun would point to a certain region of the alimentary' tract and
adjacent parts of the body" (1966, p.348). Quite often, however, the
sense of 'I feel sick' is to signify 'I experience a feeling of
sickness' and this is a statement which complies with the original
Kotarbinskian scheme. It means 'I experience this: it (my body) is
sickening' - where again, reference is made exclusively to things.
But in virtue of what are sentences of the form 'A feels ••. ', and
so on, true? Perhaps we can express Kotarbinski's view as follows. It is
as if there are certain sui generis determinations of sentient bodies in
virtue of which such bodies are directed in a quite specific way to
things. It is of course not the case that the determinations in question
could be somehow isolated, whether actually or in thought, in such a way
that they could be examined in their own right. Yet they are not simply
unknowable, either: for there exists the possibility of imitation, in
the sense that one subject can think in a way which duplicates the
thought-determinations of another. Such imitation is possible because
our mental determinations characteristically express themselves
physically in a range of familiar ways. Above all, there is an organic
relationship between a subject's feelings or thoughts and the kinds of
things he sa.ys. Hence we can come to know the former indirectly, by
coming to an understanding of the latter in a way which amounts to a
(more or less perfect) duplication of those mental determinations which
they characteristically bring to expression. Here Kotarbinski draws on
the work on meaning and expression of his teacher Twardowski 36 , as also
on Gestalt psychological ideas concerning our knowledge of other minds 37 .
Strangely, he applies these ideas even to reflexive self-knowledge. We
acquire knowledge of our own experiences, he holds, only"by 'self-
imitation,38, so that there may be a sense in which we do not know what
we think until we hear what we say.
Kotarbinski's claim, then, is that we may come to know what another
person experiences by allowing ourselves to be guided by his statements
or by other overt behaviour in such a way that we come to imitate his
experiences within ourselves. Reistically expressed, we can make
ourselves think or feel (more or less) as the other person thinks or
feels, by allowing ourselves .to be determined psychically by the things
he says. (Th~ question-begging nature of this idea becomes clear
36Ko t.arbinski <1966, p.3"'7; 1936, p .... 99).
36S && Twardowski (1912) and t.h& discussion in Smith (1989).
37S && Kohler (194.7, pp.128ff.>, Koffka <1936, pp.666ff.).
38(1966, p.34.7). Similar not.ions are present. also in t.he yrit.ings of

Theodor Lipps on t.he not.ion of empat.hy: see e.g. his (1906).
152 BARRY SMITH
immediately if we.ask ourselves in what respect we can speak of 'like-

ness' or 'similarity' here).
We try to interpret the word 'experiencing' as follows. It is
merely an announcement of the imitation of the individual spoken of
by the speaker, and it ·informs ih a summary way in what respect he
will be imitated; thus, that the individual spoken of will be imi-
tated as looking, or listening, or exploring tactually, and so on.
(1935, p.499)
For this to make sense in reist terms, therefore, it must be that our
utterances themselves are in some extended sense imitations of the very
psychic determinations they bring to expression 39 • Thus in the general
formula of the psychological statement 'A experiences this: B', the 'B'
may be seen as an imitation in this extended sense by the one who makes
the given statement of the relevant experience on the part of A. When I
say, 'John thinks this: 2 + 2 = 4', 'John feels this: they are playing
badly', 'John doubts this: to be happy', then I become a samesayer with
the way John thinks or feels. And we can even
generalize this formula so that not only a sentence, but any phrase
referring to how a given person experiences, could be substituted
for 'B'. It might even be an inarticulate exclamation, SO that a
given psychological statement would be: 'John experiences so: Oh!'
(1966, p.428)
Kotarbinski's remarks here will remind us of Davidson's analysis of

indirect discourse in his paper "On Saying That" of 1968. My assertion
of 'Galileo said that the earth moves', on this analysis, is an asser-
tion to the effect that Galileo said something, and my immediately suc-
ceeding utterance of 'the earth moves' make Galileo and me samesayers:
Galileo said that.
The earth moves.
Here it is only the first sentence, consisting of the name of a speaker,
a two place predicate 'said' and a demonstrative pronoun, that is
asserted. The second sentence is, as it were, merely exhibited or named.
For Davidson, too, therefore, there is a sense in which the best we can
do is to imitate (make ourself samesayers with) the speaker whose words
we are reporting4o • A similar idea was incidentally advanced already by
Lesniewski 41 who considers an interpretation of expressions of the type
, ~ p' in the language of Principia Matheaatica as meaning:
that which follows is asserted p.
39 A view elllployed especially by Marty as the basis of a theory of

linguist.ic communicat.ion.
40Davidson <1966, p.106'
41(1927/31; p.l0)
As Kiing points out, it is an important feature of such devices that they

allow us to talk about a sentence while employing to this end not a name
of the sentence but (a token of) the sentence itself; that is, they
allow us to avoid an ascent into the metalanguage "while at the same
time obtaining benefits usually associated with such an ascent" (Kung
1974, pp.243f.). A similar device can be used also to avoid an ascent
into set-theoretical language: instead of 'The set of men is identical
with the set of featherless bipeds', we can say: 'The following two
i terns are extensionally identical: man, featherless biped'. As Kung
argues, it is a device of this sort that lies at the basis of
Lesniewski's understanding of the quantifiers42.
§ 6. The Aristotelian Concept of Thing
Kotarbinski started out, in the Elementy, from the common-sense idea of

thing as physical body. He drew, in particular, on the clarification of
this idea that was set forth by Aristotle in his treatment of 'first
substance' in the Categories and in the Metaphysics. Thus at the
beginning of the Elementy we read:
it is in Aristotle that we can trace the distinction, within the
category of things, namely, of first and second substances. Those
second substances, universals, are the first to fall victim to
eliminating analysis as carried out by nominalism .•• On the other
hand, first substances, things in the primary sense of the word,
and for us simply things, fared in exactly the opposite way, since
the entire reduction of categories [takes] place precisely to their
benefit.
( 1966, p. 55)
What, then, are the marks of things or first substances as
Aristotle conceives them43 ?
(i) They are, first of all, individual. A substance is a "this", it
is "one in number". (Cat., 3 b 10)
(i i) They are not "predicable of a subject" nor "present in a
subject". (Cat., 2 a 11-13; Met., 1017 b 10-14, 1028 b 35f., 1029 a 1)
(iii) They are that which can exist on their own, where accidents
require a support from things or substances in order to exist. First
substances are prior in all senses: in definition, in order of
knowledge, and in time. (Met., 1028 a 30ff.)
(iv) They are that which serves to individuate the accident, to
make it the entity that it is - a feature seen by Brentano as the most
crucial element of the Aristotelian
. 44 theory. (Anal. post., 83 a 25; Met.,
1030 b 10ff.; Cat., 2 b Iff.)
42Cf . Kung and Canty (1970).
43Compare, for what follows, Novak (1963/64).
44Cf . Brentano (1933, pp.37, 108, 112, 131, Eng. pp.37, 86, 88, 109).
154 BARRY SMITH
(v) They are that which, while remaInIng numerically one and the
same, can admit contrary accidents at different times. (Cat.,.4 a 10)
(vi) They are able to stand in causal relations. (Met., 1041 a 9)
(vii) They are "one by a process of nature". A substance has the
unity of a living thing. Hence it enjoys a certain natural completness
or rounded-offness, both in contrast to parts of things and in contrast
to heaps or masses of things (Met., 1040 b 5-16, 1041 b 12, 1041 b 28-
31, 1052 a 22ff., 1070 b 36 - 1071 a 4; Cat., 1 b 5)45. Hence also, for
Aristotle, a thing is that which has no actual but only possible parts
(Met., 1054 a 20ff.). A part of a thing, for as long as it remains a
part, is not itself a thing, but only possibly so; it becomes an actual
thing only when it is somehow isolated from its environing whole. In
this sense (and also in others) the substance is the bearer of
potentiality, and it is at this point that we should have to list those
marks of substance which flow from Aristotle's hylomorphic theory, and
from his theory of act and potency.
There are further marks of substance, less easily documented in
Aristotle's texts since they were taken entirely for granted in
Aristotle's day. These are above all:
(viii) A substance is independent of thinking, a part of nature -
where no Greek would have understood what is meant by • independent of
thinking' .
(ix) A substance is that which endures through some interval of
time, however small. This means, firstly, that things exist continuously
in time (their existence is never intermittent). But it means also that
there are no punctually existing things, as there are punctual processes
or events (for example beginnings, endings, judgings, decidings, and
instantaneous changes of other sorts)46. A thing is also typically such
as to endure for such a length of time that it may acquire a proper name
for purposes of reidentification47.
(x) A substance is that which has no temporal parts: the first ten
years of my life are a part of my life and not a part of me. As our
ordinary forms of language suggest, it is events and processes, not
things, that have temporal parts.
Even leaving aside, now, the passages where Kotarbinski explicitly
allies himself with Aristotle48 , the focal instances of the concept of
thing made prominent in the Eleaenty' make it clear that he had intended
to follow Aristotle in almost all of the above. A body, as we have seen,
is bulky and lasting and such as to offer resistance. Further marks of
bodies distinguished by Kotarbinski are:
45Se .. Met.., 104:< b 16-32, for ot.her kinds of unit.y.
46Se .. Ingarden <1964/66, Yo!.I, §§28f.).
47As is clear from Arist.ot.le's t.reat.ment. of <ii), it. is possible t.hat. t.he
marks of t.he concept. of subst.ance may be est.ablished in part. t.hrough
considerations o:f the language we use "'0 re:fer i-o subst..ances t.hemselves.
See Met.., :1029 b 13.
48See ... g. (1966, pp.326f'f.).

- They are three-dimensional.

- They are all and only those entities that can be investigated by
science; every object is "knowable in principle".
- They enjoy essential perceptibility, and are further characterised
by the fact that they all exert influence upon perceptible objects.
- They are at a definite place (that is, they are at a specified
spatial distance from certain perceptible objects), and at a specified
time (that is, they are at a specified temporal distance from certain
perceptible objects)49.
Each of these marks is perfectly in conformity with the Aristo-
telian view expressed above. Kotarbinski's most important departure from
Aristotle, in the Elementy at least, is in regard to (vii). For
Kotarbinski - almost certainly under the influence of the Lesniewskian
conception of sets as concrete wholes - rejected the thesis that things
must in every case be unitary, so that he counted as things also masses
and quantities of things and even non-detached thingly parts. Bodies of
air, swarms of bees, the solar system, are 'compound bodies', in
Kotarbinski's termsd as also are society, nation, social class and all
other institutions5 •
A lack of sensitivity to the distinction between things, masses,
and parts seems to have been shared by a number of Polish philosophers,
and may in some degree reflect the fact that the Polish language, with
its lack of art icles, distinguishes less clearly than other languages
between mass and count nouns or between mass and count uses of the same
noun ('Polish logicians eat little orange during War').
One (Austro-) Polish philosopher who did not allow himself to be
swayed in this way by what is, after all, a peculiarity of language was
Twardowski, who follows Aristotle in insisting that what he calls
"objects of presentation" are characterised in every case by the fact
that they are integrated wholes, a thesis he extends even to the objects
of general presentations such as Triangle, Square, Lion, etc. 51
Lesniewski, in contrast, goes so far in rejecting the idea that to be a
thing an object must in some sense form a natural unity, that he accepts
4,9(1966, pp.327, 4,35, 34,2). Not.e t.hat., like inert.ness, t.hese marks are
held by Kot.arbinski t.o be incident.al; t.hat. is, t.hey do not. aff·ect. t.he
ext.ension of t.he concept. ~ or 6~<ty.
50The lat.t.er consist. of human beings st.anding in cert.ain relat.ions t.o

each ot.her - which is not., of course, t.o say t.hat. t.here exist. ent.it.ies
called relat.ions Cof dependence, leadership, authorit.y e~c.), in
addit.io·n to and as it. were alongside t.he human beings t.hemselves. The
various element.s of Q given inst.it.ut.ion <Ire somehow relat.ed one to
another for example in t.he. sense t.hat. some of t.hem behave in such a way
because the ot.hers behave in such a way. A similar t.reatment is offered
by Kot.arbinski for t.erms like 'funct.ion', 'disposit.ion', and so on. As
Kot.arbinski would have it., "'X has the funct.ion of t.ypist. in a bank'
means t.he same as: 'X syst.emat.ically t.ypes let.t.el"s occol"ding t.o t.he
inst.l"uct.ions of her supel"iol"'" (1966, p.4.90).
51(1894., pp.88, 105, Eng. pp.86, 100). Cf'. Ingol"den <1964./65, vol.l, p.219).
156 BARRY SMITH
a principle of the arbitrariness of thingly boundaries. This principle

is built into the axioms of his system of Mereology, which includes a
theorem to the effect that if a and b are objects, then so also is their
sum, irrespective of whether a and b are connected or contiguous or
materially related in any wal 2 • Lesniewski does not, of course, deny
that some objects (in his highly general sense) have a natural unity. It
is merely that he does not see the need to introduce this concept of
natural unity into his theories of Ontology or Mereology. The Jatter are
theories dealing with what he holds to be more primitive notions -
notions which would in any case have to be clarified before a rigorous
treatment of 'natural unit' could be attempted.
§7. Truth, Correspondence and Lesniewski's Ontology
How, now, given his essentially Aristotelian ontology of things, does

KotarbiIlski cope in the Eleaenty with the problem of accounting in
correspondence-theoretic terms for the truth of sentences such as 'John
is jumping'? Sentences of the given sort are analysed, first of all, as
what Chisholm has called concrete predications, expressing relations
between things63 . Thus 'John is jumping' is analysed as a sentence of the
form: 'John is a jumper', 'John is red' as: 'John is a red thing', 'John
desires apples' as: 'John is an apple-desirer', and so on. In each case
we arrive at a sentence containing two names of things joined together
by the copula 'is,64, so that the things picked out by the names in
question are to have the job of making true the relevant judgment.
Remember, in all that follows, that 'John is a jumper' is to be
understood as an analysis of the sentence 'John is jumping' ('John is at
present executing one or more jumps'). Thus we are to resist the natural
tendency to understand nominals like 'jumper', 'swimmer', 'bouncer', as
relating to a habitual or professional performance of the relevant
activity. This tendency derives, of course, from the already mentioned
fact that names are in normal cicumstances used for purposes of re-
identification; thus they presuppose some duration or recurrence on the
part of what they name.
'John is a jumper' accordingly analyses 'John is, on this
particular occasion, jumping', where 'is' expresses a real continuous
present. The thing picked out by 'John' seems relatively easy to
identify, at least against the background of the broadly Aristotelian
62Irrespect.ive, even, of whet.her a and .e.

exist. at. t.he sam., t.ime. Se ..
Simons (1987) for crit.icisms of t.his and relat.ed aspect.s of Lesniewski's
Mereology.
63Chisholm (1978, p.199)
64More generally we arrive at. a sent.ence cont.aining exclusively logical

const.ant.s and 'genUine names'. See Wolenski <1987, p.168).
conception described above. But what, in the light of this conception,

are we to make of the thing picked out by 'a jumper'? And what is the
relation between John and a jumper that is expressed by the copula 'is'?
Our first port of call, given the strong influence exerted by
Lesniewski on Kotarbinski's (formally much less sophisticated)
ontological views, is Lesniewski's own system, of Ontology, a theory
built up on the basis of Lesniewski's system of Protothetic or' 'theory
of deduction' by the addition of the new primitive term 'is' and the
single axiom:
"tab [a i·s b +--+ 3c (a is c & c is b)]
Colloquially:
a is b if and only if, for some c, a is c and c is b55 •
Here 'a', 'b' and 'c' are any expressions belonging to the category
name. This means, as Lesniewski sees it, that they may be either:
(1) ordinary singular designating names or nominal expressions like
'Ronald Reagan' or 'the British Prime Minister';
(2) shared or general names like 'philosophers' or 'apples in
Vermont' j
(3) fictitious or empty singular names like 'Pegasus' or 'the
largest prime number': or
(4) fictitious or empty general names like 'sirens' or 'fates ,.56.
All such expressions belong to a single category, Lesniewski argues,
since whether a name like 'man at the door' is singular or shared or
empty depends on the factually existing state of the world, and so'
cannot be regarded as basic from the point of view of logic57 • From this,
however, it follows also that we must admit as names expressions like
'jumper', whose number is in a certain sense indeterminate.
The axiom of Ontology, now, lays down simply that for 'a is b' to
be true, it must be the case that every a is b and that e~actly one
object is a. It is not difficult to show, on these terms, that if both
'a' and 'b' are singular and designating, then 'a is b' is deductively
equi valent to .' a= b'. But now, applied to what has now become the
Polish-sounding sentence 'John is jumper', this analysis of 'is' tells
55We here leave out. of account. peculiarit.ies arising from Lesniewski's

special reading of t.he quant.ifiers.
66Lesniewski's concern t.o allow fict.it.ious or empt.y names was almost.

cert.ainly inspired in part. by Twardowski, whose t.reat.m,ent. of non-
exist.ence also influenced Meinong. See Twardow'ski (1894." pp.30, 33; Eng.
pp. 28, 30f.).
67Kung (1967, p.111). We may conject.ure t.hat. such apparent. logico-

grammatical dist.inctions as that. bet.ween common ond proper names" marked
in English,by t.he presence or absence of art.icles, were overlooked by
Lesniewski, again, because of t.he lack of art.icles in t.he Polish
language.
158 BARRY SMITH
us only that, if this seritence is true, then 'John' must be a singular

designating name and 'jumper' a designating name designating (possibly
inter alia) what 'John' designates.
From this point of view it becomes clear that the system of Ontol-
ogy is in fact not an ontology at all (a theory of the different types
of' being). Rather, it is a theory of names, as is reflected in Kotarbin-
ski's use of the expression 'calculus of names' for what Lesniewski
called 'Ontology'. More precisely, it is a theory of the relations of
designation that hold between singular, shared and empty names on the
one hand and objects (of whatever variety) on the other. It reflects a
concern, therefore, not with problems of ontology, or metaphysics, but
wi t.h the issues that arise when one allows 'a' and 'b' to stand in not
merely for singular terms as straightforwardly understood but also for
any expressions wi thin the wider category thus defined. To put the
matter another way, Ontology may be seen simply as an extension of the
theory of identity to cope with a somewhat liberal view of what may
count as 'name', so that the absolute universality of '=' is inherited
by the new Ontological 'is'. This makes it compatible with any ontology
formulable by means of expressions belonging to the given category. As
Wolenski (1987, p.175) puts it, "Ontology is metaphysics free".
If, now we return to our sentence 'John is jumper' in the light of
the Lesniewskian analysis of 'is', then it would appear that two
alternati ve readings present themselves, according to whether we take
'jumper' as singular or pluraL On the first alternative, 'John is
jumper' will be equivalent to the more English-sounding 'John is
identical with a jumper', so that the referents of 'John' and 'a jumper'
will be one and the same. Now 'John', as we normally suppose, designates
an enduring object, subject at different times to contradictory
determinations (he is now jumping, now not). If, therefore, the referent
of 'a jumper' really is to be identical to the referent of 'John', then
'a jumper', too, must designate something that endures, so that 'John'
and 'a jumper' would be merely two different ways of referring to one
and the same ordinary continuant. On this account, however, the
truth-maker of 'John is jumping' would differ in no wise from the
truth-maker of 'John is John', and this is an outcome which surely flies
in the face of our intuitions to the effect that one or more jumps must
somehow be involved in making true the given sentence. If 'John' and
'a jumper' are two different ways·of referring to the same thing, then
surely, our intuitions tell us, they refer to this salle thing 'under
different aspects'. Lesniewskian Ontology is however so lacking in
discrimination in its treatment of 'things', that it is able to take no
account at all of such 'different aspects'.
What, then, as regards the second al ternati ve, which would make
'John is jumper' equivalent to: 'John is one &Bong the jumpers'. Since
from any sentence of the form 'a is one of the bs', one can in·fer wi thin
Ontology a sentence which might be rendered colloquially as 'a is this
b', where 'this b' is a singular nalle for that individual b which a is,
this second reading .ight seem to bring us back once .ore to the first
alternative, which we have seen reason to reject. We may, however, be
able to infer from 'John is one among the jumpers' also that there are
jumpers to which John himself stands in the relation of similarity. And
now, since John's circle of similars qua jumper is different from his
circle of shilars qua thinker, this- may enable the reist to distinguish
separate truth-makers for 'John is thinking' and 'John is jumping' even
in those cases where the two activities are performed simultaneously.
From this it would follow that the family of jumping things contributes'
in some way to making it true that John,. in particular, is jumping -
a consequenc&which certainly goes beyond what Lesniewski himself had to
say on these matters, but which nevertheless has advantages from the
reist point of view (to the extent that it has been found acceptable by
Lejewski). In particular, it enables the reist to distinguish the truth-
makers of 'John is Jumping' and 'John is thinking' even in those
circumstances where jumping and thinking are simultaneous: other jumpers
contribute to making true the former sentence in a way in which they do
not contribute to making true the latter.
§8. Time and Tense
This is not quite all that can be said on Lesniewski's behalf, however,
and before returning to our discussion of Kotarbinski "s own ontological
views it will be useful to look at the Lesniewskian treatment of the
phenomena of verbal tense. Recall that the 'is' in 'John is jumping' is
intended to express a real present tense. The Polish 'jest', on the
other hand, for example in 'Jan jest skaczc!lcy' (John is jumping) -
a form which sounds odd due to the absence in Polish of the continuous
aspect -does not express a present tense, and this holds too of 'Jan
skacze' (John jumps) and 'Jan jest skoczkiem' (John is a jumper).
In and of itself the Polish 'is' is timeless. In order to mark the
fact that the jumping is taking place at the moment, the speaker of
Polish must add an explicit temporal index and say, for example, 'Jan
teraz skacze' (John jumps now) or (more stiltedly) 'Jan jest teraz
skacz~cy' (John is now jumping). This timelessness, we now see, must be
characteristic also of the 'is' of Lesniewski's Ontology. This is first
of all because Lesniewski insisted that the sentences of Ontology should
be absolutely true, Le. true independently of time and occasion of
utterance58 • But it is also because, as already noted, the 'is' of
Ontology is to enjoy absolute universality of scope; it is to be
applicable to abstracta as much as to concreta, to objects past and
future as much as to objects of the present. It is in f.act the same
timeless 'is' as that which we employ when we say, e.g., 'Socrates is a
philosopher' or '3 is a prime number' or 'whales are mammals'.
Ontology is not, however, restricted to 'timeless' sentences of the
given sort. Return, for the moment, to 'Jan je-st teraz skacz~cy' (John
is now jumping). We should normally interpret the temporal index ('now'
or 'teraz') in such a sentence as governing the verb. Given the
~ere, again, he was almost certainly inf"luenced in part by Twardowski.

See the latter's def"ence _of" a'bsolute truth in his (1902),
160 BARRY SMITH
universality of scope of Ontology, however, and of the category name

with which it deals, it is open to us to allow such temRoral indices to
govern not the verb but the subject of the sentence 9. This yields
sentences of the form'Johnteraz jest skaczafY' or 'Johnnow is a jumper'
- sentences of a sort which make possible a new Lesniewskian reading of
our original 'John is a jumper'. - .
'Johnnow' is a name, like any other; but a name of what? We shall
think of it, for the moment, as designating a phase of John, remaining
neutral as to what exactly this might mean and recalling only that any
view of phases as temporal parts of things would signify a departure
from the Aristotelian conception of 'thing' outlined above. We shall for
the moment presuppose only (1) that some of the phases of John are
jumping phases, some not; and (2) that phases exist only for some
(normally relatively short) interval (or instant) of time.
Someone who asserts that John is a jumper, now, may be seen as
asserting that a present phase of John is a jumping phase of John - with
a timeless 'is', exactly as dictated by the conditions laid down by
Lesniewski on the sentences of his Ontology. The notion of a present
phase of John may be elucidated in turn as: a phase of John that is
simultaneous with the utterance in question, i.e., in reist terms, with
the relevant phase of the speaker60 • The advantage of a reading of this
sort is that we now have no need to regard 'a jumper' as the name of an
enduring object. The identity of the referents of 'John' and 'a jumper'
is assured, rather, by the fact that 'John' itself has come to refer to
an entity which enjoys a merely transient existence.
§9. Phases in the Development of Reism
Does this really help, however, in understanding what it is that makes

true the sentence 'John is a jumper'? For what is this. 'phase' of John
that is both John and a jumper? There are, it will turn out, a number of
crucial difficulties which we face in establishing what such phases
might be. Most importantly, as already remarked, it seems that however
this issue is decided the phase ontology will dictate a departure from
the broadly Aristotelian conception of 'thing', in spite of Kotarbin-
ski's apparent assumption that this ontology represents a natural and
inconsequential extrapolation of the reistic ontology of things61 •
69Bolzano, t.oo, in §46 of" t.he Wi .... en .. chaf"t. .. lehre, sees t.ime-det.ermina-
t.ions as part. of" t.he subject., so .... hat., as he put.s it., "a pair of" prop-
osit.ions such as 'Caius is now learned' and 'Caius was not. learned t.en
years ago' t.urn out. t.o have dif"f"erent. subject.s".
60See Brent.ano <1976, Part. II, Ch. VIII>.
61A similar suggest.ion is implicit. also in Lesniewski' .. work. See e.g.

Sinisi <1983, pp.57f"f".>., quot.ing f"rom t.he f"inal sect.ion of" Lesniewski
<1927/31>. cr. also Lejewski (1982).
There are, be it noted, no phases in Kotarbinski' s Elementy, and

that Kotarbinski held to a strictly Aristotelian view in this work is
seen above all in the fact that - as his examples show - he takes it for
granted there that things may change, in the sense that what is true of
a given thing at one time may be false of that same thing at another. In
his paper of 1935, "The Fundamental Ideas of Pansomatism", in contrast,
Kotarbinski embraces the phase ontology seemingly without a second
thought. Every object, he writes,
is something corporeal or something (or a whole consisting of such
components). An example of something corporeal is: a watch of the
trademark Omega No.3945614 from 1st January, 1934 to 31st December,
1934, inclusive (or any of its parts - for instance, the minute
hand from 5th March, 1934 to 7th April, 1934, inclusive). And an
example of something sentient: I, from 8 o'clock to 1 o'clock on
20th March, 1935 (or any temporal portion of this object, e.g., I,
from 9 o'clock to 10 o'clock inclusive on the same day).
(1935, p.488)
Hence at least one further stage needs to be added to the list of
• stages' in the development of reism given above. This consists in the
transition from an essentially Aristotelian ontology on Kotarbinski's
part in the period up to 1931, to a quite different ontology of phases
(or mixed ontology of phases and SUbstances) in the, years thereafter.
How this apparently unconscious change of mind on Kotarbinski's
part came about can be seen if we look at the final section of
Lesniewski's work "On the Foundations of Mathematics". Here Lesniewski
begins by expressing his gratification that so many of his own views,
especially in connection with the system of Ontology, had found support
in Kotarbinski's work62 • Lesniewski goes on to quote extensively from the
Elementy, including the famous passage in which Kotarbinski compares
Lesniewski's work to that of Aristotle, thereby providing a retrospec-
tive justification for Lesniewski's use of the term 'Ontology' - a jus-
tification which Lesniewski himself was only too willing to accept63 • For
a long time, Kotarbinski writes, the term 'ontology'
has come to designate investigations of the 'general principles of
being' conducted in the spirit of certain parts of the Aristotelian
'metaphysical' books. However, it should be noted that if the
Aristotelian definition of First Philosophy, perhaps the main
concern of these books, is interpreted in the spirit of a 'general
theory of objects', then both the word and its meaning are
applicable to the calculus of names as conceived by Lesniewski.
(1966, pp.210f.)64
62Sinisi <1983,p.66)
63S .... Wol .. nski <1987, pp.170f.).
64 I t.
should go wit.hout. saying, in light. of our discussion of Ont.ology in
§7 above, t.hat. we do not. share t.his est.imat.ion of t.he nat.ure of
Lesniewski's achievement..
162 BARRY SMITH
In the paragraphs which follow this discussion of Kotarbinski, however,

Lesniewski goes on to consider a certain difficulty for Ontology posed
by the colloquial reading of Ontological sentences of the form 'a is b',
and it is in this context that he first introduces his notion of phase
or 'temporal segment'.
Let us suppose, Lesniewski writes, that someone were to assert:
(a) Warsaw is older than the Saxon Gardens.
(b) Warsaw in 1830 is smaller than Warsaw in 1930.
(c) Warsaw in 1930 is Warsaw.
(d) Warsaw in 1830 is Warsaw.
Then, taken together with the axiom of ontology, we can derive from
these sentences the following assertion:
(e) Warsaw in 1930 is smaller than Warsaw in 1930,
which is absurd.
In his response to this objection Lesniewski insists, first of all,
that the expression 'Warsaw' be used consistently throughout. Either, he
claims, it should be used to refer to "only one object having a definite
time span, which at present we do not know", in which case it has the
sense of 'Warsaw from the beginning to the end of its existence'. Or it
should be used in such a way that it refers to indefinitely many
different objects, so that it would be possible to assert "of 'Warsaw
from the beginning to the end of its existence' as well as of 'Warsaw in
1930' and of 'Warsaw in 1830' ••• that they are Warsaws". Moreover, "it
would be possible to say with complete generality that if some object is
Warsaw, and some other object is a temEoral segment of the first object,
then the second object is also Warsaw" 6.
On the first reading, now, which sees 'Warsaw' as a singular name,
"it is not possible to call by the name 'Warsaw' any temporal segment of
temporal 'section' of the unique Warsaw referred to". In this case we
shall be able to assert neither (b) nor (c) nor (d). On the second read-
ing, on the other hand, 'Warsaw' is a plural name, which means that we
shall find it impossible to assert any sentence of the form (a). Only on
the basis of some such sentence, however, can we infer the consequence
(e). Whichever alternative is chosen, therefore, the supposed absurd
implication can be avoided.
We may infer, now, that Kotarbinski, wishing to keep in step with
Lesniewski in this, as in other matters of Ontology, simply took over
the notion of temporal phase, going so far as to accept arbitrary
temporal phases of an object as of fully equal status with that object
itself. Closer examination of the relevant passage reveals that
Lesniewski himself is here more circumspect. Thus he accepts the greater
appropriateness of talking not so lIuch of the temporal segment of the
rector of the University of Warsaw in January, 1923 but rather of the
man ('from the beginning to the end of the existence of this man') who
was in January, 1923 the rector of the University of Warsaw. Further, he
has "the inclination to use the expression 'Warsaw' as a nalle denoting
655in18i <1983, p.68), quot.ing Le';niewsk1 <1927/31>.

one object only". Since, however, he is using 'man' and similar

expressions to designate simply the relevant maximal phase ('man, from
the beginning to the end of his existence'), it seems that Lesniewski is
even here embracing the phase ontology, though in a form which seeks to
come to terms with the fact that the ontology in question threatens
conflict with our ordinary usage.
§10. What are Phases?
Certainly we refer quite naturally to Napoleon i'n his youth, to Nixon

during the period of his presidency, to the later Wittgenstein, and so
on. Normally, however, we take such e)!:pressions in their sentential
contexts, as signifying for example that Napoleon himself was such and
such in his youth. That is, we treat expressions like 'in his youth' as
adverbial modifications of the relevant verb. The phase ontologist, in
contrast, takes such forms of speech to sanction the view that there are
special objects, Napoleon in his youth, Nixon during the period of his
presidency, and so on, which are temporal parts of Napoleon and Nixon
respect i vely (' from the beginning to the end of their existence').
Objects are therefore seen as having temporal parts in just the way that
they have spatial parts like arms and legs. Thus where common sense and
Aristotle prefer a view according to which things (for example people)
exist in toto in any given moment of their· existence, the phase
ontologist seems to condone a view according to which only the relevant
temporal parts of things would exist in any given moment. He may thereby
be driven to the view that temporal parts must be in every case
instantaneous, for any temporal part of duration longer than a single
instant would have just as little claim to exist in that instant as
would the relevant temporal whole. Adoption of the phase ontology may
thereby lead to a view of ordinary things as mere entia successiva, the
separate 'momentary slices' of which would exist in successive instants
of time (as, according to some philosophers, the world as a whole has to
be recreated anew by God in each successive instant). An enduring thing,
on this view, is a mere logical construction upon the various
instantaneously existing entities that may be said to do duty for it66 •
Alternatively, however, the phase ontologist may seek to understand
'Johnnow' as signifying John himself, exactly as understood within the
Aristotelian theory, but restricted to some interval of time (t, t')
which includes the present moment67 • If, however, as is required by the
66Chisholm (1976, pp.98f.). Woodger, who independent.ly developed a phase

ont.ology similar t.o t.hat. of Lesniewski, comes close t.o a view of t.his
sort. in his (1939).
67A nat.ural longuage reading of 'John now ' may involve a necessary
indet.erminacy in t.he precise ext.ent. of t.he relevant. int.erval; since t.his
indet.erminacy can in principle be eliminat.ed, however <for example by
ut.ilising t.he resoure .. s of a formal t.heory such as t.he 'Chronology'
proposed by L·ejewski in hi~ (1982>, we can ignore t.he mat.t.er her ...
164 BARRY SMITH
Aristotelian theory, John exists in toto in every moment of his

existence, then it must surely follow, according to a process of
reasoning encountered already above, that John( t, t' ) is in fact
identical with John himself. To get round this problem the phase
ontologist might seek to regard 'John(t,t')' as referring to John as he
would have been had the universe (conceived as being in other respects
identical to the actual universe) begun at t and ended at t'. ('Napoleon
in his youth', on a view of this sort, might be understood as referring
to Napoleon as he would have been had he ceased to exist on the point of
reaching maturity.) This, however, would make of phases merely possible
existents. It would leave us in the dark as to the referent of
'John(t,t')' in this, the actual world and it would tell us nothing as
to the relation, if any, between John(t,t') and John himself.
Lesniewski's own motivation in introducing the notion of temporal
segment seems to have derived on the one hand from his timeless concep-
tion of truth, and on the other hand from those physical theories (of
'spacetime', 'world lines', and so on) which have grown out of the idea
that there is a certain analogy between the ~atial and temporal
dimensions of the entities treated of by physics . We shall assume,
then, that on the Lesniewskian view we are to regard each object as
a four-dimensional whole, capable of being cut up into parts in any of
its four dimensions. Phases result when objects are sliced in the
temporal dimension. Why, now, must such a view embody a conception of
the way in which the spatiotemporal world is parcelled into separate
entities that is in conflict with the Aristotelian ontology of things?
To answer this question we must understand what it is for an object to
change (to admit contrary accidents at different times). Consider, to
this end, the following passage from Zemach' s important paper "Four
Ontologies" of 1970:
An ontology may construe its entities as either bound or continuous

in time and in space. An entity that is continuous in a certain
dimension is an entity that is not considered to have parts in the
dimension in which it is continuous. It can be said to change or
not to change in this dimension, but what is to be found further
along in this dimension is the whole entity as changed (or
unchanged) and not a certain part thereof. The opposite is true of
an entity's being bound. If an entity is bound in a certain
68ThiS analogy is of course very rest.rict.ed (see e.g. Mellor 1981, pp.
66f., 128ff.). The accept.ance of t.he concept. of a world line in a four-
dimensional cont.inuum is moreover fully consist.ent. wit.h a cont.inued
belief in t.he ont.ology of t.hings (Sellars 1962, p.678). One might. indeed
go furt.her and argue t.hat. t.he concept.s of t.he four-dimensional ont.ology
t.hemselves presuppose t.he I-hing onl-ology for t.heir coherent. formulat.ion,
that.., for exomple, t.he idea of a world line makes senSe only if t.here is
some ident.ical ~ t.hat is tracked from one t.ime-point. on t.he line t.o
anot.her (see Simons 1987, pp.126f.). See also Brent.ano (1976, pp.296ff.)
for furt.her criticisms of t.he view of t.ime as a 'fourt.h dimension' of
space.
dimension, then the various locations along this dimension contain

its parts, not the whole entity again.
(Zemach 1970, pp.231f.)
Thus the Aristotelian substance ontology is an ontology which sees

substances as, in Zemach's terms, continuant in time and bound in space.
We see the same substance again on successive occasions, not a different
slice thereof69 • The four-dimensionalist phase ontology, in contrast, is
one which sees entities as bound both in space and in time, i. e. as
having both spatial and temporal parts. Entities so conceived are
excluded entirely from change. That a four-dimensional whole has red
phases and green phases no more signifies a change, than does the fact
that my pen is at one end red and at the other end green.
Interestingly, now, the term proposed by Zemach for the four-
dimensional wholes that are accepted by Lesniewski is the term event:
An event is an entity that exists, in its entirety, in the area

defined by its spatiotemporal boundaries, and each part of this
area contains a part of the whole event. There are obviously
indefini te ly many ways to carve the world into events, some of
which are useful and interesting (e.g. for the physicist) and some
of which - the vast majority - seem to us to create hodge-podge
collections of no interest whatsoever. Any filled chunk of space-
time is an event... When philosophers and physicists talk about
spatiotemporal worms, about point-events, or about world-lines,
when they describe material things as· 'lazy processes' and refer to
spatial and temporal slices of entities, they are using the
language of this [event] ontology.
(Zemach 1970, p.223)
Now it is no mere terminological matter to suggest that the phase

ontology brings us close to an ontology of events. Indeed, Lesniewski
himself, while critical of the specific formal treatment of the event
ontology that is gi v~n by Whi tehead7o , seems to have been not too
negatively disposed to the idea that objects and events may constitute a
single category. The whole tone of Kotarbinski's Elementy, however, is
precisely counter to an outcome of this sort, and there is not a little
irony in the fact that Kotarbinski (like Lejewski in our own day) sees
no incongruity in doing away with events via 'onomatoids', and then
(apparently) resurrecting them via temporal parts.
Certainly Lesniewski does not countenance anything like the dissol-
ution of things into events or processes that was envisioned by, say,
Heraclitus, Schopenhauer or Bergson. For not every four-dimensional
whole is such as to count as a 'thing' from Lesniewski's point of view.
His reasoning seems to have been, rather, that it is possible to
69See also t.he discussion of 'exist.ence in t.he ·present.' in Ingarden

(1964/65, § 30).
70Sinisi (1966), summarising part. of ch.IV of Le~niewski (1927/31).

166 BARRY SMITH
restrict the totality of four-dimensional wholes in such a way that the

resulting ontology will remain more or less in keeping with our presup-
posi tions corncerning things or concreta. Thus, we might say, a four-
dimensional whole, before it can be admitted by Lesniewski into his
ontology, must satisfy the two-fold condition that it be (1) 'bulky and
lasting' (Le. extended in all its four dimensions) and 'such as to
offer resistance', and also (2) such that all its (bulky and lasting)
parts are resistant in the relevant sense71 • It is, however, far from
clear that such conditions can of themselves suffice to transform an
ontology of four-dimesional wholes into an ontology of 'things' in the
sense of the reist. Indeed, there is a suspicion that they involve
a surreptitious smuggling in of the goal to be achieved (not least
because the phrase 'offer resistance' seems to belong still to the
language of continuants or substances).
Not merely does the phase ontology have no room for change. It
precludes also any account of what we might refer to as the 'stability'
of enduring substances. As W.E.Johnson points out, the ontology of four-
dimensional wholes springs from that post-Humean doctrine which regards
change as fictitious and substitutes for it "merely differently charac-
terized phenomena referred to non-identical dates." It may be that for
most scientific purposes "no more transcendental conception than that of
a whole constituted by the binding relations of time and space is re-
quired; and hence the philosophers who reject the conception of a con-
tinuant are satisfied to replace it by the notion of such an extensional
whole." What, however, is to explain, on this account, the stability of
that spatiotemporal nexus which connects, for example, the successive
'phases' of a living organism? As Johnson argues, a mere succession of
processes "offers no explanation whatever of what in objective reality
determines the stability of any given nexus." (Johnson 1924, pp.l00f.)
§11. In Defence of a Bicategorial Ontology
There is an assumption running through the thought of Lesniewski, Kotar-

binski and the followers, as also through that of Zemach, to the effect
that the most worthy aim of the ontologist is that of producing a
monocategorial ontology - and more generally of demonstrating that one
or other sort of eliminative reduction can be achieved. A more natural
resolution of the problems raised in our reflections on time and change,
and on tensed predications like 'John ·is jumping', is achieved, however,
if we abandon this concern with reduction and embrace instead a shame-
facedly bicategorial ontology of things and events, the latter being
conceived as changes in things, as dependent particulars (after the
fashion of, say, accidents in the category of action as these are
~lThis is to exclude, f'or example, the case where a .e,~-na {.u:te bulky and
resistant whole is united with, say, an empty volume of' spacetime.
conceived in the Aristotelian ontology72). Events, we can say, occur in

or between things, and are no less individual than these things them-
selves. They may be instantaneous or extended in time, and in the latter
case they have temporal parts which are themselves events. On this basis
we can now go on 'to distinguish clearly between a thing, on the one
hand, as something that is given in toto from'the very first moment of
its existence, and the 'life' or 'history' of this thing on the other
hand, as a certain type of complex event, bound up inseparably with the
thing whose history it is. The purported temporal parts of things will
then turn out to be parts of such complex events, so that the cleavage
between John as child and John as adult can be recognised, commensen-
sically, as a cleavage not in John, but in his life or history. Note,
incidentally that there is no comparable move in regard to spatial
parts. We cannot say that these are really parts of a substance's shape,
for example, or of the space a substance occupies.
Clearly, the ontology of things and events provides a pecul iarly
simple account of what makes 'John is jumping' true. More interesting,
perhaps, is the fact that this same ontology can account for certain
properties of the linguistic phenomena of verbal aspect73 , properties
which are important for us here in that they reflect a parallel on the
side of the verb to the opposition among nouns between 'mass' and
'count' (as for example between 'sugar' and 'snow' on the one hand, and
'tiger' and 'ox' on the other). The former correspond to verbs of
progressive and continuous aspect ('John knows how to jump', 'John's
been jumping all day'), the latter to verbs of achievement ('John jumped
over the ridge', 'John just jumped to victory').
What this tells us in ontological terms is that the opposition
between what is 'unitary' and what is 'mass' or 'collective' is to be
found not merely in the realm of things but in the realm of events, too,
though in an interestingly more complex form. For while events of
reddening or exploding or whistling, as well as institutional affairs
such as weddings, funerals and running of races, are all such that they
have or could have temporal parts, they may also - by inheritance from
the (moving and extended) things which support them - be extended in
space. Hence they may participate in the opposition between 'unity' and
'mass' in two distinct dimensions: events may be spread out either in
time, or in space, or of course in both.
Consider, for example, the process of jumping. This is made up, we
may suppose, of minimal unitary temporal parts - which we may call
'jumps' - comparable to • sUbstances' in the world of things. Each jump
may then be analysed in turn as a continuum of bodily movements of
a certain sort.
We can distinguish, in the world of things, not only what might be
called • substantial' parts (lumps of sugar, molecules of water), but
also atomic parts, which are marked by the fact that they have no parts
of their own. This distinction, too, can be drawn in the realm of
72See Simons (1983) for a formal t.reat.ment. of a view along t.hese lines
wit.hin a Lesniewskian framework.
73See Mourelat.os (1981), Galton (1984, Appendix II), Hoeksema (1985, ch.6).
168 BARRY SMITH
events, where we can distinguish on the one hand unitary events which
take time but have no homogeneous sub-events as parts, for example
judgings, decidings and so on, and on the other hand events which are
strictIi' punctual, such as beginnings, endings and instantaneous
changes 4. Lesniewski's Ontology and Mereology, now, have shown them-
selves surprisingly adept at coping in a formally rigorous way with
relations such as this in so far as these are manifested among things.
Truth, however, is a relation which involves not only things and the
names of things. It involves also verbs and that in reality to which
verbs correspond, which is typically an event of one or other sort.
Hence we can begin to understand why it is that the bicategorial ontol-
ogy may be particularly suited to the task of giving an account of what
makes sentences (particularly empirical sentences) true. For it allows
us to take account of just those differences in reality which are
reflected in language in the differences of verbal aspect (differences
which are preserved, incidentally, even if we move over to a language
shorn of tenses, of the sort that was favoured by Twardowski,
Lesniewski, and other proponents of the 'absolute' theory of truth).
912. Qualitative Extensions
Return, however, to the properly reistic ontology and to our original

pair of questions: what is the referent of 'a jumper' in 'John is a
jumper', and what is the relation between John and a jumper that is
expressed by the copula 'is'? Our goal was to provide answers to these
questions in monocategorial terms in a way that would tell us what
things make 'John is a jumper' true. Kotarbinski's Elementy, at a number
of points, suggests an answer to this question that comes interestingly
close to simulating the effects of the bicategorial ontology of things
and events discussed above. This answer, which is nowhere to be found in
Lesniew~ki, rests on the idea that 'is' in the given sentence expresses
a special kind of relation of part to whole. 'A jumper' - or what might
now be called 'jumping John' - is, on this reading, the name of a
special sort of transiently existing thing, in which John himself is
included as part. The idea here is that things may at certain times
exist as it were in a raw state, but that they may on occasions extend
themselves qualitatively, or become modified in certain ways (by what
the tradition called 'accidents' or 'modes'), so that John, for example,
may on occasions become jumping John or cursing John or sleeping John,
and so on. John himself will survive in each of the latter - though of
course, because of the semantic restrictions imposed by Kotarbinski, we
cannot isolate that which gets added to John to yield the various
'qualitative extensions' in which he may partake.
74It is above all Ingarden who has invested effort in the ontological
analysis of these distinctions. See his <1964/66, ch.V).
It is more than anything else Kotarbinski' s exa.mples that suggest

this qualitative extension view. Thus he tells us that "it is obvious
that only things are stimuli: burning flames, sounding strings, pressing
solids, etc." (1966, pp.434f.), and these are examples which seem to
imply not only that Kotarbinski is intending to refer to things that can
survive and acquire and lose accidental determinations in something like
the Aristotelian sense, but also that the result of a thing's acquiring
an accidental determination may be a new thing, qualitatively extending
the thing wi th which we began: a str ing becomes a sounding string,
solids become pressing solids, a match becomes a lighted match, and so
on. The qual itative extension view allows, moreover, a particularly
simple interpretation of Kotarbinski's views on psychology: a thinker is
a body that is qualitatively extended in a special (deliberating, worry-
ing, deciding) sort of way. There is not only jumping John, but also
thinking John, hoping John, dreaming John, and so on.
Even after the Elementy there are hints of this quali tati ve exten-
sion view in Kotarbinski's treatment of words and sentences as things.
Thus consider the following passage from a piece first published in
1954:
a linguistic sign is for us a physical body, whether it is a
graphic sign or an accoustic sign (in the latter case it is a cer-
tain amount of a.ir vibrating in a specified way); thus it is a
thing, and not a process in the sense of a specified changing of
something.
(1966, p.399, emphasis added)
The same amount of air, considered as enduring thing, is vibrating now

in this way, now in that, and it is a different 'accoustic sign' in the
two succesive cases. As such passages reveal, however, there are certain
consequences of the qualitative extension view that run counter to our
commonsense understanding of 'thing'. For things, now, include not only
Tom and Dick, but also Tom-the-jumper, Dick-the-thinker; not only
quantities-of-air but also quantities-of-air-vibrating-in-a-certain-way.
Note, however, that while qualitatively extended things may exist for a
very short time, so that they typically do not acquire special (proper)
names of their own for purposes of re-identification, it seems that all
the qualitatively extended things seemingly admitted en passant by
Kotarbinski do have some duration, however short (so that their
existence is never punctual, though it may be intermittent). This
reflects, perhaps, the acceptability of names like 'dying Jim', 'racing
Tom', etc., where 'ending-his-process-of-dying Jim' or 'beginning-to-
run-a-race Tom' are ungrammatical.
Of course not all complex names of the given sorts need be given
the qualitative extension interpretation. Thus it may be that Lesniewski
can entirely avoid this interpretation by means of his 'phases' (though
again, it is difficult to see how this ploy will allow us to distinguish
for example 'jumping John' from 'cursing John' when jump and curse are
simul taneous). Moreover, there are cases of expressions of the form
'---ing N' or '---ed N' where 'N' is bona fide name but the '---ing' or
'---ed' a merely modifying adjective which brings about a diminution or
170 BARRY SMITH
cancelling of the content expressed by 'N'. Consider expressions like

'missing arm', 'assassinated president', 'shattered vase', 'annihilated
electron', and so on75 •
The qualitative extension view does however give us an elegant
account of what makes sentences like 'John is a jumper' true, in a way
that involves reference exclusively to things. John himself and his
quali taU ve extension are not identical; the former is a part of the
latter. Note, however, that there is no third thing which, when added to
the 'former would yield the latter. Hence the usual mereological
remainder principle fails to hold. The idea of a mereological theory in
which the remainder principle is weakened, or even suspended tr is far
from incoherent, as a number of algebraic parallels testify 6. This
failure of the remainder principle will however suffice to render the
qualitative extension' view unacceptable in the eyes of Lesniewski's more
devoted disciples.
§13. Kotarbinski and Brentano
The interest of the view in question is for us largely historical. There

are traces of the idea of qualitative extension in Aristotle, for
example in the passages in the Metaphysics where Aristotle raises the
question whether Coriscus and musical Coriscus are one and the same77 •
Its first coherent defence is however given by Brentano:
Among the entities that have parts, there are some whose whole is
not composed of a multiplicity of parts; it appears much rather as
an enrichment of a part, though not as a result of the addition of
a second part. One example of such an entity is a thinking soul. It
ceases to think and yet remains the same soul. But when it starts
to think again no second thing is added to that entity which is the
75See Twardowski <1894-, pp.12f., Eng. pp.l1f.) and also his (1923). Twar-
dowski's t.heory was derived ill t.urn from Brent,ano <1924-/25, vo!.II, p.62
Eng. p.220). See also Husserl <1979, p.309), Mart.y <1884-, pp.179f'.;
1895, p.34-; 1908, pp.518f.) and Lesniewski <1927/31, p.4-8, n.78). There
may be a remnant. of t.his doct.rine of' modifying expressions also in
Kot..al'binski's not.ion or 'subst.itut.ive rendering's' or 'onomat..oids' dis-
cussed above.
76Thus , for example, just. as we can have a pseudo-Boolean algebra wit.h

pseudo-complement.s <see e.g. Rasiowa and Sikorski 1963, pp.52f), so also
we might, dist.inguish a family of pseudo-mereologies.
77Me t,. 1206 b 16; see also 1024- b 30, 1018 a 2, 1030 b 13. See also
Suarez's discussion of t.he acc-i4e1w CGn.c~ in his Disput.at,iones
Met,aphysicae, XXXIX, s.1,n.l0-12 and cf. Cajet.an, In De Ent,e et. Essent,ia
d. Thomas Aquinat,is, §§ 153f·f'.
soul. What we have here, then, is not like what we have when one
stone is laid alongside another or when we double the size of a
body. •.• The substance is a thing and the accidentally extended
substance is again a thing, though but a thing not wholly other in
relation to 1;he substance; hence we do not have that kind of
addition of one and one that leads to a plurality.
(Brentano 1933, pp.53f., Eng. pp.47f.)
Return, for the moment, to our list of 'stages' in the development

of reism. Stage 5 in this development was provoked by a letter Kotarbin-
ski received from Twardowski in 1929 on the publication of the first
edi tion of the Elementy, in which Twardowski pointed out that the
doctrine of reism had been propounded already some years earlier in
a series of pieces dictated by Brentano towards the end of his life and
appended to the second edition of his Psychologie78 • Kotarbinski, in
response, came to see the need to add to his reflections on reism
a certain historical dimension. Above all he began to stress the
difference between his own pansomatist views and the reistic views he
attributed to Brentano. In his paper on "Brentano as Reist" of 1966,
Kotarbinski points out further that Leibniz, too, could be viewed as
a precursor of reism, not only in the light of his monadology but also
in reflection of his principle that all formulations containing names of
abstract objects should be avoided. Kotarbinski sees himself, however,
as the only "consistent and conscious" somatist reist. He describes
Leibniz as a "spiritualist reist", Le. as one who accepts souls or
spirits as the only type of things. Brentano he describes as a dualist
reist who accepted into his ontology both bodies (res extensa) and souls
(res cogitans): "As a former priest, [Brentano] stopped at the threshold
of somatism and never crossed it." (1966, p.428) We shall investigate
below the extent to which this is an adequate account of Brentano's
reism and of its relation to that of Kotarbinski.
Brentano, familiarly, distinguishes three sorts of ways in which
a subject may be conscious of an object in his mental acts: in presenta-
tion, in judgments and in what he calls "phenomena of love and hate",
a category embracing feelings, emotions and all other value or interest
phenomena. In presentation the subject is conscious of the object, has
it before his mind, without taking up any position with regard to it.
The object is neither accepted as existing nor rejected as non-existing,
neither loved as having value nor hated as having disvalue. A judgment
arises when, to this simple manner of being related to an object in
presentation, there is added one of two diametrically opposed modes of
relating to this object, which we might call acceptance or rejection or
'belief' and 'disbelief'. A judgment is in effect either the affirmation
or the denial of existence of an object given in presentation, so that
Brentano embraces a view according to which all judgments are reducible
78Cf . Kot,orbi';ski <19660, p.4-61, Eng. p.196>, Brent,ono <1924'/26, pp.228-

277, Eng. pp.330-368).
172 BARRY SMITH
to judgments of existential form '9 .

A positive judgment is true if the
object of the underlying prespntation exists; a negative judgment is
true if this object fails to pxist.
The early Brentanian ontology of 'things' or 'objects' arises, now,
wllPll olle turns from the psych%g.v of pr-esentation to an investigation
of thp non-ps?cho logi cal corre lates of presE'nt ing acts. 'Object' is
accordingly t.o be understood as: 'possible correlate of presentation' j
it: is a t(>rm whosp meaning we can understand only by reflecting on the
meaning t)f the term 'presentation', the latter term itself being such
that we grasp its meaning directly on the basis of our own intuitive
pxperience of our acts of presentation8o •
According to Brentano's earlier view, there exist non-real objects
('entia rationis', later called 'entia irrealia') of various kinds -
mental contents or 'objects of thought', universals, states of affairs,
possibilities, lacks and so on - all of which can be given in presenta-
tion and affirmed or denied in judgment. This points to a distinction,
accepted by the early Brentano and his followers, between the existence
or non-existence of an object of presentation on the one hand, and its
reality or non-reality on the other. Thus what exists (for example
values or universals) need not be real, and what is real (for example
centaurs or chunks of wooden metal, and even the objects of simple acts
of sensation) need n~t exist. These two oppositions, Brentano held, are
independent of each other, and only the former is involved directly in
the correctness or incorrectness of a judgmentSI •
Later, however, Brentano moved to a view according to which 'real-
ity' and 'existence' would be equivalent, so that everything which
exists is an ens reale. Brentano's change of mind occurred, in fact,
~Iring the period when Kotarbi~ski was studying in Lw6w. However, it was
initially made known only to Brentano's closest associates, so that we
can rule out any influence of Brpntano's later view on Kotarbinski via
hi.s teacher Twardowski. Kotarbinski's and Brentano's thing-ontologies
may however have a common source. Thus it is noteworthy that both views
arose, in part at least, in reaction to certain apparent ontological
excesses of Brentano's students, not least of Twardowski himself, a
react ion which led in both cases to a reversion to an ontology rooted
79(1924/25, vol.II, pp,56ff., Eng. pp.213ff.) - 8rentano later adopted an

amended version of this vie,,: see his <1924/25, vol.lI, pp.164-72. Eng.
pp.295-301). The topic of existential judgments "as taken up by a number
of philosophers in the years around the turn of the century and seems to
have played a role also in inspiring the subjecL of Le~nie~ski's disser-
t.ation under T",ordo,""ski in 1911. This includes Q discussion not. only of
8rentano's Psychologie but also of Twardowski's Content and Object and
of Husserl's Logical Investigat.ions. Le~niewski may a180 1 have been
influenced by the Sketch of a Theory of Existential Judgments of Hans
Cornelius, wit.h whom he st.udied in Munich.
80Cf . Smith (fort.hcoming) for a more detailed discussion of this point..
81(1889, § §50,55); Marty (1884, pp.171ff.); T"ardo"ski (1894, p.36, Eng.

pp.33f.)
effectively in the Aristotelian conception of thing or 'first

substance'. Moreover, as Ingarden pointed out already in the early
thirties, there is a sense in which the roots of both Brentano's and
Kotarbinski's reism are ,present already in Brentano's own earlier
existential theory of judgment, since - when once 'Sachverhalte' or
'facts' have come to appear suspicious - this has the effect of reducing
each judgment to a form which asserts either the existence or the
non-existence of some object82 :
It is not that the being of A must come into being in order for the
judgment 'A is' to be transformed from one that is incorrect to one
that is correct; all that is needed is A. And the non-being of A
need not come into being in order for the judgment 'A is not' to be
transformed from one that is incorrect to one that is correct j all
that is required is that A cease to be. And if only this happened
and nothing else ••. would there not be in this fact alone, which
relates to what is real, everything that is needed for the correct-
ness of my judgment? Without doubt ••• And thus the doctrine of the
existence of such non-things has nothing whatever in its favour.
(Brentano 1930, p.95j Eng. p.85)
It is important, however, to be clear as to the precise nature of

the respective views of Brentano and Kotarbinski. Brentano came to
believe that all objects belong to a single category, the category of
ens reale or Realitaten83 • The fact that he sometimes uses the word
'thing' [Ding] to refer to the entities in this category of itself tells
us little as to the extent to which he shared with Kotarbinski tenets of
the latter's reism.
Certainly there are a number of sometimes striking similarities
between their respective philosophies:
They agree, first of all, on negatives: for both philosophers, only
concrete individuals exist. There are no abstracta, no universals or
general objects, no properties, classes, meanings or concepts84 ,
Brentano's motive for rejecting all such entities being rooted in his
conviction that all that exists is completely determined, down to lowest
differences.
Secondly, they agree as to the importance of the dimension of sem-
antic or linguistic analysis as a complement to ontology: the apparent
grammatical form of an expression is not always its actual or ultimate
form. They agree also in the view that 'this actual form is to be
821ngarden put.s t.his view t.o Kot.arbinski around 1936 when t.he lat.t.er came
t.o Lv';w t.o give a t.alk on Leibniz as a precursor of reism. (Personal
communicat.ion of W, Bednarovski)
83He was followed in t.his by his lat.er disciples, above all Oskar Kraus,
Alfred Kast.il and 'George Kat.kov, who t.ook over from Kot.arbinski t.he word
'reism' t.o describe t.he lat.er Brent.anian view. See Kraus (1937, pp.
268ff.).
84 '
Brent.ano <192~/26" vo1.ll, p.162; Eng. p.29~)
174 BARRY SMITH
achieved by translation into the language of things85 •

Thirdly, and most importantly, Brentano agrees with Kotarbinski
(and against Aristotle) in allowing collectives of things to count as
things. Organisms, for Brentano, are collectives in this sense. Further,
nei ther phi losopher takes seriollsly the requi rement that t.hings should
form a unity:
By that r·/hich is when the expression is used in the strict sense,

we understand a thing .•. j a number of things taken together may
certainly also be called a thing, though one must not suppose that
the two parts of a thing taken together constitute an addi tiona]
third thing. For where we have an addition, the things that are
added must have no parts in common.
(1933, p.4, Eng. p.16)
Similarly they agree in allowing parts of things to count as things.

Hence both are 'actualists', in the sense that they believe that a part
of something actually real is itself actually real even when it is
a part. Aristotle, in contrast, may be referred to as 'potentialist', in
the sense that. he holds that parts of things are as such only
potentially rea1 86 • Neither in Brentano nor in Kotarbinski, we can say,
do we find any trace of the Aristotelian theory of act and potency and
of the hylomorphic conception of substances to which it led.
~14. Formal vs. Material Ontology
The points of disagreement between the two philosophers derive especial-

ly from the fact that Kotarbinski starts out with the idea that physical
bodies are the prime examples of things, and sees 'resistance' or the
ability to stand in causal relations as a distinguishing mark of the
concept thing. For Brentano, on the other hand, the concept of thing, a
concept which grew out of his earlier psychological doctrine of 'objects
of presentation' does not essentially have to do with the concept of
causali ty (with the opposition between what is inert or energetic and
what is non-inert or anergetic). And even though things as Brentano
conceives them are not, except in special cases, psychological entities,
there is nevertheless a sense in which even on this latter doctrine the
term 'thing' is a psychological term.
This is shown most clearly in the different arguments the two
philosophers bring forward to defend their respective views. Brentano's
argument for the ontology of ens reale rests on the fact that the
85Cf . Brent.ono <1930, pp.87-97, Eng. pp.77-87; 1956, pp.38-4.8), Srzedni-

cki (1965, p.4.8), Chisholm (1978).
86See Met.. Z, ond t.he discussion in Smit.h (1987).

univocityof 'presentation' implies the univocity of 'thing,87. Kotarbin-

ski's argument for his own reistic ontology, in contrast, is negative in
form, resting on the unacceptability (for a variety of reasons) of
theses to the effect that there exist universals, facts, classes and the
like66 •
The Brentanian concept of thing is, we might say, a formal concept.
It is, in other words, a concept capable of applying without reservation
to objects in all material categories - si.nce, for Brentano ll objects in
all material categories may serve as objects of presentation 9. And then:
It doesn't matter at all what word we use to refer to the concept

which is common to all that is to be presented. Whether we speak of
'thing' or 'entity', it is endugh that it represents a highest
universal to which we attain by means of the highest degree of
abstraction no matter where we look.
(1930, p.108; Eng. p.96)
Aquinas, too, sees the concept of a thing or of what is real as the most
general concept to which reason can attain90 , and a broadly similar view
is present for example in Husser!' s Philosophie der Ari thmetik, where
the purely formal concept of an Etwas or 'something' lies at the basis
of Husserl's theory of number and counting.
For Kotarbinski, on the other hand, 'thing' is a term of material
ontology, to be understood by reflecting on specific sorts of examples
of thing and of the meanings of terms like 'bulky', 'extended', 'inert',
and so on, whose significance is confined to the region of physical
bodies.
Marty is in this respect closer to Kotarbinski than he is to
Brentano, since he pursued energetically the idea that the concept of
thing or ens reale should be confined to those entities which partici-
87<192.t/26, Yol.II, pp.162, 213r., Eng. pp.29.t, 321f.; 1933, p.18, Eng.
p.2.t; 1956, p.38)
88These negative arguments are supplemented by an appeal to the fact that

the language of things is psychologically more naturaL Thus one natural
way of explaining .the meanings of words is to eliminate substitutive
terms: "Should we wish i.o explain to a child what the word 'similarity'
means, should we not.. show him. in turn several pairs ~f object.s which
look alike?" <1966, p,.t23)
890n the opposition between formal and material concepts see Smith (1961>;
on Breni.ano and rorlllal concepts see Munch· (1986); on formal YS. mat.erial
ontology see Ingarden <196.t/66, esp. §9).
90Albert the Great defends a similar view or -fJeln.{r as t.he simplest

concept from which no further abstract.ion can be made, and according to
Brentano (1930, p.108, Eng. p.97) a parallel notion is als>:> present in
Aquinas.
176 BARRY SMITH
pate in causal relations (are 'energetic' in Marty's terms)91. However,

Rrentano argues, in transforming the concept of thing into the concept
of what is capable of standing in causal relations, Marty has "permitted
himself to deviate from long-established usage":
a term which has traditionally been the most simple and the most
general of all our terms has hereby been transformed into a sophis-
ticated thought-combination which has been a matter of controversy
since the time of Hume. Given Marty's sense of the term 'thing', we
wOllld have to say that according to Hume and Mill and many others,
there are no things at all!
(1930, p.l0B, Eng. p.96)
The opposition between physical and psychical things, too, is an

opposition formulated in material-ontological terms, so that to describe
Brentano as a 'dualist' is to misunderstand the formal nature of his
views. Certainly Brentano accepts spiritual substances (souls) as poss-
ible objects of presentation. And he accepts three-dimensional bodies
also. However, the psychological origins of Brentano's views imply that
it is not at all clear that he accepted as objects of presentation
physical bodies in Kotarbinski's sense. Thus material things, for
Brentano, are not restricted to three-dimensional physical bodies. They
embrace first of all 'topoids' of higher numbers of dimensions, which
conceivably exist as it were alongside the more familiar three-dimen-
sional bodies given in perception. Brentano's concept of thing embraces
further things of lower numbers of dimensions, above all boundaries
(points, lines and surfaces). Note, however, that while Brentano does
not rule out topoids of higher numbers of dimensions, he rejects the
idea that three-dimensional bodies might turn out to be boundaries of
four- or more-dimensional topoids. This is because a boundary can exist
only as the boundary of the thing which it bounds. Thus spatial and
temporal points, on Brentano's conception, never exist in isolation from
the things; extended in time and space, of which they are the bound-
aries. A body, on the other hand, is a thing in its own right, which
requires no other thing (except possibly God) in order to exist92 •
Kotarbinski, in contrast, reflects not at all on the nature of
boundaries. More generally his work, like that of Lesniewski, lacks any
topological dimension; things are seen as being arbitrarily divisible
and as arbitrarily conjoinable (without even the topological requirement
of connectedness in the latter case). This is despite the strong
tradition of topology in Polish mathematics, and despite the fact that -
as is shown for example by Tarski and Grzegorczyk (1977) ,- topological
axioms can be added very easily to the axioms of Lesniewski's Mereology.
Brentano accepts also certain sui generis zero-dimensional things,
which he calls souls. These have the capacity to comprehend intention-
ally things of all higher dimensions 93 , a notion which recalls
91<1908, §66)
92See Brent-ano (1933, p.l08, Eng. p.96; 1976, pp.38, 95).
93S <>" Br"nt-ano <1933, pp.158f., Eng. pp.119f.; 1976, pp.20,120f.).

Aristotle's dictum in De Anima (429 b 25ff., 430 a 14f.), to the effect

that the soul is somehow everything, for its nature is to be able to
know everything and therefore in a certain sense to include everything
within itself. Like the Leibnizian monad, so also the Brentanian soul is
unextended, and therefore not continuously many; yet it is for all that
continuously manifold, comparable in this respect to the midpoint of
a disc divided radially into segments of continuously varying colours94 .
The two philosophers differ further in virtue of the fact that, for
Brentano, not all things need be perceivable. Thus souls are not per-
ceivable, or at least not directly: we can apprehend intuitively at most
the activit ies of the soul (the soul as modally extended in certa in
ways)95. Further, topoids of greater numbers of dimensions would not be
perceivable; and nor, either, would the empty spaces which Brentano came
to accept at the very end of his life.
The most radical difference between the two philosophers turns,
however, on the fact that Brentano takes tense seriously in the sense
t~at for him 'exists' is in every case synonymous with 'exists now, in
the present moment', so that everything exists for Brentano only
according to a boundary {einer Grenze na.ch}. This means that every
existing object is as it were punctual from the point of view of its
temporal extension, though always in such a way as to depend for its
existence on that which has just existed or on that which will exist as
supplying the continuum which it bounds96 •
For Brentano, then, it is as if the world of things is continuously
annihilated and recreated anew with each successive passing instant.
'Past thing' and 'future thing' do not, therefore, refer to special
kinds of things, but are modifying expressions, to be compared with
'hoped for thing', 'imagined thing' and so on. Our apparent references
to the past and future are in fact in every case references to the
present as set apart temporally either as future or as past from some-
thing else. This 'something else' refers, however, merely modo obliquo
and embodies no ontological commitment (in much the same way that our
reference to a believer in demons or in fates involves no ontological
commitment to demons or fates)97.
Brentano's world is, therefore, in this respect, too, entirely dif-
ferent from that of Kotarbinski, for it is a world in which there exists
only one instant of time (even if this time is continuously changing).
Brentano in fact identifies what is real with what is subject to a cer-
tain continuous temporal transformation which is simultaneously a matter
of existing in the present, cea.sing to be future and becoming past98 •
94See Brent.ano (1976, pp.41ff.).

96AI'ist.ot.le, t.oo~ seems t.o exclude immat.eI'ial subst..onces from t.he realm
of sensible subst.ances. Cf. Met.., 988 b 26, 1040 b 31ff.
96 See
Bl"ent.ano (1976, e.g. p.37).
97 Bren t.ano (1976, Part. II, Sect.ion lID.
98 Bren t.ano (1976, Part. II, Sect.ion V).

178 BARRY SMITH
Kotarbi~ski certainly holds that all things exist in time. Yet he

reflects very little on the peculiar ontological features of things not
yet and no longer existing - though we may assume e.g. from his dis-
cussion of the 'things of history' that he accepts into his ontology
also past and future things'N.
The ontological views adopted by Brentano at the very end of his
life take him even further from reism in the Kotarbi~skian sense. Things
are now conceived hy Brentano as falling into two groups, which we might
refer to as places and sOllIs. Places may be 'empty' (lacking in all
(lualitative determination), or they may be qualitatively extended or
cnr iched (f i 11 ed by quali ties) in different ways. A three-dimensional
qual i tatively extended place is called a body100.
Brentano had earlier adopted a Cartesian view of space according to
which space and quality are mutually dependent. On his later view,
quality is one-sidedly dependent on space: space can exist without qual-
i ty but qual i ty cannot ex ist except as the qual i ty of some particular
place. A somewhat counterintuitive consequence of this view is that any
change of place or shape brings about the annihilation of the 'body' in
question, so that bodies cannot move. Movement is, rather, the becoming
modally extended in appropriate ways of a continuum of different places
in continuous temporal succession, rather like a ripple 'moving' across
the surface of a pond, no molecule of which is displaced in the horizon-
tal. There is no Chisholm (qlla. material object), but only a continuous
sequence of Chisholmy places, so that Brentano's later ontology implies
just that view which is at the basis of Leiniewski's logical grammar -
that there is a sense in which we need not distinguish between proper
names and predicates.
We can come to some understanding of the reasons why Brentano came
to choose places as the ultimate non-psychic substances if we examine
again the list of the marks of suhstance set out in li6 above. Places are
(i) individual and (ii) neither "present in a subject" nor such as to
"require a support from things or substances in order to exist".
(iii) They can exist on their own, Le. without being filled or qual-
i fied in any way. If, further, substance is identified with place, then
it becomes clear why substances underlie accidents and do not themselves
need accidents in order to exist (where it would seem that the organic
substances canvassed by Aristotle would depend for their existence on
processes of breathing, of metabolising, etc.). Places, from this point
of view, come to appear similar in this respect to the later Aristo-
telian concept of prima materia - both are infinitely plastic in the
sense that they can take on qualities ad libitllm. They come close also
to the undifferentiated Lockean I-know-not-what which would serve as the
ultimate support for the qualities given in experience.
99See (1966, pp.369f.). Cf. also <5J~. cU., p.191, "here Kotarbinski denies
the suggestion that 'is' "ould be an abbreviation of 'is no,,'.
100Brentano <1933, pp.35f., Eng. p.36)

Most importantly, as Brentano insists, places are (iv) the best

possible candidate for the role of that which individua.tes the accidents
by which they are filled. We can say that for the later Brentano a body
is the accident (qualitative extension) of a place, and that a place is
that which individuates one body from another. Two qualitatively
identical things at different places are distinct, as Brentano sees it,
only because their location is distinct101. Further, (v) places can admit
contrary accidents, being now filled by something red, now by something
black, and they are also (viii) "independent of thinking", and (ix) such
as to endure through time.
Only (vi), (vii) and (x) are less easily applied to the concept of
place. As we have seen, however, the mismatch in regard to (vi) can be
explained by pointing to the psychological origins of Brentano's views,
and in regard to (x) there is in fact no essential disagreement, since
while Brentano denies that things/places exist as a whole at all times
at which they exist, his view that they exist merely according to 11
boundary, a view quite different from the phase ontologist's view that
they exist only in one or other temporlll part, is in fact in keeping
with the views of Aristotle. Which leaves only (vii), which is indeed
rejected with great force by Brentano - as also by Kotarbinski - since
neither philosopher imposes on things a requirement of unity. From this
point of view, however, the concept pla.ce recommends itself still more
strongly as that in terms of which an account of material (non-mental)
substance is to be provided, since to regard things as merely (differ-
ently qualified) places is precisely to guarantee that arbitrary divis-
ibility and conjoinability of things on which both philosophers insist.
So strongly, indeed, does Kotarbinski identify the divisibility of
things into parts with their extension in space and time, that we may
argue that Kotarbinski, too, ought properly to have accepted the idea
that things are ultimately four-dimensional volumes of spacetime
a view that has been found attractive by not a few contemporary
phi losophers.
Note that in order properly to understand Brentano's view of (non-
mental) substances as places, and in order to make sense of its super-
ficially absurd implication that substances cannot move, nor change
their shape, it is important once again to recall the psychological
origins of Brentano's views. For Brentano had tended from the very start
to view the world of transcedent objects as something like a sensory
surface (as, say, the surface of the visual field). Objects come thereby
to be seen as similar in many respects to the images reflected on a
screen. They are capable of being demarcated as things and as parts of
things, and they are capable of being presented as moving, yet in both
cases we have to do not with autonomous properties on the side of the
objects themselves, but with mere ascriptions of properties to the
images we experience. It would take us too far afield to give a precise
101No t.e t.hat. ~ could not. be candidat.es for t.he role of t.hat. which
individuat.es one body from anot.her (as was held for example by Lot.ze);
for as we have seen, everyt.hing t.hat. exist.s is for Brent.ano such as t.o
ex i s t. .j,n lhe ~am.e U1ne.
180 BARRY SMITH
account of Brentano's views of autonomous reality. Clearly, however, his

thinking is at least in some respects comparable to that of Mach and
Einstein, both of whom sought to cast off 'metaphysical' assumptions
such as that of independent substance 102 •
~15. The Varieties of Reism
For all the divergencies between Kotarbinski's pansomatist reism and the
later formal ontology of Brentano, there is a clear sense in which they
are proponents of a common approach to ontology. This approach is shared
also by Lesniewski and his followers, as also by Quine, Goodman and
other modern nominalists. It can perhaps be characterised as an approach
which takes as its starting point in the construction of its ontology a
view of things drawing equally on examples of qua.ntities, ma.sses or
homogeneolls coll ecti \'es as on the uni tary substances of the tradition.
Thus it is contrasted with the approach to ontology of Aristotle 103 ,
Leibniz, Twardowski and Ingarden, which takes its cue primarily from the
uni tary substance and from the individual accidents which may inhere
therein.
There are, interestingly, a number of different routes taken by the
philosophers mentioned to the· homogenous collective view of things
('homogeneous' because the distinction between thing and mass is held to
reflect no fundamental ontological division). Thus Quine, for example,
seems to have been inspired particularly by those physical examples
(energy fields, liquids, gases) where arbitrary delineability does
indeed seem to hold, as also by related considerations deriving from the
semantic treatment of mass terms in· natural language. Quine, like
Brentano lO 4. and Kotarbinski I sees masses as full-fledged even though
possibly scattered individuals. Thus he regards as of no importance the
difference between what is spatially continuous and what is spatially
scattered 105 , and indeed his general approach is to view every object as a
four-dimensional section of the world, after the fashion of what Zemach
calls 'events'.
102Some clues as to Brentano's views are provided by the argument 1'or the
a fV1Ac't-iimpenetrability of bodies in his (1976, pp.1801'.). Brentano's
views in this respect seem to have been stimulated by the positivism of
Auguste Comte: see Munch (forthcoming).
103Here and in what follows we consider Aristotle exclusively as an

ontologist of first substance.
l04See Brentano's treatment of continua in Part I 01' his (1976).
106Quine (1960, pp.90-110)

ON THE PHASES OF RETSM 181
Field-theoretic physics played a role also in inspiring the later

Brentano's view of things as accidents of places, as also in securing
Brentano's acceptance of topoids of higher numbers of dimensions106.
Brentano's acceptance of the homogeneous collective view was however
motivated principally by his early work on the psychology of sensation,
and for this reason he may have resisted the idea that 'thing' involves
as one of its marks the concept of resistance or inertia. Goodman, too,
was provoked by considerations deriving from the psychology of sensation
in developing his ontology of 'individuals' in The Structure of Appear-
ance, and Quine was to some extent led to the homogeneous collective
view by psychological considerations concerning ostension 10 7 •
Lesniewski, on the other hand, was brought to his version of the
homogeneous collective view of things by formal considerations deriving
from the general theory of part and whole and from his critique of the
set-theoretic paradoxes, though such formal considerations played a role
of course also in the work of Goodman and Quine (as indeed in the work
of Whitehead).
Both the homogeneous collective ontology and the Aristotelian sub-
stance ontology are contrasted, now, with ontologies allowing general,
abstract and non-temporal entities of various kinds. Thus they may be
contrasted with the positions of, for example Bolzano and Frege, or with
Platonist ontologies of sets or classes. Bolzano, Frege and the set-
theoretical Platonists are, we might say, maximally liberal in the sense
that they impose on the entities admitted into their ontologies none of
the conditions of temporality, inertness, perceivability and so on that
have concerned us in the foregoing. Marty and the early Brentano are one
degree less liberal than this, in that, while they admit into their
ontologies entia rationis of various kinds, they insist that all such
entities enjoy a strictly temporal existence. They thereby recognise a
di vision among temporal entities between the real or energetic on the
one hand and the non-real or anergetic on the other. The bicategorial
nominalist ontology sketched in §11 above imposes the further restric-
tion that all entities be not only temporal but also energetic. The
bicategorialist does not, however, insist that all energetic objects
must also count as things, since he holds that events in his sense may
enter into causal relations. The later Brentano did however impose this
further restriction, though at the same time he abandoned the
requirement that all things must be 'energetic' in the sense of being
such as to enter into causal relations.
Kotarbinski can be said to have gone one step further than Brentano
in insisting that all things are physical bodies. Neither Brentano nor
Kotarbinski however lays any requirement of uni ty on the objects in
their respective ontologies, as contrasted with Aristotle qua
ontologist of first substance - who does impose a requirement of this
sort. All of which implies that there is a spectrum of gradually more
106See t.he "Appendix" t.o Bl'ent.ano (1933): "The Nat.ure of t.he Corporeal
World in t.he Light. of t.he Theol'Y of Cat.egol'ies", and also Bl'ent.ano
<1976, p.184).
107Cf . e.g. Quine <1960, p.6:n.

182 BARRY SMITH
restrictive positions, from the Platonism of abstract objects at tbe one

extreme to the
.
Aristotelianism of unitary• lOB
substances at the other,
a spectrum which may he represented as follows :
Platon- Bicat- Aristotle

All qua
ism egorial Bren- Kotar-
entities Marty
(Frege, Nominal- tano binski Substance
are: Ontologist
etc. ) ism
Temporal No Yes Yes Yes Yes Yes
Energetic No No Yes No Yes Yes
Things No No No Yes Yes Yes

Physical
No No No No Yes No
bodies
Unitary No No No No No Yes
We could go further, and extend the chart by taking into account opposi-
tions of other sorts, relatin§, for example to the issue as to whether
enti ties are or are not general l 9, independent of mind llO , atomic 111 , or such
as to have temporal parts 112 In addition we could think more carefully
about the different meanings of 'unitary', distinguishing for example
the requirement of connectedness of parts, the requirement of spatial
separateness from other entities, the requirement of functional inter-
dependece of parts, and so on. We could investigate further the extent
to which things may have parts which are themselves things - as an
organism may include cells, chromosomes, genes, etc. as parts113.
Already as it stands, however, the chart will enable us to see the
inadequacy of any simple-minded opposition between 'reism' on the one
hand and 'Platonism' on the other. Thus it would be wrong to go along
108Again, t.he reader should bear in mind t.hat. it. is by no means clear
where Arist.ot.le would have t.o be post.ed on t.his spect.rum if t.he whole of
his ontology ~ere laken into accoun~.
109There is a sense in which Twardowski admit.t.ed general t.hings. See

(1894., § 15), and also 1ngarden (1964./65, vol.l, p.219).
110Ingarden admit.t.ed bot.h aut.onomous and dependent. t.hings, examples of t.he

lat.t.er being creat.ures of fict.ion. See (1964./65, § 12 and ch.1X).
1111s t.he ult.imat.e furnit.ul'e of t.he universe it.self at.omic? Do all

ent.it.ies have at.omic part.s? See Sobocinski (1971), and also Bunt. (1985),
for an int.erest.ing t.l'eat.ment. of t.hese issues from t.he point. of view of
the semantics of mass and count expressions.
112Sge , again, Zemach (1970).
113See Woodger (1937), for an experiment. in t.his dil'ect.ion.

wi th Lejewski in his view that "Ontologists who oPRose reism are

believers in so called abstract entities" (1979, p.210) 4, a view dic-
tated no doubt by the fact that the principal enemies of the homogeneous
collective view in recent philosophy have been advocates of ontologies
based on the theorl of sets or of a more or less Platonistically
oriented semantics l1 • More recently, however, and especially with the
bringing to light of hitherto neglected aspects of the BrEmtanian and
Husserlian ontologies, it has become clear that reism has other, non-
Platonistic opponents. Thus for example it would not be to move into the
realm of abstract entities were one to embrace in one's ontology events
as well as things. For events may be accepted - as for example on
Dav idson' s account, and indeed on that of Whitehead - as bona fide
individual entities existing in time and space and entering into causal
relations with other events. More controversially, we may say that
Brentano (and Husserl) have shown how we may cope in a non-Platonistic
framework with those kinds of dependent, inseparable, divisive, inter-
penetrative parts which fall outside the purview of mereology as stan-
dardly conceived but which yet seem indispensable to an understanding of
the structure of minds, and of the different sorts of relations between
minds and other concrete objects.
114-Compare Lejewski's assert.ion in his (1976) t.o t.he effect. t.hat. event.s
are non-mat.erial object.s, so t.hat. anyone who admit.s event.s int.o his
ont.ology would be a 'Plat.onist.' on Lejewski's reading of t.his t.erm. This
reading is no doubt. derived from t.he fact. t.hat. event.s, for Lejewski, are
t.he ref'erent.s of abst.ract. nouns like 'swimming', 'f'alling', '\'alking'.
UOMore generally, ·t.hey have been advocat.es of what. Wolenski calls

"st.andard predicat.e logic semant.ics" in his (1987, p.170).
PHILOSOPHY OF THE CONCRETE
Klemens Szaniawski
The first formulation of the concretism, which then and afterwards has
been called reism, can be traced back to 1929. In Elementy in a para-
graph closing the discussion of ontological categories, we read as
follows:
" ... the entire reduction of categories, as outlined above, took place
precisely to their (Le. things' - K.S.) benefit. That reduction com-
pleted, it turns out that there remains only that category of objects -
that is, there are no objects other than things, in other words, every
object is a thing, whatever exists is a thing. When metaphorical, ab-
breviated, picturesque, in a word, substitutive, formulations are elim-
inated and replaced by the basic formulations, interpreted literally,
the latter will include no phrases which would appear to be names of
something other than things. They will be statements about things only.
But it must be emphasized here that by things we do not mean only inor-
ganic solids. Things are inorganic and organic, inanimate and animate,
and 'endowed with psychic li fe' - that is, they are both things in the
narrower sense of the word, and persons, too. So much for the reduction
of categories of objects to the category of things. The stand taken here
by those in favour of such a reduction might be called reism." (p. 55-6)
Even this first formulation reveals two non-equivalent variants of
the thesis of concretism. One of them is a statement about the world.
The fundamental philosophical question 'What exists?' is answered:
'Things and only things.' The other deals with language; it says that
all the statements made in a descriptive language are reducible to such
sentences which refer directly to things alone, i.e. that the only names
they contain are individual and general names of things.
The evolution of concretism, I think, consisted mainly in replacing
the ontological thesis by the semantical one and in substantial weaken-
ing of the latter; the statement about language has been transformed
into a prescription for its users; a doctrine became a programme.
The ontological variant has been abandoned (what does not mean
giving up the ideas that motivated it) mainly because of certain diffi-
culties involved in attempts to find a consistent formulation for it.
Briefly speaking, the point was that statements denying existence of
properties, relations, etc., and ascribing it to things alone, violated
185
J. Wolensld (ed.), Kotarbinsld: Logic, Semantics and Ontology, 185-190.
186 KLEMENS SZANIAWSKI
concretistic criteria of meaningfulness. For they contain, in the sub-

ject place, some expressions that are not names of things ('property',
'relation', etc.), what is inadmissible in a concretist's language, pro-
vided the sentences are to be interpreted literally. The thesis to the
effect that whatever exists, is a thing, remains true, but it appears
trivially true, since it results from a decision concerning language.
According to this decision we have to construct language in such a way
that Ii terally interpreted sentences should admit as substitution in-
stances of individual and predicate variables, only names of things.
Of course, the question arises whether the programme of concretism,
in its semant ical formulation, is workable. To put it otherw i se, is it
possible to impose upon the descriptive language - without impairing the
sense of what is said - the restriction formulated above? Or in still
another way: are scientific theorems reducible, in principle, to con-
cretistic formulations? The positive answer to these questions has been
objected to in various ways; today two of those objections seem es-
pecially important.
The first of them takes mathematics as its starting point: in the
language of mathematics the fundamental role is played by the concept of
class or set. Now it seems doubtful whether this concept can be given
the concretistic interpretation.
In fact, set can be regarded as a material aggregate of elements
(in this case, however, we had better call them parts). Stanislaw Les-
niewski belongs to the first propagators of this idea; concretism is in-
timately related to his' thought. The relation of 'being a part', how-
ever, is transitive and it becomes evident that cardinality is not an
absolute property of sets, but can be ascribed to them relative to the
way their parts are separated. That is why this concept of a set ('a set
in collective meaning') appears at least inconvenient for mathematical
construction. As is well known, the set theory makes use of another con-
cept of set ('set in distributive meaning') and gives it such an axio-
matic charact.erist.ics as to exclude the possibility of ascert.aining the
identity of any two sets having different members. If the term 'set' is
understood in this way then sets are abstract. It may be worth while to
quote here an exemplary statement formulated by Quine (1953, p. 114) in
his well known paper "Logic and the Reification of Universals", where we
read:
"The fact that classes are universals, or abstract entities, is some-
times obscured by speaking of classes as mere aggregates of collections,
thus likening a class of stones, say, to a heap of stones. The heap is
indeed a concrete object, as concret.e as the stones that make it. up; but
the class of stones in the heap cannot properly be identified with the
heap. For, if it could, then by the s~me token another class could be
identified with the same heap, namely, the class of molecules of stones
in the heap. But actually these classes have to be kept distinct; for we
want to say that the one has just, say a hundred members, whi Ie the
other has trillions. Classes, therefore, are abstract entities."
Now we should remember that the philosopher who formulated con-
cretism was aware, from the very beginning, of the fact that. sets in
di'stributive meaning were not things. In Elementy (p.12) he explains the
PHI LOSOPHY OF THE CONCRETE 187
difference between the two ways of understanding the word •set' as

follows:
"In its distributive meaning the word is an apparent term, in its
collective meaning, a real term. Its role as a distributive term is de-
fined in use when we say, for example, 'x is an element of the set of
M's' (or 'x belongs to the set of M's') means the same as 'x is M'''
And further (p.14):

"We shall ask now whether a general term denotes the set or the class
of its own designata. The answer is obvious if we have in mind the dis-
tributive meaning of the terms, if 'set' and 'class' occur as apparent
terms. In such a case there is no object which would be a set o·r a class
of the designata of a given term. Hence such a term cannot denote any
such thing, and cannot be predicated about any such thing in a true
sentence."
According to the above oplDlon, characteristic of the initial stage

of concretism, statements referring to sets in the distributive meaning
are eliminable from science, in particular from mathematics, by means of
substi tuting for them their equivalents which do not contain pseudo-
names of this sort. In the present version of concretism this opinion
has been abandoned. For the efforts to indicate a method of eliminating
from language expressions denoting sets in the distributive meaning have
failed. Even if the procedure of elimination proved applicable to
sentences of the form • x is A', where x is an individual name of the
lowest type, the problem of reducibility of sentences referring to sets
of higher types remains undecided. Yet sentences of this type are indis-
pensable in mathematics since natural numbers are defined by abstraction
with respect to a one-one relation, i.e. are regarded as sets of sets.
Corresponding higher levels of abstraction are involved in definitions
of rational, real, and compound numbers.
So much for the problems concretism must face within mathematics
and, in particular, within set theory. They still remain unsolved: "the
problem of the elimination of classes in their non-mereological (Le.
distributive - K.S.) interpretation remains a problem to be solved by
concretism." (p. 497) In this situation the original thesis about trans-
labali ty of all scientific statements into a language which admits no
names save names of things has been replaced by a postulate to aim at
the translability whenever it is possible. The question whether the
programme is universally realizable remains open.
Difficulties of different sort arise in connection with the concept
of thing, which is defined - hi conformity with common usage - as "an
object located in time and in space and having certain physical charac-
teristics." (p.495) The formulation, being of utmost importance for con-
cretism, is hardly compatible with the conceptual apparatus of modern
physics. For investigation of the structure of matter has now reached
the point where images determined by perception of objects from the
macroworld become completely useless. For example, is the field a thing
in the above sense of the word? Yet it makes - according to our present
knowledge - a fundamental constituent of physical reality •.
Concretism suspends judgment in this point and expects that physics

itself will offer a satisfactory explication of that and other concepts
which, at present, seem rather enigmatic and are susceptible to various
interpretations. Before this happens the reservations here outlined make
an additional reason for transforming the thesis of concretism into the
programme.
However, a question suggests itself, namely, what is the point in
accepting the postulate of concretism. It is obvious that ontological
position would bring a sufficient justification. If one believes that
there is nothing but concrete things (which - according to somatistic
interpretation - are bodies), then to refer to objects of other kinds,
in particular to abstracts, in such a way as if they really existed, is
to indulge in fiction. It seems, however, that the programme of concret-
istic ascetism in linguistic habits is acceptable, to a pretty fair
extent, also by those who admit a less restrictive ontology. It suffices
to affiliate a tendency directed against uncritical speculation, to aim
at the elimination of idola fori Bacon complained of. The requirement of
reducibility to concretes is the most radical weapon against the abuse
of language's control of our thought.
This position has been represented in philosophy for a long time.
We can find it in Leibniz,1 who, for that reason, may be regarded as a
precursor of concretism (although he endowed the concrete with a spiri-
t.ualistic interpretation).
"This is known also that it is the abstractions which occasion most
difficulty when it is desired to examine them minutely, as those know
who are acquainted with the subtleties of the scholastics, whose most
intricate speculations fall at one blow if we will banish abstract enti-
ties and resolve not to speak ordinarily except by concretes, and not to
admit any other terms in the demonstrations of the sciences, but those
which represent substantial subjects."
Concretism, on this interpretation, becomes an attempt to draw a
demarcation line between virtual theorems and statements that. only ap-
parently can be qualified in terms of truth and falsity. There is an air
of paradox about the fact that the possibility of a consistent realiz-
ation of the programme has been called in question by results of analy-
sis of the language of physics and mathematics, sciences of highest
rigour. After all, it is not those sciences that are in urgent need of a
cri terion which would enable them to get rid of some more ·or less evi-
dent nonsense. And yet, owing to a veto put by those sciences (to be an-
nuled in future?) the original version of concretisID, attractive due to
its radicalism, has been abandoned.
The intent of everything that has been written above is to recall,
in a concise way and not pretending to· originality, some basic ideas of
somatic concretism. There is an opportunity to do that, due to the occa-
sion of 90th birthday of the thinker who, for many years, has made this
philosophy an indispensable constituent of his Weltanschauung, as well
1"N&" Essays on Human Underst.anding", Book II, Ch. XXIII, § 1, in: The
.'hUosophical Works of Leibniz, New Haven, -1890, p. 340.
PHILOSOPHY OF THE CONCRETE 189
as an instrument of his activity as a teacher. I would like now to add a

few words on the later subject; the intimate character of what I am
going to say will be justified, I hope, by the intention of the present
paper.
I regard myself, as many of my older and younger colleagues do, a
pupil of Professor Tadeusz Kotarbinski. In what precise sense? For
within the relation of 'being somebody's pupil' one can discern several,
not equally important, layers.
T,he most common situation is that: the teacher transmits and incul-
cates a certain quantum of information, of sufficiently justified truths
belonging to a given branch of science. If he is successful in that, if
he is in a posi tion to arouse sOll\e interest in his listeners, to find a
sui table way of explaining difficulties away, then he has achieved a
great deal.
However, he may choose a more ambitious aim and strive for develop-
ing in his pupils intellectual efficiency, originality of thought, in-
quisitive ways of research. A one-sided contact proves insufficient,
what is necessary here is a systematic and competent dialogue, taking
into account the personality of the other. Some able lecturers fail to
become teachers in that sense of the word.
Finally, there are scholars - and Tadeusz Kotarbinski belongs to
them - who overstep the highest threshold of the teachership and found
schools. It is not altogether clear what the word 'school' means,
although the phenomenon itself is easily identifiable. Possibly, it may
assume different forms in different branches of science.
Sometimes it is held that the distinguishing mark of a school may
be found in a paradigm - to use a fashionable word - shared by its mem-
bers, t.hat is, an inventive conceptual apparatus or a widely applicable
method or a fruitful theory. I think, as a member of Kotarbinski's
school, that an equally important role should be ascribed, in particular
in the humanities, to the scientific ethos of the school's founder. In
Kotarbinski's case it can be easily specified: it consists in scrupu-
lous responsibility for word.
The responsibiity for word is the motivation for the concretisitic
attitude: the principle of reducibility to the concrete submits to pub-
lic control every statement that pretends to the status of a theorem
about reality. This attitude is also expressed in a systematic effort to
rebuild the concepts of philosophy, to endow them with the indispensable
univocality. It is known that Professor Kotarbinski, in his first uni-
versity lecture, recognized this objective as the most pressing. And he
succeeded in making his audience believe that this apparently unattrac-
tive work was, in fact, rewarding.
Of course, the attitude described here is not compatible with every
conception of philosophy. If one iden'tifies the aim of philosophy with
expressing one's own emotional attitude towards the world, then it is
impossible to accept any requi rements, imposed upon one's statements,
that would grant their testability regardless of individual modes of
perception. Wi thin the system of Kotarbinski' s ideas the function as-
cribed to philosophy, as well as to any system of theorem& about real-
ity, is purely cognitive. That.stimulates a far-reaching criticism to-
wards the philosophical tradition and some of its modern continuations.
If one shares those ideas he will not deny anybody's right to re-
flection upon individual existence; to emotional response, intensified
by conscious efforts, to the most general features of the world. On the
contrary: those experiences and attempts to express them are to be re-
garded as distinguishing traits of human consciousness. It is not there-
by implied that we lIay interpret as genuinely cognitive, statements
which purport to say something about the world, while in fact they are
but expressions of purely emotional attitudes. The above position is
followed in the philosophical work of the descendants of Kotarbinski's
school.
Nothing more can be achieved in science than to become the founder
of a school. Only few succeed in it; some highly distinguished scholars
fail to achieve that aim. Undoubtedly, a specific type of personality is
essential here: a strong individuality, concentrated on a clearly formu-
lated programme and, at the same time, open to others, gaining associ-
ates owing to attention paid to their development and careful observance
of the principle of partnership.
A scientific school creates peculiar values. In Kotarbinski's
school the value that has won the supreme position is intellectual
rigour. The intention of the present attempt to realize this fact is to
repay, if only in part, a debt of gratitude Kotarbinski's pupils owe to
their Teacher.
KOTARBINSKL MANY-VALUED LOGIC. AND TRUTH
Jan Wolenski
Twardowski's (1900) and tukasiewicz's (1910) are perhaps the most

important early writings in the Lvov-Warsaw School. Twardowski in his
paper offers a strong argument for the absolutness of truth and against
epistemological relativism. This argument has many ingredients but its
principal result may be condensed as:
(l) if a proposition A is true at the moment t, it is also true at the
arbitrary (past or future) moment tl,
which is, of course, equally applicable to falsity.
rhe second quoted work is devoted to an analysis of Aristotle's
principle of contradiction and its foundation. 1 Eukasiewicz' s view is
t.hat t.he principle of contradiction is not an obvious logical rule and
requires a proof. Eukasiewicz then argues that it is possible to reject
the principle of contradiction and use non-Aristotelian'logic, i.e. the
logic dropping this principle, without inconsistency resulting from such
a decision. Although his (1910) does not contain formal details of non-
Aristotelian logic, its admission as a coherent. instrument of reasoning
was tukasiewicz's first step toward many-valued logic. Eukasiewicz
(1910a) observes, at the same time, a close connection between strict
determinism and the principle of excluded middle: denying the former is,
as E'.ukasiewicz says, "probably" inconsistent with accepting the latter.
The problem posed by Twardowski and Eukasiewicz greatly influenced
discussions in Lvov philosophical circle in 1912 - 1913 and later; there
is no exagerration in the statement that Kotarbinski (1913) is the most
important contribut.ion to the debates in 1912 - 1913.
The main problem to which Kotarbinski (1.913) addresses himself be-
longs to pract.ical philosophy. This problem may be expressed by asking:
Are human creative actions possible at all? Kotarbinski points out that
strict determinism, especially predeterminism, excludes creative activ-
ity: nothing can be created in a strictTy determined world. However,
Kotarbinski continues, creative activities exist, what yields to the
conclusion that the universal validity of predeterminism is limited.
Kotarbinski is quite consious that such a factual justification is too
1See ~ukasievicz (1971) for an English summar-yo
191
J, Wolenski (ed.), Kotarbinski: Logic, Setnmltics and Ontology, 191-197.
192 JAN WOLENSKI
weak; in order to argue for the view that predeterminism is not univer-
sally valid he comes to considerationa about eternality and preeternali-
ty of truth.
The crucial concept, the one of truth, is defined in the following
way
(2) a proposition A asserting an object 0 is true if and only if 0
exists.
Now, every truth is eternal, i.e.
(3) for every A, if A is true at t, then A is true at every tl such that
t < tt.
Thus the problem consists in whether every truth is preeternal. If so,
then
(4) for any A, if A is true at t, then A is true at every tl such that
t > tl; of course, any preeternal truth is, by (3), eternal too.
According to Kotarbinski, some truths are preeternal but others are not.
To see this suppose that something, say 0, may be created by a human
action. Thus 0 does not exist until its creation and, by (2), proposi-
tions asserting 0 are not true. Let A be such a proposition. We know
that A is not true. Let us ask whether A may be false. If A were false,
its negation would be, by (3), eternally true. In that case it would be
impossible to create 0, which is contrary to the assumption. Thus A is
not a preeternal truth; there is a t, such that A is neither true nor
false at t.
Because the negative answer to (4) implies the existence of propo-
sitions which are neither true nor false, the problem of the law of
excluded middle arises. Kotarbinski considers the following forms of
this principle:
(5) for any A, either A or ,A is true;
(6) of any A, A is either true or false;
(7) for any A, if A is true, then ,A is false.
Kotarbinski points out that (5) and (6) assume the completeness of the
division of all propositions into true and false, which yields to the
equation
(8) true = not-false
On the other hand, (7) is independent of (8). Thus one may adopt (7) and
the negation of (8). An ontological counterpart of this combination is
the world with decided and undecided things. Respectively, all proposi-
tions are divided into definite (true or false) concerning decided
things, and indefinite (neither true nor false) which are related to
undecided things. (5) and (6) have to be restricted to definite proposi-
tions; (7) is a general rule. Moreover, we have the forms:
(9) for any A, A is either definite or indefinite.
This completes Kotarbinski's argument that the existence of indefinite
propositions as well as truths which are not preeternal does not violate
KOTARBINSKI, MANY-VALUED lOGIC, AND TRUTH 193
the law of excluded middle. Furthermore, Kotarbinski . links truth with

necessity, falsity with impossibility, and indefiniteness with possibi-
lity:
(10) if A is true, then A is necessary;
(11) if A is false, then A is impossible;
(12) if A is indefinite, then it is possible that A and it is possible
that ,A; possibility here means contingency.
Lesniewski's (1913) and (1913a) form the next chapter in the debate in
1912 - 1913. The general spirit of Lesniewski's (1913) arguments is not
connected with Kotarbinski's problem and, for this reason, may be
neglected in the present context. However, in the last part of Lesniew-
ski (1913) we find an argument against indefinite propositions. Lesniew-
ski defends the universal validity of (6) and tries to prove that inde-
finite propositions are impossible. Since the proof depends on a very
special convention concerning truth and falsity, I shall leave it
without further comments; it seems that Lesniewski's (1913) remarks did
not influence subsequent discussions. Lesniewski's (1913a) analyses are
much more important; in fact, that paper is entirely addressed to Kotar-
binski (1913). Lesniewski makes many significant observations on Kotar-
binski 's opinions but I shall concentrate only on Lesniewski's proof
that every truth is preeternal. He begins by explaining what the
statement that some truth, say A, is preeternal means. It means that if
A were ever uttered, it would be true at any moment t, providing that A
is true at present.
Now assume
(13) A is not a preeternal truth.
This means that
(14) A is true now, but there was a moment t such that A was not true at
t.
Further we have
(15) ,A was true at t.
Because both A and ,A are mutually contradictory, we reach (by (14) and
the principle of contradiction)
(16) ,A is actually false.
Now comes the crucial step in Lesniewski's proof. By the principle of
contadiction he concludes that
(17) ,A is always false.
Thus, by (17), ,A cannot be, contrary to (15), true at t and we must
abandon the assumption that A is not a preeternal truth. Hence, any
truth is preeternal. Of course, Lesniewski's proof presupposes an expli-
cit interpretation of tensed propositions and their values. On this in-
terpretation, the sentences 'Ceasar will cross the Rubicon in 49 B.C.'
and 'Ceasar crossed the Rubicon in 49 B.C,' have exactly the same mean-
ing because they can be transformed into 'Ceasar-in 49 B.C. is crossing
194 JAN WOLENSKI
the Rubicon'. The last sentence is timeless and may be uttered at any
moment t. The general interpretative strategy recommended- by Lesniewski
consists in putting the temporal index into the grammatical subject of a
sentence. This device enables Lesniewski to regard the sentences 'Ceasar
will not cross the Rubicon in 49 "B.C.' uttered before 49 B.C. and
'Ceasar crossed the Rubicon in 49 B.C.' uttered today as a pair of
contradictory sentences; on Kotarbinski's view the former is an indefi-
nite proposition while the latter is a definite one. Let us note that at
this point Lesniewski closely follows Twardowski's principle (1). Twar-
dowski himself also disagreed with Kotarbinski in his (1971)2. Kotarbin-
ski subsequently accepted Lesniewski's criticism as a conclusive one and
has abandoned logical indeterminism. 3
tukasiewicz himself did not take part in the outlined discussion
until 1918. On March 7th, 1918 he delivered the lecture at Warsaw Uni-
versi ty in which he gave information about three-valued logic: "In 1910
I publ ished a book on the principle of contradiction in Aristotle's
work, in which I strove to demonstrate that that principle is so evident
as it is believed to be. Even then I strove to construct non-Aristote-
lian logic, but in vain. Now, I believe, I have succeeded in this. My
path was indicated to me by antinomies which proved that ,there is a gap
in Aristotle's logic. Filling that gap led me to a transformation of
traditional principles of logic [ ••• ] I proved that in addition to true
and false propositions there are possible propositions, to which object-
ive pqssibility corresponds as third in addition to being and non-being.
This gave rise to a system of three-valued logic, which I worked out in
detail last summer. That system is coherent and self-consistent like
Aristotle's logic, and is much richer in laws and formulas [ .•• ] Poss-
ible phenomena have no causes, although they themselves can be the
beginnings of a causal sequence. An act of a creative individual can be
free and at the same fact affect the course of the world." (1918, p. 86)
Kotarbinski was, like the majority of young Lvov philosophers,
greatly impressed by tukasiewicz (1910). It is fairly evident that he
read and perhaps even participated in the meeting of the Polish Philoso-
phical Society in Lvov (February 26th, 1910) where tukasiewicz was
lecturing on the principle of the excluded middle. 4 On the other hand,
Kotarbinski (1912) goes far beyond tukasiewicz's remarks on the connec-
tion of the excluded middle and det~rminism. Thus we have every reason
to regard that paper as a very original contribution. Now, what about
the reverse influence, i. e. that of Kotarbinski on tukasiewicz? The
printed evidence is unfortunately very small - especially since Kotar-
binski's name is not mentioned in tukasiewicz' s works on many-valued
logic. Jordan (1963) claims that tukasiewicz in his treatment of strict
determinism as a semantic principle follows Kotarbinski's ideas; it is
2lecture notes based on Twardowski's course in ethics in 1923/24; note

however that the some course was also delivered by Twardowski earlier,
i.e. in 1~13/14-.
3See Kotarbinski <19157, pp. 13-14-) and <1974-).
4-~uka8iewicz <1910a) is an abstract of this lecture.

KOTARBI NSKI, MANY-VALUED lOGI C, AND TRUTH 195
not impossible, however, that Jordan is relying here on the oral tradi-
tion of the Lvov-Warsaw School. The only explicit sources are Greniewski
(1957) and (1958). He suggests on the basis of an oral report by Ajdu-
kiewicz that tukasiewicz was really influenced by Kotarbi~ski; according
to Greniewski, Ajdukiewicz was repeating tukasiewicz's own opinion. Gre-
niewski himself, impressed by Ajdukiewicz's report, claims that Kotar-
bi~ski invented a system of three-valued logic. As far as I know this is
all that one may find in the published sources.
Doubtless, some problems involved i.n tukasiewicz (1918) and his
subsequent works on many-valued logic are analogous to those involved in
Kotarbi~ski (1912), for instance, about creative human activities, prop-
ositions about future facts, propositions other than true or false,
possibility and preeternal objective contingencies. Thus it is arguable
that ideas expressed in Kotarbi~ski 's paper helped tukasiewicz to de-
velop three-valued logic. On the other hand, I do not agree with Gre-
niewski's thesis about Kotarbinski's being the inventor of a formal sys-
tem of three-valued logic. Indefiniteness may be taken as a third logi-
cal value but it is not the only possible interpretation. The second way
is to understand indefinite propositions as truth-value gaps. I see no
conclusive reasons supporting the view that Kotarbinski in his paper
prefers the first interpretation. Hence one should be cautious with
formalisation of Kotarbi~ski's ideas. If A is an indefinite proposition,
,A is one too; they both can either have a third logical value or ex-
press truth-value gaps. Even if tukasiewicz was impressed by the idea of
indefinite propositions, he interpreted them from the very beginning
within the framework of many logical values. 5 And this was not only the
interpretation of Kotarbinski's view but an entirely new proposal. One
other point should also be strongly underlined here. Kotarbinski links
indefinite propositions with the refutation of the law of excluded
middle. This is not tukasiewicz's opinion, however, since he points out
that three-valued logic is based on the negation of the law of bivalence
which is a metalogical principle. On the contrary, although (5) - (7)
are formulated in metalogical manner, Kotarbinski seems to work inside
logic, not in metalogic. And this is, in my opinion, a crucial differ-
ence between both of these approaches to indefinite propositions.
It is interesting to ask whether Lesniewski's proof is valid with
respect to tukasiewicz's conception; as a matter of fact tukasiewicz
never addressed himself to this question. As I have already said Les-
niewski's intepretation of tensed sentences is an essential ingredient
of his proof that indefinite propositions are impossible. But tukasie-
wicz could say that he is not committed to such an interpretation.
Now, let us come back to the successive steps of Lesniewski's proof
providing tukasiewicz' s three-valued logic. (15) is a premise for the
reductio ad absurdum proof. (14) should be supplemented by the observa-
tion that 1/2 is the logical value of A. Furthermore, (15) is misleading
even on Kotarbi~ski's assumptions: if A is not preeternal, both A and ,A
are indefinite in Kotarbinski' s sense j in tukasiewicz' s logic both of
these propositions are valued at 1/2. Thus ,A cannot be true. However,
5S .... ~ukasi .. wicz (1918), (1920) and <1920a).

196 JAN WOLENSKI
it is sufficient for Le~niewski's proof that ~A is not false at t; (17)

is not threatened by thi s modi fication of (15). The supposed tukasiewicz
argument against (17) may be supported by pointing out that the prin-
ei pIe of bivalence or preeternal i ty of falsi ty to save his proof, but
such a move would assume just what was to be proved. However, the Le~
niewskians still have some possibility of justifying their view. If A
and ~A have 1/2 as their logical values, the conjunction A and .,A has
the same value. This is a highly counter-intuitive result of tukasie-
wicz's logic 6 , because falsity i~ the intended value of any conjunction
of contradictory propositions. Now, if the conjunction of A and ~A is to
he false, then either A or ~A should be false. However, this principle
may be directed only against three-valued logic based on tukasiewicz's
valuations; it is possible to change valuations or extend three-valued
logic to one with more values. Thus I think that trying to prove formal
inconsistency stemming from the admission of indefinite propositions is
hopeless. The real problem is to argue for or against the principle of
bivalence. An important consequence of Le~niewski's "proof" is that it
exhibits a close connection between (1), bivalence and the timeless
interpretation of tensed sentences. In particular, it is difficult, i f
possible at all, to generalize (1) to
(la) for any logical value v, v is eternal and preeternal.
For Twardowski, Le~niewski and the later Kotarbinski real v's must sa-
tisfy (la), however for young Kotarhinski and tukasiewicz it is not
necessary. The first party observes that truth and falsity are the only
proper v's, the second one that an extension of v's requires a weakened
version of (la). But this is not all that should be said here. Twardow-
ski, Le~niewski and "the second" Kotarbinski want to have semantical re-
lations which are independent from time. They do not admit the question:
is A true at t? Truth has no temporal stance. For "the first" Kotarbill-
ski and tukasiewicz, the phrase 'A is true at t' is quite a legitimate
assertion. Tarski's semantic conception of truth is based on Twardowski
and his allies' assumptions. Thus the discussed debate shows that the
semantic nature of 1/2 and other v's "between" truth and falsity is not
clear. 7
It is commonly known that Kotarbinski' s formulation of the classi-
cal conception of truth was the starting point for Tarski's considera-
tions in his famous monograph (1933).8 However, another point in Kotar-
binski (1929) and (1934) deserves much more attention. 9 Kotarbinski dis-
tinguishes in these writings a real and a verbal use of 'true'. Consider
now
6 See Prior (1967, p. 135) and Borkowski (1977).
7 S .... Bo .. kowski (1981).
8Tarski begins his analysis wit.h a formulat.ion of t.h .. classical t.rut.h

definit.ion d .... ived from Kot.orbi';ski (1929). See Wole';ski & Simons (1989)
for furth~r details concerning the development of ~he classical truth
I.heory in Poland.
9Kol.arbi';ski (1934-) is a review of Tarski (1933).

KOTARBINSKI, MANY-VALUED lOGIC, AND TRUTH 197
(18) the theory of relativity is true,

and
(19) it is true that Warsaw is a town.
The adjective 'true' occurs in (19) in the verbal sense and its ommlSJ.On
does not reduce the content of this assertion. Such an observation moti-
vates so called nihilistic theories of truth: the adjective 'true' is
superfluous and can even be dropped because the phrases 'A' and 'A is
true' have exactly the same meaning. Kotarbinski's view is that 'true'
in sentences like (18) has meaning and cannot be omitted. Tarski (1944)
and (1969) gives quite .a general formulation of this view. If we consi-
der
(20) all consequences of true sentences are true,
we can easily see that 'true' cannot be eliminated without essential
change of the content of (20). The nihilist could argue that 'true' in
(18) plays an inessential role since one might reduce the theory of
relativity to its axioms. But (20) which is a general metalogical
principle cannot be analysed in such away. Thus, I think, Tarski' s
argument, doubtless inspired by Kotarbinski, is devastating for the
nihilistic theory of truth.
The problem of truth was a continuous subject of Kotarbinski' s
philosophical research. Except for his early inclinations to weaken'ing
(la), Kotarbinski consequently defended the absolute conception of
truth. I am inclined to think that even in his (1913) truth is a more
essential question than an eventual attempt to revise logic towards
many-valuedness.
CONCERNING REISM
Boguslaw Wolniewicz
1. Once in the course of a conversation - it was in 1967 - Professor

Kotarbinski had told me that of the many metaphysical oppositions and
contrasts one seemed to him to be the sharpest. I t is that between, as
he put it, the fact-mongers and the friends of things. (Actually he used
two Polish neologisms of his own making - 'fakciarze' i 'rzeczowcy' -
and much more telling than the crude English equivalents we have been
able to produce here.) The latter sounds solid and reliable, whereas the
former reminds one rather of peddling and shady deals. No doubt, this
exactly was the Professor's point. I asked him yet, a bit rashly,
whether he would not think Heraclitus to have been the first of all the
fact-mongers. He did not answer directly, but with an impatient gesture
he just snapped: "I don't like these old idols with their broken off
noses".
The metaphysical opposition mentioned is that between an ontology
of facts and an ontology of things. The former is expressed most em-
phatically in the famous dictum of Wittgenstein: "The world is the to-
tality of facts, not of things". The latter is what Kotarbinski called
reism, and its thesis may be put as follows: there are only things, and
the world is their totality; moreover, the world itself is a thing too.
We are not up to engage in this fundamental controversy here. What
we have to offer are just a few marginal remarks on it, very simple and
rather loose.
2. Let us say that a language L satisfies the postUlate of reism - or

that L is a reistic language - if for any proposition a of L there is in
L also a proposition a' which is strictly equivalent to it, and which
either contains no names at all (like the proposition "it's raining",
where 'it' is not a variable, but a purely grammatical expedient), or
contains names of concrete objects only.
Now let L be a reistic language, and let us view it with Wittgen-
stein simply as a totality of propositions. Moreover, let the sublan-
guage Lc c L be a language of concreta, i. e. let Lc consist of those
propositions of L which are either entirely name-free, or which deal ex-
clusively with concrete objects like dogs, neutrons, or galaxies. Then
the postulate of reism may be given the following form:
199
J. Wolelfski (ed.), Kotarbilfski: Logic, Semantics and Ontology, 199--204.
200 BOGUS tAW WOLNIEWICZ
(R) Va e: L 3a 'e: Le
,
a Cn a ,
~~
where the formula "a c~ a'" is to say that the proposition a is equi-
valent to the proposition a' under the consequence operation Cn fixed
for L. Hence Cn cannot be simply the logical consequence operation of L,
being based on at least some additional rules of inference or defini-
tions. (Clearly a proposition like "if it's raining, then there is a
rain going" is not a thesis of logic by itself.)
The postulate (R) applies to single propositions of L. But we may
express it also in such a way as to apply to sets of them, i. e. as:
(R' ) VA c L 3A'c Le Cn A = Cn A'.
The formulae (R) and (R') are equivalent. Indeed, (R) is implied by
(R'). For take an arbitrary proposition a e: L, and set A = {a}. Then by
(R') there is a set A'c Lc such that Cn {a} = Cn A'. Hence a e: Cn A' and
so - with Cn finitistic - there is also a finite subset B'c A' such that
a e: Cn B'. Now taking the conjunction fl' of all members of B' we get:
fl'e: Lc and a e: Cn {fl'}. But B' c A' c Cn {a}, and so - with each con-
junct of fl' following from a - we have: 8'e Cn {a}. Thus we get Cn {a} =
= Cn {8'}, for some fl'e Le, which is what was to be shown.
The converse implication holds too. For take an arbitrary set Ac L.
By (R) there is for each a e A an a 'e Le such that a c~ a'. Now let
r(A) c Le be the set of all such propositions, i.e.
r(A) = { a'e: Le: a c~ a', for some a e: A }.

Thus
(1) Vae:A3a'e:r(A) ac~a'.
Let us see, to begin with, that Cn A C Cn r(A). Indeed, take an ar-

bitrary B e: L, and suppose B e: Cn A. Hence also B e: Cn B for some finite
Be A. Thus B e: Cn {r}, where r is the conjunction of all propositions
in B. By (1), moreover, for any Bi e: B there is an Bi'e: r(A) such that
~~ B '
Bi Cn i.
Taking now one Bi'e: r(A) for each Bi e: B, we get a finite set of
propositions B'c r(A). So let r' be the conjunction of all its items,
i.e. Cn{r'} = Cn B'. By construction, the members of Band B' are pair-
wise equivalent, with the relation Cn being a logical congruence.
Hence the conjunctions rand r' must be mutually equivalent too: Cn{r} =
Cn{r'}. Thus we have: Be: Cn{r'} = Cn B' c Cn r(A) which is what was to
be shown.
The converse inclusion is obtained by a parallel reasoning since in
view of the definition of r(A) we have the formula:
(2) Va'e: r(A) 3a e: A a' c~ a
symmetric to (1).
CONCERNING REISM 201
Clearly, postulate (R') applies also to sets of propositions which

are theories under the consequence operation Cn, i.e. to sets of the
form T = Cn T. The set r(T) then consists of all propositions of the
theory T which may be expressed in the language of concreta, i.e.
r(T) = T n Lc. The postulate of reism guarantees that each proposiion
a E T has its equivalent counterpart a' in r(T). In the theory T there
may be talk of numbers, functions and classes; or of forces, voltages,
and electromagnetic oscillations. In the propositions of r(T) there is
only talk of concrete things, though not necessarily of the observable
ones only.- So the set r(T) should not be mistaken for the purely obser-
vational sublanguage of the theory T. These are quite different matters,
and their mutual relation is a separate question.
By assumption we 'have the identities: Cn r(T) = Cn T = T. Each pro-
position a' E r(T) is exp'ressed in the language of concreta, the set
r(T) being also - in view of the identities just mentioned ~ an axioma-
tics for the theory T. Consequently, one might say here that the theory
T is axiomatizable reistically.
This suggests a generalization of the reistic position, but still
very close to Kotarbinski's intentions. It is certainly conceivable that
even if not all the propositions of T had their counterparts in the lan-
guage of concreta Lc, the theory T could have stich a counterpart as a
whole. In that case one might say that a language L is weakly reistic if
all its theories are such. I.e. if, the following holds:
(R" ) VT c L (T = Cn T -- 3A 'c Lc T = Cn A).
It is easily seen that (R') entails (R"), but not conversely. Thus the
weak postulate of reism might be put as follows: all' theories expres-
sible in our language should be axiomatizable reistically. From the
reist's point of view this might do.
3. The program of a reistic reduction gives rise,to all, kinds of objec-

tions. One is to be considered here, though it is in fact of a very gen-
eral nature, and applicable to many similar programs, or similarly jus-
tified ones. In the case of reism 'the objection in question takes a
particularly clear form, which - by the way - is hardly a fault of that
position, but ratehr one of its medts. .
Let K be a class of propositions .in L, some of them containing
onomatoids, i.e. expressions with the syntactic appearance of names, but
denoting no concrete objects. (Like the word 'sake' in the phrase 'for
the sake of'.) Suppose we are able to indicate a general method of re-
ducing reistically all propositions in K. Thus with each aE K we can
correlate a definite proposition a'= r(a) belonging,to the language of
concreta Le, and in view of some reductive assumptions strictly equival-
ent to the former. An example stemming from Kotarbinski himself is the
following (cf. Eleaenty, p.53): let a be the proposition 'John stands to
his son in the relation of seniority', and let 8 be' the proposition
'between 'Gerlach and Giewont there holds the relation of superiority'.
(Gerlach and Giewont are two peaks in the Tatra mountains.) In these
proposi tions the phrases 'the relation 'of' seniori ty' and ,'the relation
of superiority' are onomatoids, since neither denotes a concrete object,'
However, these onomatoids are dispensable ones, for it is easy to indi-
202 BOGUStAW WOLNIEWJCZ
cate two propositions a' and /3' which are strictly equivalent to the
former, and which do not contain any onomatoids; i.e., the propositions
'John is older than his son' and 'Gerlach is higher than Giewont'. The
latter contain no onomatoids, for - as Kotarhinski puts it - phrases
like 'older than' and 'higher than' are not names of relations. To put
the point in general terms, the difference hetween t.he two couples of
propositions at hand is like the difference between the formulae
'R(x,y)' and '(x,y) E R'. In the latter the symbol 'R' is an onomatoid,
in the former it is not.
The two phrases mentioned are certainly no names, for they are both
predicates. But does that mean that they do not refer to anything, nor
denote it? The proposition 'it is a fact that ,Jack and Jill went up the
hill' is true if and only if Jack and Jill went up the hill, i.e. if the
proposition ',Tack and .Till went up the hill' is true. But why should we
take this as an indication that there are no facts in the world, only
couples of people walking up hills? And how does one justify the view
that participial expressions like 'Jack and Jill walking up the hill'
denote things, and not some syntactically reified facts? Or to put it
generally: why should the circumstances that to each proposition a con-
taining the expression 1'; one may indicate an equivalent proposition a'
not containing t; be regarded as proof that the expression l; is devoid of
the semantic function of denoting?
With regard to reism this very objection was raised already half a
centllry ago by Ajdukiewicz (1930), in his review of Kotarbinski' s Ele-
menty: "the fact that propositions including 'apparent terms' may be re-
placed with equisignificant ones which do not include them proves only
that [ ..• J we can do without apparent terms." (see this volume, p. 12).
This objection seems very important to us, and difficult to meet. The
task of meet iIlg it, however, rests in equal parts wi th the reist and
with his opponent. For it represents their common cause, though
approached by them from opposite quarters.
4. Propositions containing onomatoids may be called hypostatizing ones,

and we shall refer by the same term to the use of such propositions. Now
the question arises: why do we hypostatize at all? In Elementy (p. 53)
Kotarbinski ventured the answer that we do that "for the sake of im-
agery, of brevity, and out of habit". This is hardly satisfactory. The
fountain-head of hypostases should be looked for at a far deeper level
than that.
To our opinion the phenomenon of hypostatizing is connected with
the situation of a finite mind confronted by an infinite world. To map
that world somehow we have to fall back upon all kinds of tricks and
expedients, and hypostases might be one of them. Thus they are perhaps
indeed a matter of brevity, but not of that distinguishing an inch from
a mile; rather of that which distinguishes a sum from a series. And they
are also a matter of imagery, but again not of the one we find in
metaphors; rather of that which is found in homomorphisms. And habits
are hardly of any moment here.
Suppose we cannot do without hypostatizing. Nevertheless we might
try to subject it to certain logical rules and restrictions. This is
even urgently needed, for hypostatizing gets easily out of control,
CONCER.NING REISM 203
turning into a kind of linguistic neoplasm. Here is an example of such

unhridlefl hypostatizing, one out of many. We read in Derrirla (1972):
"The structure of deferr Ing makes it virtually Imposs i ble to reduce
temporation t.o a simple dialectic complication of the living present,
conceived as an original and incessant synthesis of the retentive traces
and the protentive openings. Along with the ot.herness of t.he 'uncon-
scious' we have to deal here not just with the modified horizons of the
present., but with a 'past' which never was present, and never will.
Consequently, the concept of a trace is incommensurable with that of re-
t.ention. [ ••. ] Traces - and hence di fferance t.oo - cannot be grasped if
we start from the present, i.e. from the presence of what. is there."
('Differance' is a French neologism coined by Derrida for calling to
mind the notions of deferring, of difference, and of confl ict, simulta-
neously. )
And so the text goes, on and on, in Derrida and in others, through
hundreds and thousands of printed pages, s1 iming up minds with always
t.he same 'transcendental' stuff. Reism was to be a dam against such
sliming. For it is easily seen that the main compound of that stuff are
multiple hypostases, like the 'differance' mentioned. To curb such
dealings reism suggests a simple rule:
For each hypostatizing proposition used one should bE' able to

(R'" indi cate its truth-condit. ions, not even full ones, but formulated
entirely in the language of concreta.
The variant of reism confined to that rule only might be called moderate
reism. What is demanded now is not any more the complete elimination of
hypostases. We wish them only subjected to some limitations. If you want
to talk about abstracta like 'temporation' or 'the structure of defer-
ring', please do it at your convenience. But we should appreciate it if
you took at least into account that if something is to hold in the realm
of abstracta - if, e.g., the structure of deferring makes it impossible
to reduce temporation to a simple complication of the living present -
then something else should hold in the realm of concreta. What exactly?
For reism answering this question is a test of meaningfulness. And if no
answer is coming forth, then it becomes clear that hypostatiz ing has
already trespassed safe limits.
The reist's test would be effective if we had any general principle
for correlating particular hypostatizing propositions with their con-
crete truth-conditions. No such principles are extant so far, however.
And so the transcedentalists could easily satisfy the reist's request by
correlating his dicta with arbitrary true propositions referring to con-
creta, e.g. with the consecutive items of the Warsaw telephone direc-
tory.
So even moderate reism, as a recipe for doing philosophy, is merely
a programme, a declaration of intent. In fact, all its efforts went in a
different direction: to oppose the abstractions of set theory, not the
products of neoromantic lucubrations. But the former are heavily bridled
by set theory itself, while the latter do not know any bridle at all.
204 BOGUStAW WOLNIEWICZ
5. The glllf between reism and an ontology of facts is wide enough. But
it looks less unbridgeable if one tltkes into consideration that they
have something important in common: their respect for logic. Tndeed,
both were inspired by logic, growing out of it as its metaphysical
extensions: one from the logical systems of Le~niewskj, the other from
those of Frege and Russell. The real philosophical abyss opens up only
where the consensus on logic comes to an end. When we hear war-cries to
the effect that "defining impairs the theoretical efficiency of con-
crepts", so down wi th defining; or t.hat "the ideal of univoci ty is to Iw
rf·jPcled as a mattpr of principlp"; then we see that what one tries to
do herp is replacing the rules of logic by those of euphony. It does not
matter an) more whether something in philosophy is consistent with some-
thing plse, or not. What matters is merely that it should have It good
sound, "'ith 'g,)od' meaning either 'elated', or 'scientific', or 'human-
istic', or just 'swell'.
H"ism is not a barrier massive enougb to withstand the concerted
onslaught of all those - as Professor Kotarbinski said on occasion -
"soulful men forming a trend semi-rational only, and steering towards
goals still less rational" (1958a, p.731). Such a barrier, however was
the Reist himsplf, that grand seigneur of Polish philosophy. And while
he was there, he gave a warrant by his own person that nothing really
bad may happen to that philosophy; the standards will be kept up. But
now he is no more.
THE VOICE OF THE PAST IN KOTARBINSKI'S WRITINGS
Ewa iarnecka-BiaIy
Tadeusz Kotarbillski contributed to the history of logic in many ways: as

the author of works devoted to issues of the history of logic in the
broad sense of the term l ; as the chronicler of his times, reviewing the
works of scholars who would later achieve a prominent place in the
history of logic 2 ; as an editor3 ; as a teacher interpreting contemporary
logic in connection with its historical roots. In fact, teaching logic
was a paradigm underlying all his works devoted to logic and its devel-
opment. "I reached the chair of philosophy," wrote Kotarbhlski in 1965,
"via logic. Teaching logic became the field of my activity as a
university professor of philosophy, a member of other humanistic
faculties. Emphasis is here placed on the words 'teaching' and
'humanistic'; For my lectures and classes were conceived as an organon
in the classical sense of the term, for phi losophers as well as for
those who, having completed their course of study, would espouse the
cause of disseminating humanistic knowledge and thinking, particulary
future secondary school teachers." (this volume, p. 3)
When studying the history of logic one can learn a lot about the
nature of human cognitive processes, of science, of the world. Kotarbin-
ski did it. Observing Kotarbinski's unusual fascination for the history
of logic one can learn a lot about his philosophical lievelopment.
The relevant facts are in this paper presented just to let them
speak for themselves.
lSee Kotarbinski <1967a), <1968b) and (1979).
~.g. Kotarbinski (1934) reviewed Tarski (1933). The review was quite
enthusiastic and stressed the need or rendering this book accessible
also to the roreign reader - which was done thorough its German (1936)
LranslaLion (English translation - 1966).
3Kotarbinski was one of" the initiators of the Library of Philosophy

Classics (in Polish) and headed its ediLo~i'ial board. Polish translations
or books by Condillac, Bacon, Mill and Bolzano were published wiLh his
thoroughgoing, erudite introduotions. E.g. he dnalyzed Bolzano's inter-
pretation or the notion or inf"erence.
20S
I. Wolerfski (ed.), Kotarbilfski: Logic, Semanlics and Ontology, 205-211.
" 1990 Kluwer Academic Publishers.
206 EWA ZARNECKA-BIAf.Y
1. An unwavering admirer of Aristotle.
As far as historical studies of a more detailed kind are concerned, it

seems that Kotarbi~ski was first of all a faithful follower of
Aristotle. He taught a course called 'Aristotelian logic' at Warsaw
University as early as the academic year 1919/1920, i.e., immediately
after he had begun working there. It is a pity that the Author's lecture
not('s were not published and - deposited in a Warsaw library - they were
lost during the Second World War.
KotarbiTlski's penetrating observations conce.rning Aristotelian
logic and its links with ontology, methodology and philosophy of the
nClturill sciences clln be found in variolls texts written by him. Much
space is devoted to Aristotle in his (1957a) and many other Kotarbin-
ski's writings.
As an important ami stimulating contribution to the Polish research
on ancient logic, one should also mentione<1 Kotarbinski's early comments
on Ar.istotelian logic included in his two textbooks, (1925) and (1926).
These wpre among the first Poli sh textbooks of log ic wri tten in the
spirit of contemporary logic and explaining traditional logic in modern
terms. Besides, Kotarbirlski's (1925) is probably the first text con-
taining the exposition of Le~niewski's ontology and the first to place
the idea of reism in a historical context. It was here that Kotarbi6ski,
after he had discussed Aristotle's ontological categories, came to for-
mulate his reistic postulate to the effect that all these ontological
categories are in fact reducible to the basic one. He wrote (1929,
p. 24): "Each object is a thing; there exist only things." Let us add
that we find in Kotarbinski (1925) also information about the sentential
variants of syllogisms, about logical connectives in Stoic logic; we
encounter an analysis of the Aristotelian classification of sentences
according to modality, and a generalization of the so called logical
square.
Let us use some quotations from Kotarbinski (1925): "Besides cate-
gorical syllogisms and their variations, traditional logic distinguished
also hypothetical, hypothetico-categorical, disjunctive and disjuntive-
hypothetical syllogisms." (p. 76). About hypothetical syllogisms he
wrote, "we substitute for the variables here not names, but the whole
sentences." (p.76) "Aristotle mentioned this form as one of the forms of
reasoning; he called it reasoning with the help of three - obviously,
three sentences." (p.77) While stressing the importance of tukasiewicz's
discovery of the early stage of sentential calculus in Stoic logic,
Kotarbinski failed to realize that he himself was probably the first one
to notice the beginnings of sentential calculus in Aristotle's writings.
In Kotarbi~ski (1926) among other changes, the material had been
rearranged: traditional logic was now discussed only after sentential
calculus and more room was devoted to standard modern issues (following
Principia Mathematica). Yet there are some new historical remarks: we
find an observation that the implicational form is basic for Aristotle's
categorical syllogism, while the traditional transcription "is not
correct enough." (p.209) There is also an accurate characterization of
reduction of syllogistic moods using dirett or indirect proofs. Formal
THE VOICE OF THE PAST IN KOTARBINSKI'S WRITINGS 207
description of the rules of deduction (suppositional proofs) as well as

the establishment of the relationship between the rules of inference and
logically true implicational formulas were only a short step away from
here.
Leon Chwistek, famous for his aggressive polemics, blamed Kotarbin-
ski for being "an unwavering admirer of Aristotle" (1935, p. 230) and
even for his promoting Lesniewski's calculus of names as consistent with
Aristotelian tradition. One should note that what Chwistek arbitrarily
considered to be a fault - this admiration of Aristotle - was a trait
very essent ial for Kotarbirlski' s philosophical development - and one
dating from the very beginning of his teaching activity.
2. Kotarbinski's Lectures on the History of Logic.
Kotarbinski t s interest in the history of logic emerged in his Lectures

on the History of Logic that appeared in 1957, preceeding by five years
the well known work of W. and M.Kneale The Development of Logic. Kotar-
binski himself treated Lectures as a textbook for philosophy students;
for further studies he recommended Bochenski' s Formale Logik (1956). a
book he could only have seen at the moment when his Lectures were ready
for printing. 4
"II n'est pas un specialiste de la logique symbolique, si fami-
lier qu' i l puise etre avec elle, mais un logicien et un philosophe au
sens Ie plus large et Ie plus autentique," said R. Poirier about the
author of Lectures in the introduction to its French translation (p.V).
Kotarbinski aims to interpret the main lines in the development of the
theory of deduction and induction as well as of semantics. And even
t.hough the mathematically oriented reader may see the work as somewhat
elementary - it is this deliberate simplicity that makes the mechanisms
of the development of logic easy to trace. Modern logic is -presented in
the Lectures not only as continuation of the previous systems, for
instance Aristotelian syllogistic or Stoic logic, but also as the result
of broadly understood art of efficient thinking, indebted to the works
of de la Ramee, Descartes, Arnauld and Nicole. Kotarbinski stresses the
natural relationship between the development of logic and philosophical
thought together with natural science and m~thematics. In such a context
Leibnizian logical ideas are described as well as the origin of the
algebra of logic. Here the most place is devoted to George Boole, but
independent results of Platon Porecki (a Kazan mathematician said to be
of Polish origin) are also mentioned.
4Hence Lectures takes the position of one of the pioneering publications

on the history of logic in this half-century.
208 EWA ZARNECKA-BIAty
3. The problem of induction. Francis Bacon.
The problem of induction, regarded in connection with the development of

the natural sciences, receives special treatment in Kotarbinski's work.
The author discussed the beginnings of the theory of induction
(Socrates; Aristotle; Ancient medicine; the Epicurean school) and pro-
vided a concise outline of its modern development - without, however,
involving himself in the current controversy concerning the justifica-
tion of induction. Kotarbinski's main concern here, and one reflecting
his earlier interests and predilections, was induction by elimination.
According to Kot.arbinski, Bacon was the dominant figure in thi s
development. 5 It was as soon as in 1929 that Kotarbinski noticed (see
his 1929) t.he importance of Baconian logic for the development. of
J.S.Mill's canons of eliminative induction. There he took up as well the
problem of relationship between induction by elimination, ontological
assumptions of induction, enumerative induction and deduction. In
Lectures he refers to this analysis and adds to it remarks about
J.Herschel's results on induction and C.Well's treatise about. dew. 6
Shortly before the Second World War Kotarbinski translated Novum
Organum into Polish, with notes and an introduction. This unfort~nately
was lost during the war, a fate of far too many Polish works.
4. Writings about the beginnings of logic in Poland; periodization of

the history of Polish logic.
There are no comments on the beginning of Polish logic in Kotarbinski's

earlier writings. His synthetic presentation of the early logic in
Poland appeared under the title La logique en PoloSDe, son originalite
et les influences etrangeres (see Kotarbinski, 1959)7. This was in fact
the first attempt to show all the development of the Polish logic from
a contemporary point of view.
6Kotarbinski had high regard f"or Bacon as a Renaisance man and humanist,
representing an epoch which, as he pointed out, originated in England
later than on the Continent but which was to rea~h here a more mature
f"orm. Probably Kotarbinski was f"ascinated by Bacon not only as a philos-
opher but also as a praxiologist.
6 1 remember Kotarbinski lecturing about this treatise - inspiring and

evoking poetic associations - f"or Warsaw University students in the year
1964./66; now the image of" the Professor comes back to me whenever I walk
QC~OS8 dew-covered fields.
7 The Polish version of" this work appeared as Kotarbinski (1966), sup-
plemented by the author with some new inf"orma'ion included in the two-
paged Afterword, dated .;July 19th, 1966.
I t would appear that Kotarbinski did not use the oldest books or
manuscripts. He made reference only to the accessible literature on this
subject, mainly Struve (1911) - see Kotarbinski (1966b, p.15). On the
other hand, Kotarbinski's conception of Polish logic since the 18th
century (Kotarbinski 1959, p. 3) was based on his own research, e.g.,
concerning the history of a textbook written by Condillac specially for
the Polish schools. 8
The synthesis in Kotarbinski's outline of the history of Polish
logic is visible in e.g. the clear-cut periodization of this history. He
suggested five periods: the Renaissance; the Jesuist period; the
Enlightenment; the pre-logistic period of modern times; and the contem-
porary, characterized by the primacy of mathematical logic. This is a
sufficiently detailed division, especially with respect to the earlier
periods; Kotarbinski here accepts Struve's general scheme. It should be
also not.ed that Kotarbinski's acount of Polish logic begins with the
year 1499 - the date of the publication of the first Polish book on
logic, Exercitium novae logicae, a commentary on Organon written by Jan
of Glogow, and not at the commencement of the teaching of logic in
Poland which would take us further back.9
Let \IS add a digression here. Writing about Jan of Glogow
Kotarbinski crit.icised C. Prantl who had said mentioning one edition of
Exercitium: "die Worte 'Tobie Mily Boze Chwala' am Ende der vorletzten
Seite bedeuten doch wohl der Drucker" (1870, p. 29). Perhaps influenced
by Struve's interpretation of Prantl, Kotarbinski (1959, p. 2) remarks:
"The famous erudite Prantl, while reading a later, posthumous edition of
Jan's Exercitium printed in Cracow in 1511, in Latin of course, found
there on the page before last a couple of words unitelligible to him,
which he thought were the name of the printer. These were the Polish
words 'Tobie Mily Boze Chwala' [Glory to Thee, Good God]. They are
evidence of the Polish nationality of the editor, printer or the
author." Really, Prantl did not realize, as he could have said if he had
known it, that the quoted phrase is meaningful in Polish. Nonetheless
Prantl was right associating these words with the printer: it had been
customary for printers to include their inscriptions in the books they
set. It could be a pious invocation, it could also be a cryptographic
text encoding the printer's name.
8 I t. was t.he t.ext.book "La Logique ou les premiers developpement.s de l'art.

de pencer" - writ.t.en in 1779 on t.he special invit.at.ion of a Polish st.aff
aiming at. school reforms U:omisja Edukacji Narodowej, supposed t.o be
t.he first. European Minist.ry of Eduoat.ionl. The Polish t.ranslat.ion by
J.Znosko was published in 1802.
9Markowski (1980) dat.es t.he beginnings of t.eaching logic in Poland on

1400.
210 EWA ZARNECKA-BIAty
5. Historical studies of contemporary Polish logic - sympathetic, if not

quite impartial.
In his account of logic in Poland Kotarbinski concentrated on the latest

period, paying the most attention to tukasiewicz. He stressed the orig-
inality of tukasiewicz's studies in the theory of probability and of his
parenthesis-free notation as well as his pioneering works on many-valued
logics, of which more below (see Kotarbinski, 1956, 1959 and 1967).
In Kotarbinski (1979) we find a comprehensive report of the works
of Lesniewski whose logical ideas he had himself adopted. Further he
mentions Petrazycki, whose interests he shared and work on the methodol-
ogy of law by less known E. Waskowski. There is a kind of a synthesis in
it - but it could be noted that Kotarbinski selects only some of the
available material. The closer Kotarbinski gets to modern times the more
manifest his own preferences become. Szaniawski defined Kotarbinski' s
historiography as 'Russellian' , i.e., seeking to take a stand on every
issue under discussion; as being "sympathetic but not impartial" (1979,
p.20, Foreword). This sympathetic lack of partiality characterises all
Kotarbinski's work in this area.
As one of the achievements of Polish logic Kotarbinski regarded
tukasiewicz's construction of a three-valued logical system, stressing
that this was inspired by tukasiewicz (1910). On many occasions Kotar-
binski accentuated the importance of tukasiewicz' s results in
constructing three- and many-valued systems of logic. Incidentally,
there has always been speculation with respect to the authorship of
these systems and their dating. Much confusion has been caused by Kline
(1965) who argued that many-valued logics were founded by Vasilev from
Kazan University and presented in his 1910-1912 works. 10 Influenced by
Kline, Kotarbinski (1966b) was ready to attribute this authorship to
Vasilev. However, it should be noted that Vasilev did not create any
formal system and given that we find an anticipation of non-Aristotelian
(i.e. non-Chrysipian, many-valued) logic already in tukasiewicz (1910),
it would be more correct to say that the two scholars started developing
their ideas, similar only in the sense of looking for a new logic while
not at all identical, simultaneously and independently.
Some authors have criticised Kotarbinski for praising tukasiewicz,
while forgetting about himself. For instance, Greniewski (1957, p. 135)
reminds us that Kotarbinski in 1913 formulated quite clearly in everyday
language the principles of a three-valued propositional logic.
Greniewski (1958, p.I71) remarks: "T.Kotarbinski the historian of logic
must be criticized for having ignored a discovery made by T.Kotarbinski
the logician."
10Vasil .. v called f'or 'non-Aristot .. lian' logic, hoy .. ver Yithout abandoning
the id .. a of tyo-yaluedn ..... Kline's int.erpretation i . misleading. On the
other hand, Klin .. is not. the only author respon.ible for this conf'usion.
He yas pr .. ce-ded by ChYist.ek and also by Gr·enieyski. See tarnecka <1986,
1988).
6. Final remarks.
No account is given here of other issues of the history of logic taken

up by Kotarbinski, connected with his interests in semantics, methodol-
ogy, and logic in the most general sense of the term. It is difficult in
any case to distinguish between Kotarbinski's history of logic and his
history of philosophy, or between his treatment of the development of
the idea of logical efficiency of thought and the idea of efficiency of
action. As a historian Kotarbinski did not cease to be a reist-philoso-
pher and never ceased to be a praxiologist.
According to Kotarbinski, logic should not be an isolated disci-
pline - its place is among other branches of science. Logic should serve
the purpose of integrating other disciplines. We ought also to strive to
know the past in order to understand the present. Writing about tukasie-
wicz's research into the history of logic Kotarbinski (1958, p. 57)
reminds us about the challenge tukasiewicz found in Aristotle's
writings; Kotarbinski quotes tukasiewicz as saying: "Not everyone hears
the voice of the books of old" and praises him for having this disposi-
tion for listening. Kotarbinski himself was also an attentive listener
to that voice of the past.
REFERENCES
Ajdukiewicz K. (1930) Review of Kotarbinski (1929), Przegl~d Filozo-

ficzny 33, 140-60. Eng. trans. in Kotarbinski (1966), 515-36; part.
repro in this volume
Ajdukiewicz K. (1935) "W sprawie uniwersali6w" (On the problem of uni-
versals), Prze!1l~d Filozoficzny 38, 219-34. Eng. trans. by J. Giedy-
min in Ajdukiewicz (1978), 95-110
Ajdukiewicz K. (1937) "Problemat transcedentalnego idealizmu w sformulo-
waniu semantycznym" (A semantical version of the problem of ideal-
ism), Przegl~d Filozoficzny 40, 271-81. Eng. trans. by J.Giedymin in
Ajdukiewicz (1978), 140-53
Ajdukiewicz K. (1951) "On the Concept of Existence", Studia Philosophica
4, 7-22; repro in Ajdukiewicz (1978), 209-21
Ajdukiewicz K. (1963) J~zyk i poznanie (Language and cognition), vol. I,
PWN: Warszawa
Ajdukiewicz K. (1978) The Scientific World-Perspective and Other Essays
1931-1963, ed. by J. Giedymin, Reidel: Dordrecht
Bochenski I.M. (1956) Formale Logik, Alber: Munchen; Eng. trans.'A His-
tory of Formal Logic, Univ. of Notre Dame Press: Notre Dame, 1961
Borkowski L. (1977) "W sprawie intuicyjnej interpretacji logiki wielo-
wartosciowej tukasiewicza" (On the problem of intuitive interpreta-
tion of tUkasiewicz's many-valued logic), Roczniki Filozoficzne 25,
no. 1, 61-8
Borkowski L. (1981) "Kilka uwag 0 zasadzie dwuwarto~ciowo~ci i logikach
wielowarto~ciowych" (Remarks on the principle of bivalence and many-
valued logics), Roczniki Filozoficzne 29, no. 1, 9-14
Brentano F. (1874) Psychologie vom empirischen Standpunkt, vol. I,
Duncker & Humblot: Leipzig, reproduced with additional material as
vol. I of Brentano (1924/25)
Brentano F. (1889) HUber den Begriff der Wahrheit" , first publ. in
Brentano (1930), 3-29
Brentano F. (1911) Klassifikation der psychischen Phanomene, Duncker &
Humblot: Leipzig; reproduced with additional material as vol. II of
Brentano (1924/25)
Brentano F. (1915) Von den Gegenstanden des Denkes, in Brentano
(1924/25), vol II, 213-225
Brentano F. (1924/25) Psychologie vom empirischen Standpunkt, 2nd ed.,
in 2 vols. ed. by a.Kraus, Felix Meiner: Leipzig. Eng. trans. ed. by
L.L.McAlister, Psychology from an Empirical Standpoint, Routledge
and Kegan Paul: London, 1973
Brentano F. (1930) Wahrheit und Evidenz, ed. by a.Kraus, Felix Meiner:
Leipzig. Eng. trans. ed. by R.M.Chisholm, The True and the Evident,
Routledge and Kegan Paul: London, 1973
214 REFERENCES
Brentano F. (1933) Kategorienlehre, Felix Manner: Leipzig. Eng. trans.

The Theory of Categories, by R.M.Chisholm and 'N.Guterman, Martinus
Nijhoff: The Hague/Boston/London, 1981
Brentano F. (1976) Philosophische Untersuchungen zu Raum, Zeit und
KontinuulI" ed. by S. Korner and R.M.Chisholm, Felix Meiner: Hamburg.
Eng. trans. Philosophical Investigations on Space,' Time and the
Continuum, by B.Smith, Croom Helm: London and Sydney, 1987
Bunt H.C. (1985) Mass Terms and Model-Theoretic Semantics, Cambridge
University Press: Cambridge
Chwistek L. (1935f Granice nauki (Limits of the science), Ksi~znica
Atlas: LwDw. Eng. trans. London 1946
Cantor G. (1887/88) "Mitteilungen zur Lehre vom Transfiniten", as repro
in Cantor's Abhandlungen mathellatischen und philosophischen Inhalts,
E.Zermelo (ed.), Olms: Hildesheim, 1966, 378-439
Carnap R. (1932) "Uberwindung tier Metaphysik durch logische Analyse der
Sprache", Erkenntnis 2, 219-41
Carnap R. (1950) "Empiricism, Semantics and Ontology", Revue Internatio-
nale de Philosophie 4, 20-40
Chisholm R.M. (1978) "Brentano's Conception of Substance and Accident"
in Chisholm and Haller (1978), 197-210
Chisholm R.M. and Haller R. - eds. (19i8) Die Philosophie Franz Brenta-
nos, Rodopi: Amsterdam (also as Grazer Philosophische Studien, 5)
Cohen R.S. and Schnelle T. - e,ds. (1986) Cognition and Fact. Mater'ials
on Ludwick Fleck, D.Reidel: Dordrecht/Boston/Lancaster/Tokyo
Cornelius H. (1894) Versuch einer Theorie der Existentialurteile,
Rieger: Munich
Davidson D. (1968) "On Saying That", as repro in his Inquiries into
Truth and Interpretation, Oxford University Press, 1984, 93-108
Derrida J. (1972) Marges de la philosophie, Editions de Minuit: Paris
Frege G. (1893) Grundgesetze der Arithlletik, begriffsschriftlich abge-
leitet, Jena (repr. Olms: Hildesheim, 1966). Partial Eng. trans. by
M.Furth, The Basic Laws of Arithlle~ic, University of California
Press: Berkeley and Los Angeles, 1964,
Galton A. (1984) The Logic of Aspect. ,An Axiomatic Approach, Oxford
. University Press: Oxford
Ge~~h P. (1972) Logic Matters, Blackwell: O~ford
Geblewicz E. (1983) ,"An Analysis of the Concept of Goal" in 'Praxiologi-
cal Studies, PWN-Reidel: Warszawa-Dordrecht
Goodman N. (1951) Structure of Appearance, Harvard University Press:
Cambridge, Mass.; 3rd ed., D.Reidel: Dordrecht/Boston, 1977
Greniew~ki H. (1957) "2n+1wartosci logicznych" (2 n+1 logi~al values),
Studia Filozoficzne I, 82-115
REFERENCES 215
Greniewski H. (1958) "Tadeusza Kotarbinskiego Wyklady z dziejow logiki"

(Tadeusz Kotarbinski's Lectures on History of Logic), Studia Filozo-
ficzne 3(6)
Grzegorczyk A. (1955) "The System of Lesniewski in Relation to Contempo-
rary Logical Research", Studia Logica 3, 77-97
Grzegorczyk A. (1960) "Axiomatizability of Geometry without Points",
Synthese 12, 228-35
Grzegorczyk A. (1977) "On Certain Formal Consequences of "Reism",
Dialectics and Humanism I, 75-80
Harris Z. (1982) "Grammar of English on Mathematical Principles" , John
Wiley & Sons: New York
Hoeksema J. (1985) Categorial Morphology, Garland Publishing: New York/
London
Hintikka J. (1959) "Existential Presuppositions and Existential Commit-
ments", JQurnal of Philosophy 56, 124--37
Hofler A. (1890) Logik. Philosophische Prop8deutik, written in collab-
oration with A.Meinong, Tempsky: Vienna, 2nd enlarged ed., 1922
Husserl E. (1891) Philosophie der Arithmetik. Psychologische und
logische Untersuchungen, vol.I, Halle a.d. S., critical edition as
Husserliana XII
Husserl E. (1900/01) Logische Untersuchungen, Max Niemeyer: Halle
a.d.S., critical edition as Husserliana XVIII, XIX/I, XIX/2. Eng.
trans. of 2nd ed. (1913/21) by J.N.Findlay, Logical Investigations,
Routledge and Kegan Paul: London, 1970
Husserl E. (1979) Aufsatze und Rezensionen (1890-1910), ed. by B.Rang,
Martinus Nijhoff: The Hague/Boston/London (Husserliana XXII)
Ingarden R. (1964/65) Der Streit um die Existenz der Welt, 2 vols., the
2nd in 2 parts, Max Niemeyer: Tubingen (original Polish edition,
1947/48)
Ingarden R. (1969) "Le Concept de Philosophie chez Franz Brentano",
Archives de Philosophie 32, 458-75 and 609-38
Johnson W.E. (1924) Logic, Part III, The Logical Foundations of Science,
Constable: London, repro Dover: New York, 1964
Jordan Z. (1963) "Logical Determinism" Notre Dame Journal of Formal
Logic X, 1-38
Kneale W.& M. (1962) The Development of Logic, Oxford University Press
Kline M. (1965) "N.A. Vasiliev and the Development of Many-Valued Logic"
in A.T. Tymieniecka - ed. Contributions to Logic and Methodology in
Honour of I. M. Bochenski, North Holland: Amsterdam, 315-26
Koffka K. (1935) Principles of Gestalt Psychology, Kegan Paul, Trench,
Trubner: London
Kohler W. (1947) Gestalt Psychology, Liveright: New York
216 REFERENCES
Kotarbinski T. (1913) "Zagadnienie istnienia przyszlosci" (The problem

of the-existence of the future), Przeglad Filozoficzny 16, 74-92.
Eng. trans. The Polish Review (New York) 13, no. 3, 1968, 7-22
Kotarbinski T. (1920) "Sprawa istnienia przedmiotow idealnych" (The
problem of the existence of ideal objects), PrzeglaQ Filozoficzny
23, 149-70
Kotarbinsld T. (1922) "0 istocie doswiadczenia wewn~trznego" (On the
essence of inner experience), Przegl~d Filozoficzny 25(2), 184-96
Kotarbinski T. (1925) Logika. Skrypt z wYkladow w trymestrze jesiennym
1924-5 (Logic. The autumn trimester 1924-5 lecture notes), Warszawa
Kotarbinski T. (1926) Elementy logiki formalnej, teorii poznania i meto-
dologii. Skrypt. (Elements of formal logic, theory of knowledge, and
methodology. Lecture notes.), ed. by D. Steinberzanka, Warszawa
Kotarbinski T. (1929) Elementy teorii poznania, logiki formalnej i
metodologii nauk (Elements of the theory of knowledge, formal logic
and methodology of the sciences), Ossolineum: Lwow, revised edition:
Wroclaw, 1961. Eng. trans. by O. Wojtasiewicz in Kotarbinski (1966)
Kotarbinski T. (1929a) "Analiza materializmu", Slowo Polskie 69, 7-8;
repro in Kotarbinski (1958a), 90-6
Kotarbinski T. (1930-31) "Uwagi na temat reizmu" (Comments on reism),
Ruch Filozoficzny 12, 7-12; repro in Kotarbinski (1958a), 104-13
Kotarbinski T. (1931) "Le r~alisme radical", Proceedings of the 7th In-
ternational Congress of Philosophy, 1930, London, 488-500
KotarbiIlski T. (1931a) "Zasady reizmu" (Principles of reism), Sprawozda-
nia Poznanskiego Towarzystwa Przyjaciol Nauk 5, 16-8; repro in
Kotarbinski (19S8a), 114-6
Kotarbinski T. (1931b) "Oroznych znaczeniach slowa 'materializm'" (On
various meanings of the word 'materialism'), Przeglad Filozoficzny
34, 16-30; repro in Kotarbinski (1958a), 69-89
Kotarbinski T. (1934) "w sprawie pojecia prawdy" (On the concept of
truth), Przegla~ Filozoficzny 37, 85-91
Kotarbinski T. (1935) "Zasadnicze mysli pansomatyzmu" (The fundamental
ideas of pansomatism), Przegl~ Filozoficzny 38, 283-94 Eng. trans.
by A.Tarski and D.Rynin, Mind 64, 1955, 488-500
KotarbiIlski T. (1935a) "The Development of the Main Problem _in the Me-
thodology of Francis Bacon", Studia Philosophica I, 107-17
Kotarbinski T. (1949) "0 postawie reistycznej, czyli konkretystycznej"
(The reistic, or concretistic approach), Mysl Wspolczesna 4, No. 10,
3-11. Eng. trans. in Kotarbinski (1966), 420-8
Kotarbinski T. (1955) Kurs logiki dla prawnikow (A course of logic for
lawyers), Gebethner i Wolff: Warszawa
Kotarbinski T. (1955a) Traktat 0 dobrej robocie (Treatise on good work),
PWN: Warszawa. Eng. trans.: Kotarbinski (1965)
REFERENCES 217
Kotarbinski T. (1956) "La logique en Pologne", Les Etudes Philosophiques

11, 234-41; repro in R.Klibansky, ed. Philosophy in" the Mid-Century.
A Survey. La Nuova Italia Editrice: Firenze 1958, 44-52
Kotarbins],i T. (1957) Wybor pism (Selected Works), vol. I, PWN: Warszawa
KotarbiIlski T. (1957a) Wyklady z dziejow logiki (Lectures on history of
logic), Ossolineum: todi. French trans. Lecons sur l'historie
logique by A.Posner, PWN: Warszawa 1965
Kotarbinski T. (1958) "Fazy rozwojowe konkretyzmu" (The development
phases of concretism) , Studia Filozoficzne 4(7), 3-13. Eng. trans.
in Kotarbinski (1966), 427-37
Kotarbi nski T. (1958a) Wybor pism (Selected Works), vol. II, PWN:
Warszawa
Kotarbinski T. (1958b) "Jan tUkasiewicz's Works on the History of
Logic", Studia Logica 8, 57-62
Kotarbinski T. (1959) La logique en Pologne, son originalite les
influences etrangeres, Accademia Polacca di Scienze e Lettere: Roma
Kotarbinski T. (1961) "The Basic Postulate of Concretism", Schriften zur
Phonetik, Sprachwissenchaft und Kommunikationsforschung 3, 79-87
Kotarbinski T. (1965) Praxiology: An Introduction to the Science of
Efficient Action; trans. of (1955a) by O. Wojtasiewicz, PWN-Pergamon
Press: Warszawa-Oxford
Kotarbinski T. (1966) Gnosiology. The Scientific Approach to the Theory
of Knowledge, Pergamon Press-Ossolineum: Oxford-Wroclaw. Eng. trans.
by O.Wojtasiewlcz of 2nd ed. of Kotarbinski (1929), with appendices
Kotarbinski T. (1966a) "France Brentano comme reiste", Revue interna-
tionale de Philosophie 20, 459-76. Eng. trans. in McAlister (1976),
194-203
Kotarbinski T. (1966b) "Glowne wiadomo~ci 0 dziejach logiki w Polsce" ,
Studia i Materialy z Dziejow Nauki w Polsce, A/9, 9-24; repro in
Kotarbinski (1979), 229-57
Kotarbinski T. (1966c) "Slowo wst'ipne" (Foreword), in B. Bolzano Parado-
ksy nieskonczono~ci (Paradoxes of infinity), PWN: Warszawa, XI-
XXIII; repro in Kotarbinski (1979), 52-64
Kotarbinski T. (1967) "Notes on the Development of Formal Logic in
Poland in the Years 1900-39", in McCall - ed., 1-14
Kotarbillski T. (1968) "Reism: Issues and Prospects", Logique and Analyse
11, 441-58
Kotarbinski T. (1968/69) "The Task and Problems of Praxiology", Prakseo-
logia 1968/69, no. 31, 7-25
Kotarbinski T. (1974) "Determinism and Fat~lism on the Face of Activ-
ity", Dialectics and Humanism 1, no. 2, 3-11
Kotarbinski T. (1979) Szkice z historii filozofii i logiki (Essays in
history of philosophy" and logic), ed. by K.Szaniawski, PWN: Warszawa
218 REFERENCES
Kraus O. (1937) Die Werttheorien. Geschichte und Kritik, Rohrer: Brunn/

Vienna/Leipzig
Kung G. (1967) Ontology and the Logistic Analysis of Language, Reidel:
Dordrecht
Kung G. (1974) "Prologue-Functors", Journal'of Philosophical Logic 3,
241-54
Kiing G. and Canty J.T. (1970) "Substitutional Quantification and Les-
niewskian Quantifiers", Theoria 36, 165-82
Kuratowski C. (1948) Topologie I, Monografie Matematyczne: Warszawa-
Wroctaw
Lejewski C. (1958) "On Lesniewski's Ontology", Ratio 1, 150-76, as repro
in J.J.T.Srzednicki et a1. - eds. (1984), 123-48
Lejewski C.(1976) "Ontology and Logic", in S.Korner, ed., Philosophy of
Logic, Blackwell: Oxford, 1-27
Lejewski C. (1979) "On the Dramatic Stage in the Development of Kotar-
binski's Pansomatism", in P.Weingartner and E.Morscher, eds.,
Ontology and Logic, Duncker & Humblot: Berlin, 197-218, repro in
this volume
Lejewski C. (1982) "Ontology: What Next", in W.Leinfellner et al., eds.,
Language and Ontology, Holder-Pichler-Tempsky: Vienna, 173-86
Leonard H. and Goodman N. (1940) "The Calculus of Individuals and its
Issues", Journal of Sy.bolic Logic 5, 45-55
Lesniewski S. (1911) "Przyczynek do analizy zdan egzystencjalnych"
(Contributions to the the analysis of existential propositions),
Przegl~ Filozoficzny, 14, 329-45
Lesniewski S. (1913) "Krytyka logicznej zasady wytll;czonego srodka" (Cri-

tique of the principle of excluded middle), Przeglad Filozoficzny,
16, 315-52
Lesniewski S. (1913a) "Czy prawda'jest tylko wieczna czy tez wieczna i
odwieczna?" (Is truth only eternal or both eternal and preeternal?),
Nowe Tory 18, 493-528
Lesniewski S. (1914) "Czy klasa klas, nie podporzll;dkowanych sobie, jest
podporz~dkowana sobie?" (Is the class of classes not subordinate to
themselves subordinate to itself?), Przegl~ Filozoficzny 17, 63-75
Lesniewski S. (1927/31) "0 podstawach matematyki" (On the foundations of
mathematics), Przeclad Filozoficzny, vols. 30-34. Cited according to
the abridged Eng. trans. by V.~.Sinis~, "On the Foundations of
Mathematics", Topoi 2, 1983, 7-52
Lesniewski S. (1927) "0 podstawach ~atematyki" (On the foundations of
mathematics), Przegl~ Filozoficzny 30, 164-206
Lesniewski S. (1928) "0 podstawach matematyki" (On the foundations of
mathematics), Przeglod Filozoficzny 31, 261-91
REFERENCES 219
Lesniewski S. (1929) "Grundzuge eines neuen Systems der Grundlagen der

Mathematik", Fundamenta Mathematicae 14, 1-81
Lesniewski S. (1930) "tiber die Grundlagen der Ontologie", Comptes rendus
des Seances de la Societe des Sciences et des Lettres de Varsovie,
Cl. III 23, 111-32
Lesniewski S. (1931) "0 podstawach matematyki" (On the foundations of
mathematics), Przegl~d Filozoficzny 34, 142-70
Lipps T. (1905) "Das Wissen von fremden Ichen", in Lipps, ed., Psycho-
logische Untersuchungen, Engelmann: Leipzig, 1/4
Luschei E.C. (1962) The Logical System of Lesniewski, Amsterdam
tukasiewicz J. (1910) 0 zasadzie sprzecznosci u Arystotelesa (On the
principle of contradiction in Aristotle, PAU: Krakow; 2nd ed. PWN:
Warszawa 1985
tukasiewicz J. (1910a) "0 zasadzie wylaczonego srodka" (On the principle
of the excluded middle), Przeglad Filozoficzny 13, 372-3. Eng.
trans. by J. Wolenski and P. Simmons History and Philosophy of Logic
8,1987,69
tukasiewicz J. (1912) "0 tworczosci w nauce" (On creative elements in
science), in Ksi~ga pami~tkowa ku uczczeniu 250 rocznicy zalozenia
Uniwersytetu Lwowskiego, Lwow. Eng. trans. by O. Wojtasiewicz in
tukasiewicz (1970), 1-15
tukasiewicz J. (1918) Wyklad pozegnalny w auli Uniwersytetu Warszawskie-
go, 7 marca 1918 r. (A farewell lecture delivered in the Warsaw
University Lecture Hall on March 17, 1918), Warszawa. Eng. trans. by
o. Wojtasiewicz in tukasiewicz (1970), 84-6
tukasiewicz J. (1920) "0 poj~ciu mozliwosci" (On the concept of possibi-
lity), Ruch Filozoficzny 6, 169-70. Eng. trans. by H. Hiz in McCall
(1967), 15-6
tukasiewicz J. (1920a) "0 logice trojwartosciowej" (On three-valued
logic), Ruch Filozoficzny 6, 170-1. Eng. trans. by O. Wojtasiewicz
in tukasiewicz (1970), 87-8
tukasiewicz J. (1970) Selected Works, North-Holland/PWN: Amsterdam/
Warszawa
tukasiewicz J. (1971) "On the principle of contradiction in Aristotle",
trans. from German by V. Vedlin, Review of Metaphysics 24, 485-509
McAlister L.L. - ed. (1976) The Philosophy of Brentano, Duckworth:
London
McCall S. - ed. (1967) Polish Logic 1920-1939, Oxford University Press
Markowski M. (1980) "Poczatki nauczania logiki w Uniwersytecie Krakow-
skim (The beginnings of teaching logic at Cracow University), in E.
Zarnecka-Bialy - ed., Logika i jej nauczanie w dziejach Uniwersytetu
Jagiellonskiego (Logic and its teaching in the history of the
Jagiellonian University), Uniwersytet Jagiellonski: Krakow, 11-26
220 REFERENCES
Martin R.M. (1958) Truth and Denotation, Routledge and Kegan Paul:
London
Marty A. (1895) "tiber subjektlose Satze und das Verhaltnis der Grammatik
zu Logik und Psychologie", 7th article, Vierteljahrsschrift fur
wissenschaftliche Philosophie 19, 19-87
Marty A. (1908) Untersuchungen zur Grundlegung der allgemeinen Grammatik
und Sprachphilosophie, vol.I (only volume published), Niemeyer:
Halle a.d.S.
Mehlberg H. (1963) "The Present Situation in the Philosophy of Mathema-
tics", in B.H.Kazemier and D.Vuysje eds. Logic and Language. Studies
Dedicated to Professor Rudolf Carnap on the Occasion of His Seven-
tieth Birthday, Reidel: Dordrecht, 69-103
Mellor D.H. (1981) Real Time, Cambridge University Press: Cambridge
Mostowski A. (1965) Thirty Years of Foundational Studies (Acta Philoso-
phica Fennica XVII), Basil Blackwell: Oxford
Mourelatos A.P.D. (1981) "Events, Processes and States", in P.J.Tedeschi
and A.Zaenen, eds., Tense apd Aspect (Syntax and Semantics, vol.14),
Academic Press: New York, 191-211
Mulligan K., Simons P.M. and Smith B. (1984) "Truth-Makers", Philosophy
and Phenomenological Research 44, 287-321
Munch D. (1986) "Brentanos Lehre von der intentionalen Inexistenz", in
J.C.Nylri, ed., From Bolzano to Wittgenstein. The Tradition of
Austrian Philosophy, Holder-Pichler-Tempsky: Vienna, 119-127
Munch D. (forthcoming) "Brentano's Soul", to appear in Grazer Philoso-
phische Studien
Novak M. (1963/64) "A Key to Aristotle's 'Substance''', Philosophy and
Phenomenological Research 24, 1-19
Pelc J. - ed. (1979) Semiotics in Poland 1894-1969, Reidel: Dordrecht/
Boston
Prantl C. (1870) Geschichte der Logik in Abendlande, vol. IV, Leipzig
Prior A. (1967) Past, Present, Future, Oxford University Press: Oxford
Prior A. (1976) "Intentionality and Intensionality", Papers in Logic and
Ethics, Duckworth: London
Przel~cki M. (1982) "The Law of Excluded Middle and the Problem of
Idealism", Grazer Philosophische Studien 18, 1-16
Pszczolowski T. (1978) Mala encyklopedia prakseologii i teorii organiza-
cji (Mini-encyclopedia of praxeology and theory of organisation),
Ossolineum: Wroclaw
Quine W.V.O. (1939) "Designation and Existence", Journal of Philosophy
36, 701-9
Quine W.V.O (1953) From a Logical Point of View, Harvard University
Press
REFERENCES 221
Quine W.V.O. (1960) Word and Object, MIT Press: Cambridge, Mass.
Rasiowa H. and Sikorski R. (1963) The Mathematics of Metamathematics,
PWN: Warsaw
Reichenbach H. (1957) Experience and Prediction, 5th edition, The Uni-
versity of Chicago Press: Chicago
Schnelle.T. (1982) Ludwik Fleck - Leben und Denken. Zur Entstehung und
Entwicklung des soziologischen Denkstils in der Wissenschaftsphilo-
sophie, Hochschulverlag: Freiburg i. Br.
Sellars W. (1962) "Time and the World Order", in H.Feigl and G.Maxwell,
eds., Scientific Explanation. Space and Time (Minnesota Studies in
the Philosophy of Science, vol. III), University of Minnesota Press:
Minneapolis, 527-616
Simons P.M. (1983) "A Lesniewskian Language for the Nominalistic Theory
of Substance and Accident", Topoi 2, 99-109
Simons P.M. (1987) Parts. An Essay in Ontology, Oxford University Press:
Oxford
-Sinisi V. (1962) "Nominalism and Common Names", Philosophical Review 71,
230-5
Sinisi V. (1966) "Lesniewski's Analysis of Whitehead's Theory of
Events", Notre Dame Journal of Formal Logic 7, 323-27
Sinisi V. (1983) "The Development of Ontology", Topoi 2, 53-61
Srupecki J. (1955) "S. Lesniewski's Calculus of Names", Studia Logica 3,
7-70
Srupecki J. (1958) "Toward a Generalized Mereology of _Lesniewski",
Studia Logica 8, 131-54
Smith B. (1981) "Logic., Form and Matter", Proceedings of the Aristotel-
ian Society, Supplementary Volume 55, 47-63
Smith B. (1987) "The Substance of Brentano's Ontology", Topoi 6/1, 37-47
Smith B. (1989) "Kasimir Twardowski: An Essay on the Borderlines of
Psychology, Ontology and Logic", in Szaniawski - ed., 313-~3
Smith B. (1990) "Sachverhalt", in J.Ritter and K. Grunder, eds., Histo-
risches Worterbuch der Philosophie, Schwabe: Bas~l/Stuttgart, vol.
VIII
Smith B. (forthcoming) "Brentano and Marty: An Inquiry into Being and
Truth", in K.Mulligan, ed., Mind, Meaning and Metaphysics. The
Philosophy and Theory of Language of Anton Marty, Martinus Nijhoff:
Dordrecht/ Boston/Lancaster
SoboGinski B. (1954-5) "Studies in Lesniewski's Mereology", Rocznik
Polskiego Towarzystwa Naukowego na Obczyznie 5
Sobocinski B. (1971) "Atomistic Mereology", Nortre Dame Journal of
Formal Logic 12, 89-103, 203-13
222 REFERENCES
Struve H. (1911) Historia logiki jako teorii poznania w Polsce (History

of logic as the theory of knowledge in Poland), 2nd ed., Warszawa
Srzednicki J.J.T. (1965) Franz Brentano's Analysis of Truth, Martinus
Nijhoff: The Hague
Srzednicki J.J.T. et ai. - eds. (1984) Lesniewski's Systems: Ontology
and Mereology, Martinus Nijhoff: The Hague/Boston/Lancaster
Szaniawski K. (1977) "Philosophy of the Concrete", Dialectics and
Humanism I, 67-72; repro in this volume
Szaniawski K. - ed. (1989) The Vienna Circle and the Lvov-Warsaw School,
Kluwer: Dordrecht/Boston/London
Tarski A. (1933) Pojecie prawdy w jezykach nauk dedukcyjnych (On the
concept of truth in formalized languages), Towarzystwo Naukowe War-
szawskie: Warszawa. Germ. trans. by L. Blaustein, Studia Philosophi-
ca I, 1935, 261-405. Eng. trans. (from German) by J.H.Woodger in
A.Tarski, Logic, Semantics, Metamathematics. Papers from 1923-1939,
2nd. ed., Hackett: Indianapolis 1983, 152-278
Tarski A. (1944) "The Semantic Conception of Truth and the Foundations
of Semantics", Philosophy and Phenomenological Research 4, 341-76
Tarski A. (1956) Logic, Semantics, Metamathematics. Papers from 1923
to 1938. Oxford University Press: Oxford
Tarski A. (1956a) "Foundations of the Geometry of Solids", in Tarski
(1956), 24-30
Tarski A. (1969) "Truth and Proof", Scientific American 220, 63-77
Tegtmeier E. (1981) Komparative Begriffe. Eine Kritik der Lehre von
Carnap und Heapel, Dunker & Humblot: Berlin
Twardowski K. (1894) Zur Lehre vom Inhalt und Gegenstand der Vorstel-
lungen. Eine psychologische Untersuchung, Holder: Vienna, repro
Philosophia: Munich/Vienna, 1982. Eng. trans. by R.Grossmann, On the
Content and Object of Presentations, Martinus Nijhoff: The Hague,
1977
Twardowski K. (1900) "0 tzw. prawdach wzgl~dnych" (On so called relative
truths), Ksiega Pamiatkowa Uniwersytetu Lwowskiego,' Lwow. Germ.
trans. by M. Wartenberg, Archiv fur systematische Philosophie 8,
415-47, repro in D. Pearce and J. Wolenski; eds. Logische Rational-
is.us. Ausgewahlte Schriften der Lemberg-Warschauer Schule,
Athenaum: Frankfurt 1988, 38-58
Twardowski K. (1902) "tiber sogennante.n relative Wahrheiten", Archiv fur
systeaatische Philosophie 8, 415-447: repro in D.Pearce and
J.Wolenski - eds., Logische Rationalismus. Ausgewahlte Schriften der
Lemberg-Warschauer Schule, Athenaum: Frankfurt, 1988, 38-58
Twardowski K. (1912) "Actions and Products. Comments on the Border Area
of Psychology, Grammar and Logic", as extracted in Pelc (1979),
13-27
REFERENCES 223
Twardowski K. (1923) "Issues in the Logic of Adjectives", as extracted

in Pelc (1979), 28-30
"-
Twardowski K. (1971) "0 sceptycyzmie etycznym" (On ethical scepticism),
Etyka 9, 173-222
Whitehead A.N. and Russell B.A.W. (1910/l3) Principia Mathematica,
Cambridge University Press: Cambridge
Wierzbicka A. (1969) Dociekania semantyczne (Semantic investigations),
Ossolineum: Wroc!aw
W~lenski J. (1987) "Reism and Lesniewski's Ontology", History and
Philosophy of Logic 7, 167-76
Wolenski J. (1989) Logic and Philosophy in the Lvov-Warsaw School,
Reidel: Dordrecht/Boston/Lancaster
Wolenski J. (1990) "Kotarbinski, Many-Valued Logic and Truth", in this
volume
Wolenski J. (forthcoming) "Marty and the Lvov-Warsaw School", in K.
Mull'igan - ed., Mind, Meaning and Metaphysics: Philosophy and Theory
of Language of Anton Marty, Nijhoff: Dordrecht
Wolenski J. and Simons P. (1989) "De veritate: Austro-Polish Contribu-
tions to th~ Theory of Truth from Brentano to Tarski", in Szaniawski
(1989), 391-442
Woodger J.H. (1937) The Axiomatic Method in Biology, Cambridge
University Press: Cambridge
Woodger J.H. (1939) The Technique of Theory Construction (International
Encyclopedia of Unified Science, vol. II, no.5), University of
Chicago Press: Chicago
Woodger J.H. (1951) "Science without Properties", The British Journal
for the Philosophy of Science 2, 193-216
Wundt W. (1906) Logik, T. I., 3th ed., Stuttgart
Zemach E. (1970) "Four Ontologies", Journal of Philosophy 67, 213-47,
repro in F.J. Pelletier, ed., Mass Terms. Some ,Philosophical
Problems. Reidel: Dordrecht/Boston/London 1979, 63:-80
Zarnecka-Bidy E. (1985) "w zwia~ku z pismami Kotarbinskiego z zakresu
historii logiki" (Concerning Kotarbinski's writings in the history
of logic), Ruch Filozoficzny 42, 64-69
Zarnecka-Bia!y E. (1988) "Truth and Logic in Vasil' ev hiaginaryWorlds",
Types of Logical Systems and the Problem of Truth (Logic and its
Application). Bulg. Academy of Sciences: Sofia, 11-19
INDEX OF NAMES
Ajdukiewicz K. VII, ix, 7, 33, 47, Einstein A. 180

58, 62-4, 69-83, 85, 89-90, 93, Espinas 5
107, 141-2, 195, 202
Albert the Great 175 Fall J. vii
Aquinas Thomas 170, 175 Frege G. 4, 24, 32-7, 139-40,
Aristotle 3, 7, 10, 25, 31-3, 18 -2, 204
39-41, 48-9, 63, 99, 143-4,
153-6, 160-5, 169-70, 172, 174, Galton A. 167
177-82, 191, 194, 206-8, 210-11 Gawronski A. ix, 23, 43
Arnauld A. 207 Geach P. ix, 31
Geblewicz E. 98
Bacon F. 4, 9, 188, 205, 208 Goodman N. 54, 180-1
Bednarowski W. 173" Greniewski H. 195, 210
Bergson H. 165 Grzegorczyk A. ix, 39, 42, 44, 176
Berkeley G. 66
Bernard C. 103 Harris Z. 49, 51
Bochenski I. 207 Heraclitus 165, 199
Bolzano B. 160, 181, 205 Herschel J. 208
Boole G. 42-5 Hintikka J. 113
Borkowski L. 196 Hiz H. ix, 47
Brentano F. 1, 4, 97, 137-41, 145, Hobbes T. 108
148-9, 153, 160, 164, 170-83 Hoeksema J. 167
Broughton L. vii Hofler A. 148
Bunt H. 182 Hume D. 175
Burali-Forti C. 4 Husserl E. 73, 137, 139, 144, 170,
172,175,183
Cajetan 170
Carnap R. 62, 75, 141 Ingarden R. 154-5, 165, 168, 175,
Cantor G. 139 180, 182
Canty J. 153
Le Chatelier 103 Jan of Glogow 209
Chisholm R. 156, 163, 174, 178 Johnson W. 63, 166
Chwistek L. 207, 210 Jordan Z. 194
Cohen R. 137
Comte A. 180 Kant I. 13, 25, 36
Condillac E. 205, 209 Kastil A. 173
Cornelius H. 172 Katkov G. 173
Coutu rat L. 4 Kline M. 210
Kneale M. 207
Davidson D. 152, 183 Kneale W. 207
Derrida J. 203 Kofka K. 151
Descartes 149, 178, 207 Kohler W. 151
225
INDEX OF NAMES 226
Kotarbinska J. vii, ix, 53 Pascal B. 43

Kotarbinski T. passim Peano G. 4
Kraus O. 173 Pelc J. vii
Kung G. 153, 157 Petraiycki L. 99, 210
Kuratowski K. 44 Plato 31-3, 36, 53, 55, 81-3,· 98,
180, 182-3
Lange O. 103 Poirier R. 207
Leblanc A. 137 Porecki P. 207
Leibniz G. 4, 171, 173, 177, 180, Prantl C. 209
188 Prior A. 34, 196
Lejewski C. vii, ix, 69, 137, Przel~cki M. ix, 85
139-43, 148, 159-60, 163, 165, Pszczolowski T. vii, ix, 97, 99
182-3
Leonard H. 54 Quine W. 32, 134-5, 180-1, 186
Le~niewski S. 4, 8, 32, 33, 41-4,
49, 54-6, 58, 64, 72-3, 75, de la Ramee P. 207
100-1, 107, 110-11, 115-7, 119, Rasiowa H. 170
125, 137, 139-40, 143, 149, Reichenbach H. 62
152-3, 155-70, 172, 176, 178, Rosch E. 28
180-1, 193-6, 204, 206, 210 Russell B. 4, 8, 12, 31, 32, 37,
Lipps T. 151 54, 63, 108, 113, 129, 140,
Locke J. 178 144, 204, 210
Lotze 179
Eukasiewicz J. 4, 8, 191, 194-6, Se llars W. 164
206, 210-11 Schnelle T. 137
Luschei E. 4, 54 Schopenhauer A. 165
Schumann K. 137
Mach E. 99, 180 Sikorski R. 170
Mackie J. 35 Simons P. 137, 145, 147, 164, 167,
Markowski M. 209 196
Martin R. 108 Sinisi V. Vll, ix, 107, 109,
Marty A. 137, 139-40, 144, 149, 119-20, 160-2, 165
152, 170, 172, 175-6, 181-2 SIupecki J. 41, 54, 110
Mehlberg H. 66 Smith B. ix, 137, 145, 147, 151,
Meinong A. 144, 157 172,174-5
Mellor D. 164 Sobocinski B. 114, 182
Mill J. 9, 123, 175, 205, 208 Socrates 208
Moore E. 144 Sokal R. 28
Morscher E. vii Spinoza B. 149
Mostowski A. 66 Srzednicki J. 174
Mourelatos A. 167 Strawson P. 32
Mulligan K. 145, 147 Struve H. 209
Munch D. 137, 175, 180 Suarez 170
Suszko R. 34
Newton I. 40 Szaniawski K. vii, ix, 146, 185,
Nicole P. 207 210
Novak M. 153
Ockham W. 40, 93-6, 108

INDEX OF NAMES 227
Tarski A. 44, 140, 176, 196-7, 205 Wittgenstein L. 24, 31, 35,144,
Taylor E. 103 199
Tegtmeier E. 147 Wole6ski J. vii, ix, 137, 139,
Twardowski K. 1, 137-4D, 144, 148, 156, 158, 161, 183, 191, 196
151,155,157,159,168,170-2, Wolniewicz B. 34, 148, 199
180, 182, 191, 194, 196 Woodger J. 108, 163, 182
Wundt W. 121
Vasilev N. 210
Zarnecka-Bia!y E. vii, ix, 205,
Waskowski E. 210 210
Weingartner P. vii Zemach E. 164-6, 180, 182
Well C. 208 Zermelo E. 140
Whitehead A. 165, 181, 183 Znosko J. 209
Wierzbicka A. 26
INDEX OF SUBJECTS
action, efficient 5, 97, 99 existence 4, 10, 40-2, 51, 53, 61,

aesthetics 9 65-6, 72, 75, 82-3, 85-7, 92-7,
107, 113-4, 134, 138-9, 143-4,
bixalence 195-]96 ]48, 154, 160, 162-5, 167, 169,
Boolean algebra 42, 44, 170 171-3, 177-8, 181
extension 28
category 10, 31, 33-4, 41, 50-I,
57, 60, 64, 69-76, 81, 94-6, falsity 5, 47, 188, 191, 193, 197
98-9, 107, 121-3, 128-31, 139,
141-2, 146, ]53, 157-8, 160, goal 98, 101-4
165-6, 171, 173, 175, 185
category, semantic 12, 14, 70, hypostases 4, 202-3
72-5, 82, 98, 141
category, syntactic 63 idealism 8, 85
class 58, 65, 88, 100-1, 104, idola fori 4
107-8, 114-8, 121, 124-5, 128, indeterminism 194
132-4, 139, 155, 173, 175, 181, intension 28
]86-7, 201 intentionality 23
conceptualism 53
concretism 5, 56, 65-6, 77, 141, logic 2-5, 7-8, 32-4, 37, 41, 78,
146, 185-8 97, 107, 113, 140, 194-5, 197,
conventionalism 107 200, 204-11
copula 10, 17, 21, 32, 56, 58-9, logic, history of 205, 207, 211
61-4, 70, 78, 109-10, 119-21, logic, many-valued 191, 194-6, 210
124-6
materialism 9, 97, 110
definite description 108, Ill, matter 2, 187
113, 129 mereology 44, 54, 110, 116, 143,
determinism 191, 194 156, 168, 170, 176
dogmatism 8 metaphysics 100, 119
methodology 3, 7-9, 98, 103
effectiveness 101-2 name 4-5, 10, 17, 28, 31-5, 37,
efficiency 99, 101, 103, 183 42, 47-9, 51-2, 56-60, 63-6,
empiricism 40 76, 93, 109-13, 118-123, 141-2,
Epicureanism 2 144, 146, 150, 152-3, 156-8,
epistemology 3, 7, 8, 100 161-2, 168-9, 171, 178, 185-7
ethics 2-3, 5, 99-100 name, apparent 34, 57-61, 63, 65,
event 10, 34-6, 50, 69, 82, 98, 146
101, 104-5, 107, 121-2, 129, name, common 32, 108-10, 114, 117,
134, 139, 141-3, 145-7, 154, 120, 125, 128,
165-8, 180-1, 183 name, complex 32-3, 169
229
INDEX OF SUBJECTS 230
name, empty 157-8 process 34-5, 139, 143, 145-7,

name, genuine 31, 57-8, 63, 76, 149-50, 154, 164-7, 169, 178
81, 98, 105, 108-9, 118-21, property 10, 12-3, 39, 41,
124-5, 127-8, 156 56-8, 60-2, 69, 71-3, 79-80,
name, non-denoting 111 82, 86, 107, 112-3, 121-3, 129,
name, plural 162 138-9, 141-2
name, proper 32, 50, 135, 154, 169 protothetic 110, 157
name, pseudo 32, 60, 69-70, 76, psychological statements 9, 16,
93,98,107,112,117,119-21, 19-20, 149, 150, 152
123-5, 127-8, 130, 132-5, 187 psychologism 23-6, 29
name, sham 31
name, shared 32-3 realism 5, 8, 53, 149
name, singular 31-2, 109-10, 114, realism, practical 2
119-20, 129-30, 157-8, 162, realism, radical 8
name, vacuous 109, 113, 114, 120, reism 4-5, 8, 10-1, 13, 37, 39-40,
130 42, 47-8, 50, 56, 69-70, 72,
nominalism 8, 53, 55-6, 75, 107-8, 74-5, 78-3, 93-5, 96-8, 100,
119, 124, 134-5, 137-9, 141, 104, 107-8, 113-4, 119-20, 123,
153 131, 134-5, 137-8, 140-42,
number 36-7, 107, 187, 201 144-5, 148,160-1, 171-3, 178,
180, 182-3, 185, 199, 201-4
object 4-5, 8-10, 12, 21, 28, yeism, material 47, 51
33-4, 3i, 40, 47, 49, 5-54, reism, ontological 56, 58, 61, 63,
56-7, 60, 64-6, 66, 69, 72-4, 70-2, 75-7, 142
76-7, 81-3, 85-8, 90-6, 99-100, reism, real 10, 12-3
]04,107-9, 113-7, 119-21, reism, semantical 10-1, 47, 56-7,
123-4, 127-30, 134-5, 138-9, 59, 66, 70, 72, 76
143-4, 149-0, 155-6, 158-61, relation 4, 10-3, 34-5, 41, 142,
]63-5, 171-6, 178-83, 185-8, 145-8, 150, 154-8, 164, 1fi6,
199-201 168, 171, 174, 176, 181, 183;
onomatoid 10-1, 33-4, 49-50, 57-8, 201
69, 71, 75-7, 80-2, 146, 165, relativism 104, 191
170, 201-2 reliable guardian 2
ontology 7, 40-2, 44, 59, 138-41,
143, 146, 148, 150, 156, 178, scepticism 8
180-3, 188, 199, 204 semantics 3, 7, 14, 23-6, 29, 43,
ontology, Lesniewski's 8, 41-2, 53, 66, 77, 85, 88, 108, 118-9,
56, 64, 108, 110-1, 113, 115, 134, 140-2, 145-6, 148, 182-3
156-62, 168, set 40, 42, 51,.53-5, 60, 66,
99-101, 111, 114-8, 120, 130,
pansomatism 69, 76-7, 79, 138, 134, 139-41, 153, 155, 181,
141, 161 186-7
Platonism 82-3, 139, 182-3 set theory 4-5, 37, 40, 53-4, 66,
praxiology 5-6, 97-9, 103-5 99, 101, 139-40, 181, 183,
pragmatics 28 186-7, 203
principle of contradiction 191, somatism 138, 141, 161
193 speech act 25-7, 29
principle of excluded third 191-2, states of affairs 100, 121, 139,
194 144-5, 147, 149, 172
stoicism 2
INDEX OF SUBJECTS 231
thing 5, 10-3, 36, 39-42, 45,

47-8, 50-1, 53, 56-60, 62-5,
75-83, 86-92, 93, 95, 97-8,
101,103-5,107-8,111,116,
1]8-34, 138-9, 141-51, 153-6,
158-61, 163-74, 176-83, 185-8,
199, 201-2
truth 5, ]8, 27, 34, 47, 51,
58-69, 61, 64, 97, 92, 110,
111, 125, 131, 138, 140-1,
144-5, 150, 156, 164, 168, 188,
191-3, 195-7
truth-makers 145, 158-9
universals 8, 53, 63, 72, 75, 134,

139, 153, 172, 173, 175, 186
utilitarianism 2
1. N. Rotenstreich: Philosophy, History and Politics. Studies in Contemporary

English Philosophy of History. 1976 ISBN Pb 90-247-1743-4
2. J.TJ. Srzednicki: Elements of Social and Political Philosophy. 1976
ISBN Pb 90-247-1744-2
3. W. Tatarkiewicz: Analysis of Happiness. Translated from Polish by E. Rothert
and D. Zielinski. 1976 ISBN 90-247-1807-4
4. K. Twardowski: On the Content and Object of Presentations. A Psychological
Investigation. Translated from Polish with an Introduction by R. Grossmann.
1977 ISBN Pb 90-247-1926-7
5. W. Tatarkiewicz: A History of Six Ideas. An Essay in Aesthetics. Translated
from Polish a.o. by C. Kasparek. 1980 ISBN 90-247-2233-0
6. H.W. Noonan: Objects and Identity. An Examination of the Relative Identity
Thesis and Its Consequences. 1980 ISBN 90-247-2292-6
7. L. Crocker: Positive Liberty. An Essay in Normative Political Philosophy.
1980 ISBN 90-247-2291-8
8. F. Brentano: The Theory of Categories. Translated from German by R.M.
Chisholm and N. Guterman. 1981 ISBN 90-247-2302-7
9. W. Marciszewski (ed.): Dictionary of Logic as Applied in the Study of
Language. Concepts - Methods - Theories. 1981 ISBN 90-247-2123-7
10. I. Ruzsa: Modal Logic with Descriptions. 1981 ISBN 90-247-2473-2
11. P. Hoffman: The Anatomy of Idealism. Passivity and Activity in Kant, Hegel
and Marx. 1982 ISBN 90-247-2708-1
12. M. S. Gram: Direct Realism. A Study of Perception. 1983
ISBN 90-247-2870-3
13. J.T.J. Srzednicki, V.F. Rickey and J. Czelakowski (eds.): Leiniewski's
Systems. Ontology and Mereology. 1984 ISBN 90-247-2879-7
For 'The Lesniewski Collection' see also Volume 24.
14. J. W. Smith: Reductionism and Cultural Being. A Philosophical Critique of
Sociobiological Reductionism and Physicalist Scientific Unificationism. 1984
ISBN 90-247-2884-3
15. C. Zumbach: The Transcendent Science. Kant's Conception of Biological
Methodology. 1984 ISBN 90-247-2904-1
16. M. A. Notturno: Objectivity, Rationality and the Third Realm: Justification
and the Grounds of Psychologism. A Study of Frege and Popper. 1985
ISBN 90-247-2956-4
Volumes 1-8 previously published under the Series Title: Melbourne 1nternational Philosophy Series.
17. i. Dilman (ed.): Philosophy and Life. Essays on John Wisdom. 1984
ISBN 90-247-2996-3
18. J. J. Russell: Analysis and Dialectic. Studies in the Logic of Foundation
Problems. 1984 ISBN 90-247-2990-4
19. G. Currie and A. Musgrave (eds.): Popper and the Human Sciences. 1985
ISBN Hb 90-247-2998-X; Pb 90-247-3141-0
20. C. D. Broad: Ethics. Lectures given at Cambridge during the Period 1933-34
to 1952-53. Edited by C. Lewy. 1985 ISBN 90-247-3088-0
21. D.A.J. Seargent: Plurality and Continuity. An Essay in G.P. Stout's Theory of
Universals. 1985 ISBN 90-247-3185-2
22. J.E. Atwell: Ends and Principles in Kant's Moral Thought. 1986
ISBN 90-247-3167-4
23. J. Agassi, and I.C. Jarvie (eds.): Rationality: The Critical View. 1987
ISBN Hb 90-247-3275-1; Pb 90-247-3455-X
24. J.T.J. Srzednicki and Z. Stachniak: S. LeSniewski's Lecture Notes in Logic.
1988 ISBN 90-247-3416-9
For 'The Lesniewski's Collection' see also Volume 13.
25. B.M. Taylor (ed.): Michael Dummett. Contributions to Philosophy. 1987
ISBN 90-247-3463-0
26. A. Z. Bar-On: The Categories and the Principle of Coherence. Whitehead's
Theory of Categories in Historical Perspective. 1987 ISBN 90-247-3478-9
27. B. Dziemidok and P. McCormick (eds.): On the Aesthetics of Roman Ingarden.
Interpretations and Assessments. 1989 ISBN 0-7923-0071-8
28. J.T.J. Srzednicki (ed.): Stephan Korner - Philosophical Analysis and
Reconstruction. Contributions to Philosophy. 1987 ISBN 90-247-3543-2
29. P. Brentano: On the Existence of God. Lectures given at the Universities of
Wiirzburg and Vienna (1868-1891). Edited and translated from German by
Susan P. Krantz. 1987 ISBN 90-247-3538-6
30. Z. Augustynek: Time. Past. Present and Future. Essays dedicated to Henryk
Mehlberg. 1990 (in preparation) ISBN 0-7923-0270-2
31. T. Pawlowski: Aesthetic Values. 1989 ISBN 0-7923-0418-7
32. M. Ruse (ed.): What the Philosophy of Biology Is. Essays dedicated to David
Hull. 1989 ISBN 90-247-3778-8
33. J. Young: Willing and Unwilling. A Study in the Philosophy of Arthur
Schopenhauer. 1987 ISBN 90-247-3556-4
34. T. Z. Lavine and V. Tejera (eds.): History and Anti-History in Philosophy.
1989 ISBN 0-7923-0455-1
35. R.L. Epstein: The Semantic Foundations of Logic. Volume 1: Propositional

Logics. With the assistance and collaboration of W.A. Carnielli, I.M.L.
D'Ottaviano, S. Krajewski and R.D. Maddux. 1990 ISBN 0-7923-0622-8
36. A. Pavkovic (ed.): Contemporary Yugoslav Philosophy. The Analytic Ap-
proach. 1988 ISBN 90-247-3776-1
37. A. Winterbourne: The Ideal and the Real. An Outline of Kant's Theory of
Space, Time and Mathematical Construction. 1988 ISBN 90-247-3774-5
38. K. Szaniawski (ed.): The Vienna Circle and the Lvov-Warsaw School. 1989
ISBN 90-247-3798-2
39. G. Priest: In Contradiction. A Study of the Transconsistent. 1987
ISBN 90-247-3630-7
40. J. Wolenski (ed.): Kotarbinski: Logic, Semantics and Ontology. 1990
ISBN 0-7923-0865-4
Further infonnation about our publications on Philosophy are available on request.

Kluwer Academic Publishers - Dordrecht I Boston I London

Kotarbinski Logic PDF

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

Kotarbinski Logic PDF

Caricato da

Copyright:

Formati disponibili

KOTARBINSKI: LOGIC.

SEMANTICS AND ONTOLOGY

General Editor: JAN T. J. SRZEDNICKI

Editorial Advisory Board:

Kotarblnskl logic. semantics. and ontology / edIted by Jan WOlenski.

1. Kotarbinskl. Tadeuz. 2. LoglC. Symbolic and mathematical.

Published by Kluwer Academic Publishers,

Kluwer Academic Publishers incorporates

Sold and distributed in the U.S.A. and Canada

In all other countries, sold and distributed

Printed on acid-free paper

All Rights Reserved

No part of the material protected by this copyright notice may be reproduced or

Index of names 225

Index of subjects 229

Tadeusz Kotarbinski is one of towering figures in contemporary

Tadeusz Kotarbinski (1886-1981) was a Professor of Philosophy in

The proposal to sketch an outline of my philosophical profile originates

1. Woletfsld (ed.), Kotarbitfsld: Logic, Semantics and Ontology, 1-6.

philosophy at Warsaw University, reopened in independent, united Poland.

i.rhose int.erest.ed in t.he .,st.em are referred t.o Lusch .. i (1962).

world to work on praxiological problems was organized. In my later years

1. A GENERAL CHARACTERIZATION OF THE WORK AS A TEXTBOOK AND AS A SCIEN-

The importance and variety of the issues in which the author

Its new achievements consist apparently in that the author penetratingly

The penultimate chapter, entitled On practical abilities, provides

Let us begin with reisa. We are not in a position to formulate, on the

instance, the sentence 'Seniority is a transitive relation' is equlslg-

In a reist's literal language it may meaningfully (though falsely)

In a critical review of the author's opInlons on semantic issues we

Now a number of objections might be raised against such a system of

state directly as to content, in language L, a person's thought that q."

of cancer, and the experience: this is a case of carcinoma. The first

To begin with let us pose a question: what do I aean by psychologism in

Before embarking upon a critical analysis of the theses embodied in

the word 'Si vi' I experience something particular or that I imagine

Unlike strong psychologism in semantics, moderate psychologism does not

of their truth values, because their relevance is not cOllprehensible.

tion wi th what is happening at that moment. I mean, in a relatively

communicative disturbances, because I cannot narrow down someone's

In the present essay I shall be largely concerned with the recognition

naaing role of such common naaes was maintained continuously in

sentences. After one of ay lectures delivered to an adult.audience, one

events and processes shall be eliminated in favour of narrative proposi-

'The news o( Queen's Anne's death causes Lord Bolingbroke to swear.'

What does speak strongly in favour of the reduction to a structure

think the 'demythologizing' of language about events can be

thus attach not directly to things but to Begriffe, Le., to kinds of

take Frege to have meant that he seemed to glimpse a difficulty, but

The philosophical programme of Tadeusz Kotarbinski has been called some-

1. Linguistic requirements and some general ideas

Exallining this experience I find it still useful. The argument for a

should be formulated in utterances containing the nouns of human beings

theoretical description of the totality of facts. Modern physics pro-

2. Formal advantages of the curtailaent of the problem of existence

The first philosophical advantage we get from the reistic attitude is

is .... L .. ~ni .. w .. ki <1930, SIup .. cki (1966),

This interpretation makes us aware of the essential insufficiency of our

3. The reistic semantics

The Elementy, the chief philosophical work of Tadeusz Kotarbinski begins

28 .... the present volume.

The other semantic notions should be introduced in a similar way.

4. Inspirations in geometry and topology