Proceedings of

the Boston Area Colloquium

in Ancient Philosophy
 Proceedings of
the Boston Area Colloquium
in Ancient Philosophy

Volume XXIII, 2007

Edited by

John J. Cleary
Gary M. Gurtler, s.j.

LEIDEN • BOSTON
2008
This book is printed on acid-free paper.

ISSN 1059-986X
ISBN 978 90 04 16685 1 (Pbk)
ISBN 978 90 04 16686 8 (Bound)

© Copyright 2008 by Koninklijke Brill NV, Leiden, The Netherlands.

Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing,
IDC Publishers, Martinus Nijhoff Publishers and VSP.

All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without prior written permission from the publisher.

Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV
provided that the appropriate fees are paid directly to The Copyright Clearance Center,
222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA.
Fees are subject to change.

printed in the netherlands

 CONTENTS

Preface ...................................................................................... ix

COLLOQUIUM 1
Misology and Truth
RAPHAEL WOOLF .................................................................................... 1
Commentary on Woolf
JAMES L. WOOD ...................................................................................... 17
Woolf/Wood Bibliography ..................................................................... 24

COLLOQUIUM 2
Method and Evidence: On Epicurean Preconception
PIERRE-MARIE MOREL ............................................................................ 25
Commentary on Morel
DAVID KONSTAN ..................................................................................... 49
Morel/Konstan Bibliography .................................................................. 55

COLLOQUIUM 3
Rhetoric, Refutation, and What Socrates Believes in Plato’s Gorgias
HENRY TELOH ......................................................................................... 57
Commentary on Teloh
DAVID ROOCHNIK ................................................................................... 78
Teloh/Roochnik Bibliography ................................................................ 82

COLLOQUIUM 4
Plato’s Question of Truth (Versus Heidegger’s Doctrines)
FRANCISCO J. GONZALEZ ........................................................................ 83
vi CONTENTS

Commentary on Gonzalez
GARY M. GURTLER, S. J. ......................................................................... 112
Gonzalez/Gurtler Bibliography .............................................................. 118

COLLOQUIUM 5
Plato’s Anti-Hedonism
MATTHEW EVANS ................................................................................... 121
Commentary on Evans
VERITY HARTE ........................................................................................ 146
Evans/Harte Bibliography ...................................................................... 154

COLLOQUIUM 6
The Good is Benefit: On the Stoic Definition of the Good
KATJA MARIA VOGT ............................................................................... 155
Commentary on Vogt
STEPHEN MENN ....................................................................................... 175
Vogt Bibliography ................................................................................... 185

COLLOQUIUM 7
On Names and Concepts: Mythical and Logical Thinking in Plato’s
Symposium
GÜNTER FIGAL ........................................................................................ 187
Commentary on Figal
DENNIS J. SCHMIDT ................................................................................. 199
Figal/Schmidt Bibliography ................................................................... 204

COLLOQUIUM 8
Yet Another Way to Read the Republic?
ALASDAIR MACINTYRE ........................................................................... 205
MacIntyre Bibliography .......................................................................... 224
 CONTENTS vii

About our Contributors ........................................................................... 225

Index of Names ......................................................................................... 229
 PREFACE

JOHN J. CLEARY

All those readers of previous volumes, who have been either bemused or
annoyed by my numerological reflections, may feel relieved on realizing
that the number 23 is a prime number, which will prevent me from
connecting it with the number 7 which has been one of the guiding threads
for my prefatory remarks about the significance of volume numbers. So I
must lapse into a temporary silence on this topic.
In the present volume we are following the common practice of other
academic journals by including at the head of each paper an abstract
written by the author. These abstracts now serve to introduce the reader to
the topics discussed in these colloquia, so that we can dispense with the
editor’s Introduction which served that function in previous volumes.
In contrast to Volume 22, which was predominantly about Aristotle, this
volume is completely dominated by studies of Plato’s dialogues. Even the
colloquium which discusses Heidegger’s appropriation of Greek
philosophy, is concerned exclusively with his interpretation of Plato’s
concept of truth. The other colloquia focus on topics like Plato’s
discussion of the dangers of misology in the Phaedo, the Socratic use of
rhetoric in the Gorgias, Plato’s anti-hedonism in the Philebus, and the
relationship between mythical and logical thinking in the Symposium. The
only two colloquia that buck this general trend are devoted, respectively,
to the Epicurean notion of preconception, and to the Stoic conception of
the good.
These papers were all presented during the academic year 2006-7 with
the exception of the final colloquium on Plato’s Republic, that was
presented by Alisdair MacIntyre in a previous year but was held back at
the request of the author. This is one reason why it appears here without a
commentary, unlike the other papers in the volume.
These colloquia represent a significant part of the activities of BACAP
at the following participating institutions: Boston College, Boston
University, Brown University, Clark University, College of the Holy
Cross, Dartmouth College. Reflecting the spirit of a colloquium, we have
tried to retain something of the dialogical character of these meetings by
publishing both the main presentation and the commentary. However, it
should be noted that the original presentations have been revised by their
authors in the light of these discussions, and particularly in response to
anonymous reports from external readers. Therefore I would like to
express gratitude to the following referees for their advice and expertise:
x JOHN J. CLEARY

Marcelo Boeri, Francisco Gonzales, Drew Hyland, Anthony Long, Joel

Martinez, Marina McCoy, Daniel Russell.
I would also like to thank the members of the BACAP committee who
host the visits of speakers at the different participating universities, and
generally contribute to the collegial vitality of the BACAP colloquium. As
usual, my sincere thanks goes to Gary Gurtler, my co-editor, and editorial
assistant, Michael J. Smith, for their indispensable cooperation in
preparing this volume for publication.
Finally, I want to express profound gratitude to the enlightened
administrators who continue to provide financial support for the
production of these Proceedings at Boston College.
BOSTON COLLEGE & NUI MAYNOOTH (IRELAND)
 COLLOQUIUM 1

MISOLOGY AND TRUTH

RAPHAEL WOOLF

ABSTRACT
In Plato’s Phaedo, Socrates warns against the dangers of ‘misology,’ or ha-
tred of argument, claiming that it threatens to deprive us of truth and knowl-
edge. In the same passage he tells us that the mark of the philosopher is to
care above all for the truth. His remarks invite us to ask why we should care
about the truth. The Phaedo proposes that truth can be valued either for its
practical utility or because it has a content worthy of engaging us. Neither
model recognises truth for its own sake as a goal, despite its apparent status as
the philosopher’s special interest. How, then, should we value truth? In what
ways can a concern for truth motivate enquiry? And what is the relation be-
tween the search for truth and the beliefs we are especially committed to? The
Phaedo, I argue, both raises and offers a framework for exploring these ques-
tions.

What is the value of truth? The question may be an intriguing one, but
posed thus starkly does not seem particularly amenable to an answer. I
want to make things slightly more tractable by considering one text in par-
ticular which, I think, both poses and reflects on the question. In so doing,
it offers some concrete ways of thinking about how we may value truth,
thereby providing, if not a definitive answer, then at least a framework for
further enquiry. The text I have in mind is Plato’s Phaedo, with special
(though not exclusive) reference therein to Socrates’ discussion of ‘mi-
sology,’ or hatred of argument, at 89d-90d.
I want to suggest that there are at least two ways in which the dialogue
proposes that truth may be valued: (1) for its practical utility, or (2) be-
cause its content expresses a state of affairs that we value. Position (1)
belongs to Simmias, while (2) more closely resembles a position that can
be attributed to Socrates. What I want to argue is that, as set out, neither of
these models recognises a third possibility, namely truth for its own sake
as a goal. Rather obviously, this is the case with a position like that of
Simmias. But the point is applicable to Socrates’ outlook as well. For he
acknowledges, in effect, that he will fight to defend the thesis of the soul’s
immortality not out of a love of truth for its own sake but because of the
value he places on the state of affairs that would obtain if the thesis were
true. The truth is as it may be; and it may not coincide with the outcomes
we are most invested in. In battling to make these two elements coincide,
Socrates invites us to wonder where his deepest allegiance lies.
2 RAPHAEL WOOLF

I think that the richness of the Phaedo’s stance on truth has been insuffi-
ciently recognised. Pursuit of truth for its own sake certainly has some
presence in the dialogue—it is, after all, what philosophers are (or should
be) engaged in. But even on the Phaedo’s own terms it is a radical notion.
Being a philosopher is hard; and reflection on some of the dialogue’s al-
ternative approaches to the value of truth may help us understand why.

As is well-known, in the Phaedo Socrates attempts to prove that the soul is

immortal. But by 88c his efforts look to be in some difficulty, and indeed a
point of crisis has been reached. His main interlocutors, Simmias and Ce-
bes, have each issued stern challenges. Simmias has proposed that the soul
is a kind of harmony or attunement of the elements of the body, which,
though beautiful and incorporeal, will be dissolved on death. Cebes has
argued that even if the soul does outlive a particular body, as a weaver
may outlast a particular cloak, it may yet not be strong enough to outlive
all the bodies it might successively occupy. Perhaps the soul that occupies
my body has been so worn out by its various incarnations that it is due for
destruction when I die.
Phaedo, the dialogue’s narrator, then breaks off to tell Echecrates (to
whom he is reporting Socrates’ final hours) of the despondency that
gripped the audience at this point. But he makes it clear that in fact Socra-
tes succeeded in rallying his companions and managed to re-establish his
case for the soul’s immortality. A large portion of the remainder of the
dialogue is devoted to showing how he did this—against Simmias by a
direct refutation of his theory and against Cebes via what has become
known as the Final Argument.
But Socrates’ first move is to issue a warning against what he calls ‘mi-
sology,’ the hatred of argument. He claims that this arises when one lacks
skill () concerning argument (
); and this in turn seems to mean
a propensity to put too much trust in the truth of particular arguments, just
as misanthropy (according to Socrates’ own analogy, 89d-90b) is caused
by excessive trust in people’s character. Thus an accumulation of disap-
pointments, in which those considered particularly close and reliable turn
out to be dubious, may lead one to trust nobody. So too an overly credu-
lous attitude towards argument may lead one to consider no argument at
all to be reliable and even to conclude that there is no determinate reality,
no fact of the matter about anything (90b-d).
Is it precisely arguments that the misologist comes to hate? One might
wonder whether ‘argument’ (rather than an alternative such as ‘theory’) is
 MISOLOGY AND TRUTH 3

the best rendering of 

 in this context, given that one would not,
strictly, speak of arguments, as Socrates does here of 
, as being true
or false. 1 Socrates’ discussion of misology should probably be character-
ised as concerned to some degree with the relation between theory and
argument. For the rational enquirer, faith in a theory (and ultimately, if one
is not careful, in all theories) starts to dwindle as one loses confidence in
the ability of argument to sustain it. Socrates’ disparaging reference at
90b9-c1 to those who deal in opposed arguments (the ‘contradiction-
mongers,’ 

) makes it clear that a principal source of the prob-
lem of misology is exposure (for one who is not properly prepared) to suc-
cessive arguments for and against a given position. It is not implausible
that over time this might engender scepticism about finding, or even about
there being, determinate truth.

II

That being so, the question I want to press is the following: why should it
matter if one is a misologist? What, if anything, is regrettable about losing
faith in argument? Consider Socrates’ account in a little more detail. He
says that it would be a ‘bitter outcome’ (
 . . . 
, 90c8) if there
were some true and sound argument capable of being discerned, which
one failed to discover through having given up on argument. We would, as
a result, be deprived of ‘the truth and knowledge of the things that are’
(90d6-7). Now one might suppose that being deprived of truth and knowl-
edge is, as Socrates seems to imply, a very bad thing. On this view, mi-
sology is bad because truth and knowledge are good and misology, by
causing us to abandon the search prematurely, threatens to prevent us from
securing them.
Why, then, assuming their attainment is possible, is it good to have truth
and knowledge? Initially two kinds of response can be gleaned from the
text of the Phaedo. The first I shall label ‘theoretical,’ the second ‘practi-
cal.’ On the theoretical reading, truth and knowledge are good because
their possession constitutes by itself a good state, indeed the best state, to
be in. This, or something like it, seems to be Socrates’ position throughout
the Phaedo; and he thinks that for those who have lived rightly it is the
state that awaits us on death (a point that will become relevant shortly).
But there is also a somewhat different strain of thinking in the dialogue,
propounded by Simmias in the preface to his exposition of the harmony

_________
1 See Gallop 1975, 154.
4 RAPHAEL WOOLF

theory (85c-d). On this model, which corresponds to what I am calling the

practical reading, it is not the mere possession of truth and knowledge that
gives them value; rather, their goodness lies in their utility. They provide a
vehicle to guide us on our hazardous journey through life (d1-4).
Simmias in fact is somewhat sceptical about the possibility in our pre-
sent lives of attaining certain knowledge (85c1-4), equating the latter with
a divine account (
 
, d4) that would provide a vehicle by which
we could travel ‘more safely’ ( 
, d3). In its absence he rec-
ommends that we get hold of the best and least refutable of human ac-
counts that we can (c8-d1). The language Simmias employs here turns out
to be quite similar to Socrates’ talk, in the later discussion of cause or ex-
planation (), of the Forms hypothesis as representing a ‘second sail-
ing’ and offering a ‘safe’ answer. 2 The resonance is perhaps closest when
Socrates refers to his second sailing as arising out of a failure to ‘discover
or learn’ (99c9) teleological explanations, which recalls in virtually identi-
cal terms Simmias’ description of resorting to the best human argument
available if one cannot ‘learn or discover’ (85c8) how things are. Simmias
himself even claims that Socrates probably shares the view he is express-
ing (85c2-3).
Such artless appeal to a common perspective might put us on alert, how-
ever. The differences between the positions are as notable as the similari-
ties. Firstly, what Simmias says is humanly attainable is avowedly less
safe than the divine account would be. Socrates, on the other hand, does
not offer his own second sailing (whatever may be the precise import of
that phrase) as comparatively lacking in safety. He refers to it as ‘safest’
( 
) at 100d8. Its safety, given by its air of formal unimpeach-
ability (e.g., beautiful things are beautiful through beauty), is its strong
point. Moreover the objectives of the two still diverge somewhat. What
‘safety’ qualifies for Socrates is, after all, a type of answer (cf.

, d9) in a project of explanation. This project engaged him,
he tells us, because he thought it would be ‘magnificent’ (
,
96a7) to know how the universe worked. In spite of the (misleadingly)
trivial appearance of his ‘safe’ answer, 3 his inspiration was the grandeur of
the project itself, rather than any further benefit that might accrue (which
is not to say that he would scorn such benefit). What Simmias speaks of,
by contrast, is the need to provide a bulwark against the dangers of life.
Knowledge he presents as of value not because its attainment is itself a
_________
2My thanks to Gisela Striker for drawing my attention to the parallel.
3 The answer emerges as a response to a sophisticated rejection of other kinds of expla-
nation; and in turn paves the ground for the enquirer’s chief task: to attain knowledge of the
relevant Form.
 MISOLOGY AND TRUTH 5

wonderful thing (though he need not deny this), but because it (or the
nearest one can get to it) offers a basis on which to structure our lives. In
the case of a merely human account, this may be no sturdier than a ‘raft’
(85d2), but is equally capable of preserving us. One might capture the core
of the difference as follows: for Socrates, our journey is over when knowl-
edge is attained; for Simmias, it will have just begun.

III

What underlies this difference in attitude, or at least in emphasis, between

Socrates and Simmias? One important point is that Socrates’ conception of
truth and knowledge is what one might call ‘ideological.’ I mean that he
has a specific, contentful idea of what the objects of truth and knowledge
are, which provides the grounds for his advocacy of truth and knowledge
as the highest goods. 4 When he recommends the attainment of knowledge
in the Phaedo, he generally has in mind a certain relation between our
souls and the immutable Forms. Even if Simmias shares, to some extent,
Socrates’ metaphysical viewpoint, his advocacy of the practical utility of
knowledge seems quite independent of it. As far as Socrates is concerned,
though, it is just in knowing Forms that we have attained what is most
valuable. This is not to deny that Socrates may consider Forms the only
kind of entity that, properly speaking, can be known. But knowledge of
Forms is not valuable just because they occupy this formal role, but be-
cause of certain substantive features they possess (these features may of
course be what qualify them for the role). Thus it is a further question why
Socrates should think that knowledge of Forms is the highest condition
one can attain. But at the heart of it seems to be the view that such a state
is the realisation of the soul’s natural affinity, or qualitative similarity,
with the Forms, whose characteristics (such as stability, e.g., 79d) are con-
sidered independently to be of the highest value.
Simmias, however, is about to present an argument which will suggest
that the human soul is an entity that dissolves at the end of one’s mortal
span. If this is right, there is no chance of immortality and, in particular,
no chance of our dwelling with the Forms forever, or perhaps at all. There
is no question, I think, that Simmias finds Socrates’ vision of immortality
an immensely appealing one. Nor is it necessarily the case that he regards
the harmony theory as the least refutable of human doctrines. Socrates is

_________
4 I do not mean that his position is not, or could not be, supported by argument. That is a
separate issue.
6 RAPHAEL WOOLF

shortly to elicit from him the view that the Recollection Argument is the
more firmly grounded of the two, and that if the latter is true the former
cannot be (92b-e). Notwithstanding, Simmias takes the harmony theory
seriously at the outset, and what he takes it to imply is that the soul will
fail to outlast its body. This, of course, has important practical conse-
quences. Most obviously, if the theory is correct then Socrates’ idea that
one’s embodied life should be a preparation for the life to come must be
abandoned. Unless the theory is put to rout, the thesis of the soul’s immor-
tality will be too fragile to serve as a framework for shaping one’s life. To
this extent, the challenge issued by the harmony theory gives concrete em-
phasis to Simmias’ insistence that only doctrines that can withstand testing
should serve as one’s life’s guide.
Equally, Socrates’ own account of the nature of the soul and its relation
to the Forms gives backbone to his emphasis on knowledge as itself the
highest condition one can aspire to. But this opens the possibility that his
strictures against misology are themselves ideological. That is, when he
laments those who, thanks to misology, deprive themselves of access to
truth and knowledge of ‘the things that are’ ( , 90d6), one can
read the latter phrase as more than a place-holder. The reference may be
specifically to Forms (as at, e.g., 66a3, 83b1). The text does not compel
such a reading; !  can refer quite generally to what there is (as at,
e.g., 79a6), and this must be the sense a few lines earlier at 90c4. But in
the light of Socrates’ overall position in the Phaedo, the specific reading
fits well here. If we adopt it, then misology starts to matter for Socrates
when viewed against the background of his metaphysics, in particular his
conception of Forms and the soul’s relation thereto. One who gives up on
argument is deprived of the only path to knowledge of the Forms. For
Simmias, by contrast, misology will matter in any world in which humans
have to make decisions about how to live.
This reading of Socrates coheres well with the somewhat surprising fact
that he is evidently a partisan of certain arguments, most notably the ar-
guments for the immortality of the soul. 5 One can best approach this point
by considering again the framing comments of Phaedo and Echecrates
mentioned briefly above. In reflecting on the story so far, the pair initially
evince a rather non-partisan outlook, at least as far as the soul’s immortal-
ity is concerned. In the wake of Simmias’ and Cebes’ attack, Phaedo avers
not that the participants worried that the soul might not be immortal, but
that they might not be good judges of argument (88c6), or even that things
themselves (!  " #) may be untrustworthy (c6-7). The latter

_________
5 This is noted by Dorter 1982, 93-4.
 MISOLOGY AND TRUTH 7

anticipates an element of Socrates’ own sketch of the symptoms of mi-

sology (90c); and for all that Phaedo has said, the worry is just that one
may become mired in the thought that there is, quite generally, no getting
things right. Echecrates’ response, though, rather changes the tone. He
asks how Socrates came to the aid of the argument (88e2-3) and Phaedo
mentions in reply that Socrates rallied his companions to its defence
(89a6-8). He then narrates Socrates himself speaking of fighting against
and defeating the arguments of Simmias and Cebes (89c3-4), the martial
picture reinforced by the comparison drawn with Heracles and Iolaus at
89c5-10, which immediately precedes the remarks on misology. By the
time we encounter these, the suggestion of partisanship with regard to ar-
gument is firmly before us.
It may be that the language of partisanship functions as no more than a
vivid expression of the desirability of rescuing the idea that there is such a
thing as a trustworthy argument and restoring the notion that truth and
knowledge (whatever their objects) are attainable. At the very least,
though, one must ask whether the concrete topic of the argument under
threat has any bearing on the desirability of its rescue. It seems to me
highly implausible to deny this. It is manifestly desirable in its own right
that the thesis of the soul’s immortality turn out true. This is something
agreed upon by the interlocutors throughout the dialogue. It is not, surely,
a coincidence that this very thesis is the one to prompt the language of
battle and victory when under threat. On it rests more than the formal pos-
sibility of truth and knowledge. This is a fight the winning of which will
sustain Socrates’ own vision of the philosophical afterlife, serving at the
same time as a powerful antidote to the interlocutors’ fear of death. 6
Truth and knowledge are, in this regard, heavily loaded concepts before
battle is even joined. To say this is just to admit that the dialogue presents
what it takes to be a highly attractive picture of immortal souls in com-
munion with everlasting Forms; and that this picture is so vigorously de-
fended, at least in part, because it is such an attractive one. Some truths,
perhaps, are more equal than others. Yet we are still entitled to see present
the idea that the possession of truth, whatever it may be, is something
worth striving for—and not just because of its practical utility as envis-
aged by Simmias. Confirmation is to be found in some intriguing remarks
by Socrates himself (91a-c) that represent a kind of confession. He con-
_________
6 As I have noted elsewhere (Woolf 2004, 126), the Final Argument, in considering the
soul simply as bringer of life, does not itself show that its rational faculties are immortal.
However, in seeing off Cebes’ challenge, and in combination with the refutation of the har-
mony theory, it indirectly upholds the more substantial vision of communion with Forms
promulgated most notably in the Affinity Argument.
8 RAPHAEL WOOLF

cludes his discussion of misology by admitting that his own attitude may
not be the correct one. He says that he himself may not be acting philoso-
phically, but from a love of victory, where the distinction turns on whether
one cares for how things really are with respect to the matters under dis-
cussion, or whether one is instead more concerned that one’s own view—
in this case on the immortality of the soul—be thought true, thought true
in fact by Socrates himself (a9-b1), who is after all about to die. Socrates
ends by exhorting his companions not to care for Socrates more than the
truth, and to resist argumentatively by every means if they think he is not
speaking the truth.
Here Socrates both advocates a disinterested pursuit of truth (as the
mark of the philosopher) and at the same time admits that he might not be
practising what he preaches. His partisanship threatens to get the better of
him. If we read his remarks on misology ideologically, then the truth and
knowledge that misology threatens to deprive us of will be exactly what he
fears will slip away should the arguments for immortality fall foul of his
interlocutors’ challenges. His subsequent self-criticism will then be a
poignant acknowledgement that facing argument squarely poses as much
of a threat to his vision as misology does. The ideological reading, to re-
peat, is not compelled, but is given substance by the fact that the warning
against misology is precisely one issued by Socrates, who we know inde-
pendently has a specific conception in this dialogue of what truth and
knowledge are about. The same words read ‘disinterestedly’—without
their particular attachment to speaker—would give us back the idea that
the goal of truth free from such preconceptions is being promoted all
along. 7

_________
7 Let me offer here some brief remarks on James Wood’s reading of Socrates. Wood is
surely right to emphasize the value that Socrates places on a life of seeking the truth and his
citation of the Meno is pertinent. Indeed it is hard to over-estimate how radical is Socrates’
labeling of the failure to strive intellectually (as opposed to, say, on the battlefield) as a
defect in virtue. However, it is a further step to claim that as objects of one’s striving Forms
are ‘merely upheld as a possibility, a goal, or a regulative ideal’ (22), such that one should
live ‘as if it were true’ that they exist, whether or not they do. By analogy, one might say
that it does not matter whether the dragon one is out to slay actually exists; what matters is
that one act as if it does—this might be dubbed, somewhat tongue-in-cheek, a Don Quixote
ethics. More seriously, the misology passage, as we noted earlier, sees as the main cost of
giving up on argument that we would be deprived not of the search but of the attainment of
‘truth and knowledge of the things that are’ (90d6-7). The natural inference from this, it
seems to me, is that the value of the search is predicated on the existence of what one is
searching for.
 MISOLOGY AND TRUTH 9

IV

Yet the idea of a disinterested pursuit of truth is, in one sense at least, an
odd one. Here one should distinguish disinterest with regard to outcome of
enquiry from disinterest with regard to process. In the first case, where I
take the oddity to lie, one is not predisposed to favour any particular out-
come of the enquiry, just so long as it reaches the truth, whatever that may
be. In the second case, one is concerned that the process of enquiry incor-
porate no distorting factors that may militate against the discovery of the
truth. Evidently these two kinds are related, insofar as concern for a truth-
ful outcome come what may ought to motivate concern for a sound proc-
ess. Conversely, one who favours a certain outcome is in danger, even if
unconsciously, of infecting the process with that bias, on issues of inter-
pretation of data, evaluation of arguments, and so on. Of course in princi-
ple one can have strong preference as to outcome without allowing this to
infect the soundness of the process. But it introduces an extra possibility of
infection missing where one has no such partiality.
Simmias remains perfectly sensible in claiming that possession of truth
will be our best guide to life. But in characterising what it means to be
philosophical, Socrates says nothing about this. The philosopher, accord-
ing to him, just cares for how things really are; we are not told that this
concern is predicated on any practical utility. Socrates, by contrast, cares
that things should really be in a highly specific way, and suspects this may
compromise the process of enquiry. He is not, if his own words are to be
believed, quite a philosopher. He wants very much that things be a certain
way, and he is going to devote all his formidable argumentative powers to
the attempt to demonstrate that they are that way. Disinterestedness as to
outcome is notably lacking and its dangers for the pursuit of truth are rec-
ognised.
Which truth? We have noted that if, when truth is pursued, Socrates’
conclusions about immortality turn out to be wrong, he may no longer be
able to look forward to contemplating the truth of the things that are.
Whether or not we read that phrase ideologically, there is no question that
the ideological reading captures what in fact Socrates regards as the high-
est goal of our existence. Caring for truth come what may seems to be the
mark of the philosopher, and Socrates is nothing if not an advocate of fol-
lowing the argument wherever it leads, as reiterated in this very passage.
We are not obliged to speak of a ‘thick’ version of truth, as represented by
Socrates’ grand vision, and a ‘thin’ version—truth as whatever may be the
case. But we can understand why Socrates depicts himself as liable to wa-
ver at this point, and see his wavering raise the question in a particularly
10 RAPHAEL WOOLF

pointed way of what role the pursuit of truth is playing for him, and what
role it should play: for him and for us.
We might compare and contrast two models of enquiry: one in which
the search is guided by a vision of the truth antecedently considered desir-
able; the other in which there is no such vision. Vigorous dialectical ex-
change is not precluded by the first model, as the Phaedo amply testifies.
It is, rather, a question of motivation. 8 The second model leaves it rather
unclear why anyone should care about pursuing the truth. As a practical
guide to life, perhaps. But as an end in itself it seems something that not
even Socrates can quite manage. Truth as a merely formal end seems too
thin (from a psychological point of view) to get enquiry going. Aristotle
tells us that we all desire to know. But he would be happy to admit that
most of us have no desire to know (for example) the exact number of hairs
on one’s head. 9 Such cases suffice to indicate that ‘truth for its own sake’
is, in itself, a dubious motivation and may not be all there is to Socrates’
conception of being philosophical. To care simply that one reaches truth,
whatever it may be, is to be disinterested as to outcome. But what moti-
vates is something about the putative object of enquiry—some aspect (or
perceived aspect) that strikes us as fascinating, mysterious or noble. After
all, in other parts of the Phaedo, not to mention the Republic and else-
where, being a philosopher is intimately connected with having and pursu-
ing a certain definite and purportedly inspiring, vision of reality.
Now there do seem to be cases in which one might want to discover the
truth even while suspecting that what one discovers will be unpleasant or
worse: a spouse’s adultery might be an example. 10 Perhaps this could be
categorised as a seeking of the truth for its own sake, but the main point is
unaffected. For one would have to acknowledge that the urge to discover
would presumably be lacking if one were quite indifferent to one’s spouse.
Of course one might have motives despite the indifference: to gain an ad-
vantageous divorce settlement, for example. But then one could no longer
characterize the case as one of seeking truth for its own sake. The truths
one is motivated to seek for their own sake need not be ones whose dis-
covery one expects to welcome, but they must at minimum be ones whose

_________
8 See here (Heal 1987/88, 97-108). In his (2002, 286 n.8) Williams argues that a posi-
tion such as Heal’s does not refute the idea of truth pursued for its own sake. But Williams
regards it as motivationally necessary that ‘the matters in question are interesting’ (66),
which concedes, for my purposes, the essential point.
9 One can easily conjure scenarios in which such knowledge would be sought: the win-
ning of a bet, for example. But then it would be lucre, not truth, that motivated us.
10 Adapted from (Lynch 2004, 502).
 MISOLOGY AND TRUTH 11

content or subject-matter one has some concern with. Otherwise we have

simply failed to capture the idea that one is going to be motivated at all.

I do not want to appear to be begging the question here. After all, the issue
is precisely whether truth can be valued for itself independently of con-
tent, whether it can be a motivation on its own. My point at root is a psy-
chological one. It is extremely difficult for us to care about truths whose
content we do not care about one way or the other. And, in a rather differ-
ent sense, it can be hard to confront the possibility that, notwithstanding
examples such as the adultery case, content we had taken to be true may
not be so, where that content represents features without which the world
would be, to a significant degree, drained of meaning or interest for us.
One might, though, argue that being motivated by, say, the nobility or
challenge of one’s field is quite different from being partial about specific
outcomes of the enquiries that these features motivate us to undertake. An
astronomer, for example, may find the celestial universe fascinating with-
out favouring one view of it over another, except in the innocuous sense of
favouring whichever accords with the best evidence. Here there seems no
tension between having, in the familiar sense, a keen interest in one’s topic
of enquiry and pursuing enquiry in a disinterested fashion.
Yet this is to some extent an illusion. Astronomy would quickly lose its
appeal if it were discovered that we had been suffering from a mass hallu-
cination and the extra-terrestrial universe were nothing more than a piece
of cardboard with tinsel decoration. The astronomer is partial to outcomes
that confirm his view of the universe as an object worthy of enquiry. For
this is no more than to say that the (apparent) depth and complexity of the
universe is presumably a motivating factor in the undertaking of astro-
nomical enquiry in the first place. So, with a different example, interpret-
ers of Plato will typically perceive his ideas and arguments to have a cer-
tain richness that makes them worth interpreting. Generalising, it seems
that we humans want and need things to engage and absorb us. We are
biased in favour of the world being a place of interest and the bias is a
healthy and necessary one. Fields of enquiry entice us with the challenges
they present, the richness of the worlds they offer, and the prospect of dis-
covery commensurate with the challenge. We tend, thankfully, to be bi-
ased as to outcome. The astronomer who does not prefer making great
discoveries to trivial ones is an object to puzzle at not admire. But this bias
brings, in every walk of life, the danger of reading more into our data than
12 RAPHAEL WOOLF

is strictly justified. We need the , the discipline of good method, to

counter the tendency. 11
Socrates’ self-criticism shows that he recognises the dangers. And it
would be a mistake to tie his vulnerability too closely to the very particular
fact that he is about to die. His vision of the afterlife is precisely one in
which the philosopher has, in his view, the objects of contemplation that
are most worthy of engagement in death or in life. His partisanship be-
longs recognisably to the class I have just tried to describe, even if it is the
occupation of eternity rather than a lifetime that is now at stake. As I sug-
gested earlier, it is surely no coincidence that what prompts the audience’s
spirits to flag is the challenge issued by Simmias and Cebes to Socrates’
appealing vision. It is psychologically plausible that despair about the pos-
sibility of having good grounds to believe at all should arise from the un-
dermining not of anything about which we happen to have been persuaded
argumentatively, but of the category of conclusions we have found espe-
cially attractive or uplifting. So too Socrates’ discussion is creditable in
discerning how our investment in such conclusions may lead to opposite
(though related) extremes: a kind of nihilism, of which Phaedo is at risk,
versus, in his own case, a stubborn desire to remain convinced at all costs.
Clearly neither alternative is ideal, though perhaps a Pyrrhonian sceptic
might bridle at a pejorative characterisation of the first limb. For the Pyr-
rhonian, it is just by giving up on the idea that truth can be discovered that
peace of mind, and so happiness, ensues. Yet it seems to me that, even if it
were possible for a person to exist without (in some sense) believing any-
thing, it would be miserable to exist without believing in anything. And
the ideas that we believe in we must ultimately take to be true. Conversely,
ideas that have deep appeal—ideas which, if you will, we would wish to
be true—are always liable to get a grip on us, and part of what this means
is that we find ourselves invested in their being true, and hence liable to
allow what we actually believe to come apart from what we may be justi-
fied in believing.
We do of course recognise, more or less explicitly, that there are norms
of belief. A distinction is acknowledged (even if it cannot always be ar-
ticulated) between adequate and inadequate bases for belief. Other things
being equal, we would rather be convinced by the truth than just con-
vinced. Since other things rarely are equal, it is not clear that the goal of
getting to truth rather than falsehood need override other considerations.
_________
11 The method itself is apparently to be extracted from the knotty remarks on hypothesis
(100a-b, 101d-e), which seem to underpin the ‘safe’ answer and in which the 


make a return appearance as the enemy. I shall not here offer a view on what exactly the
method consists in.
 MISOLOGY AND TRUTH 13

But if we ignore or downplay this goal, wittingly or otherwise, part of

what we demand is not being answered to. What emerges, then, filtered
through the lens of Socrates’ self-criticism, are the following conditions
for successful enquiry: a vision of the truth worthy of commanding our
attention, combined with an unflinching devotion to argumentative rigour.
There is an evident (if creative) tension between these elements, but the
model suggested seems an approximately attainable one. In recognising
both our need for content that engages (or even inspires) us and our re-
spect for reason, it represents a version, though a rather idealised one, of
how humans—in science, in politics, in everyday negotiations—actually
behave. 12

VI

Socrates’ self-criticism occurs towards the end of the Phaedo, and the lack
of specific challenge either to his critique of the harmony theory or then to
the Final Argument artfully avoids further exposure of the inherent tension
in his partisan stance. 13 Maybe rightly, the idea of pursuit of truth for its
own sake remains something of a still, small voice beside the grand So-
cratic vision and its ingenious dialectical defences. The notion of truth
come what may as the supreme (philosophical) value stands, nonetheless,
as the most radical element of the dialogue, one that quietly subverts and
reproaches, from within, the noisy business that surrounds it—the more so
given that the Phaedo’s outlook is already subversive of much of what we
might take to be regular human values. The explicit presence of Socrates’
self-criticism means that one should not label as ironical the fact that the
ultimate test of his philosophical credentials would be a readiness (in ap-
propriate circumstances) to yield up his philosophical vision to the claim
of truth. If we nonetheless find it difficult to imagine Socrates dying with-
out the commitments that sustain his equanimity, then perhaps, if it is not
too whimsical, we may regard his self-criticism as a promissory note to a
younger, fresher self. The place of his redemption is the world to come:
_________
12 For a famous (or notorious) case of commitment in the scientific realm, one might
consider Einstein’s conviction that ‘god does not play dice’ and its influence on his later
work. To declare that, as a good scientist, Einstein should not have allowed his work to have
been informed by such convictions is to place a corresponding burden on the motivational
force of enquiry for its own sake.
13 Thus the text leaves it instructively unclear whether we should regard Socrates as
having reconciled the tension, or still in its grip. He is ready to affirm at the end of the Final
Argument that ‘the first hypotheses’ should be more clearly examined (107b5-6), a task left,
I suggest below, for the Parmenides.
14 RAPHAEL WOOLF

not Hades but Plato’s Parmenides, in which a youthful Socrates suffers as

a variety of conceptions of the Forms comes under attack from the
eponymous, albeit fictionalised, Eleatic.
Mention of the Parmenides opens up large questions about the devel-
opment of Plato’s thought that it would not be appropriate to address here.
But it is worth noting that even in this most stringent of works the role of
truth remains somewhat equivocal. Parmenides suggests, in the wake of
his attack, that without Forms there can be no philosophical discourse and
the neophyte Socrates agrees (135b-c). In the Phaedo, if the thesis of the
soul’s immortality at least receives serious challenge, the same cannot be
said for its chief prop, the theory of Forms. The positing of Forms is re-
garded as a ‘hypothesis’ within the dialogue (100b), whose methodology
certainly allows that a hypothesis may itself be in need of defence (101d).
But as a matter of fact the Forms hypothesis is not called into question. If
the Socrates of the Phaedo concurs with his young counterpart in treating
the existence of Forms as a prerequisite of philosophical discourse (let
alone of truth and knowledge), that in its turn should be a position open to
critical scrutiny. Yet the Parmenides, while laying siege to the hypothesis
that there are Forms (and that they can be known), leaves this thesis un-
touched.
Speculation as to why is not necessary for present purposes. Perhaps the
thesis is considered true by definition; if so, then Parmenides has a princi-
pled reason for holding fire. Still, it is not obviously true; careful unpack-
ing of ‘Form’ and ‘discourse’ is required before we can be equipped to
affirm or reject the thesis. So it is striking that in a dialogue much con-
cerned with the probing of assumptions, here we seem to have reached
bedrock. Appearances can mislead; one must be properly sensitive to the
possibility that a somewhat changed conception of what a Form is may
emerge from Parmenides’ further enquiries, so that the commitment to
Forms would turn out to be a rather fluid one. On the other hand, Par-
menides himself would be a little misleading if his project were not to up-
hold, in some intelligible sense, those very objects whose credentials have
been questioned but which are said to have such a necessary role in dis-
course. Either way, it is too easy to declare that the truth will out. Cer-
tainly, every substantive commitment is revisable. But that does not mean
every commitment is dispensable. The bottom line, so the Parmenides
suggests, is that what remains should allow, or enable, a life to be worth
living.
Parmenides states (135b5-c2) that one who does not grant that there are
Forms will have nowhere to turn his thought and will thus altogether de-
stroy the power of discourse ($ ). It has been debated whether he
means to rule out not just philosophical discourse (‘dialectic’) here but
 MISOLOGY AND TRUTH 15

meaningful communication of any sort. His language permits either read-

ing, but the indeterminacy only adds force to the way he now starts to get
personal. He switches from the indefinite third person to address Socrates
in the second, noting that the latter seems quite aware of what will follow
if one denies Forms (135c2-3), and then asking pointedly, ‘So what will
you do about philosophy?’ if the critique of Forms goes unanswered (c5-
6). Socrates concedes to Parmenides that he does not see what to do at the
moment (c7). The effect of the exchange is to indicate that, whatever the
broad implications, one aspect of the impasse is of particular consequence.
Philosophy is already what matters most to Socrates (even if, as Par-
menides laconically remarks at 130e, it has not yet gripped him as he be-
lieves it eventually will). So the pair determine not that philosophical dis-
course cannot proceed but that more work needs to be done to establish
that it can.
Parmenides’ prescription is methodological: the intellectual ‘gymnas-
tics’ without which, he says, the truth will elude Socrates (135d6, cf.
136c5). 14 The text does not tell us how to interpret ‘truth’ here. If I am
right, the overall aim of the gymnastics is to vindicate the possibility of
philosophy in the face of arguments that threaten to render it insecure.
Structurally, Parmenides’ attack on the Forms is analogous to the role
played in the Phaedo by the challenges of Simmias and Cebes to the thesis
of the soul’s immortality. In each case the response is complex and distin-
guished by a refusal to indulge in either despair or pig-headedness; but to
call it open-minded would be a misrepresentation. What plays out, rather,
is a search for new ways of bolstering vital commitments.
That death is final; that philosophy is impossible—what would be un-
bearable if true must therefore be shown to be false. To put things thus is
not to criticise the response as somehow impure but to acknowledge its
enduring insight. The scrutiny of a Parmenides or a Simmias, while un-
questionably a motor of intellectual progress, earns its place not by forcing
a choice between our commitments and the truth, but by marking a tension
that will always, and for good reason, be with us. Plato hopes, perhaps
expects, that we will not have to jettison either the truth or the beliefs that
give our life meaning. Debate about the value of truth must proceed from

_________
14 As in the Phaedo, the method is described as one of hypothesis (135e-136c), thence
to be illustrated in the remainder of the Parmenides.
16 RAPHAEL WOOLF

his recognition that reconciliation of the two is a necessary goal, but a far
from inevitable achievement. 15
KING’S COLLEGE LONDON

_________
15 I would like to thank the audience at Boston University for stimulating discussion,
and my BACAP commentator, James Wood, for his thoughtful and challenging response; I
am grateful also for the helpful remarks of an anonymous BACAP referee. An earlier ver-
sion of this paper (entitled ‘Why Does Misology Matter?’) was read at the 2004 Arizona
Colloquium in Ancient Philosophy. My thanks to the audience on that occasion (and in
particular my commentator Roslyn Weiss) for their exemplary philology.
 COMMENTARY ON WOOLF

JAMES L. WOOD

ABSTRACT
This commentary takes issue primarily with the contention that Socrates in
the Phaedo, values the content of his truth claims above truth itself. As a
secondary matter it also disputes the claim that Simmias values truth for prac-
tical reasons and Socrates does not. Both value truth primarily because of its
inherent worth, though also for its practical effects. Socrates has practical
reasons to advance the truth claims that he does, and it is likely that he finds
value in the claims themselves. But there are highly suggestive textual indi-
cations that Socrates does not dogmatically believe the claims he advances.
Due attention to the “second sailing” passage suggests that he believes them
as plausible hypotheses, which are of theoretical value so long as ceaseless
examination continues to validate them and of practical value so long as they
support a noble and philosophical existence. Maintaining a hypothesis in this
sense is living “as if” it is true, which simultaneously acknowledges the
“ideological” and “practical,” as well as the “inherent” values of truth, while
always according primacy to the latter.

I would like to begin where Dr. Woolf ends, by citing the rather
Nietzschean remark that begins his last paragraph: “That death is final;
that philosophy is impossible—what would be unbearable if true must
therefore be shown to be false” (17). It was precisely Nietzsche’s convic-
tion that the illusory nature of truth is no obstacle to its usefulness (and
hence value) to us, and that (false) belief in truth and in certain particular
truths is often necessary for preserving our sanity and for simply getting
by in life. As I gather, Woolf finds a similar strain of thought in Plato,
though juxtaposed dialectically with the philosopher’s preeminent com-
mitment to “truth for its own sake” (1). What it seems Plato and Socrates
cannot accept is the notion that there is no truth, that therefore every belief
is equally grounded in illusion, at most subjectively or relatively true, ex-
cept perhaps for the belief in the one truth that there is no truth. The re-
fusal to accept this notion is grounded in a belief in the value of truth and
truth-seeking, and more basically a belief in the existence of truth and the
possibility of obtaining it, however incompletely. It is not difficult to see
why they hold this belief, for without the reality of truth which can be
sought and obtained, philosophy would indeed be impossible, and the phi-
losopher a fool.
The question is, however, whether this basic commitment to the exis-
tence and value of truth and truth-seeking entails a commitment to some
particular truth content, most relevantly to the existence of forms and the
18 JAMES L. WOOD

soul’s immortality. Must the philosopher believe in such things even if

they turn out not to be true? Does the value of truth rest upon the truth of
the soul’s immortality and the truth of the existence of forms? Woolf sug-
gests that for Socrates, or at least the Socrates of the Phaedo, to a large
extent it does. He further suggests that Socrates’ belief in the value of
truth does not include any commitment to the practical value of truth,
which he sees upheld by Simmias in the dialogue. I disagree completely
with the latter point; as I shall argue, Simmias, like Socrates, upholds the
inherent value of truth, and Socrates’ commitment to the truth, like Sim-
mias,’ is highly practical in its orientation. As for the former point, while I
do agree that Socrates’ commitment to the inherent value of truth is placed
(by Plato) in dialectical tension with his commitment to the particular truth
claims which he advances in this dialogue, I am not convinced that, in the
end, Socratic philosophy is dependent upon the existence of the forms or
the immortality of the soul. There is a sense in which Socrates’ valuing of
the truth depends on his belief in the truth of his philosophical “vision,”
but only if one qualifies the sense in which he believes in it. I shall ex-
plain the qualification I have in mind later on. Bringing these points to-
gether is my basic contention that Socrates’ fundamental commitment is to
the life of truth-seeking, that is, to philosophy as a practical rather than a
purely theoretical affair, and that, consequently, his commitment to any
particular truth claim is less dogmatic than might at first appear to be the
case.
Let us begin with the practical value of truth. Woolf points out the re-
markable similarity of the language in Simmias’ 85c-d speech and in Soc-
rates’ “second sailing” speech, but goes on to emphasize what he takes to
be the crucial differences. In particular, Simmias is said not to value
knowledge for its own sake but because it “provide[s] a bulwark against
the dangers of life” and “a basis on which to structure our lives,” while
Socrates seems to value knowledge itself, inspired by the “grandeur” of
what he is investigating (4-5). Woolf concludes: “One might capture the
core of the difference in the way knowledge is valued as follows: for Soc-
rates, our journey is over when knowledge is attained; for Simmias, it will
have just begun” (5).
I do not deny that there is a difference here, but I take it to be a differ-
ence in emphasis and formulation, not a difference in the way knowledge
is valued by each. Simmias’ predominant concern in this passage is that
one not be a “soft or weak man” (85c6) in the face of the elusiveness of
certainty, particularly about the fate of the soul, but continue to press for-
ward with one’s investigations (note: translations of this passage are mine;
subsequent translations are Grube’s). The relevant journey here is the
journey in search of the truth. One might have to make use of the “best
 COMMENTARY ON WOOLF 19

and least refutable” of “human accounts” (85c9) along the way; but their
usefulness, like that of the raft, is purely instrumental: tools to get one to
the journey’s goal and endpoint: truth and knowledge. Similarly, for Soc-
rates the “second sailing” takes place upon a less-than-ideal vessel: human
hypotheses, and specifically the hypothesis of forms. This hypothesis may
be the “safest” among the available options, but only because Socrates’
desired mind-governed teleology does not appear to be an option. That
teleological account echoes Simmias’ “divine account” as the best and
most desirable option—but one which unfortunately remains unavailable
for the present. For both Simmias and Socrates, then, truth is the goal of
the journey, and one makes use of opinions or hypotheses only when one
must. As a goal, the value of truth is not primarily instrumental but final;
truth is useful in one’s life precisely because it has value in itself.
It is true, however, that Simmias places an emphasis on the relation be-
tween truth and life, and on the practical value of truth and truth-seeking,
that seems to be lacking in Socrates’ second-sailing speech. Does this
mean that Socrates neglects the practical importance of truth? Not at all.
We should remember that everything Socrates says and does in this dia-
logue is colored by a very practical set of concerns centered around his
own impending death. He is concerned with pursuing philosophy up to
the last moments of his life, with convincing his friends and companions
not to fear death, and with inspiring those friends to pursue a philosophical
existence themselves so far as they are able. These concerns are con-
nected by Socrates’ basic convictions that it is good to be philosophical
and that it is not philosophical to fear death. If these are Socrates’ preemi-
nent goals in the Phaedo, both for himself and for his friends, then the
eventual fate of the soul would seem to be a distinctly secondary matter—
of importance mainly because of the possibly deleterious effect that belief
in the finality of death might have on philosophical living and fearless
dying (cf. 95c; Woolf 13). If this is the case, the “ideological” value of
truth is in fact subordinate to the practical value of truth for Socrates,
while its practical value in turn is subordinate to its inherent value.
But if it is true that Socrates values truth and the life of truth-seeking
above the content of any particular truths, then his commitment to phi-
losophy does not depend upon the truth of the particular claims he de-
fends: namely, the existence of the forms and the immortality of the soul.
Socrates himself acknowledges that it would be unphilosophical of him to
desire to convince himself and others of the truth of his claims more than
to desire for them all to obtain the truth, whatever that may be (91a-c; cf.
Woolf 8). Yet as Woolf points out, Socrates is never forced to choose be-
tween truth for truth’s sake and his particular truth claims because he ap-
parently manages to convince everyone that his claims are true (14). Cer-
20 JAMES L. WOOD

tainly Socrates acts very much as the partisan in defense of his claims, and
it is indeed difficult to imagine Socrates dying defeated at the end, having
renounced his philosophical “vision” (cf. 13). But must we conclude then
that Socrates is more committed in practice to his ideology than to the
truth, in spite of his lip service to the truth at the conclusion of his mi-
sology speech? Not necessarily. In fact, we need not even conclude that
Socrates himself is completely convinced of the truth of what he is saying.
For if ridding his companions of their fear of death and inspiring them to
live philosophically requires that he convince them of the soul’s immortal-
ity and the existence of the forms, it may be that Socrates’ valuing of truth
and truth-seeking leads him to sacrifice the “contentful” value of truth: or
in other words, to lie on behalf of the truth.
This is perhaps putting it too strongly, though one could refer to the Re-
public’s talk of “noble lies” in this context. But I would refer instead to
the beginning and end of this dialogue: the substantive philosophical dis-
cussion is framed by a myth about the afterlife at the end and by Socrates’
admission at the beginning that in response to a persistent dream he means
to practice “music” and “poetry” in the remaining days of his life (60d-
61b). A few lines later, Socrates remarks that “it is perhaps most appro-
priate for one who is about to depart yonder to tell and examine tales about
what we believe that journey to be like” (61d-e). We are thus led to won-
der: Is the entire defense of the soul’s immortality a tale or myth, an act of
poetry? Is it the hymn to Apollo which Socrates says at the beginning he
has written, his version of the song sung by swans to Apollo as they cele-
brate joyously and prophetically, says Socrates, the day of their death (cf.
84e-85b)? At the end of his concluding myth Socrates cautions:
No sensible man would insist that these things are as I have described them,
but I think it is fitting for a man to risk the belief—for the risk is a noble
one—that this, or something like it, is true about our souls and their dwelling
places, since the soul is evidently immortal, and a man should repeat this to
himself as if it were an incantation, which is why I have been prolonging my
tale. (114d)
Of course Socrates is referring here to the myth itself, not to his previous
arguments on behalf of the soul’s immortality, but he uses similar lan-
guage much earlier. In response to Cebes’ request that Socrates try to per-
suade them, like children, not “to fear death as a bogey,” he instructs them
to “sing charms” over their scared inner children, seeking among them-
selves and others for a suitable charmer, “since there is nothing on which
you could spend your money to greater advantage” (77e-78a).
Let us consider also Socrates’ repeated admission of the inadequacy or
incompleteness of his arguments (cf. 84c-d, 107b), his praise of and evi-
dent delight in the vigorous arguments of his interlocutors against his posi-
 COMMENTARY ON WOOLF 21

tions (cf. 63a, 86d, 89a), and his final declaration to Crito that above all he
wishes for them to do what he has always said, to care for themselves,
even if they don’t believe his present arguments (115b-c). Taken together
with Socrates’ open admission of his partisanship and defense of the value
of truth at 91a-c, these passages indicate that Socrates’ preeminent desire
is for his companions to live good and philosophical lives, even if they
should come to reject his proofs for the immortality of the soul and his
hypothesis for the existence of the forms. The total effect of his speeches
is indeed to convince them of the truth of his claims, but also and more
importantly, to inspire them to live philosophically, which may require
mitigating their sorrow over Socrates’ death and their fear of their own
eventual deaths, but need not require a life-long dedication to Socrates’
arguments or hypotheses.
It is admittedly difficult to accept that Socrates may not believe his own
extensive and intricate arguments on behalf of the soul’s immortality, and
there is no way to prove such a claim. One might refer outside this dia-
logue to Socrates’ considerably more agnostic remarks about the fate of
the soul in the Apology (40c ff.), and to the many dialogues in which Soc-
rates does not put forward his hypotheses about the forms or the soul, yet
nevertheless presses forward with his truth-seeking investigations. Most
notably, he does not even refer to the forms in his defense of his philoso-
phical life in the Apology, where his mission could again be described
most accurately as concerned with seeking the truth himself and inspiring
others to live good and philosophical lives.
Yet even if the total evidence of the dialogues is at best inconclusive on
Socrates’ (and Plato’s) substantive philosophical theories, it is worth re-
flecting on Socrates’ remark near the end on the “nobility” of risking be-
lief. Does he really think it is “noble” to believe in the soul’s immortality
and eventual fellowship with the forms after death, even if it is not true?
What could he mean by this? One is reminded here of Socrates’ remark in
the Meno, in response to Meno’s paradox, that even if his recollection ar-
gument, which purports to prove the immortality of the soul and its pos-
session of truth about reality, is not flawless in every respect, he is still
utterly convinced that “we will be better men, braver and less idle, if we
believe that one must search for the things one does not know” than to
believe knowledge impossible and give up before one begins (86b-c). If
we translate this principle into the terms of the Phaedo, Socrates may be
saying that it is better to posit and seek the forms as the ultimate object of
philosophical investigation, to believe in their kinship with the soul, and to
believe in the possibility of eternal communion with them after death, than
to reject from the outset such a lofty view of reality and the soul in favor
of a more easily defensible but also more mediocre and restrictive vision
22 JAMES L. WOOD

of human possibility. This belief is good and noble because it supports an

elevated way of life, a virtuous and philosophical life, even if it is merely
posited and never proved, and even if it is not literally believed, but
merely upheld as a possibility, a goal, or a regulative ideal. In other
words, one should live as if it were true, at the same time as one continu-
ally subjects its consequences and applications to investigation and never
loses sight of its hypothetical status (cf. 101d). The qualification here is of
particular importance, and should be stressed, for failure to maintain con-
stant awareness that one’s hypothesis is a hypothesis, and failure to main-
tain due diligence in investigating it, would transform an ennobling prem-
ise into deadening dogma. And as I have been arguing, one of the most
distinctive features of Socrates’ philosophy is his unwillingness to regard
any premise as settled and above further investigation, even on the verge
of his own death. Socrates may occasionally play the partisan, but he is
never an ideologue.
It is in the above sense, I believe, that Plato suggests the indispensability
of the forms to philosophical discourse in the Parmenides (135b-c), even
as he subjects them to intense and potentially devastating criticism at the
hands of Parmenides. The forms represent the highest possibility of truth,
being, and goodness: they transcend opinion, determine the essence of
beings, and govern acting and living. In other words, they are epistemo-
logical, ontological, and ethical ideals. Living as if such ideals exist is
noble, even if proof of their actual existence remains elusive. The same is
true of the immortality of the soul. Even if the soul is mortal, so that the
philosopher never gains his ultimate objective of communion with the
forms, he will come closest to this objective by living as if it were possi-
ble. Socrates, in fact, indicates as much early on, when he articulates the
fundamental dilemma facing the philosopher:
Either we can never attain knowledge, or we can do so after death. . . . While
we live, we shall be closest to knowledge if we refrain as much as possible
from association with the body and do not join with it more than we must, if
we are not infected with its nature but purify ourselves from it until the god
himself frees us. (66e-67b, my emphasis)
In other words, if we overcome a diminishing and paralyzing fear of death
and a bestial and degrading slavishness to the body, by subordinating the
body’s drives and desires to the eros of the soul—its desire to gain truth,
know being, and become good and noble—we shall be living philosophi-
cally, in the way Socrates describes, perhaps unfortunately, as “training for
death” (81a). I say unfortunately, because in spite of the calumny of crit-
ics like Nietzsche, Socrates (or Plato) is not advocating an otherworldly
escapism in this dialogue, but a certain way of living in the light of exalt-
 COMMENTARY ON WOOLF 23

ing and ennobling possibility, which Plato represents throughout the dia-
logues in the character of Socrates.
So what if the Socratic (or Platonic) vision turns out not to be true? In
his description of the hypothetical method, Socrates recommends that
when it becomes necessary to defend the hypothesis (of forms) itself, pre-
sumably because the examination of its consequences has revealed prob-
lems in the hypothesis, one should adopt another hypothesis, the best of
“the higher ones,” and defend its consequences in turn (101d-e). Review-
ing the Platonic dialogues as a whole, we do indeed see modifications,
revisions, and additions to the “Socratic vision” taking place, but we never
see its decisive rejection, even in the Parmenides. As Woolf puts it, “Cer-
tainly, every substantive commitment is revisable. But that does not mean
every commitment is dispensable” (14). I would add that the Socratic vi-
sion remains indispensable not because Socrates is an ideologue, unable
finally to tear himself away from an attractive dogma, but because it re-
mains until the end the best among available hypotheses—best both be-
cause theoretical examination continually confirms its truth, even if a pro-
gressively revised and refined truth, and because practical action continu-
ally confirms its nobility as the basis for living one’s life.
It is in any case probably impossible either to prove or to disprove the
reality of the Socratic vision, in its broadest sense, theoretically. But liv-
ing as if the Socratic vision were true is validated if in fact it results in the
realization of a noble, virtuous, and philosophical existence. To live this
way it is necessary both to believe in the highest possibilities of human
nature and to subject one’s beliefs constantly to critical examination, to
live a good and noble existence while always questioning what it means to
be good and noble, and to uphold to oneself and others the preeminent
value of truth and knowledge while at the same time maintaining an inde-
monstrable and possibly false vision of the nature of truth that makes its
pursuit most worthwhile. One may regard this delicate balancing act as
ultimately incoherent or unsustainable, but one may not finally reject it
without putting it to the test on its own terms, as put forward by Plato
through Socrates in the dialogues, by making it, practically, the basis for
one’s own life.
BOSTON UNIVERSITY
 WOOLF/WOOD BIBLIOGRAPHY

Cooper, John M. and D. S. Hutchison. 1997. Plato: Complete Works. Indianapolis.

Dorter, K. 1982. Plato’s Phaedo: An Interpretation. Toronto.
Gallop, D. 1975. Plato, Phaedo. Oxford.
Heal, J. 1987/88. The Disinterested Search for Truth. Proceedings of the Aristotelian
Society 88: 97-108.
Lynch, M. 2004. Minimalism and the Value of Truth. Philosophical Quarterly 54: 497-
517.
Williams, B. 2002. Truth and Truthfulness. Princeton.
Woolf, R. 2004. The Practice of a Philosopher. Oxford Studies in Ancient Philosophy
26: 97-129.
 COLLOQUIUM 2

METHOD AND EVIDENCE: ON EPICUREAN PRECONCEPTION 1

PIERRE-MARIE MOREL

ABSTRACT
In this paper it is argued that preconception (prolêpsis), i.e., the general no-
tion derived from sensation according to Epicurus, is the ‘key concept’ of the
Epicurean methodology. Scholarly discussions have so far mainly focused on
issues about the psychological status of prolêpsis, and the two main points of
view traditionally held—preconception as a representation and preconception
as a movement of thought—have seemed to be incompatible. I argue here that
they are not and that preconception must be considered under both aspects, as
a mental image as well as a movement of thought. However the most impor-
tant point in Epicurus’ agenda is the methodological status of preconception.
It is not reducible to the single function of a basic concept that is necessary
for any subsequent investigation. Preconception, in many occurrences, con-
tinues to operate as a criterion throughout the process of discovery, and not
just as a point of departure. Thus, Epicurean preconception reconciles the
immediateness of sensation and, more generally, of self-evidence with the ra-
tional mediation of method.

Epicureanism, like any empiricist philosophy, must confront the problem

of the status of ‘ideas,’ whether these are understood as general notions,
abstract representations, or simple thoughts. Since we grasp not only indi-
viduals (this cat, this tree here or that one there), but also classes or species
(cats, of which that cat is an instance, trees in general), or again abstract
notions (such as values), we necessarily grasp also ‘ideas,’ in the very
broad sense that I am employing. This poses no problem of principle,
since it is not necessary that such ‘ideas’ are innate ideas, or that they exist
as such, separately from the mental act that grasps them, in the manner of
Platonic ideas. It is enough that we agree on some use of the term. 2
_________
1 Many thanks to David Konstan for his translation of the first version of this paper and
for the stimulating discussions we have had on the epicurean preconception. I would also
like to thank Mary Louise Gill, Erin Roberts and Dimitri El Murr for their remarks and the
Anonymous Referee for her/his accurate reading and useful comments.
2 As does John Locke at the beginning of the Essay Concerning Human Understanding
(I, 1, Intr., § 8) : “What ‘Idea’ stands for. Thus much I thought necessary to say concerning
the occasion of this Inquiry into human Understanding. But, before I proceed on to what I
have thought on this subject, I must here in the entrance beg pardon of my reader for the
frequent use of the word idea, which he will find in the following treatise. It being that term
which, I think, serves best to stand for whatsoever is the object of the understanding when a
man thinks, I have used it to express whatever is meant by phantasm, notion, species, or
26 PIERRE-MARIE MOREL

The problem begins when we have to define the status of an idea. Let
us distinguish between psychological status and logical status. In respect
to psychological status, we may inquire whether an idea is a kind of im-
age, that is, a kind of accessible mental trace—in our case, i.e., the epicu-
rean theory of knowledge, a representation derived from sensation—, or
else a movement, an act of thinking, and whether an idea is a proposition
or reducible to a proposition. As for logical status, we may ask whether an
idea can be in itself true or false, or is only true insofar as it depends on
other terms, for example as logically connected with other terms in a
proposition. We may equally wonder about its methodological function: is
an idea simply a linguistic convention, a pre-knowledge which, because it
derives from earlier experiences, may anticipate experiences yet to come,
or else a criterion of self-evidence that can confirm, after the fact, the va-
lidity of our opinions concerning a given experience?
Let us begin with the problem as it is stated in the epicurean texts. It is
clear that ancient Epicureanism is empiricist. For the Epicureans, sensa-
tion is the first criterion of truth and the origin of all knowledge. 3 Sensa-
tions are, accordingly, in themselves irrefutable. 4 But Epicureanism also
allows for the existence and for the epistemological use of ‘ideas,’ and it is
not ‘anti-intellectual’ in this sense. The Epicureans did not at all seek to
reduce the knowledge of hidden entities to a direct extension of the per-
ception of phenomena, for example via a simple addition of sensible ex-
periences. Our eyes see shade and light, but they do not instruct us as to
the difference between them: “this falls to the mind’s reason (ratio animi)
to discern. The eyes cannot discover the nature of things (natura rerum).” 5
The natura rerum, the ‘nature of things,’ which constitutes the very object
of Lucretius’ poem, only reveals itself truly, then, to the eyes of reason.
E. Asmis, in her fundamental book of 1984, showed that the Epicurean
canon was not just an epistemology (a theory “which proposes sense per-
ceptions and concepts as criteria for testing the truth of beliefs”), but also a
methodology, that is a theory “which proposes two rules that govern the
conduct of an inquiry from the beginning.” 6 The subtlety of Epicurean
methodology resides precisely in the explanation of different modes of
_________
whatever it is, which the mind can be employed about in thinking; and I could not avoid
frequently using it.”
3 Cf. Diog. Laert., X, 31 ; Epicurus, KD (Key Doctrine) 24.
4 Lucretius, DRN (De rerum natura), IV, 469-521 ; Diog. Laert., X, 32.
5 DRN, IV, 384-385 (transl. Long & Sedley).
6 See Asmis 1984, esp. 24. The two rules, according to E. Asmis, are : “a requirement
for initial concepts to demarcate the problem,” and “a requirement for empirical facts to
provide a solution.”
 METHOD AND EVIDENCE 27

inference, that is of the move from sensations to concept (%

): by con-
frontation, analogy, similarity or combination. 7 The procedures of infer-
ence, which permit the verification and, ultimately, the validation of opin-
ions, provide in principle sufficient guaranties for us to make use of repre-
sentations of things that are not immediately perceivable by the senses.
Moreover, the Epicureans have available a very rich arsenal of terms, of-
ten difficult to distinguish clearly from one another, to designate the vari-
ous kinds of ideas or general notions.
The most important of these terms is &: ‘preconception’ (in
Long & Sedley’s translation), or ‘presumption.’ It seems that this word,
which, Cicero affirms, was introduced into philosophy by Epicurus him-
self, 8 is a generic term, which includes others that designate abstract no-
tions or mental operations, in contrast to sensations and affects. We read
at the beginning of a fundamental text on this question:
[1]
Preconception, they [the Epicureans] say, is as it were a perception
(&), or correct opinion ($' (), or conception ()
), or uni-
versal “store notion” (
*  %
"), i.e. memory of that
which has frequently become evident externally: e.g. “such and such a kind
of thing is a man”. For as soon as the word “man” is uttered, immediately its
delineation also comes to mind by means of preconception, since the senses
give the lead. Thus what primarily underlies each name is something self-
evident. And what we inquire about we would not have inquired about if we
had not had prior knowledge of it. For example: “Is what’s standing over
there a horse or a cow?” For one must at some time have come to know the
form of a horse and that of a cow by means of preconception. Nor would we
have named something if we had not previously learnt its delineation by
means of preconception. Thus preconceptions are self-evident (% +).
And opinion depends on something prior and self-evident, which is our point
of reference when we say, e.g., “How do we know if this is a man?” (Diog.
Laert., X, 33) 9
The generic character of the preconception is not explicitly asserted. Nev-
ertheless, it is quite clear that preconceptions are the basic material of all
other notions, as they are also for the Stoics. These latter notions, as I have
said, come in several forms:

_________
7 Diog. Laert., X, 32.
8 Cicero, De natura deorum, I, 44.
9 Long & Sedley translation (as below), but in the first sentence, punctuation is mine.
28 PIERRE-MARIE MOREL

[2]
Also, all notions (%
) arise from the senses (   ) by
means of confrontation, analogy, similarity and combination, with some con-
tribution from reasoning too. (Diog. Laert., X, 32)
It is possible, as some have suggested, that preconceptions, because they
are the most natural and immediate or most basic notions, correspond to
cases of ‘confrontation’ (). The question is very difficult to re-
solve. However that may be, since all %
 derive from sensations, we
can say that each constitutes a kind of “memory of that which has fre-
quently become evident externally,” which is just what a preconception is.
It is thus difficult to establish a clear distinction between preconceptions
and other notions. In addition, although certain opinions, thoughts, or
judgments are false, it seems that, for Epicurus, preconceptions are always
true: (text [1]) “preconceptions are self-evident. And opinion depends on
something prior and self-evident, which is our point of reference when we
say, e.g., ‘how do we know if this is a man?’” Preconception in this sense
is a fundamental or primary ‘idea’ that is always true, because it is abso-
lutely clear (% ), and that is common to all men. It is thus not surpris-
ing that the concept of preconception plays a central role in Epicurean
doctrine, sometimes implicitly, but also explicitly, as the correct concept
of the divine or of justice.
Unfortunately, Epicurus did not bequeath us a general treatment of pre-
conception, and Lucretius’ De rerum natura, which generally translates
the Greek & by notitia or notities, offers only rather dispersed
comments on the matter. 10 What we have is just a few paradigmatic cases
(the gods, the just) where Epicurus makes use of the word &, but
without giving a clear definition of it. Furthermore, the psychological
status of preconception is not entirely clear: the questions that I have
posed above in a general way are relevant as well to the particular frame-
work of Epicurean philosophy. They have given rise to highly divergent
interpretations and there is no current consensus on the matter.
I would like to show that the main problem with respect to preconcep-
tion is not that of its psychological status, on which discussion has gener-
ally focused, so much as that of its logical and, related to this, methodo-
logical status. The question is, then, to understand what the methodologi-
cal function of preconception is. Prolêpsis has, in fact, several different
functions, which are not reducible to the single function of a basic concept

_________
10Occurrences of these terms in Lucretius are : II, 124 ; 745 ; IV, 476 ; 479 ; 854 ; V,
124 ; 183 ; 1047.
 METHOD AND EVIDENCE 29

that is necessary for any subsequent investigation. 11 Each of these func-

tions consists in making some particular use of the self-evidence that is
specific to preconception. This variety in the uses of preconception per-
haps explains the differences among the texts in which it is discussed, and,
correspondingly, the divergence of modern interpretations. At a deeper
level, I would like to show that prolêpsis is the ‘key concept’ in Epicurean
methodology, and that the texts that have survived, even if they do not
confirm it directly, allow us to reconstruct what amounts to a proleptic
method. The logical status of preconception, in the activity of direct infer-
ence but also in the process or technique of confirmation or ‘witnessing,’
perfectly illustrates the advantage that Epicureanism seeks to derive from
first ‘ideas’: preconception is not just a representation endowed with in-
trinsic self-evidence, like sensation and affect, but is equally a mental act
in which thought is related to sensation. It is the condition without which
one could neither establish nor confirm the connection between the invisi-
ble ( $) and the manifest ("
), whether in simple processes
or in more complex ones. It thus reconciles the immediateness of self-
evidence with the rational mediation of method. By referring to prolêpsis
as a ‘key concept,’ then, I mean that preconception links the various acts
or states of knowledge together, so that it is not only the generic term,
which includes the other abstract notions, but also that which correlates
thought with direct experience. This does not mean that preconception
would be a better criterion than sensation, which is, as we shall see, the
first criterion of truth.

I. The Psychological Status of Preconceptions

Let us begin with the difficulties posed by the psychological status of pre-
conception and the problem of the connection between preconception and
sensation.
Two types of argument allow us to affirm that sensation is the primary
criterion of truth: not only negative arguments, for example those that Lu-
cretius proposes to establish the irrefutable nature of sensations, but also
positive arguments, which have to do with the physical status of aisthêsis
itself. As may be seen in the physiological account of sensations in the
Letter to Herodotus, we do not even have to establish that the truth of sen-
sation corresponds to reality: it is reality itself, or at all events a part of
reality. Knowledge, prior to being a relation of correspondence with what
_________
11 According to Asmis 1984.
30 PIERRE-MARIE MOREL

is known, or an equivalence to what is real, is a relation of belonging, of

inherence in what is known. Thus, vision results from the reception of
replicas (-
) or images (simulacra; .$) that are naturally emitted
by the object that is seen. Since they are directly transmitted by effluences
which, in ideal conditions, preserve the structure and properties of the ag-
gregate from which they come, these replicas allow us to form a represen-
tation or impression () which remains in “sympathy”
(") with the object. 12 This same principle of sympathy is equally
valid for the other senses. 13 The impression is thus not strictly subjective
and still less entirely mental: we perceive something that the object pro-
duces of itself, so that the impression is the shape of the body itself:
[3]
And whatever impression we get by focusing our thought or senses, whether
of shape or properties, that is the shape of the solid body, produced through
the image’s concentrated succession or after-effect. (Epicurus, Hrdt., 50)
Under non-standard conditions, it is true, this sympathy will only be par-
tial, a consequence, for example, of air that wears down the simulacra and
is responsible for the effect that, seen from a distance, a tower that is in
fact square seems round to us. 14 But it nevertheless remains the case that
the impression is constituted via an immediate sympathy with the flow of
simulacra or images, and thus that it is constituted in sympathy with the
objective conditions of their production.
But sensation is not just the act of receiving a physical imprint: it also
includes an act of attention or projection (%/
) 15 toward this condi-
tion of passive reception. This interior act, by which we apprehend the
thing perceived within ourselves and relate to it, although it is in its own
right strictly mental, may also be described as a kind of natural process.
Unfortunately, the Epicureans have not given us a clear physical account
of %/
, nor again of other mental operations. Thus, it is difficult to
know whether they followed up on their physical explanation so as to in-
clude %/
. In any case, on the epistemological level, their position is
clear: whatever its exact nature, the %/
 that is included in sensation is
direct. It is therefore different from the judgment, which can be false, that
is applied to this sensation and its objective correlate. Focusing on the

_________
12See Hrdt. (Letter to Herodotus), 50, l. 2.
13Hrdt., 49-53 ; DRN, IV, 462-468.
14Lucretius, DRN, IV, 353-363.
15Long & Sedley: “focusing”; Asmis: “application.” Both seem acceptable to me. In
this context, since I am using Long & Sedley’s translation, I chose “focusing.”
 METHOD AND EVIDENCE 31

affect of blue or red is not the same as the opinion or judgment that this
thing is blue or red.
Sensations are, in any case, true in themselves, because they testify di-
rectly and physically to the actual presence of the thing sensed. As Lu-
cretius puts it, “whatever impression the senses get at any time is true.” 16
In fact, sensation does not require logos—reason or discourse—or even
memory in order to establish the truth of what it expresses. According to
Diogenes Laertius, “all sensation, he [Epicurus] says, is irrational (0

)
and does not involve memory.” 17 Epicurean epistemology thus depends at
bottom on what I would call a ‘principle of immediacy’ or ‘currency.’ The
word ‘immediacy’ here embraces both the unmediated nature of an ex-
perience, and its direct connection with objective reality.
However, can mental states that are distinct from sensations, and which
refer to a past or future sensation, satisfy this principle, to the extent that
they no longer have the advantage of the immediacy of sensation? The
problem poses itself all the more urgently when these states are presented
as criteria, on the same level as sensations and affects. This is precisely
the case with preconceptions. How can they be true in themselves, al-
though their objective correlate (a man, a cow, justice, the divine) is no
longer or not yet present?
There are some texts that may relieve our doubts as to the intrinsic va-
lidity of preconceptions. Thus, the summary that Diogenes offers of the
Epicurean ‘canon’ associates them directly with sensations and affects:
[4]
Thus Epicurus, in the Kanôn (Yardstick), says that the sensations, preconcep-
tions, and the feelings are the criteria of truth. The Epicureans add the “focus-
ings of thought into an impression.” (Diog. Laert., X, 31) 18
An easy solution to the problem, then, would be to recall that preconcep-
tions are not radically distinct from sensations. This is doubtless true.
Long and Sedley hold, quite rightly, that Gassendi's insertion of the defi-
nite article before prolêpseis is needless: 19 even though they constitute a
distinct class among the several criteria, preconceptions are closely tied up
with sensations. There is indeed a natural continuity between sensation
and preconception. That is why, as Diogenes Laertius specifies (text [1]):

_________
16“Proinde quod in quoquest his visum tempore, verumst”, DRN, IV, 499.
17Diog. Laert., X, 31.
18See also Cicero, Academica, II, 142.
19See the Greek text edited by Long & Sedley, who—like Hicks, H.S. Long, Arrighetti,
Marcovich—delete the “” inserted by Gassendi.
32 PIERRE-MARIE MOREL

For as soon as the word ‘man’ is uttered, immediately its delineation also
comes to mind by means of preconception, since the senses give the lead.
Thus what primarily underlies each name is something self-evident. (Diog.
Laert., X, 33)
However, even if the association between the prolêpsis and the word is
direct, it does not have the same type of immediate presence that sensation
does. We immediately grasp in thought the preconception of ‘man,’ but
this operation cannot take precedence over an objective immediacy: it is
not presence, and still less the physical inherence in us of the thing that is
seen, that testifies to the validity of the preconception. I can certainly
judge that this here thing is true, because it results from a sensible impres-
sion whose physical traces are still present in me. That is a necessary con-
sequence of Epicurean physicalism, which Diogenes of Oinoanda ex-
presses quite clearly:
[5]
and after the impingements of the first images, our nature is rendered porous
in such a manner that, even if the objects which it first saw are no longer pre-
sent, images similar to the first ones are received by the mind […]. (Diogenes
of Oinoanda, fgt 9.III.6-14 Smith) 20
Nevertheless, preconception always occurs after the sensation—or the set
of sensations—from which it derives. It is even true that a preconception,
notably that of the gods, may occur in us without any previous perception,
as an innata cognitio, 21 which poses the problem of how to explain gener-
ally the origin of preconceptions. 22 More globally, from an epistemologi-
cal point of view, that which constitutes the basis of proleptic self-
evidence is not the sensation from which it derives: it is rather, on the one
hand, the spontaneity of the association between a preconception and, on
the other, the word or the object that approaches me, for example a horse
or cow, to take the examples given by Diogenes Laertius. Text [1] is en-
tirely clear on this score. What testifies indeed to the truth and self-
evidence of a preconception is not its physical and sensible origin but:
- (a) the fact that a preconception appears to us “as soon as (1") the
word ‘man’ is uttered” and “immediately” (#-);
_________
20"! $2 !  3 %3 $3 


+ 4"
5 4 - 6,
7 "*  )   " 1  
 8$, ! 9"
 
+ 3
 : $
;
$[<] " [. . .]. Smith’s translation.
21 Cicero, De natura deorum, I, 44.
22 We can partly solve this problem if we assume (for example with Goldschmidt
1978, 157-158), that preconception results, in any case, from a material and external event:
the fact that images enter the body through his pores. In the case of gods, it seems that the
.$ act directly on the mind (see in this sense Modrak 2006, 655).
 METHOD AND EVIDENCE 33

- (b) the very principle of the signifying relation: “Nor would we have
named something if we had not previously learnt its delineation by means
of preconception.” In other words: there is a signifier of x, if and only if
there is a preconception of x. Seen this way, the following sentence, which
gives as a consequence the self-evidence of a preconception, is quite clear:
“Thus (
=) preconceptions are self-evident.”
Thus, it is perhaps not true that what appears to me at a distance is a
horse; perhaps it is a cow; but if I am thinking of a horse, then it consists
in spontaneously representing to myself, at the moment of the perception,
a correct preconception of what a horse is. The truth of the preconception
does not reside, then, in contact with the object, really present or merely
named, with which it is related; it resides rather in the spontaneous asso-
ciation of what is actually present with what no longer is (the past sensa-
tion) and/or with what is not yet. Still more simply, although sensation
“does not accommodate memory” or “is incapable of memory,” precon-
ception (text [1]) is a “memory of that which has frequently become evi-
dent externally.” 23 Let us add that preconception is a certain kind of
doxa—a correct one—, and that accordingly it has a ‘propositional struc-
ture’ of the type ‘such a thing is a man’ or ‘the gods are happy and inde-
structible creatures.’ We will have to deal more precisely with this point
later but, in any case, the same is not true of sensation.
Defined this way, preconception seems to constitute a kind of represen-
tation, that is a mental image that is simultaneously distinct both from its
original source and from the object to which it can be applied. Now, not
only does its quality as memory (the fact that it is a recollection of some-
thing past) contrast with the immediacy that gave the sensation its force,
but, once again, it is not a substitute for the direct grasp of a real, external
thing. In no case can the prolêpsis of a sensible object (a man, a horse, or
a cow) take precedence over the actual perception of the thing when it is
actually present. How could it be a criterion of truth, that is, something
that is immediately true in itself, if it depends originally on the truth of
sensation? 24 In other words, of what value is the recollection of the actual
condition (the recollection of the sensation) if only the actual condition is
a guarantee of truth?
Mustn’t we, then, again question the notion, according to which a pre-
conception is a simple representation or a mental image? Understood not
only as that which persists after repeated sensations of a single object, but
_________
23 Diog. Laert., X, 33.
24 As Manuwald 1972, 114, points out, sensible perception is the guarantee of the value
of the preconception as criterion.
34 PIERRE-MARIE MOREL

as a cognitive operation or a movement of thought, it might have the same

kind of actuality (or currency) as the sensible experience, in the strict
sense. In fact, & sounds like an active substantive: the act of
‘grasping in advance.’ 25 To support this hypothesis, we may consider the
possibility, as David Glidden has done, that preconception is a form of
‘apprehension of thought,’ an %/
* < $
. 26 This hypothesis has
been severely criticized by Jürgen Hammerstaedt, 27 who has well illus-
trated the difficulties that it bumps up against, and to which I shall return.
Let us begin by analyzing the terms and presuppositions of this debate.
There is no doubt that prolêpsis is, to some extent, a certain type of rep-
resentation, insofar as it is a stable term of comparison, to which we can
refer particular instances that we encounter. Several texts that are authenti-
cally Epicurean clearly suggest as much. Thus, when Epicurus, in para-
graph 72 of the Letter to Herodotus, contrasts the perception of time to the
way in which we process other things (doubtless he means bodies here, as
Anke Manuwald maintains), he specifies:
[6]
We should not inquire into time in the same way as other things, which we
inquire into in an object by referring them to the preconceptions envisaged in
ourselves (%7 ! /
" > 4"+ #
+ 
&). (Hrdt., 72) 28
The preposition %7 indicates clearly that we relate to something that we
already contain within ourselves and which we can even ‘perceive’ or ‘en-
visage’ (/
") 29 in ourselves. Although there is no objective sub-
stratum of time, preconceptions determine the permanent properties of
stable substrata, or at least those that are relatively permanent. The same
preposition %7 is used to express a relation to preconceptions in a rather

_________
25See “
"/” in Liddell-Scott-Jones, Greek-English Lexicon.
26Glidden 1985. See also Annas 1992, 166-168.
27Hammerstaedt 1996.
28Long & Sedley’s translation modified. About the knowledge of time in Epicureanism,
see my Morel 2002. According to Sedley 1973, it could be that Hrdt., 72-73, on time, “was
not included in the original version of the Letter, but was added at a later date, following a
controversy in which Epicurus perhaps replied to the charge that his theory of properties
could not account for our understanding of a term like ‘time’” (15). In Sedley’s view, the
inclusion of preconceptions as truth-criteria comes later than the Letter to Herodotus.
Accordingly, Hrdt., 37-38 would contain “only the germ of the notion of prolêpsis” (14). On
the latter point, see my following footnote. Regarding the former, I confess that I don’t put
any new hypothesis forward about the very difficult and controversial problem of the
chronology of Epicurus’ works.
29 In Hrdt., 37-38, the same verb / designates the perception of first notions,
which are probably preconceptions. On this text, see further, pp. 46-47.
 METHOD AND EVIDENCE 35

difficult passage in book XXVIII of the Peri Phuseôs of Epicurus, the

purpose of which is to expose human error, an error that does not reside in
preconceptions and appearances ("), but is produced in relation
to preconceptions and appearances:
[7]
Supposing that in those days we thought and said something equivalent, in
the terminology which we then employed, to saying that all human error is
exclusively of the form that arises in relation to preconceptions and appear-
ances because of the multifarious conventions of language […]. (Epicurus,
On nature, Long & Sedley 19 D ; Arrighetti 31.10.8-9)
Again, the Letter to Menoeceus, in paragraph 123, invites us not to add to
the ‘common notion’ (
* ) of God (a notion of which we have
the outline in ourselves) an opinion that contradicts it, that is to say, an
opinion that is contrary to the idea that the gods are happy and incorrupti-
ble. Now, we learn next that this common notion is a preconception, 30 as
opposed to the false assumptions that the majority of men make:
[8]
For there are gods—the knowledge of them is self-evident. But they are not
such as the many believe them to be. For by their beliefs as to their nature the
many do not preserve them. The impious man is not he who denies the gods
of the many, but he who attaches to gods the beliefs of the many about them.
For they are not preconceptions but false suppositions, the assertions of the
many about gods. (Epicurus, Letter to Menoeceus, 123-124)
It is clear, from this text, that we have in ourselves a stable representation
of the gods, however we may have acquired it, a representation to which
we must refer in order to compare with it the various opinions that we may
have concerning the gods. All these texts that insist on the presence of
preconceptions in us go to show the same thing. This is particularly true
of the exposition by Velleius, who insists repeatedly on the inherence of a
notion of the gods, as nature itself has inscribed them in every mind, 31
whence the proposition: “we have ingrained, or rather innate, knowledge
of them” (insitas eorum vel potius innatas cognitiones habemus). 32 In all
these cases, a prolepsis appears as a mental given, as knowledge consti-
tuted in advance, and it is in this sense comparable to a representation.

_________
30 See Philodemus’s use of & and 
"/ about preconception of gods in
Philodemus, On Piety, I, 441-443 ; 1300 ; 1887 (Obbink).
31 I, 43 : in omnium animis eorum notionem impressisset ipsa natura.
32 I, 44.
36 PIERRE-MARIE MOREL

The situation is analogous to the case of the prolêpsis of justice. We

have two maxims of Epicurus on the topic, which are essential to my ar-
gument, since they make explicit reference to the preconception of justice:
[9]
What is legally deemed to be just has its existence in the domain of justice
whenever it is attested to be useful in the requirements of social relationships,
whether or not it turns out to be the same for all. But if someone makes a law
and it does not happen to accord with the utility of social relationships, it no
longer has the nature of justice. And even if what is useful in the sphere of
justice changes but fits the preconception for some time, it was no less just
throughout that time for those who do not confuse themselves with empty ut-
terances but simply look at the facts. (Epicurus, Key Doctrine 37)
Where without any change in circumstances the conventional laws, when
judged by their consequences, were seen not to correspond with the precon-
ception of justice, such laws were not really just; but wherever the laws have
ceased to be useful in consequence of a change in circumstances, in that case
the laws were for the time being just when they were useful for the social re-
lationships of the citizens, and subsequently ceased to be just when they
ceased to be useful. (Epicurus, Key Doctrine 38)
The idea we have of political utility (that is, of what is useful to a given
political community in a given place and time, so that people do not do
each other harm) must adapt itself to the preconception of justice. This
must be, then, sufficiently stable to serve as an invariant and as a point of
comparison. 33
The following question now poses itself: if a preconception is a repre-
sentation, what kind of representation is at stake and what does it show
about the thing it represents? As I have said, we desperately lack system-
atic texts which can help us answer this question. The variety of terms and
concepts that Cicero offers in connection with preconception (notio, an-
ticipatio, informatio, opinio, innata cognitio, praenotio), 34 just where he
states that Epicurus introduced the term and specified its sense, only ac-
centuate the problem.
Anke Manuwald has clarified the situation neatly, by electing—it is
true—to privilege certain texts of Epicurus over the doxographical tradi-
tion and other later sources. This author has shown that what characterizes
prolêpsis, apart from its function as a criterion which it shares with other
modes of knowledge, is that its content is always something general—a
god is always incorruptible and happy; body is not conceivable without a
determinate number of constant properties, such as shape or size—and that
_________
33 On this difficult question, see Goldschmidt 1977 and, more recently, my Morel 2000.
34 De natura deorum, I, 43-45.
 METHOD AND EVIDENCE 37

this content consists in the essential determinate idea that corresponds to

the word, thanks to which the prolêpsis is apprehended. 35 This point is
nicely confirmed by Philodemus, who specifies that preconception takes
the place of definition:
[10]
There is also the meaning that this is the particular definition of that, and this
is the preconception, as when we say that body as body has bulk and resis-
tance, and man as man is a rational animal. (Philodemus, De signis, 52 ;
xxxiv-xxxv De Lacy) 36
Let us say, in a very general way, that prolêpsis is a primary concept, be-
yond which we must not go (since it is self-evident) at the risk of finding
ourselves in an infinite regress toward some supposed prior self-evident
thing. 37
It is just here, however, that we begin to see the inadequacy of the repre-
sentational approach, taken alone. If preconception is only the representa-
tion or mental image of a person or a horse, what guarantees that this im-
age is the best and final one—irreducible—of the reality that it looks to?
In other words, for prolêpsis to be a criterion and self-evident in a final
way, it must be something other than one mental image among others, an
image which could otherwise be confused with illusory representations,
such as a chimera or a centaur. This is why Gisela Striker 38 distinguishes
two points of view in her analysis of prolêpsis: insofar as it is ‘seen,’ it
naturally appears to us as an image; but insofar as it can be described as
something demonstrated or as an indemonstrable, it functions as a criterion
which, according to her, turns it into a proposition. Thus, for justice, the
preconception associated with the word ‘just’ would have for its content
the proposition, ‘what is useful for a human society is just’; the prolêpsis
of god would have for its content the proposition, ‘gods are happy and
immortal.’ 39 By privileging their epistemological or logical function over
their psychological character, G. Striker assigns to preconceptions, accord-
ing to her, a status comparable to that of first premises in Aristotle’s the-
ory of science.

_________
35 Manuwald 1972, 103-105.
36 [?][7 ] 
 .$
 8 $[ ]@$ 7 - &, 6[] 9
."  "  "  
 ) 7 [ ], 7  0
 A 

CD
 
. De Lacy’s translation.
37 Therefore I think that preconception corresponds to the “primary concept” in Letter to
Herodotus, 37-38. See text [13] below.
38 Striker 1996.
39 Striker 1996, 41.
38 PIERRE-MARIE MOREL

It is undeniable that prolêpsis fulfills this function, if we associate it

with Epicurus’ mention of primary notions which allow us to dispense
with demonstration. And I too believe, like Striker, that what most inter-
ests Epicurus is not the psychological explanation of the origin of precon-
ceptions, but their logical function. That having been said, as G. Striker
herself indicates, in contrast to first premises in Aristotle’s theory of sci-
ence, prolêpseis do not constitute the point of departure for demonstrative
syllogisms. They serve rather to support the method of evaluating opin-
ions that consists in comparing opinions with the self-evident truth.40 For
example, the prolêpsis of justice, even if it bears on what is useful to soci-
ety, does not permit us to deduce the best means of encouraging economic
growth while preserving social safeguards. It would permit us to test and
evaluate, according to the situation, the different opinions or options that
might present themselves on the matter, for example a Keynesian eco-
nomical politics vs. the option of spontaneous regulation via the competi-
tive development of the free market.
One may, nevertheless, wonder whether the assimilation of preconcep-
tion to its propositional content really solves the problem. Let us note,
first of all, that the Epicureans, and Epicurus in particular, insist on the
immediate significance of words, as opposed to definitions and ways of
speaking that distance us from their manifest sense. 41 We know, further-
more, that they criticized the value of definitions and that Epicurus, ac-
cording to an anonymous commentator on the Theaetetus, held that the
names are clearer than definitions and ridiculed the absurdity of saying,
instead of “Hello Socrates!,” “Hello, rational mortal animal!” 42 More
radically still, I can certainly associate with a preconception a proposition
that will enunciate its attributes, but that does not tell me in what way this
proposition will be better than any other proposition concerning the same
subject, for example, “the gods are mortal, greedy, and blood-thirsty.”
There again, one must assume something more than the simple ‘content’
of a prolêpsis in order to understand how it can serve as a criterion of
truth, since the content of a prolepsis, as opposed to the content of a sensa-
tion, is not immediately guaranteed by the actual presence of its objective
correlate. That is why the term prolêpsis seems to me to contain as well a
certain movement of assent, which represents the %/
, and must in this
respect be taken as an active substantive.
_________
40 See Long 1971, 120 : “(…) prolêpseis are necessary for the formation and testing of
all assertions and objective judgements.”
41 See especially Peri Phuseôs, Book XXVIII ; Long & Sedley19 D-E.
42 Anonymous Commentary to Plato’s Theaetetus, 22, 39-47 ; Long & Sedley 19 F.
 METHOD AND EVIDENCE 39

David Glidden 43 has emphasized, from this perspective, the implicit

connection that Cicero proceeds to draw, in Book I of De natura deorum,
between preconception and the process that the Epicureans traditionally
designated by the expression %/
* < $
. 44 The connection is in
fact quite explicit in Lucretius. He maintains that there can be an %/
*
< $
, which he translates as injectus animi, toward invisible things,
in the case of colorless bodies such as atoms. He calls this %/
, pre-
cisely, a notitia. 45 In fact, as Glidden notes, the %/
* < $
, like
prolêpsis, requires something more than a simple passive presentation of
what is carried by the flow of simulacra: they require the organization of
this material, an organization that demands the active intervention of the
spirit. 46 Preconception, then, would be a certain kind of %/
* <
$
, characterized by the general nature of its content.47 A testimony
by Clement of Alexandria clearly goes in this direction, specifying that
Epicurus understands preconception as “a focusing on something evident
and on the evident notion (%
) of the thing.” 48 Preconception here is
a movement of thought, an active perception and no longer a simple repre-
sentation.
This solution is supported by the presence of %/
* < $

among the criteria of truth, at least if one trusts Diogenes Laertius, who
attributes this doxa to Epicureans whom he does not identify. 49 In fact,
%/
* < $
 seems to constitute, like sensation, a self-grounding
act of thought, true by itself, and in this sense a criterion of truth. This act

_________
43 Glidden 1985, 188-194.
44 See especially De natura deorum, I, 49.
45 DRN, II, 739-745: “And if by chance it seems to you that the mind cannot project
itself into these bodies <e.g. the atoms>, you wander for astray. For since those born blind,
who have never descried the light of the sun, yet know bodies by touch, never linked with
color for them from the outset of their life, you may know that for our mind too, bodies
pointed with no tint may enter our comprehension <or : ‘preconception’>” (Bailey’s
translation).
46 Glidden 1985, 191.
47 Glidden 1985, 194.
48 Clement of Alexandria, Stromates, II, 4, 157.44; H. Usener, Epicurea, Leipzig 1887
(quoted below: Us.), 255.
49 Diog. Laert., X, 31. Assuming that %/
* < $
 has been inserted later in
the list of the criteria, one could think that it is a distinct criterion, and then, that it is distinct
from the preconception. But this addition—which Sedley 1973, 16, calls a “mystification,”
actually due to Diogenes Laertius—is probably neither so crucial, nor so rigorous. It could
be explained by the aim to extend the list of the criteria, in order that all the true ideas
should be included in it.
40 PIERRE-MARIE MOREL

is distinct by nature from judgments that add non-confirmed opinions to

our initial impressions:
[11]
If you are going to reject any sensation absolutely, and not distinguish opin-
ions reliant on evidence yet awaited from what is already present through
sensation, through feelings, and through every focusing of thought into an
impression, you will confound all your other sensations with empty opinion
and consequently reject the criterion in its entirety (…). (Epicurus, Key Doc-
trine 24; first sentence) 50
Preconception, if we assimilate it to a certain type of %/
* < $
,
thus recovers an immediacy or presence which turns it into a kind of ge-
neric perception. Its epistemological value, as I have said, is guaranteed
by the immediacy of the first movement of thought, in response to a given
stimulus. That can be a word, or the sensory perception of an outline that
appears in the distance. This means that it is not true in virtue of being
assimilated to something real or of a physical sympathy with its object, nor
again in virtue of a capacity, necessarily random, for the physical preser-
vation of past experiences, but that it is true in virtue of the spontaneity by
which the mind associates with an exterior stimulus bits of knowledge that
have been previously acquired.
I see at least one textual indication that favors, if not the assimilation
pure and simple of preconception to an %/
* < $
, then at all
events the active interpretation of the substantive, in a formula in the Let-
ter to Menoeceus, § 124 (text [7]): “For they are not preconceptions but
false suppositions, the assertions of the many about gods.” By this con-
trast, & is placed on the same level as &, judgment or sup-
position. Now, the latter is here a kind of , that is, a declarative
act, which is surely not the same thing as a simple representation. 51 That
does not mean that & and & are two species of 
nor that & is equivalent to a declarative act, but it confirms that
preconception is a certain form of thinking about the object that it denotes.
Glidden’s thesis, nevertheless, has some difficulties which, as I have
said, were exposed by J. Hammerstaedt in the 1996 article that he devoted
to the role of Epicurean preconception. First of all, if, as he believes,

_________
50 About the distinction between, on the one hand, the spontaneous and immediate
movement of %/
* and, on the other hand, the movement we produce in addition by
ourselves, see Long 1971, 118.
51 In X, 34, Diogenes Laertius points out that $' is an & which is true or false.
Now $' is not reducible to a mental image: it’s certainly more of a movement of thought
than a static image.
 METHOD AND EVIDENCE 41

prolêpsis must be identified with the 

 %" mentioned in para-
graph 37 of the Letter to Herodotus, that means that it can be designated
by a term that is certainly not an active substantive: it is a notion that we
‘contemplate’ or ‘look at’ (/), from the moment when we hear
the word that corresponds to it. Besides, Hammerstaedt believes that, in
the text of Cicero’s De natura deorum, the perception of divine images
mentioned in § 49 cannot be assimilated to prolêpsis, which is described in
§ 43 as ‘an already-formed notion.’ 52 He rejects, more generally, the in-
terpretation of prolêpsis as an act of perception. 53
Without entering into detail concerning these two positions, we should
at least ask whether these two points of view—the traditional idea of pre-
conception as a representation, and the idea of preconception as a move-
ment of thought—are really incompatible. Nothing prevents us from con-
sidering preconception as a way of making use of stabilized traces of past
sensations, that is, a way of actualizing or re-actualizing a memory in ac-
cordance with a specific situation. We can further suppose that prolêpsis
is structured like sensation, which contains at the same time a purely pas-
sive affect, the effect of an impression coming from outside, and an active
focusing (%/
) on this affect. Preconception very probably embraces,
at the same time, representation—the trace that is preserved of past ex-
periences—and attention or the focusing on this representation.

II. The Logical Status of Preconceptions

The question hence becomes the following: what types of knowledge de-
liver preconceptions to us? The answer, in my view, is given both in the
testimony of Diogenes Laertius and in the texts of Epicurus where precon-
ception is, if not defined, at least operative. This reply touches simultane-
ously upon both the nature and the function of prolêpsis.
We may note first of all (text [1]) that it is that in virtue of which an out-
line (-
) presents itself to us, whether of a man, a horse, or a cow. One
point here is not very clear: we may be tempted to assimilate prolêpsis to
an outline, to the degree that there is not a doubling of the image—the pre-
conception and then the outline—but a single image. Nevertheless, the

_________
52 “una nozione già formata,” p. 235.
53 See in this sense Manuwald 1972. Nevertheless, when Hammerstaedt (236, n. 64)
quotes the most relevant texts—i.e., Men. (Letter to Menoeceus), 124 and Diog. Laert., X,
33, he doesn’t give real arguments: he just maintains that “è difficile interpretare prolêpsis
come un atto di percezione.”
42 PIERRE-MARIE MOREL

-
 presents itself to us “in virtue of the preconception” (!
&), which seems to presuppose that it is distinct from it. We must,
I think pause a moment to consider the significance of -
 in Epicurean
epistemology and physiology. To that end, we may refer to two passages
in the Letter to Herodotus, paragraphs 35-36 on the one hand, and para-
graph 46 on the other. The first, as we know, mentions “an outline of the
totality of the doctrine,” that is a condensed epitome of Epicurean philoso-
phy that can be of help to us in all circumstances, whatever our level of
competence in the area of natural philosophy. The second explains the
transmission by means of simulacra of the solid shapes from which they
emanate. More precisely, Epicurus says: “these delineations we call ‘im-
ages.’” It seems, then, that the physico-epistemological category of -

looks to its origin as well as it designates an immediate representation. It
is a trace that preserves the immediacy of its origin thanks to its density,
whether it is a question of the transmission of teachings or of simulacra.
The term -" appears elsewhere in two contexts, in paragraphs 36
and 50 of the Letter to Herodotus. Returning now to preconception, we
may suppose, then, that the -
—the manifestation of which occurs in
virtue of preconception—is the actualization or the putting in place of the
content of the preconception. The density of the outline thus accounts for
the fact that the qualities that define this content cannot be dissociated, for
example the essential attributes of divinity: incorruptibility and blessed-
ness.
The question, finally, is to understand what precisely “in virtue of pre-
conception” means: is it a pure initial representation that serves simply as
a point of departure 54 for other mental operations, or does it play an active
role in the presentation of the -
? The answer is, I think, given in the
first passage of the Letter to Menoeceus which appeals to the preconcep-
tion of the gods:
[12]
First, think of god as an imperishable and blessed creature, as the common
idea of god is in outline, and attach to him nothing alien to imperishability or
inappropriate to blessedness. (Epicurus, Letter to Menoeceus, 123)
There is no doubt, as I have said, that the subject here is that of prolêpsis,
which appears explicitly in paragraph 124. Now, this text shows that
_________
54 The Epicurean theory of preconception stands probably for a response to the question
of Plato’s Meno concerning the possibility of choosing a starting point in the search of
knowledge, if we don’t know anything. See, in this sense: Diog. Laert., X, 33 ; Cicero, De
natura deorum, I, 43 ; Clement of Alexandria, Stromata, II, 4, 157.44 and, generally, Usener
255.
 METHOD AND EVIDENCE 43

prolêpsis has two essential and inseparable functions: on the one hand, it
produces an outline of that with which it is associated, 55 because it
sketches it in a general and so necessarily approximate way; on the other
hand, it establishes a rule for the attribution of predicates to the subject, in
this case, god. In that very particular case of the preconception of gods,
preconception serves first of all as a negative criterion for characterizing
its content: it defines what the gods are not. Conversely, it has, by virtue
of its very imprecision, a positive function: a prolêpsis is the condition for
the validity and legitimacy of our statements concerning its object.
Thanks to it, I can accept various representations of the gods, and concede
some points to the traditional, imagistic theology, for instance that the
gods have this or that shape and in particular a human shape, 56 or that they
live in a community and speak Greek, 57 or even that some things in nature
may be called by the names of gods. Lucretius makes this quite clear in
Book II:
If anyone is resolved to call the sea Neptune and corn Ceres, and likes rather
to misuse the title of Bacchus than to utter the true name of the vine-juice, let
us grant that he may proclaim that the world is the Mother of the gods, if only
in very truth he forbear to stain his mind with shameful religious awe. (Lu-
cretius, De Rerum Natura, II, 655-660)
The fact that we may spontaneously attribute a human shape to the gods
poses no problem in itself, provided that we do not ascribe to them any-
thing that is incompatible with their essence, which we grasp in the
prolêpsis. The prolêpsis of the gods is at once both sufficiently imprecise
and sufficiently constraining so as to define the limits within which we can
allow for different representations of the gods. The same obtains in con-
nection with the prolêpsis of justice, following KD 37-38 (text [9]): the
conception of what is useful can change according to the circumstances—
better, it must change—even as it remains consistent with the prolepsis of
justice. This does not suffice, in and of itself, to define justice in a way
that applies to all times and all places, but it constitutes a framework for its
variation or a rule of evaluation for particular situations in which the ques-
tion of legal justice may arise. Because the preconception outlines and
traces its boundaries, it corresponds well to the -
, to which it adds a
regulatory function or capacity.

_________
55 For this reason, Goldschmidt 1978 interestingly compares the Epicurean pre-
conception with the Kantian schema (‘schème’).
56 See Cicero, De natura deorum, I, 46.
57 Philodemus, De Dis, III, Kol. 14 (H. Diels, Philodemos über die Götter, 1917).
44 PIERRE-MARIE MOREL

Now, if we are speaking of a rule, we are no longer speaking merely of a

mental image, but also of an active principle, whether it is a matter of a
purely mental activity or of a practical activity. We are speaking of a
way—at once depictive and active—of producing a representation and
making use of it. Thus, a prolêpsis is, as a sensation and an affect, a
3, a criterion, that is, a rule, in the sense in which it serves both, on
the one hand, as a point of departure and of comparison and, on the other
hand, as an active principle in the use of our cognitive faculties. At all
events, it is in this respect, according to Lucretius, that sensation is a
regula: it is like a carpenter’s square in architecture, 58 in that it is a in-
strument used at the beginning—but an absolutely necessary one—for the
design of a building which, when accomplished, no longer has need of it.
Now, sensation is surely not reducible to a pure representation, stripped of
activity. Prolêpsis, if it is a representation, is thus a representation that is
at once approximate and active, and not one that is finally fixed and per-
fectly adjusted to its object. Correspondingly, it is at once a representation
and the primary use—a regulatory use—of this representation.59 At the
same time, preconception satisfies the ‘principle of immediacy,’ like sen-
sation, because it is an act that refers to some experience, whether this is
internal (for instance, contemplating within oneself the nature of the gods)
or external (recognizing that the silhouette that is coming from a distance
is that of a man and not of a cow).
It is not necessary to refer to later conceptions—such as the Kantian cri-
tique—of schematizing and of the regulatory function of the faculties of
knowledge: the Epicureans themselves provide sufficient signs that point
clearly in the direction that I have just outlined, and they have no need to
analyze the a priori conditions of knowledge of diverse phenomena. In
spite of its cleverness, there is a basic mistake in the interpretation pro-
posed by Goldschmidt, 60 namely that of imposing on Epicurean empiri-
cism a Kantian solution: he must have been persuaded that a concept
wholly derived from experience could not play, actually, the role of an
epistemological rule.
It remains to understand why Epicurus and the texts relating to his ca-
nonic treat prolêpsis in such different ways: simple recollection, anticipa-
tion and rule of variation, linguistic criterion of self-evidence, and crite-

_________
58 DRN, IV, 513-514.
59 See, on that point, the illuminating conclusion of Goldschmidt 1978, 160 : “la
prénotion ne vaut que dans et par son application.”
60 See Goldschmidt 1977 and 1978. Goldschmidt claims that the use of the prolêpsis is
something like a ‘subsomption’ (‘subsumption’ in Kant’s terminology).
 METHOD AND EVIDENCE 45

rion of confirmation or witnessing. As we have seen, this diversity is im-

plied and presupposed by the psychological status of preconception, which
in itself is not a problem. The answer may be found, I believe, in what I
have called the ‘proleptic method’: the various uses of preconception in
respect to their several logical functions. We may distinguish five func-
tions (f) or kinds of use, often interconnected, of preconceptions:
x (f1) preconception as recollection: the ‘natural’ use of precon-
ception as a recollection or retention of previous experiences:
texts [1], [5], [6].
x (f2) linguistic function of preconception: the ‘conventional’ use
of linguistic self-evidence, the self-evidence of the connection
between the thing and the preconceptions: texts [1], [7], [10]
and probably [8] and [12].
x (f3) preconception as an indemonstrable principle: the first
principle of discovery or beginning, which avoids a regressus
ad infinitum: texts [1], [6]. 61
x (f4) the regulatory function of preconception: preconception as
a principle concerning variation in sensory experiences: texts
[4], [8], [12].
x (f5) preconception as mean of confirmation: preconception as a
criterion of witnessing or attestation of our opinions and infer-
ences, on the basis of sensory experience: [9]. On this point, the
basic text is that by Sextus Empiricus on the Epicurean method
of witnessing. 62
Let us pause a moment on this last point. When beliefs are related to the
object of a direct sensory experience, their truth is established by attesta-
tion (%"-) and their falsity by non-attestation (
#
%"-). Thus, when I believe that Plato is coming toward me, I
still need attestation or its opposite, non-attestation, which sensory experi-
ence will provide me when the man I see has come near. When beliefs
relate to hidden things, they can be the subject of a non-disconfirmation
(
# "-) or a disconfirmation ( "-). In this case
I must establish a relation of consequence between the invisible and the
_________
61 See, also, Clement of Alexandria, Stromata, II, 4, 157.44 (Us. 255). Asmis 1984 has
focused on this function, which is, according to her, the “first rule” of Epicurus’ method:
“Epicurus demands that at the very beginning of an inquiry the investigator have concepts
corresponding to the words that are used” (20). According to her, this is the specific
function of the preconception : “literally, a prolêpsis is a ‘grasp’ that has been obtained
‘before’ an inquiry” (22). I hope I have begun to show, in what precedes, that preconception
is not only useful “before an inquiry”.
62 Sextus Empiricus, Adversus Mathematicos,
, VII, 211-216.
46 PIERRE-MARIE MOREL

sensory evidence. Take for example the existence of the void. This cannot
be directly confirmed, but it can be established by non-disconfirmation.
We assume the existence of motion. But this implies the existence of the
void. 63 Therefore we posit the existence of the void. Thus the contrary
hypothesis is disconfirmed and the conclusion is warranted. Sextus makes
no direct reference to preconceptions, and with good reason, since he is
considering here the possibility of false opinions or judgments, whereas
preconception is always true. Let us note, however, that when Diogenes
Laertius alludes to this method of attestation, 64 it comes right after the
mention of preconception (text [1]). Besides, it is clear, from the summary
I have given of Sextus, that every anticipatory judgment is based on a pre-
conception, at least in respect to function (f1): for example, that of a man,
which is the class under which I locate Plato. Also, the notion of move-
ment, directly derived from sensible experience, is clearly considered here
to be a preconception. It permits us, in turn, to establish the existence of
the void, which is in itself imperceptible. There we encounter function
(f5), which is operative in text [9] as in those texts relating to the gods,
because it serves as a point of comparison for our opinions, whether about
laws or about representations of gods, with a primary notion that is inher-
ent in our minds. The proof of this lies in the explicit reference at the be-
ginning of text [9] to what “is attested ( %"
-"
) to be useful
in the requirements of a social relationship.” Preconception thus continues
to operate as a criterion throughout the process of discovery, and not just
as a point of departure.
The best way to conclude, now, is to refer to Epicurus himself, who, in
one text, that in which he defines ‘primary concepts,’ explains what one
must expect of those ‘ideas’ that are the preconceptions. This text is very
dense and elliptical, and it has been the object of numerous commentaries.
In one way, the book already mentioned by E. Asmis is wholly dedicated
to elucidating it. I have inserted into the text possible connections with my
table of the methodological functions of prolêpsis:

_________
63 See Hrdt., 40
64 Diog. Laert., X, 34 (Long & Sedley 18 B).
 METHOD AND EVIDENCE 47

[13]
First, then, Herodotus, we must have grasped () 65 the things which
underlie words [(f1)-(f2)], so that we may have them as a reference point
against which to judge matters of opinion, inquiry and puzzlement [(f5)], and
not have everything undiscriminated for ourselves as we attempt infinite
chains of proofs [(f3)], or have words which are empty [(f2)]. For the primary
concept (
 %") corresponding to each word [(f2)] must be seen
and need no additional proof, if we are going to have a reference point for
matters of inquiry, puzzlement and opinion. Second, we should observe
everything in the light of our sensations, and in general in the light of our pre-
sent focusings whether of thought [probably: (f4)-(f5)] or of any of our dis-
criminatory faculties, and likewise also in the light of the feelings which exist
in us, in order to have a basis for sign-inferences about evidence yet awaited
and about the non-evident. (Epicurus, Letter to Herodotus, 37-38)
There is much to say about details in this passage. But it suggests, at all
events, that ‘primary concepts’ are preconceptions. Although this point has
been variously interpreted, it seems undeniable to me, 66 in light of the ta-
ble of the various functions of Epicurean preconception. Accordingly, it
seems to me highly probable that prolêpsis is included, at the end of the
passage, in the formula “or of any of our discriminatory faculties”—if not
in the “focusings of thought.” If that is true, it confirms that preconception
is required, not only as a fresh start, at the beginning of an inquiry, but
also during the inquiry.
A preconception is not only a mental image, whose immediacy is sec-
ond-order or derived. Nor is it just a cognitive act. It must be considered
under both aspects, mental image and movement of thought. From this
perspective, it satisfies the ‘principle of immediacy,’ like sensation, but in
a different way: like every rule, it is effective at the time of its application.
It is in a given situation, at the moment that we refer and adapt our experi-
ences and our judgments to it, that it really plays the role of a criterion. It
is in this sense a regulatory schema, that is a minimal representation, ap-
proximate but sufficient. It is characterized by an incompleteness that is
both necessary and positive. Thus, what is crucial to it is less its psycho-
logical nature and its content than its methodological function. The Epicu-
rean theory of preconception permits us to identify at least five functions,
which constitute what I have called the proleptic method. The table of
_________
65 Here, the verb is "/, the very verb that is used to form the word &. In
the first sentence, Long & Sedley translate “we must grasp.” I follow the referee’s
suggestion. I agree with her/his idea that the perfect indicates that “we must have acquired a
knowledge,” for example of what a man is, before we ever recognize that what is ap-
proaching is a man.
66 I agree, on that point, with Asmis 1984, 22 and note 9.
48 PIERRE-MARIE MOREL

these different functions allows us to understand why the relevant texts

treat preconception from such different angles. It shows as well what Epi-
curean empiricism expects of an ‘idea’: not a simple, memorized trace of
previous sensory experiences, and still less an intelligible entity distinct
from the sensible world, but a way of referring to our mental images that
connects them, as closely as possible, with the observation of phenomena.
The Epicurean conception of an idea, in the case of preconception, is thus
not a simple faculty psychology. It is above all a methodology, because
the general function of prolêpsis is to articulate, in a single cognitive act,
the particular sensible object and its general character. It is, for this rea-
son, the fundamental condition for any connection between the invisible
and the manifest. Is this too much to ascribe to a primary and basic form
of knowledge such as preconception? It seems to me, on the contrary,
proper to a rigorous empiricism to be able to return, at each moment in the
process of discovery or recognition, to the primary self-evidence of basic
forms of knowledge.
UNIVERSITY PARIS I—PANTHÉON-SORBONNE
 COMMENTARY ON MOREL

DAVID KONSTAN

ABSTRACT
It is argued here that Epicurean prolêpseis, as a criterion of truth, are neces-
sarily incorrigible, like perceptions and the pathê or sensations of pleasure
and pain. Prolêpseis are the result or precipitate of successive perceptions or
aisthêseis, but may represent complex ideas, including a notion of the gods as
immortal, that are not reducible to simple images.

It is clear that prolêpsis was a fundamental concept in Epicurean episte-

mology, but it is notoriously difficult to pin down just what its function
may have been. Pierre-Marie Morel explains why: prolêpsis has multiple
uses, and comes in two different forms—it is both a “mental image” and a
“movement of thought.” Morel’s careful analysis has without a doubt ad-
vanced the discussion greatly, and any interpretation of prolêpsis from
now on can safely take his treatment as a starting point. Nevertheless, by
way of engaging with Morel’s argument I shall attempt to restore a certain
unity to the idea—with what success the reader will have to judge.
Prolêpsis was included by Epicurus among his so-called “criteria of
truth.” Thus, in the passage from Diogenes Laertius (10.31) cited by Mo-
rel, we read: “And so, Epicurus appears in the Canon as saying that sensa-
tions [aisthêseis], prolêpsis, and pathê are the criteria of truth, and Epicu-
reans add imaginative projections of thought [or projections of thought
capable of producing images: ! ! %/
! < $
]”
(my translations throughout). Let us begin, then, by seeing whether we
can infer something about prolêpsis by the company it keeps. In the same
paragraph, Diogenes quotes Epicurus as saying that “every sensation is
non-rational [0

], and is receptive of no memory whatever.” Dio-
genes reports (10.66 = fr. 311 Usener) that, according to Epicurus, “one
part of it [i.e., the soul] is non-rational [0

], and dispersed throughout
the rest of the body, whereas the rational part [ 
] is in the chest,
as is evident from fears [/
] and from joy [khara].” It is a reasonable
inference that sensations are located in the non-rational part of the soul.
What about pathê? Diogenes tells us (10.34) that, according to the Epicu-
reans, “there are two pathê, pleasure [4$
] and pain [  $3], which
exist in every animal, the one pertaining to what is one’s own [
+
],
the other pertaining to what is foreign [ 
], by which choices and
avoidances are distinguished.” That these two pathê exist in all animals,
and not just in human beings, makes it likely that they too pertain to the
50 DAVID KONSTAN

non-rational part of the soul. True, fears and joys clearly are located in the
rational part, that is, in the chest, according to Epicurus (in Principal Doc-
trines 10 they are said to pertain to dianoia or thought; cf. also 18); but
they are not pathê in the special sense that Epicurus gives to the term. For
him, the pathê are just pain and pleasure, not more complex forms of
awareness such as fears and joys. The latter have some cognitive content,
which is why they partake in rationality or logos; the former—pleasure
and pain—do not: they are simply responses to the affective quality of
things in the world, just as sensations are responses to the sensory qualities
of things—their redness, hardness, and so forth. As a result, they are in-
corrigible: just as “all sensations are true” according to the Epicureans (cf.
Lucretius 4.499), so all experiences of pain and pleasure are correct. You
cannot tell me that I am not seeing red (though you can tell me that my
attribution of red to a particular object is mistaken), and likewise you can-
not tell me that I am wrong in saying that I feel pain. The information of
our five senses, together with our capacity to feel pleasure and pain, gives
us the basic information with which we know the world (see Konstan 2006
for further discussion).
What, then, is the status of prolêpseis, which are sandwiched in between
sensations and pathê in Diogenes’ report? Unlike sensations, they involve
memory: they are, Diogenes tells us, “a memory of what has appeared of-
ten from outside” (10.33). This would suggest that prolêpseis do not per-
tain exclusively to the alogon or non-rational part of the soul. And yet, as
Morel notes, “preconceptions are always true,” or rather, to use Epicurus’
own vocabulary, they are “clear” (% +): the Epicureans apparently
reserved the terms “true” and “false” for what they called doxa and hu-
polêpsis, that is, belief and supposition, and belief evidently pertains to the
rational part of the soul. A matter of belief ($
'), according to Epi-
curus, “depends upon a previous thing that is clear”—this is no doubt a
prolêpsis—to which we refer it when we say, ‘How do we know whether
this is a human being?’” So, beliefs—which may be true or false—depend
for their truth value on prolêpseis, which are clear; and prolêpseis, as we
have seen, result from repeated sensations, which are incorrigible. What is
more, while a prolêpsis depends on sense impressions, it is typically trig-
gered by a word or name: for when we hear a word, then, in accord with a
prolêpsis, we conceive of an imprint (-
) of a thing, and in this proc-
ess, moreover, “the senses lead.” This is getting complicated, but the
whole operation would seem to work as follows. Something—a horse or a
cow—is standing in the distance. How do I know what it is? Let us say I
believe that it is a horse: this may be true or false. At the word “horse,”
which I either say or think, I conceive of an imprint of horse, which I refer
to the prolêpsis of “horse” which I have formed from repeated sense im-
 COMMENTARY ON MOREL 51

pressions of horses—impressions which, in themselves, are always accu-

rate, in that they are just what they are. Now, I could be wrong and dis-
cover that, on close inspection, it really was a cow. No problem: my ini-
tial belief was false, in that I matched the impressions I was receiving with
the wrong imprint and prolêpsis. So I correct my belief, and all is well.
Now, several questions arise. First, how do sensations produce those
prolêpseis that are clear and are the basis of our knowing what things are?
Here again, I think Morel has got it right. Sensations are not just random;
they correspond to the way things are constituted in the world. A cow
really is different from a horse or a human being, and the eidôla or films
that are emitted from it, and which preserve its essential properties in re-
gard to one or another of the senses, likewise differ from those emitted by
horses or human beings (cf. de Lacy 1969). So the prolêpseis registered in
memory as a result of successive sensations are clear and distinct from one
another, and can be trusted to discriminate things properly—provided we
stick to the prolêpseis and do not add to them extraneous suppositions—
and (this is a second matter) provided we stick to the primary concept as-
sociated with each word or name.
But what belief could we add that would get in the way of our recogniz-
ing a cow by reference to the prolêpsis that we have formed on the basis of
sensation? How do we go wrong in such identifications? Here, I think,
we see the weakness of conceiving of a prolêpsis or an imprint strictly in
the form of an image. When I hear the word “cow,” I do not simply con-
ceive of a thing that has the shape of a cow; if that were the case, I might
well confuse a cow with a statue of a cow; worse, I would have no basis
for distinguishing the cow from the statue. I have to know rather more
about what a cow is, and the imprint and prolêpsis of a cow need to con-
tain a good deal of this information. And yet, it must all ultimately derive
from the senses, for it is through repeated sense impressions that the
prolêpsis is formed in the first place. The answer must be that sense im-
pressions carry a lot more information than the mere shape of a thing, or
smell or feel. But what?
Now, I must confess that Epicurus speaks (according to Diogenes) of
recognizing “the shape ["
] of a horse or a cow by way of prolêpsis”
(10.33). And perhaps this is enough, at least in some circumstances. But
Philodemus tells us, as Morel points out, that the prolêpsis of a human
being involves the quality of being a rational animal (On Signs 52), and
that the prolêpsis of body involves its having bulk and resistance. These
qualities cannot be inferred from a static image. The repeated sense im-
pressions that result in a prolêpsis of a human being must include evidence
of rational behavior, not just of the human form. Just how successive sen-
sations produce the conception of a human being as rational, or of a cow
52 DAVID KONSTAN

as whatever a cow essentially is, is difficult to say; but I think that we can
assume that the prolêpsis of a cow includes, for example, the fact that a
cow does not possess reason. If we add to the prolêpsis of a cow the sup-
position that it is rational, then we have a false belief about cows. And in
that case, we must return to the prolêpsis and eliminate the false belief.
The way it works becomes clearer when we proceed to a more abstract
prolepsis, namely that of the gods. According to this prolêpsis, the gods
are blessed and immortal, and we know this because our recognition of
them is “clear” (Letter to Herodotus 123)—the same word that is applied
to imprints and prolêpseis. How could we derive this knowledge from
sense impressions—especially since we cannot readily perceive immortal-
ity? For we do have sensations of the gods—if not through our usual
sense organs, then via images that impinge directly on the soul (I am not
sure whether it is the rational or irrational soul that such simulacra stimu-
late, but I incline to think it is the irrational: they enter us particularly
when we are asleep, and cause dreams)—and as a result of successive ex-
periences of these sensations we form a prolêpsis of them. Whatever the
process, acquiring a clear conception of the gods as immortal by way of
the senses does not seem to me to be in principle different from acquiring
a prolêpsis of human beings as rational—or of a cow as being whatever it
is that defines a cow. One can at least imagine how complex sequences of
moving images might lead to the formation of such concepts (Santoro
2000, 37 argues that our idea of the gods’ immortality is based on infer-
ence, but it is unlikely that inference enters into the formation of prolêp-
seis). We must not, however, attach to the gods opinions or beliefs that
are not derived directly from the sensations themselves, for example, that
they are perturbed by human behavior, have passions, and the like, any
more than we should ascribe reason to non-rational animals; Epicurus em-
ploys the terms doxa and hupolêpsis as opposed to prolêpsis for such sup-
positions, just as he did when speaking of humans and cows (124).
Now, not only are there prolêpseis of cows, human beings, and gods,
but also of such general concepts as justice. Thanks to this prolêpsis, we
are in a position to recognize what is and is not just, not only in specific
acts but also in respect to entire law codes. Such codes may be just in
some social contexts, but not in others; it is thanks to the prolêpsis of jus-
tice—one that we must have formed on the basis of sensation—that we
can evaluate when the laws of our own society, for example, have ceased
to be just. This is a highly sophisticated view; but what kind of prolêpsis
is at stake here? Certainly, it cannot take the form of a simple image, such
as we might have imagined in the case of cows, human beings, or even
gods. Are we dealing then with two different kinds of prolêpsis? I am
inclined to think not, just because, as I have indicated, I do not think that
 COMMENTARY ON MOREL 53

the prolêpsis even of a cow is merely an image. We form a conception of

what a cow essentially is, just as we do in the case of justice. I leave aside
the question of whether a prolêpsis necessarily takes the form of a propo-
sition, e.g., “a human being is a rational animal,” “gods are immortal,”
“justice is what is advantageous to society,” and the like, since the status
of propositions in Epicurean epistemology is highly uncertain. Whatever
the answer to this question, I would argue that the prolêpseis of a cow and
of justice are not fundamentally different.
Morel adds a further stipulation concerning prolêpseis: “one must as-
sume something more than the simple ‘content’ of a prolêpsis in order to
understand how it can serve as a criterion of truth, since the content of a
prolêpsis, as opposed to the content of a sensation, is not immediately
guaranteed by the actual presence of its objective correlate. That is why
the term prolêpsis seems to me to contain as well a certain movement of
assent, which represents the epibolê, and must in this respect be taken as
an active substantive” (38). The nature of the epibolê is even more vexed
than that of prolêpsis, and I am not at all confident that I understand it.
Nevertheless, I think that the two concepts must be kept more distinct than
Morel suggests. First, the roots indicate two contrary kinds of activity:
prolêpsis derives from pro- and lambanô, that is, to “seize” or “grasp” in
advance; epibolê is composed of epi- and ballô, to “throw” or “cast” upon.
The one is a gesture of receiving, the other of tossing out. What, then, is
cast forth? Most often, it is thought or dianoia, although Epicurus affirms
that one can perform this epibolê also with other criteria (Letter to He-
rodotus 38; cf. 51). We may note in passing that if what is cast out is a
criterion, then the epibolê or casting forth itself is unlikely to be one, and
the idea that it is a criterion may really belong to later Epicureans, unless
Epicurus was speaking loosely here. What are the other criteria, apart
from dianoia, that might be cast forth? Not sensations or pathê, I would
say, since they are mentioned independently in this passage, although Mo-
rel takes a different view: “sensation is not just the act of receiving a
physical imprint: it also includes an act of attention or projection (epibolê)
toward this condition of passive reception” (30). A better candidate is
prolêpsis itself, since we know that it is a criterion, and it is not otherwise
mentioned here. But why should a prolêpsis be projected? What we are
told is that the name of a thing causes us to conceive of the imprint in ac-
cord with the prolêpsis: there is no hint of casting the prolêpsis forth here.
We would do better, I think, to focus on the sense of dianoia, which we
know to be the chief thing subject to epibolê.
Now, Epicurus sometimes treats dianoia as the mental parallel to sen-
sory experience, such as opsis or sight (Letter to Herodotus 49); thus, at
Letter to Herodotus 50, Epicurus says that we can grasp (lambanô) a
54 DAVID KONSTAN

phantasia of a thing’s shape ("

) or of its attributes ("//)
either by way of dianoia—in this case via epibolê (he uses the adverb
%/)—or else by the sense organs (). If this analogy
were the entire story, then dianoia should be incorrigible, like the senses,
and pertain to the alogon or non-rational part of the soul: it would not
mean “thought,” but rather something like “mental sensation.”1 Is dianoia
a criterion, like sensation, the pathê, and prolêpsis itself? If so, how does
it differ from prolêpsis, and why is it projected?
The best I can suggest is that dianoia or thought is at a somewhat higher
cognitive level than prolêpsis, perhaps involving the processes of “collid-
ing, analogy, similarity and synthesis” that Epicurus says are associated
with epinoiai or concepts (the root noi-, from nous, is common to both
terms). In this respect, dianoia is part way toward supposition and belief,
which may, as we have seen, be false as well as true: indeed, one of the
terms that Diogenes (10.33) tells us is equivalent to prolêpsis is “correct
belief,” as well as ennoia and “general conception” (
* ). If
dianoia too involves, or may involve, such combinatory mental processes,
then it would not be a direct product of sensation, the way prolêpseis seem
to be, and hence it would not automatically correspond to some object in
the world and the effluences it emits. Rather, it would be a notion that we
project. In one fragment of the Peri phuseôs (fr. 26.42 Arrighetti), Epicu-
rus seems to allow for such a distance between sensation and dianoia:
“. . . defined [or divided] by some distance; in this way thought will more
securely grasp stability for the earth, and in a way more in tune with what
appears to our senses” ([] [] $"
 HC
"I 9J !

 4 $
 * "
* : : &, [7] "


+ ! !  
"
). But the precise status of dianoia in
Epicurus’ epistemology, and consequently the nature of its projection or
epibolê, seem to me to be still in need of clarification.
BROWN UNIVERSITY

_________
1 I had formerly supposed that the casting forth of thought was a way of explaining how
we can attend to a particular object at will (cf. Lucretius 4.779-817): among the innumerable
simulacra pouring in upon us, we project our minds onto those we wish to think about, and
this is just the epibolê or projection of our thought. But the notion of projection is not in
fact alluded to in this context.
 MOREL/KONSTAN BIBLIOGRAPHY

Annas, J. 1992. Hellenistic Philosophy of Mind. Berkeley-Los Angeles-Oxford.

Asmis, E. 1984. Epicurus’ Scientific Method. Ithaca-London.
Brunschwig, J. (ed.) 1976 ; 20062. Les Stoïciens et leur logique. Paris.
De Lacy, P. 1969. Limit and Variation in the Epicurean Philosophy. Phoenix 23:
104-13.
Giannantoni, G. and Gigante, M. (eds.) 1993. Epicureismo Greco e romano. Atti del
congresso internazionale. Napoli.
Gill, M.L. and Pellegrin, P. (eds.) 2006. A Companion to Ancient Philosophy
(Blackwell Companions to Philosophy). Malden-Oxford-Victoria.
Glidden, D.K. 1985. Epicurean Prolepsis. Oxford Studies in Ancient Philosophy 3: 175-
217.
Goldschmidt, V. 1977. La Doctrine d’Épicure et le droit. Paris.
_______. 1978. Remarques sur l’origine épicurienne de la ‘prénotion’. In Brunschwig
(ed.): 155-169.
Hammerstaedt, J. 1996. Il ruolo della Prolêpsis epicurea nell’ interpretazione di
Epicuro, Epistula ad Herodotum 37 SG. In Giannantoni and Gigante (eds.): 221-
237.
Konstan, D. 2006. Epicurean “Passions” and the Good Life. Reis and Haffmans
(eds.): 194-205.
Long, A.A. 1971. Aisthesis, Prolepsis and Linguistic Theory in Epicurus. Bulletin of
the Institute of Classical Studies 18: 114-133.
Modrak, D. 2006. Philosophy of Language. In Gill and Pellegrin (eds.): 640-663.
Morel, P.-M. 2000. Épicure, l’histoire et le droit. Revue des Études Anciennes 102:
393-411.
Morel, P.-M. 2002. Les ambiguïtés de la conception épicurienne du temps. Revue
Philosophique de la France et de l’Étranger: 195-211.
Manuwald, A. 1972. Die Prolepsislehre Epikurs. Bonn.
Reis, B. and Haffmans, S. (eds.) 2006. The Virtuous Life in Greek Ethics. Cambridge.
Santoro, M. (ed. and tr.) 2000. [Demetrio Lacone]: [La forma del dio] (PHerc. 1055).
Naples.
Sedley, D. 1973. Epicurus, On Nature Book xxviii. Cronache Ercolanesi 3: 5-83.
Striker, G. 1996. Essays on Hellenistic Epistemology and Ethics. Cambridge.
 COLLOQUIUM 3

RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES IN

PLATO’S GORGIAS

HENRY TELOH

ABSTRACT
I argue that Socrates employs rhetoric on Gorgias, Polus, and Callicles in the
Gorgias. Moreover, in each conversation, Socrates distinguishes good from
bad rhetoric, and he uses good rhetoric on his interlocutors. Socrates, I main-
tain, believes that good rhetoric is a technê because it aims at the health of the
soul. Finally, I defend the view that good rhetoric is a technê against the
criticisms of Brickhouse and Smith, and Roochnik, as well as the fact that in
the Gorgias Socrates does not turn his interlocutors to a life of philosophy.

I aim to show that Socrates extensively uses rhetoric in the Gorgias; in

fact, it is reasonable to categorize a large part of what Socrates does in that
dialogue as rhetorical. Socrates uses rhetoric as part of his philosophical
and educational task to attempt a care of his interlocutors’ souls (&). I
further aim to show that Socrates recognizes that he uses rhetoric, and that
he implies that the sort of rhetoric which he uses is a real art () of
rhetoric. To support this claim I argue that in each of his conversations
with Gorgias, Polus, and Callicles, Socrates distinguishes good from bad
rhetoric where the former aims at the good of an interlocutor’s soul, and
the latter is only a repertoire of techniques for persuasion, which serve
whatever aim a rhetorician may have. To accomplish my task I will argue
the following for each of Socrates’ three conversations: (a) that Socrates
uses rhetoric, (b) that he thinks his rhetoric to be an art, and (c) he believes
that his use of this art aims to improve the souls of his interlocutors. A
result of my interpretation is that Socrates’ words and deeds “chime in
harmony together” (Laches 193d-e). The art he talks about is the art he
uses.
A number of interpreters believe that the Gorgias constitutes a condem-
nation of all rhetoric, and it is one of my aims to show that this position is
mistaken 1 Martha Nussbaum says the following in The Fragility of Good-
ness. 2
_________
1 Many commentators believe that Socrates condemns rhetoric. See among others,
Thompson 1894, xiv-xviii; Zeller 1876, 190; Pater 1920, ch. 4; Rendall 1977, 165; Cushman
1958, 231; Jaeger 1943, 1851. Kaufman 1979, 115-125 states that Plato’s condemnation of
58 HENRY TELOH

The last part of the dialogue [Phaedrus] breaks with the Gorgias’ general
condemnation of rhetoric, describing a ‘true’ rhetorical art in which a central
place is given to the knowledge, through experience, of the souls of individu-
als. (268a-b)
Presumably Nussbaum would agree that Socrates states or implies the
“general condemnation of rhetoric” in the Gorg. If we conjoin Nuss-
baum’s claim with my arguments that Socrates uses rhetoric, and knows
that he uses rhetoric, then Socrates is a hypocrite who condemns what he
uses. Socrates is no better than any other relativistic rhetorician who tries
to outmuscle his opponents. 3 Can we save Socrates from this charge? I
think we can. My distinction between good rhetoric (the art of rhetoric)
and bad rhetoric (no art at all) will exonerate Socrates from the charge of
hypocrisy; Socrates practices the art of rhetoric which aims at the care of
the soul, while he condemns a rhetoric which is only concerned with tech-
niques for persuasion. 4 Moreover, we can explain why Socrates implies
praise of one type of rhetoric with Gorgias, namely, the real art, and con-
demns another with Polus, namely, the empirical knack (%") which
only consists of the techniques for persuasion (see Gorg. 450b-c, 452e). 5

_________
rhetoric is conditional, but he also believes that the rhetoric Socrates practices is defective.
Kaufman says that dialectic and rhetoric are complementary, but he does not say what this
means. Black 1958, 365-369 argues that Socrates only attacks Gorgias’ view of rhetoric; a
position with which I agree. He goes on to assert that the same view of rhetoric is found in
both the Gorgias and the Phaedrus, a view which I deny. Among those who realize that the
Gorg. is mostly rhetoric are Lewis 1986, 195-210, and Tarrant 2002, 61-78. Roochnik
1995, 82 denies that there is a sharp distinction between rhetoric and philosophy, and that
Socrates is ambiguous in his hostility to rhetoric. As we will see, I disagree with the latter
claim. A. Spitzer 1975, 15-21 holds the position closest to mine. She distinguishes true
from false rhetoric, and correctly sees that Socrates is the true rhetorician. Moreoever, she
sees that Socrates addresses different characters differently in order to care for their souls.
She does not, however, analyze Socrates’ use of rhetoric or address the issue of technê.
Stauffer 2006, 80, 133 mentions once that Socrates uses rhetoric on Polus, and he mentions
that Socrates uses rhetorical punishment on Callicles, but neither of these claims is devel-
oped, and on p. 160 he states that Socrates is unwilling to take up a new good rhetoric.
Stauffer is too concerned with his claim that Socrates wants Gorgias to embrace the new
rhetoric, to see clearly that Socrates does it.
2 Nussbaum 1979, 227.
3 The connection between rhetoric and relativism is the following: the best way to per-
suade the people is to tell them what they want, but what they want changes from context to
context.
4 I say that Socrates aims at care because it is clear to him that he faces an uphill battle
which he well might lose. The reason he might lose is that his interlocutors are corrupted by
society, and filled with imperialistic desires for power.
5 This is in opposition to Roochnik’s position. See Roochnik 1995, 87 and 1996, 189.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 59

One implication of my position is that Socrates is radical about how we

should understand art. A real art of rhetoric must aim at the good of the
soul, just as a real art of cobbling must aim at the good of the feet. Any-
thing which is a real art must aim at its proper teleological end. 6 Thus
Socrates aims to legislate a new, ethically saturated, account of what art is,
one which emphasizes the good ends to be achieved. 7
I believe, but do not argue for, the claim that in the Gorg. Socrates in-
tentionally blurs the distinction between rhetoric and dialectic. Two crite-
ria recommend themselves to distinguish these activities: first, length and,
second, the type of assent. Dialectical questions are short while rhetorical
speeches are long; moreover, one must assent with belief to a dialectical
question or else there cannot be a genuine conversion, while rhetoric does
not require assent at all. Both criteria are vague, and neither criterion
works. Sometimes Callicles assents with belief to a long speech (493c,
494a), and at other times he does not assent with belief to a short question
(495a-b, also see Republic 349b). Sometimes a long speech is a question
(459c-460a), and at other times short questions are parts of long speeches
(507a-509c). Plato mixes the types of speeches to which we give catego-
ries, and there is no clear and sharp divide between rhetorical and non-
rhetorical performances. More generally I do not believe that we should
approach Plato’s text with a preconceived scheme about how to classify
texts: for example, some sections consist of refutation, others amount to
dialectic, and others contain myths which exhort or threaten, and banter
that fills in background information, shaping the drama of the dialogue.
Furthermore, some characters such as Polus and Callicles make rhetorical
speeches. I believe that these distinctions are not entirely useful because
refutation, dialectic, myths, and drama overlap one another. Also, Socra-
tes seldom does just one thing in a passage, and even “background” banter
educates, persuades, horrifies, satirizes, and so on. Of course, on the dif-
ferent ends of a spectrum some texts are clear examples of rhetoric or refu-
tation, and for these texts there are sufficient conditions, but many texts
_________
6 This also is how Socrates distinguishes good from bad art in the Republic. A real work
of art, like the Rep. itself, images the divine paradigms, and a bad one does not. This is why
the Rep., itself, is likened to a work of art on numerous occasions.
7 It is not noted enough how revolutionary Socrates is; he often tries to change the very
meaning of key terms. For example, “nature” does not refer to birth, but to what is innate
within us, and “teach” does not mean put in, but pull out. “Practice” does not refer to
memorization, but to dialectic, and “want” must refer to the good. Socrates uses many dif-
ferent methods to forge these new connections, including the claim that we already believe
them!
60 HENRY TELOH

occupy a blurry middle position. This means that my task of arguing for
Socrates’ use of rhetoric is all the more difficult, since neither necessary
nor sufficient conditions for the application of this concept need be avail-
able for all contexts.
In the second section of this essay I need to defend my claim that Socra-
tes practices an art. Many years ago I argued for what I call the Phaedrus
principle—the principle that Socrates tailors his refutations to the particu-
lar character of each respondent. 8 I continue to endorse this claim. An
element of the Phdr. principle is that Socrates aims to test the lives of his
interlocutors by questioning their core beliefs—that is, beliefs which guide
and organize their lives (e.g., Gorgias’ belief about what rhetoric is, La-
ches’ belief about courage, and so on). 9 Though Socrates’ tests take myr-
iad forms, the point of them all is to tell whether an interlocutor can de-
fend her belief. If she cannot, then she is not an expert to consult, ideally
she will come to see that she is not, and this humbling experience should
leave her with a desire to seek wisdom.
The Phdr. principle, though, may seem to have a certain implication
which is more contentious and which I have come to see only in more re-
cent years. Namely, since each of Socrates’ claims is designed for a spe-
cific situation involving a particular interlocutor, we cannot rightly draw
inferences from these claims about what Socrates believes, much less what
Plato’s views are. 10 Socrates might make claims which he himself does
not believe—and if he does, his ends do justify his means—because if an
interlocutor fails to defend a belief successfully, even if the counter argu-
ment is unsound or invalid, that interlocutor still is shown to be ignorant.
Socrates has absconded; we cannot know what, if anything, he believes,
because we cannot know when he endorses a claim. 11
_________
8 See Teloh 1986.
9 See Talisse 2002, 51. Talisse says, “Rather, the elenchos is a method of challenging
an interlocutor’s ‘claim to know’.” Also see Brickhouse and Smith 1994, 12-14.
10 Brickhouse and Smith 1994, 14-16 assert that Socrates need not endorse all of the
premises he uses in elenctic arguments.
11 Socrates’ claim of only human wisdom in the Apology rebukes both our, and the ju-
rors, pretensions to a wisdom greater than human, and the personified laws in the Crito
flatten Crito’s certainty that Socrates should escape. Each claim is designed for a specific
situation, and we cannot infer from them about what Socrates, the personified laws, or even
Plato believes. These are the sorts of claims commentators who emphasize the drama of the
dialogues like to assert. See Roochnik 1995, 90 and 1996, 181, 194. Roslyn Weiss 2001,
53-57 makes a similar point for the recollection passage of the Meno. Socrates does not
literally believe in recollection but, rather, uses the theory to exhort Meno to seek for wis-
dom. Also see Gary Alan Scott 2000, 17-21 who argues that Socrates’ denial that he is a
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 61

David Roochnik argues for a similar claim about Socrates’ employment

of the technê analogy in the Gorg. Roochnik’s claim is that Socrates uses
the technê analogy to refute and to exhort an interlocutor. 12 For Roochnik,
Socrates argues that rhetoric is, for example, like medicine, and medicine
has a particular external aim, so also rhetoric should have a particular ex-
ternal aim. This analogy forces Gorgias to find a particular external aim
for rhetoric and to abandon the view that rhetoric is just a group of tech-
niques for persuasion and that this persuasion can be about anything. And
once Gorgias tries to say what specific thing rhetoric is about, he gets him-
self into trouble. Roochnik also claims that Socrates holds out technê as
an ideal to strive for, but that Socrates himself does not practice a technê,
since he engages in conversations which are at a particular time and mo-
ment, and which address a particular character. 13 Moreover, Roochnik
asserts, Socrates does not believe that any component of care for the
soul—rhetoric, refutation, and so on—can be a technê because, if they
were, their object would be the soul, but the soul is too indeterminate to be
a proper object of technê. 14
I will argue that Socrates believes in a technê of care for the soul, and
that this art has rhetoric as a component. Socrates uses rhetoric in the
Gorg. to address souls which are best addressed in this way; his aim is to
improve these souls through his use of rhetoric. Moreover, Socrates wants
to transform prevailing conceptions of technê by infusing them with an
ethical aim. A necessary condition of a real art is that it aim at the good.
Thus, in the Gorg. he focuses on what rhetoric aims at, but pays little ex-
plicit attention to rhetorical technique. 15

Let us start with Socrates’ interaction with Gorgias. Socrates asks Gorgias
a series of questions which are, I will argue, an example of rhetoric. I
should quote the passage in full:

_________
teacher must be “heard in light of the wider political and legal objectives he might be sup-
posed to have….”
12 Roochnik 1995, 88-89; 1996, 189-190, 192.
13 Roochnik 1996, 193, 194, 201, 209, 211.
14 Roochnik 1996, 209; 1995, 91.
15 Likewise in the Phdr. Socrates brushes away the niceties of the rhetoricians. See
Phdr. 267bff.
62 HENRY TELOH

Whether the orator is or is not a match for the rest of them by reason of that
skill, is a question we shall look into presently, . . . let us consider first
whether the rhetorician is in the same relation to what is just and unjust, base
and noble, good and bad as to what is healthful, and the various objects of all
the other arts; he does not know what is really good or bad, noble or base, just
or unjust, but he has devised a persuasion to deal with all of these matters so
as to appear to those who like himself, do not know, to know better than he
who knows. Or is it necessary to know, and must anyone who intends to
learn rhetoric have a previous knowledge of these things when he comes to
you? Or if not, are you, as the teacher of rhetoric, to teach the person who
comes to you knowing nothing about them—for it is not your business—but
only to make him appear in the eyes of the multitude to know things of this
sort when he does not know, and to appear to be good when he is not?… For
Heaven’s sake, as you proposed just now, draw aside the veil and tell us what
really is the function of rhetoric.
Gorgias: Why, I suppose, Socrates, if he happens not to know these things he
will learn them from me. (459b-460a)
Prior to this admission Gorgias pushes the position that rhetoric is only a
repertoire of techniques for persuasion (453a, 459b-c), that the rhetorician
can overpower all other arts by learning just the single art of rhetoric
(452e, 459c), and that a rhetorician need not be concerned with how her
students turn out, because it is not the rhetorician’s purpose to transmit
aretê to her students (456e-457a). These claims form Gorgias’ preferred
position about the practice of rhetoric because Gorgias himself freely
states them in order to exhibit his wares to his potential students. Gorgias
is always giving exhibitions, and here he exhibits what his real view about
rhetoric and its power is.16 But the above-quoted questions coerce Gorgias
to a different conception of rhetoric, one which is ethically saturated, and
which is incompatible with Gorgias’ preferred view.
Why is Socrates’ series of questions an example of rhetoric? The ideal
for dialectic is that questions and answers are brief (449b-c), and that the
questioner should ask only one question at a time (466c-d). But Socrates
asks many questions at once, and he talks at length as if giving a speech.
Second, Socrates’ phrasing leaves Gorgias with little choice but to give the
reply he gives; for who would prefer the apparent good over the good?
Third, Socrates’ questions appeal to Gorgias’ sense of embarrassment:
potential students are observing Gorgias, and he would lose face before
them unless he replies that he will teach a student who lacks it, aretê.
_________
16 The Gorg. opens with three mentions of displays or exhibitions, and Socrates’ first
request is to find out from Gorgias what the power (dunamis) of his art is, thus focusing
Gorgias on the crucial, and for Gorgias, fatal, issues.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 63

Gorgias’ bragging prepares him for this fall; since rhetoric, he claims, is
the supremely powerful art, how can it avoid ethical regulation?
Socrates continues to use rhetoric in other parts of his conversational
exchange with Gorgias. Against Gorgias’ technique conception of rheto-
ric he employs the following analogy: as medicine is to health, so rhetoric
is to X (450aff). 17 Socrates’ manipulation of this analogy runs all the way
from 449a to 454b. But the analogy Socrates chooses already defeats
Gorgias’ account of rhetoric. For just as health is external to medicine so,
too, X is external to rhetoric. Thus, whenever Gorgias tries to stick with
his preferred position that rhetoric comprises only techniques for persua-
sion, Socrates always asks what this persuasion is about. He chooses a
loaded analogy, and then manipulates Gorgias to the preconceived view
that rhetoric must have an external end. 18 Socrates is not a dispassionate
inquirer who asks simple questions and receives uncoerced answers but,
rather, a passionate manipulator who persuasively coerces his interlocutors
into assenting to certain views.
Gorgias comes to conclude that rhetoric is about right and wrong things
($  7 0$, 454b) in the law courts and in other public meetings.
And this brings us to the third point at which Socrates uses rhetoric. Gor-
gias may mean no more than what it is prudentially right or wrong to do in
some public circumstance. But near the end of their discussion, Socrates
asks whether Gorgias agreed earlier that “rhetoric dealt with speech, not
on even and odd, but on the just and unjust” (460e), and Gorgias says he
did. Socrates continues:
Well then, I supposed at the time when you were saying this that rhetoric
could never be an unjust thing, since the speeches it made were always about
justice….
Gorgias agrees, and his concession turns into a full-blown ethical use of
dikaios. But where does this meaning come from? It comes from Socra-
tes. He slips it into the conversation, and the fact that he manipulates the

_________
17 In addition to medicine Socrates also uses analogies with gymnastics, and even ge-
ometry and calculation (450d). The object of all these analogies is to persuade Gorgias that
rhetoric must have an object since medicine, gymnastics, and calculation have objects
(451a-c). Although the medical analogy is one among several, it is the medical analogy
which keeps recurring throughout the dialogue (see 452a-b, 456a-b, 460b, 504b-c, 514d-e,
521e).
18 It is not just the analogy which refutes Gorgias, but both the analogy and the end
Gorgias picks for rhetoric. But, as we shall see, it is always the end which refutes Gorgias,
Polus and Callicles.
64 HENRY TELOH

meaning of this term is not uncharacteristic of a general topos which he

employs. 19
Socrates’ last two uses of rhetoric when conjoined together, completely
trounce Gorgias. The analogy forces Gorgias to concede an external end
for rhetoric, and Socrates manipulates the end Gorgias stumbles upon into
a full-blown ethical conception of the rhetorical art. Socrates concludes
the conversation with Gorgias by stating that “we agree once more that it
is impossible for the rhetorician to use his rhetoric unjustly or consent to
do wrong” but that to determine the truth will take some time (461a-b).
Why does Socrates use so much rhetoric on Gorgias? The answers lie in
the character of Gorgias, and the circumstances of their conversation.
Gorgias’ self-image is that of an inventor of rhetorical techniques, a
teacher of them and of set speeches; he is wedded to this identity. His
posture is that of one who gives authoritative exhibitions; when the con-
versation begins he has just completed an exhibition (447a), and he hopes
to give another with Socrates (447b). Moreover, Gorgias is arrogant; he
can answer any question (447c), and nobody has asked him anything new
in many years (448a). Gorgias also has a crowd of listeners, the young
men who are trying to decide whether to sign on with him, and he wants to
impress this crowd, particularly with the power of his art. What is Socra-
tes to do in this context? Arguments to convert Gorgias likely are not
available; so in this circumstance Socrates employs rhetoric on Gorgias.
After all, what would impress a professional rhetorician and his potential
students more than to be beaten at their own game? But the end Socrates
has in view is to push Gorgias to a better conception of rhetoric and, in the
process, improve Gorgias’ soul. For it is essential that a teacher of rheto-
ric thematize what she is doing, precisely because rhetoric is so powerful.
Gorgias seems even to take a genuine interest in the remainder of the con-
versation, and to wonder at what Socrates means (463a, 463d) or at least
he acts to facilitate the conversation (497b). Perhaps he even begins to see
new possibilities for rhetoric beyond the raw exercise of power, although
we do not know if he does. Nor is there any evidence that Gorgias is con-
verted to philosophy and, indeed, how could he be since the real Gorgias is
not a philosopher. What we do know is that Socrates leaves Gorgias with
a question at 461a-b about two conceptions of rhetoric, while at the same
time he lobbies for the ethically saturated conception. I conclude that Soc-
rates practices a real art of rhetoric on Gorgias while at the same time he
proselytizes for the same view.
_________
19 I will cite a second example with Polus.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 65

Socrates uses rhetoric with Polus, too. When given the option to ask or
to answer questions, Polus chooses the former (462b), but he does not un-
derstand Socrates’ responses to Socrates’ satisfaction (463d), so Socrates
explains his meaning with a long speech. This speech extends from 464b
to 466a, and in it Socrates develops an extended analogy. Toward the end
of his speech Socrates provides a justification for his ‘unusual’ procedure:
“For when I spoke briefly you did not understand me; you were unable to
make any use of the answer I gave you, but required a full exposition”
(465e).
In his speech Socrates develops the distinction between empirical
knacks (%") and arts (). First he distinguishes the body from
the soul, and then he distinguishes the good condition of each from what
appears to be good but is not. Arts aim at the former and knacks at the
latter. 20 As the arts part of the analogy goes, medicine and gymnastics are
to the body as justice and legislation are to the soul. These aim at the good
of the body and soul respectively, and the former member of each pair
attempts to restore the body or soul to good functioning, while the latter is
an excellence of the active exercise of body or soul. As the knacks part of
the analogy goes, cookery and cosmetics are to the body as rhetoric and
sophistry are to the soul. Each of these appears to be good for the body or
soul, but is not really so. Moreover, cookery provides a false appearance
of medicine, cosmetics of gymnastics, rhetoric of justice, and sophistry of
legislation.
Socrates develops his analogy in a long speech, and in his speech he
condemns rhetoric as a mere knack. But is Socrates not condemning what
he himself does? For certainly a long persuasive speech counts as rhetoric
if anything does. Is Socrates a hypocrite? I think he is not. Notice, first,
that whether something counts as an art or, instead, a knack, depends on
whether it aims at the good of its end. Arts do, and knacks do not. But
even cooking, for example, can be an art by being aimed at the correct
end; e.g., masking restorative medicine, or providing healthy food. So can
rhetoric: it admits of good and bad manifestations.
What would a good form of rhetoric be? Socrates pairs rhetoric with
justice as its semblance, and this pairing is an important clue about the

_________
20 If one aims at just pleasure, then one might hit what is good, but it would be an acci-
dent, and not the result of contrivance. If one aims at the good, but misses it, then this is a
failure of contrivance which does admit of some sort of explanation. Socrates provides us
with such an explanation for why he cannot educate Callicles; Callicles’ love of demos
makes him not quite believe Socrates (513c-d).
66 HENRY TELOH

nature of good rhetoric. Just as justice aims to restore a sick soul to health
so, too, good rhetoric has the same function. 21 I contend that it is this
view of good rhetoric which governs the discussions in the Gorg. And it
is this conception of rhetoric which Socrates uses on Polus. Polus is brash,
arrogant, and conceited because of his belief in his own knowledge (462a).
He desires liberty to say as much as he wants (461d), but he has no ethics
of responsibility for what he says. Finally Socrates emphasizes both be-
fore and at the end of his speech that Polus does not understand Socrates’
answers. In this context Socrates gives a speech to try and make Polus
understand. The necessity for this speech is shown by the breakdown in
dialectical discussion which precedes it (462b-463a). I submit that Socra-
tes practices the art of rhetoric on Polus.
At a second critical point Socrates employs rhetoric on Polus, and he
again manipulates what words mean. Polus believes that the rhetori-
cian/tyrant is the most powerful person because she can do whatever she
wants. But Socrates responds that the tyrant is powerless because she does
not know what good to aim at, and thus she does not do what she really
wants (466c-e). Polus responds that, nevertheless, the tyrant does what
seems best to her (467b), but Socrates denies that what seems best to
someone is what that person really wants. Polus exclaims that this is a
shocking, monstrous answer (467b), and it is; for normally our best crite-
rion for what people really want is what they say they want.
Socrates means to legislate new meanings for key terms like ‘power,’
‘good,’ and ‘want.’ Ways of legislating new meanings by means of sound
argument are frequently unavailable, so Socrates uses rhetorical strategies:
for example, he refuses to consent to standard connections between words,
and he offers persuasive definitions of words. An example of the latter
occurs when he connects ‘want’ with ‘the real good’ where the latter is
detached from demotic beliefs, and remains obscure and undefined.
Socrates’ rhetorical maneuvers aim to pry Polus away from a cluster of
debased beliefs: that the power to commit injustice is what is good, that
the life of a tyrant is what we should emulate, and that no special care is
needed to discern what is good. To remove these beliefs would help to
produce health in Polus’ soul.

_________
21 Rhetoric is remedial because it can, at most, generate true beliefs in a soul. Only dia-
lectic can tie down these beliefs and thus transform them into understanding (see Meno
98a). Thus Callicles’ lack of “knowledge,” good will, and frankness might be ameliorated
by a Socratic rhetorical beating, but such a beating would not make Callicles wise; at best it
would prepare him to do dialectic.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 67

Polus has already assented to the claims that doing wrong is more harm-
ful than suffering it, and that escaping punishment is worse than suffering
it, when Socrates hits him with two separate speeches (480b-481b). What
Socrates says in these speeches follows, he insists, from Polus’ previous
assents. The emphasis on what follows from what, and what Polus previ-
ously assented to, harkens back to Polus’ interpretation of what Socrates
does with Gorgias: Socrates, Polus asserts, led Gorgias into some inconsis-
tency, and this is why Socrates is rustic (0 

, 461c). Polus does not
value consistency, and Socrates rubs his face in it (480b-e). Consistency
is a minimal condition for rationality, and Socrates forces Polus to be con-
sistent. Polus, Socrates assents, is committed to “the great use of rhetoric”
(481b), which is to ensure that oneself, one’s relatives, and one’s friends
who commit wrong receive punishment, and that one’s enemies who do
the same are never punished. Socrates inverts the demotic use of rhetoric.
Polus is incredulous (480e), and Callicles believes that Socrates is jesting
(481b). But to what purpose does Socrates put this inversion of Gorgian
rhetoric? I contend that these paradoxes are meant to stun Polus, and to
get him to reevaluate what he believes and does.
Socrates extensively uses rhetoric with Callicles, too, and a reason for
this opens their conversation. Callicles wants to know whether Socrates is
serious about the “great use of rhetoric;” for if he is, all of life will be
turned upside down (481b-e). We are to decide who is serious, Socrates or
Callicles, and the choice is between polar opposites. Each of them has two
loves: Socrates loves Alcibiades and philosophy, and Callicles loves De-
mus and the city (demos, 481d). Although each has two loves, they treat
their loves in opposite ways: Socrates always says the same things about
the same things, and Callicles never is consistent (481d-482b, 490a-491b).
So how is Socrates to converse with someone with whom he shares noth-
ing in common? Arguments will not faze Callicles for he will shift his
position, refuse to answer, and so on. Nevertheless, Socrates can beat Cal-
licles; he can defeat him in the war of words, thus overturning Callicles’
self-image as the strong and superior person. Callicles will suffer a prag-
matic defeat, an inconsistency between his words and his deeds, and such
a defeat will influence others not to imitate him. 22 Only if Callicles’ self-
image is shaken will there be even a remote chance that he will reevaluate
his life. Callicles recognizes how the attack will proceed; he states that
Socrates is a demagogue ($" 
, 482c).

_________
22 I am indebted to Ken Casey for this point.
68 HENRY TELOH

Socrates ironically praises Callicles for his knowledge, good will, and
frankness (487a). The praise is ironic because Callicles exhibits none of
these character traits, each of which is necessary for a good dialectical
discussion. By ‘knowledge’ Socrates does not mean anything technical;
someone must think about a position enough to give it a good defense.
Without good will a conversation degenerates into a fight, and without
frankness one won’t test one’s real position, since that position is not put
forth for investigation. Callicles lacks all three traits, and so Socrates em-
ploys rhetoric.
Callicles’ position proves hopelessly incoherent. On the one hand he
claims to be stronger by nature than the many, on the other he caters to the
whims of the city. Socrates collapses the nature/convention distinction by
showing that the many are the stronger (489a-b). Callicles loses his pur-
ported wisdom.
Callicles asserts that the strong person should let her desires grow as
great as possible, and then satisfy them. Socrates counters with the notion
of self-rule. He hits Callicles with two images, the soul as a sieve and as a
leaky jar. After each story Socrates asks “Do I persuade you?” (493c,
494a), and Callicles says “No.” These images are instances of rhetoric
since they are coercive speeches. Then Socrates, in what is an overt agôn,
employs reductios on Callicles: anyone who maintains that all pleasures
are good must approve of the life of a catamite (494e). This conclusion is
a shocking attempt at coercive persuasion.
The remainder of the discussion first alternates between Socrates’ ques-
tions to Callicles and his preaching to him, with the preaching rapidly
gaining the ascendancy. The reasons for this ascendancy are not hard to
discern: at 489b, e, 494d, 495b, 497b, 501c, 505c, d, 506c, 510a, and 522e
Callicles either insults Socrates or refuses to answer him or both. Callicles
has no good will and frankness. Socrates hits Callicles with one long
speech after another, culminating with his encomium on, and exhortation
to, cosmic harmony at 507a-509c. Socrates admits that he is a dema-
gogue, because Callicles will not answer (519d-e). Callicles’ defeat is
total; by his silence he is forced from the field. In conclusion Socrates
gives a very long logos/mythos from 523a to 527e which returns a threat
for a threat: Callicles threatens Socrates with malicious accusations in this
life, and Socrates threatens Callicles with condemnation in the next, when
our naked souls are judged by the just gods.
At 503a-b Socrates announces that there is a noble rhetoric which en-
deavors to make souls as good as possible. Then at 503e he says that the
noble rhetorician is just like any other craftsman who has the purpose of
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 69

giving a form (8$

) to what he is working upon. Socrates again empha-
sizes the medical analogy (504a-b): just as a physician brings order
("
) to the body so, too, the good rhetorician engenders order and
regularity (') in the soul (504d). 503aff is itself an instance of Socra-
tes’ unrelenting use of rhetoric on Callicles: by this rhetoric Socrates at-
tempts to purge Callicles’ belief in his natural superiority. I conclude that
Socrates practices the medicinal rhetoric he describes at 503eff. And to
further confirm this conclusion, Socrates uncharacteristically claims that “I
think I am one of the few, not to say the only one, in Athens who attempts
the true art of statesmanship (K  
: L) and the only man
of the present time who manages affairs of state” (521d). Socrates at-
tempts the true art, but he realizes that the love of the city shapes Callicles’
desires (513c-d) and that he is unlikely in a short time, or at all, to reshape
them.
I will now turn to observations which apply to all three discussions.
Technê erupts in each conversation. Socrates confronts Gorgias with two
types of rhetoric: one, a set of techniques for persuasion, and the other, an
art that aims at the good of the soul. With Polus Socrates overtly distin-
guishes knacks from arts, and with Callicles he declares that not all rheto-
ric is base mob oratory, for there is a noble rhetoric which is like the other
arts that aim to improve their subjects. I conclude that Socrates does not
just practice rhetoric, he also thinks of the rhetoric he practices as an art.
He is unconcerned with technique; he does not discuss it even if he illus-
trates rhetorical technique. His main focus is on the end to be achieved.
Why does Socrates use so much rhetoric in the Gorgias? Each conver-
sation exhibits the character and intellectual defects of its interlocutor.
Gorgias is fixated on his view of rhetoric, and on his own mastery of it and
of other people; Polus is brash, uncomprehending, and does not listen; and
Callicles believes in his own natural superiority. Arguments are fre-
quently inefficacious to ameliorate such defects, so Socrates turns to rheto-
ric. His aim is to engender psychic health, and it is not unreasonable to
connect what he attempts with each interlocutor as a movement toward
health. If Gorgias were to entertain a better view of rhetoric, then his de-
sire for power would ameliorate; if Polus came to value consistency, then
he could reconsider his desire for power; and if Callicles lost his self-
image as the naturally better person, then his unjust desires for self-
aggrandizement would abate.
Why does Socrates think that rhetoric is medicinal? I believe that he
holds a sharper distinction than is justified between what dialectic and
what rhetoric can achieve, even though dramatically these two activities
70 HENRY TELOH

tend to merge. Dialectic is the method to tie down beliefs rationally; it

tests claims one by one, and forces a thesis holder to respond to all known
objections. But even if a belief is tied down, one should not be dogmatic,
because new objections can occur. Rhetoric finds its purpose by instilling
beliefs through exhortation, harangue, praise, story, and so on. These be-
liefs may become habituated even if they are unjustified. Rhetoric, in
other words, has the capacity to train. Character training must precede
dialectic, but proper character training presupposes an understanding of
the soul, and its excellence. Thus good rhetoric presupposes philosophical
understanding of virtue, 23 and philosophical understanding on a large scale
presupposes a civic project of proper training. 24
Socrates’ use of rhetoric may connect with a certain model of human ac-
tion. On one model of desire all desires aim at the overall good, and only
ignorance makes us miss the mark. On the second model only reason
looks out for the overall good, and spirit and appetite have their own re-
spective goods. Which model does Socrates presuppose in the Gorg? Not
exactly either model. Socrates’ images of the soul as a leaky jar and a
sieve—his conception of an ordered soul—and his division of the soul into
an ordering and ordered part, suggest the distinction between rational and
non-rational desires. Perhaps what Socrates is getting at is that dialectic
appeals to rational desires, and those who are rationally dominated, while
rhetoric appeals to non-rational desires and those who are non-rationally
dominated. 25
I conclude that rhetoric is one among a large repertoire of moves which
a philosopher/teacher can choose from in attempting the true political art
of education.
But why do I claim that Socrates is rhetorical? Let us gather up the
moves which he makes. First he is a speech maker, and he makes
speeches to all three of his interlocutors. Then Socrates employs manipu-
lative analogies, persuasive definitions of words, paradoxical and unde-
fended conclusions, humiliating reductions to absurdity, and even prem-
ises he may not subscribe to in his ad hominem attacks. Let us add to this
list Socrates’ use of images and threats. None of these moves is a neces-
sary condition to conclude that Socrates is rhetorical, but one or more of
these tactics is a sufficient condition. Socrates uses a large repertoire of

_________
23
This is the main theme of the Phdr.
24
This is the major theme of the Rep.
25
Moline 1978, 1-26 argues that the parts of the soul in Rep. IV are agents, and that rea-
son uses different types of appeals to control spirit and appetite.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 71

non-rational modes of manipulation, not unlike what the rhetoricians do,

but what is common to these modes is that he aims them at the care of the
psychê.

II

I argue that Socrates extensively uses rhetoric in the Gorg. Moreover, I

claim that Socrates would recognize his use of rhetoric to be an art. But
Socrates does not, I believe, have any short list of techniques or methods
for psychic care because he adjusts his methods to the condition of his
interlocutor. Also, in the Gorg. Socrates does not, so far as we know,
convert his interlocutors, if by “convert” we mean “turn them to philoso-
phy.” One might believe that the last two claims are not compatible with
the practice of an art, for how could an art involve so much experimental
conjecture and be so unsuccessful? Finally, commentators lodge serious
arguments against Socrates practicing an art.
Brickhouse and Smith argue that Socrates does not have a method or a
craft. 26 To develop their argument they list several criteria for a craft.
First, craftsmen exhibit a rationality that results in regular and orderly pro-
ductions; they do not act by conjecture. Second, a craft is teachable and
learnable. Third, craftsmen are inerrant—a craftsman “does not err in his
work or in his judgments about the subject matter of his expertise;….”
Fourth, a craft has a “distinct subject matter.” Finally, a craftsman has a
knowledge or wisdom, and it is in virtue of this knowledge that the
craftsman is inerrant, and can teach his craft. Brickhouse and Smith have
several other criteria, but these are the main ones, and they claim that Soc-
rates’ elenchos meets none of them. By extension, rhetoric would not be
an art either, nor would any other method of care for the soul.
David Roochnik develops, in addition to his adherence to the objection
above, a second set of objections against Socrates practicing an art. First,
medicine aims at the value-neutral end of health, while the good condition
of the soul is value-laden. But arts are not value-laden. 27 Second, an art
permits a gap between theory and practice, but Socrates believes that if
one knows the good, then one will aim at it. 28 Third, the soul is not a de-

_________
26 Brickhouse and Smith 1994, 3-10.
27 Roochnik 1996, 195.
28 Roochnik 1996, 125.
72 HENRY TELOH

terminate end like a ship; a soul is self-conscious, and always in motion.29

Finally, at Gorg. 503eff, where Socrates says that good rhetoric is an art,
he cannot mean what he says, because the human soul is pictured as pas-
sive, like the materials for a ship. But surely, the human soul is not pas-
sive in its own education. 30
I will reply to the most important of these arguments seriatim but, first, I
have three observations. Socrates does not theorize about his methodol-
ogy. He is unconcerned about enumerating or regimenting his techniques;
rather, his eye is always on the end to be achieved. Second, Socrates does
do something: he cares for the souls of his interlocutors, and how he cares
for them is evidence for his method. Last, let us develop the analogy
‘medicine is to health’ as ‘rhetoric (a part of care for the soul) is to X.’
Does the physician have exact knowledge so that she can bring about in-
evitably healthy results? We all could wish so much! Do physicians op-
erate by sure and certain knowledge? So far as I can tell they often oper-
ate more by conjecture and hypothesis. They try out and test various op-
tions in order to see what happens.
Is medicine easily teachable to someone else? If it were, then there
would be no need for a lengthy period of apprenticeship. Is health a value-
neutral concept? I think not, and later I will show that Socrates does not
draw the fact/value distinction as we do, especially with respect to the no-
tion of cosmos whether in the body, soul, or ouranos.
The main analogy Socrates uses in the Gorg. is the medical analogy, al-
though he also refers to cobblers, shipbuilders, and the like. 31 But given
the centrality and frequency of the medical analogy, it is much more likely
that Socrates thinks he practices a medicine for the soul. And since medi-
cine is a paradigm case of an art so, too, is what Socrates practices an art.
Once we clarify the medical analogy, there is a response to Brickhouse
and Smith. Just as a physician does not have sure knowledge which re-
sults in inerrant ends so, too, Socrates does not have sure knowledge about
how to care for the soul. And just as a doctor will experiment with many
techniques to heal the body so, too, will Socrates attempt numerous tech-
niques in therapeia of the soul. Finally, just as a physician may lose most
or all of her patients and still be a doctor who practices the art of medicine
so, too, Socrates will improve only a small number of his patients, and still
he practices an art of care. Imagine a nineteenth century physician who
_________
29 Roochnik 1996, 125.
30 Roochnik 1996, 207; on this point also see Hall 1971, 203-204.
31 See note 17.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 73

loses most of his patients who suffer from yellow fever, and still she is a
physician. Socrates, I submit, faces an analogous situation when he at-
tempts to cure the self-aggrandizing desires of those who inhabit imperial
Athens (cf. Gorg. 516aff). 32
But someone might object that a reasonable success rate is necessary to
the practice of a technê, and Socrates certainly does not convert any of his
interlocutors in the Gorg. to the pursuit of philosophy. But consider the
following example: a group of M.I.T. mathematicians works all of their
lives on solving some very difficult equation, and they fail. Do they still
not practice the technê of mathematics? I believe that this example shows
that success is not a necessary condition for the practice of a technê. But
the example also shows something else: certain conditions may obtain
where we would be surprised if a technê did succeed, and I claim that Soc-
rates faces just such conditions in imperial Athens, a place where human
desires run amok.
When we look at the so-called early dialogues there is no unambiguous
evidence that Socrates improves anyone except the young. He does seem
to use successful protreptic with Lysias in the Lysias and Cleinias in the
Euthydemus, but the other interlocutors, who are older, are beset with a
litany of character defects. In fact part of the mimetic educational impor-
tance of the so-called early dialogues is to see what is wrong with these
interlocutors and for the reader or audience to avoid these defects.
Now the aim of the rhetorical technê is to restore psychic health on the
model of justice. This means that Gorgias, Polis, and Callicles must relin-
quish their self-image and desire structures, and replace them with new
images and structures as a means to psychic health. Rhetoric must bring
about the means to its end, just as medicine must produce the means to
health. But just as it is hard to produce bodily health in a polluted envi-
ronment so, too, it is hard to engender psychic health in a socially corrupt
environment, and imperial Athens is a socially corrupt environment. So
an explanation for the desperation of Socrates’ attempt, and for his failure,
_________
32 Roochnik 1996 makes a distinction between technê 1 and technê 2, where the former
is like what Brickhouse and Smith say technê is, and the latter is more like medicine as I
describe it. But Roochnik insists that Socrates employs a technê 1 in his refutations, and
that it is this strict kind of technê that he exhorts others to (see 198). But why is Roochnik
so sure that Socrates always employs a strict technê? Socrates is generally unconcerned
about the epistemology and techniques of what he does, and most concerned about the end
in sight. Perhaps Socrates’ refutation of Thrasymachus would suggest a strict technê, but it
is Thrasymachus who commits himself to the idea of a strict ruler even though Socrates
finds this position strange (see Rep. 340 d-e).
74 HENRY TELOH

is not hard to find: it lies in the imperialistic desire structure of Gorgias,

Polus, and Callicles; i.e., in their love of demos.
Roochnik argues that bodily health is value-neutral while psychic health
is not. But the object of an art should be value-neutral, he claims. Thus a
care for the soul cannot be an art like medicine. But I deny that Socrates
employs a fact/value distinction which puts these two healths on opposite
sides of that divide. At Gorg. 507e-508a Socrates says, “And wise men
tell us, Callicles, that heaven and earth and gods and men are held together
by communion and friendship, by orderliness, temperance, and justice;
and that is the reason, my friend, why they call the whole of this world by
the name of order, not of disorder or dissoluteness.” Order ("
) is a
basic category of the whole cosmos, and it is rather like a fusion of fact
and value. We might say that it is an ultimate value which Socrates takes
as a fact! This position he inherits from some pre-Socratics, for whom
justice or order is both a fundamental moral and an empirical category of
the cosmos. 33 Now order is not just found in the cosmos, it is also in the
soul, the body, and the community. Of course the last three of these can
become disordered, and when they do, we should try to reestablish order.
Furthermore, Roochnik’s position ignores how revolutionary Socrates
is. Socrates wants arts to become ethical; or to put it a different way, only
what is ethical counts for him as an art. I take this to be the main point of
the Gorg.: a real art aims at the good of its subject. But how could this
possibly work for cobbling, even if it works for psychic care? A real cob-
bler makes not just a shoe, but a good shoe, one which helps a specific
foot to accomplish what it needs to accomplish. We are right at a funda-
mental characteristic of Platonism: only a good X is a real X: for example,
only a good painting—one which serves the proper purposes of art—is a
real painting. This sense of good fuses both fact and value as well as pru-
dential and ethical phenomena.
Roochnik claims that Socrates does not practice an art because the soul
is not a determinate end. The word “determinate” is unfortunate because
of its imprecision, but apparently Roochnik means that as self-conscious
beings we mutate ourselves so as to defy study. Roochnik says, “He [Cal-
licles] is erotic, spirited, ever in motion toward conquest. He is alive, and
his desires are unleashed. He cannot be refuted by Socrates, and thereby
shows how, in some sense at least, the soul is indeterminate.” 34 All of this
_________
33 Think of Anaximander’s single fragment which makes ethical rightness a category of
the cosmos.
34 Roochnik 1996, 209.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 75

is true, but Callicles has a character, a set of executive habits for action
and belief, and this character is both what Plato depicts and Socrates ad-
dresses. In fact, if Callicles’ character were not at all reliable and predict-
able, then there could be no drama to the dialogue; for events would hap-
pen at random, and neither Callicles’ nor Socrates’ moves would be intel-
ligible.
Roochnik must disarm the implications of Gorg. 503eff; for Socrates
says there that the true rhetorician is H  (504d). His argument is
that Socrates treats matter in these passages as passive, but no person can
be passive in their ethical education. He focuses on Socrates’ example of
a shipwright who gives a certain form to the materials he works upon. The
shipwright forces one part to fit and to agree with another until a cosmos is
achieved (503e-504a). Moreover, the true rhetorician engenders justice in
a soul (504d-e). But just after the shipwright example, Socrates talks
about doctors and trainers bringing order into a body (504a). Would their
patients also be passive? I think not. Certainly a physical trainer engen-
ders order in a body by prescribing a regimen and so, also, does a doctor.
But the subject is active in following their prescriptions. Talk about en-
gendering order does not imply that the matter is passive. Nevertheless,
why would Socrates talk about engendering justice rather than drawing
answers out from within his interlocutor? Two aspects of the dialogue
context explain this. First, the topic is rhetoric, and rhetoric is remedial
like medicine; it attempts to restore a soul to health. Second, Socrates is
besting Callicles with rhetoric, and the recipient of a rhetorical beating
often is passive. So while 503eff does not imply passivity, the language of
the passage sometimes suggests it.
Finally, Roochnik argues that Socrates practices a non-technical wis-
dom, one which is caught in inevitable particularity and kairos—the criti-
cal particularized moment. But, then, are not most arts caught in particu-
larity? Think again, of medicine as an example, or even cobbling, but not
mathematics.
What are the consequences of Roochnik’s position? Socrates uses the
technê analogy and the strict knowledge involved in an art to refute others
and to exhort them to seek such wisdom, Roochnik maintains. Socrates
proffers art and its strict knowledge as an ideal for others to pursue. 35 But

_________
35 See Roochnik 1996, 191-194. What ideal does Roochnik attribute to Socrates? He
formulates it as that of rigorous knowledge, the determinate, a strict technê, and a vision of
order and harmony. All of these formulations refer to identical or closely related phenom-
ena. A strict technê would employ rigorous knowledge about a determinate object, and this
76 HENRY TELOH

why would Socrates mislead others by exhorting them to seek for what he
does not believe in? Moreover, on Roochnik’s own account Socrates
could not do philosophy without it being particularized with respect to a
specific interlocutor, with all her unique characteristics, a specific topic,
time, location, and so on. Socrates would, then, on Roochnik’s view, urge
his interlocutors to seek for the impossible—a contextless philosophical
conversation. Finally, Socrates again misleads his interlocutors when he
claims to be one of the few in Athens who attempts the true art of ruling
(521d); for he tries to do what he believes cannot be done. Why not, then,
take Socrates at his word that he attempts such an art, but that he might
fail because of the recalcitrant matter of human nature, in this case, Calli-
cles’ love of demos (513c). 36
Roochnik polarizes rhetoric and philosophy. The former is particular-
ized and temporal, and dreams of only the critical moment for victory,
while the latter is a dream of the eternal. The paradigm of the eternal is
mathematics, and Roochnik connects mathematics with strict technê. But
much of Roochnik’s evidence for this connection is from the Rep. where
Forms certainly occur. But are there eternal Forms in the Gorg. or just a
vague appeal to cosmic order? Does Socrates, in the Gorg. dream of rig-
orous knowledge, or of an account which at some dialectical moment
withstands refutation? It is true that Socrates dreams of a vague cosmic
order (507a-509c), but why is this determinate and its instantiations in the
human soul indeterminate? As far as I can tell, Socrates dreams of estab-
lishing an order here in the cosmos; he dreams of this, hopes for this, and
works toward this. In the Gorg. the difference between philosophy and
bad rhetoric lies in their ends, and not in an eternal/temporal divide.
What are the positive reasons to accept my interpretation? First, Socra-
tes’ care for the soul is like a doctor’s care for the body. Each aims at en-
gendering order within their respective subjects. Both Socrates and the
doctor act by trial and error; they both employ many techniques, and their
_________
technê would produce an ordered and predictable result. Roochnik does not believe that any
of these ideals can be brought into existence when we are dealing with the human soul; in
other words, neither rhetoric nor care for the soul can be a technê. For we do not possess
rigorous knowledge about the soul, the soul is not determinate, and we cannot inevitably, or
even for the most part, produce the desired results. Thus Socrates’ dream, which he exhorts
others to have, cannot be implemented; his dream breaks the harmony of word and deed.
36 Roochnik 1995, 91 and 1996, 209 interprets Gorg. 521d as Socrates focusing on an
ideal which, nevertheless, it is impossible for him to realize. Moreover, Roochnik believes
that the passage may be a reductio; since Socrates does not practice an art, nobody does. I
see the passage as a straight out assertion that Socrates practices an art, and that he is prac-
ticing it on Callicles, but not with the success he would like.
 RHETORIC, REFUTATION, AND WHAT SOCRATES BELIEVES 77

respective arts are compatible with numerous failures. Moreover, on my

interpretation, what Socrates does images what he says; the Gorg. is a self-
reflexive drama. Socrates forges an ethics-laden form of pedagogical
rhetoric and, then, he employs it on his interlocutors. His words and his
deeds chime in harmony together, unlike those of Callicles. 37 Finally,
Socrates asserts or implies the same ethics-laden art in each of his conver-
sations, and this increases our confidence that this is his position.38
I have argued that Socrates practices rhetoric, and that this rhetoric is
part of a philosophical art of care for the soul. This art employs a large
repertoire of techniques—think of those Socrates uses with Callicles—and
there are numerous subordinate goals under the general heading of psychic
excellence. These goals include both character and belief changes. For
example, Gorgias must think about what he teaches, and Callicles must
change his desire for conquest. Finally, we do know something important
that Socrates believes. He believes in an art of care, and might this not
turn out to be the human wisdom which in the Apology he claims to pos-
sess? 39

VANDERBILT UNIVERSITY

_________
37 This is a major theme of all the dialogues and especially the Laches. The two gener-
als, Laches and Nicias, cannot make their words and deeds match because they have the
wrong conceptions of what good words and deeds are. See Teloh 1986, 41-56.
38 Indeed, we cannot infer from what Socrates says to what he believes on a statement-
by-statement basis, but when we look at a whole dialogue and its repeated themes, then we
can, I believe, make reasonable inferences to what Socrates believes.
39 I wish to thank Mason Marshall for his significant help on this essay. I also want to
thank Robert Talisse, and Wesley DeMarco for his example of the mathematicians who fail.
 CCOMMENTARY ON TELOH

DAVID ROOCHNIK

ABSTRACT
Teloh argues that in Plato’s Gorgias Socrates speaks about, and “aims to leg-
islate a new, ethically saturated, account of what technê is.” Precisely be-
cause it is ethically saturated, it is also revolutionary. This commentary sug-
gests that Teloh has insufficiently appreciated the ironic aspects of Plato’s
treatement of technê.

Professor Teloh has articulated, and articulated well, an important thread

or strand or line of thought that is present in Plato’s Gorgias. Socrates
speaks about, and “aims to legislate a new, ethically saturated, account of
what technê is” (59). Because it is ethically saturated his conception of
such a technê is “revolutionary” (74). Equipped with this new concept
Socrates endorses the technê of rhetoric. “Socrates’ primary aim in the
Gorgias,” Teloh writes, “is to persuade his interlocutors, and Plato his
readers, that a real technê of rhetoric must aim at the good of the psyche”
(58-59). Not only that, but because his arguments against Gorgias, Polus,
and Callicles aim at the good of their psychai, Socrates himself puts into
practice this genuine technê of rhetoric. The Gorgias thereby exhibits a
harmony of word and deed. Socrates practices what he propounds on a
theoretical level.
On the one hand, Teloh is not wrong in his analysis of the dialogue. The
problem is, from my point of view, that he has over-simplified it. He has
accurately uncovered one of its threads, but ignored the many ways in
which Plato manages to weave together that thread with a quite different
one. In other words, the Gorgias is not a vehicle composed simply to
communicate the single line of thought Teloh attributes to it; namely, that
Socrates both theoretically affirms and puts into practice the true art of
rhetoric. Instead, it is more like a tapestry made from threads of different
colors, some of which clash. The result is a work that often undermines
itself. A technical, but dangerous, word to describe this process of concep-
tual undermining is “irony.” This is what Teloh ignores, for his Plato is a
straightforward and rather earnest moralizer. By focusing on a few of Te-
loh’s claims I will try to show how Plato himself actually undermines
them.
First and foremost is the assertion that in conversing with Gorgias, Po-
lus, and Callicles, Socrates not only endorses but also practices the (value-
saturated) technê of rhetoric, the one which “restores a psyche to health”
 COMMENTARY ON TELOH 79

(65). This seems wrong to me because in each of these three cases Socra-
tes fails. Gorgias and Polus are historical figures; they are professors of
(conventionally understood) rhetoric. Socrates did not convert or trans-
form them, or ameliorate their souls. And he certainly failed with the char-
acter of Callicles. How, then, could Socrates be described as a technitês,
an “expert,” a possessor of a technê?
Teloh anticipates this sort of objection and so he argues that sometimes
a technê, especially medicine (which is the key analogue Socrates em-
ploys), fails: “Does the doctor have exact knowledge so that she can bring
about inevitably healthy results? We all could wish so much!” (72). Teloh
also says “Socrates does not have sure knowledge about how to care for
the psyche” (72), “Socrates will improve only a small number of his pa-
tients” (72), and “both Socrates and the doctor act by trial and error; they
both employ many techniques, and their respective technai are compatible
with numerous failures” (76).
But Teloh is too generously allowing for failure. Yes, the medical
technê is “stochastic;” it is not modelled on mathematics and so is com-
patible with failure. But it can’t fail all the time. Indeed, success has to be
the expected outcome. If I go to the doctor with a medical problem that is,
objectively speaking, curable—e.g,. a broken arm or a mild bacterial in-
fection—I expect it to be cured. If the doctor fails, she’s a bad doctor. She
doesn’t really have the medical technê. When it comes to very complex
diseases, the well trained doctor may fail. Still, they have to be successful
on some sort of regular basis in order to claim a technê. By contrast, Soc-
rates is depicted as an utter failure. Not a soul in the dialogue is improved
at its end.
This comment leads to my next. I think Teloh puts too much emphasis
on the “aim” of a technê. He says, “Whether or not something counts as a
technê or an empeiria depends on whether or not it aims at the good of its
end” (65). Having a good end is obviously not a sufficient criterion of the
sort of technê Socrates describes. A moron can be animated by the good
aim of improving my soul, but unless he has knowledge he surely has no
technê. Socrates says this at 465a: “I do not call [flattery] a technê but an
empeiria because it has no logos of that which it applies and what sort of
things they are with respect to their nature. As a result, it is unable to ar-
ticulate the cause () of each thing. I do not call that which is bereft of
a logos a technê.”
Teloh has put, not incorrectly, a great deal emphasis on the ethical com-
ponent of a genuine technê. But he has said little about the epistemic side
of Socrates’ demand. The true technites must be able to give a logos. A
logos of what? The human soul? Various character types? The virtues?
This is left ambiguous. And its ambiguity is meant, in my reading, to sig-
80 DAVID ROOCHNIK

nal the altogether peculiar nature of what Socrates is propounding. My

complaint against Teloh is that he doesn’t render these Socratic claims
peculiar enough.
Of course, Teloh understands that technê must have a epistemic compo-
nent. He says, “Good rhetoric presupposes a philosophical understanding
of virtue” (69-70). I suppress the question of the relationship between vir-
tue and the psychê, but reiterate my questions: what is this knowledge?
Where exactly is it exhibited in the dialogue? Teloh would have us think
that Socrates practices a technê. I have already mentioned the fact that he
seems only to fail. Now I add he never presents the logos that would seem
to be required of him were he genuinely to warrant the title, technitês, or
“expert.”
Teloh’s strategy seems to be to minimize the epistemic demands re-
quired of a true technê in order to allow Socrates to have it. For example,
he says this: “Are not most technai caught in particularity? Think, again,
of medicine as an example, or even cobbling, but not mathematics” (75).
To reiterate, he says this because he wants Socrates, whom we readers
observe thrust into the particular situation of conversation with a Gorgias
or a Polus, to retain the title technites. But the last three words of his sen-
tence, “but not mathematics,” hit me as wrong. Teloh seems to want to
drive a wedge between mathematics and the properly understood rhetori-
cal technê because the latter requires a fluency with the particulars, while
the former treats only universals and does not tolerate failure. But even in
the Gorgias Socrates uses mathematics as the key example of technê.
When introducing Gorgias to the concept of a technê at 450c-d, he begins
by dividing the technai into two kinds: there is the productive branch,
which includes painting and sculpture, and the branch which is $! 
,
“through logos.” The latter has no ergon, product or result issuing from it.
The examples of this sort of technê, under which rhetoric itself is classi-
fied, are arithmetic, logistic, geometry and draughts. As I have argued
elsewhere, this is altogether typical of how Socrates divides the technai
throughout the Platonic corpus. 1 He regularly bifurcates them, with one
branch being, to use Aristotelian terminology, “productive,” and the other
“theoretical.” And with great consistency, the standard example for the
latter is mathematics. This suggests, as I have argued, that the mathemati-
cal arts have a strong, perhaps the strongest, claim for being a technê ac-
cording to Plato. (And why technê should not be translated as “craft.”)

_________
1 See my Of Art and Wisdom: Plato’s Understanding of Technê (University Park: 1996),
pp. 271-282.
 COMMENTARY ON TELOH 81

This is a complicated argument whose steps I cannot rehearse here. In-

stead, I can only reiterate my basic point about Teloh’s argument. He em-
phasizes the ethical aims of a real technê but tells us next to nothing about
its epistemic content. It seems to me that when attention is paid to this lat-
ter task, Socrates’ own claim to having a technê begins to be undermined.
Socrates offers no logos of an . If he has knowledge it does not re-
semble at all a mathematical art. He does indeed seem to know how to
navigate his way around particular human beings, but, as mentioned, he
regularly fails to improve human souls. Thus, I resist Teloh’s invitation:
“Why not take Socrates at his word that he attempts such an art, but that
he might fail because of the recalcitrant matter of human nature” (75-76). I
remain suspicious of the claim that Socrates possesses and attempts to put
into practice a technê.
In a sense, Teloh himself seems to acknowledge the possible cogency of
my line of interpretation. At the beginning of his paper, he says, I think
correctly, that Socrates tailors his remarks in order to address the character
of a specific interlocutor. Therefore, Teloh continues, “because each claim
[Socrates makes] is designed for a specific situation, we cannot infer from
them about what Socrates … or even Plato believes” (60). This seems
right. So too does the following line: “Socrates has absconded; we cannot
know what, if anything, he believes, because we cannot know when he
endorses a claim” (60). Again, this seems quite right. But Teloh does not
seem to abide by this statement as rigorously as I think he should, for he
also says—and of course this is his main thesis—“Socrates does believe in
a technê of care for the psyche, which involves rhetoric. …He lobbies for a
real technê of rhetoric while he employs this technê on Gorgias, Polus, and
Callicles” (61). What happened to Socrates absconditus, the master iro-
nist? In other words, what is the philosophical significance of the fact
that, as Teloh rightly puts it, we cannot know when Socrates himself en-
dorses a claim?
BOSTON UNIVERSITY
 TELOH/ROOCHNIK BIBLIOGRAPHY

Anton and Kostas (eds.) 1971. Essays in Ancient Greek Philosophy. Albany.
Black, E. 1958. Plato’s View of Rhetoric. The Quarterly Journal of Speech XLIV:
361-374.
Brickhouse, T. and Smith, N. 1994. Plato’s Socrates. New York and Oxford.
_______. 2002. The Socratic ‘Elenchos.’ In Scott (ed.): 145-157.
Carpenter, M. and Polansky, R.M. 2002. Variety of Socratic Elenchi. In Scott (ed.):
89-100.
Cushman, R. 1958. Therapeia. Chapel Hill.
Gonzalez, F. (ed.). 1995. The Third Way: New Directions in Platonic Studies.
Lanham, Maryland.
Hall, R. 1971. Techne and Morality in the Gorgias. In Anton and Kostas (eds.): 202-
218.
Jaeger, W. 1943. The Ideals of Greek Culture. New York.
Kahn, C. 1996. Plato and the Socratic Dialogue. Cambridge.
Kaufman, C. 1979. Enactment as Argument in the Gorgias. Philosophy and Rhetoric
12: 114-129.
Lewis, T.J. 1986. Refutative Rhetoric as True Rhetoric in the Gorgias. Interpretation
14: 195-210.
Moline, J. 1978. Plato on the Complexity of the Soul. Archiv für Geschichte der
Philosophie 60: 1-26.
Nussbaum, M. 1979. The Fragility of Goodness. Cambridge.
Pater, W. 1920. Plato and Platonism. London.
Rendall, S. 1977. Dialogue, Philosophy, and Rhetoric. In Philosophy and Rhetoric
10: 165-179.
Roochnik, D. 1995. Socrates’ Rhetorical Attack on Rhetoric. In Gonzalez (ed.): 81-
94.
Roochnik, D. 1996. Of Art and Wisdom: Plato’s Understanding of Techne.
University Park.
Scott, G.A. 2000. Plato’s Socrates as Educator. Albany.
Scott, G.A. (ed.). Does Socrates Have a Method? University Park.
Spitzer, A. 1975. The Self-Reference of the Gorgias. Philosophy and Rhetoric 8: 1-
22.
Stauffer, D. 2006. The Unity of Plato’s Gorgias. Cambridge.
Talisse, R. 2002. Misunderstanding Socrates. Arion 9: 111-121.
Tarrant, H. 2002. Elenchos and Exetasis. In Scott (ed.): 61-78.
Teloh, H. 1986. Socratic Education in Plato’s Early Dialogues. South Bend.
Thompson, W.H. 1894. The Gorgias of Plato. London.
Weiss, R. 2001. Virtue in the Cave: Moral Inquiry in Plato’s Meno. Oxford.
Zeller, E. 1876. Plato and the Older Academy. London.
 COLLOQUIUM 4
PLATO’S QUESTION OF TRUTH
(VERSUS HEIDEGGER’S DOCTRINES)

FRANCISCO J. GONZALEZ

ABSTRACT
Heidegger’s thesis that truth underwent in Plato a fateful transformation from
unconcealment to correctness is well-known because defended in an essay
that Heidegger published during his lifetime and that has been widely read
since: Plato’s Doctrine of Truth. One goal of the present paper, however, is to
show how Heidegger’s interpretation of the Republic’s Cave Analogy in the
1931 course from which the essay is supposedly derived, as well as his inter-
pretation of the Myth of Er in the Parmenides course delivered shortly after
the essay’s composition, are in tension with, and even undermine, the essay’s
thesis. The other, and more important, goal of my paper is to show that this
richer interpretation offers a fruitful approach to the question of truth in Plato
that is missed by both Heidegger’s detractors and his defenders.

When one thinks of the topic of Plato and Heidegger on truth, one usually
1
thinks of Plato’s Doctrine of Truth, a 1940 essay (first published in 1942)
supposedly deriving from a course Heidegger first gave in 1931/32 and
2
repeated in 1933/34: On the Essence of Truth. Here, as is well known,
Heidegger argues that the essence of truth underwent in Plato a fundamen-
tal and fateful transformation from unconcealment to correctness. What is
not well known is that when Heidegger closely interprets Plato’s texts, as
he arguably does not in the 1940 pamphlet, he discovers a much richer
conception of truth sharply at odds with the mentioned thesis. The com-
plex history of Heidegger’s interpretation of truth in Plato is far beyond
the scope of this or any other paper. 3 I therefore focus on only two key
interpretations that are at odds with Heidegger’s thesis, namely, the inter-
pretation of the Cave Analogy in the 1930’s, and the interpretation of the
_________
1 First published in 1942, though Heidegger tells us that it was “zusammengestellt” in
1940 (1978, 477). Dostal strangely gives 1930/31 as the date the essay was written (1992,
61), presumably on the basis of Heidegger’s claim that the “Gedankengang” goes back to
the 1930/31 (an error for 1931/32) course (Heidegger 1978, 477). Obviously, to say that the
path of thinking pursued in an essay goes back to 1930/31 is not to say that it was written in
1930/31.
2 Heidegger 1988, translated as Heidegger 2002. The 1933/34 version can be found in
Heidegger 2001.
3 I attempt a fairly exhaustive account of this history in my book, Plato and Heidegger:
A Question of Dialogue (Penn State University Press, forthcoming).
84 FRANCISCO J. GONZALEZ

Myth of Er in the 1942 Parmenides course (Heidegger 1982). From the

perspective of these two interpretations, not only the thesis of the essay
Plato’s Doctrine of Truth, but even more Heidegger’s later retraction of
this thesis in favor of the claim that the Greeks always experienced truth
only as correctness, will prove untenable.
But why should Heidegger’s interpretation of truth in Plato, especially
given its contradictions and inadequacies, concern us at all? One reason is
that reflection on Heidegger’s distinction between truth as unconcealment
and truth as correctness can, I will argue, shed much light on Plato’s con-
ception of truth, and therefore also on the understanding of philosophy at
work in the dialogues. Here Heidegger can be defended against the cri-
tiques of Jonathan Barnes (1990), Jan Szaif (1998, especially pp. 145-
147), and Enrico Berti (2005), among others. The other reason is that if
Heidegger’s thesis of a ‘transformation’ proves ultimately untenable, this
is not because Plato’s ‘unsaid doctrine’ was that truth is unconcealment,
but rather because truth in Plato is the problematic, tense, and inseparable
coexistence of unconcealment and correctness and as such is always an
open question. Faced with Heidegger’s own insistence on an opposition
between unconcealment and correctness so fundamental that only the pos-
tulation of a historical transformation in the essence of truth can account
for it and only ‘another beginning’ can redeem it, we can find in Plato a
corrective to such an insistence. In short, both Plato and Heidegger stand
to gain from their Auseinandersetzung on the question of truth.

I. The Essence of Truth in the Cave Analogy: WS 1931/32 and WS 1933/34

A. Truth as the Play of Concealment and Unconcealment in the Cave

In the courses of 1931/32 and 1933/34 Heidegger’s interpretation of the
Cave analogy emphasizes the extent to which it is not a static image but a
story, i.e., the story of prisoners being freed from their bonds, gradually
making the difficult ascent out of the Cave, and then gradually adjusting
their eyes to the light outside the Cave. His account thus divides the anal-
ogy into its different stages and focuses our attention on what is happening
at each stage. But what he sees these progressive stages as exhibiting is a
conception of truth not as correctness and not as a property of assertions,
but as a “property” of being, specifically, being in its unconcealment. Hei-
degger can thus claim in 1931-32 that the central unifying theme of the
analogy is “the true” ( ): “and this has nothing to do with imita-
 PLATO’S QUESTION OF TRUTH 85

tion and correctness and correspondence” (und dabei gibt es nichts von
Angleichung und Richtigkeit und Übereinstimmung; 1988, 30).
Crucial to Heidegger’s reading is Socrates’ claim that when a prisoner is
freed from his chains and forcibly turned around, he will, because blinded
by the light of the fire, believe that the shadows he saw on the wall when
in his chains are truer ( ) than the things now shown to him,
even though the latter are more being ("M
 ) than the former.
Heidegger interprets the comparative  as meaning “more un-
concealed” (unverborgener, 1988, 32). Jonathan Barnes has strongly ob-
jected to this interpretation, asserting that even the “objective sense” of
truth, which he acknowledges, has nothing to do with unconcealment:
In the first place, being true is not at all the same, from the semantical point
of view, as being un-hidden and un-veiled. One can hide the true cows. The
true houris are always veiled. Being true, even in its objective usage, never
signifies being un-hidden. (Being  never signifies being 
).
(192; my translation) 4
This is certainly true of our objective sense of truth as used of cows, etc.,
but misses completely what is distinctive of the Greek conception: as Aris-
totle explicitly claims, those things that are most true ( , Met.
993b28-29) are by nature most manifest (! : - 3 
993b11) even when, due to the limitations of our own vision, they are hid-
den or invisible to us. To return to the Republic, Socrates only a few lines
after the cited passage repeats what appears to be the same claim about the
prisoners who have been turned around, but now in the significantly dif-
ferent form of saying that they will judge what they saw earlier to be “in
reality more manifest” (D  , 515e3-4) than what they are
shown now:  appears to be substituted here for .
Furthermore, what the prisoner judges (mistakenly) to be “more true” are
not statements but the very things he takes to be (again, mistakenly) “more
real”: the shadows on the wall. Clearly the prisoner does not see the shad-
ows as “more true” in the sense of “more correct” since, not only are the
shadows not statements about anything, but from the perspective of the
prisoner they are not even shadows of anything, but rather the only beings
there are.
_________
4 Jan Szaif, in contrast to Barnes, does not rule out an etymological interpretation of
 as unconcealment within the context of the analogy (see especially 319), but does
not see this as the sole interpretation at work there (1998, 145-152). One of his arguments,
however, appears based on a misunderstanding: he suggests that the etymological interpreta-
tion would incorrectly identify  with “being-known” or “knowability” (151); Hei-
degger at least sees “unconcealment” as what makes possible “being-known” or “knowabil-
ity.”
86 FRANCISCO J. GONZALEZ

As Heidegger notes, the comparative “more true” goes hand in hand

with the comparative “more being” (2001, 137-8; see also 1988, 33). The
degrees of truth characterize not statements but degrees of being. The
prisoner considers the shadows “more true” because, compared to the
bright light of the fire that blinds him, they are much more unconcealed
and manifest; but precisely because they are more unconcealed and mani-
fest, they are also for him more real. From the perspective of Socrates or
someone who has emerged from the Cave, in contrast, the fire and the ob-
jects placed before it are, once the eye of one’s soul adjusts to them, more
manifest and more real than the shadows on the wall. Of course the
sunlight and the objects it illuminates outside the Cave are in turn more
manifest and more truly being than the artificial light and objects within
the Cave. It is because the ascent out of the Cave is an ascent both from
what is less to what is more unconcealed, and from what is less to what is
more being, that Heidegger considers unavoidable the conclusion that
“what is at issue in the entire analogy is predominantly ” (1988,
42), where  is understood in what Heidegger takes to be the origi-
nal sense of unconcealment. 5
There is another important dimension to Heidegger’s account of the
Cave analogy as a history. Since the analogy describes the turning of the
prisoner towards what is more true and more real as a freeing of the pris-
oner, Heidegger interprets it as showing that “the happening and existence
of unconcealment as such goes hand in hand with the freeing of man, more
precisely: with the success of this freeing, that is, with genuine being-free”
(1988, 37-38). This first attempt to free the prisoner fails only because, in
merely being turned around towards a light to which his eyes are not ad-
justed, the prisoner is not really freed: he wants to return to the shadows
on the wall and remain in his bonds because he cannot yet recognize the
shadows as shadows nor therefore the bonds as bonds. It is only when the
prisoner is dragged out of the Cave altogether and made to undergo a
gradual habituation () to the light outside that he is genuinely
freed (1988, 41-42). In other words, genuine freedom requires a genuine
dwelling and existing in what is most true and most real. Therefore, what
is at issue in the Cave analogy for Heidegger is not only the essence of
truth but also, and inextricably, “the history of man’s essence” (1988,
114). In case this interpretation appear arbitrary or forced, Heidegger
points out that the analogy is introduced by Socrates as analogous to our
_________
5 But in the margin next to the cited claim Heidegger some time later wrote: “yes and
no” (1988, 42). This is because he will later, and only later, interpret the Cave analogy as
largely eclipsing the original meaning of  in favor of correctness.
 PLATO’S QUESTION OF TRUTH 87

nature (4 4" -) with respect to $ and $: in other

words, our nature not as something static, but as something characterized
by a history occurring between the poles of $ and $.
As Heidegger well recognizes, the Cave analogy describes not only an
ascent out of the Cave but also a descent back into the Cave. But it is not
immediately clear how the account of the descent can fit into Heidegger’s
interpretation. If the ascent is an ascent from what is less unconcealed to
what is most unconcealed as well as from the bonds of a prisoner to the
most genuine freedom, then it seems that with the conclusion of the ascent
everything there is to say about truth as unconcealment and freedom has
already been said, so that the account of the descent would be nothing but
an inessential add-on. But this is precisely what Heidegger denies when he
maintains that the descent is a necessary stage in the process of becoming
free, that it is in fact the completion of this process. “Freedom is neither
merely the being-freed from the fetters nor also only having-become-free
for the light, but rather genuine being-free is being-a-liberator from the
darkness” (Befreier-sein aus dem Dunkel, 91). In other words, genuine
being-free is a continual setting-free, not only of others but presumably
also of oneself. Freedom is not a stable possession but a continual strug-
gle. The reason is that truth
is not a calm possession [ruhender Besitz] in the enjoyment of which we
come to rest at some particular standpoint in order from there to lecture at the
rest of humanity, but rather unconcealment happens (geschieht) only in the
history (Geschichte) of constant setting-free. (91; see also 2001, 184)
But that both freedom and truth must be continuously conquered presup-
poses something else: that concealment belongs to the very essence of un-
concealment.
Therefore, truth is not so simply the being-manifest of beings [Offenbarkeit
von Seiendem], whereby the previous hiddenness [Verdecktheit] would be
left behind somewhere, but rather is necessarily and in itself the overcoming
of a concealment [Überwindung einer Verbergung]; concealment belongs to
unconcealment in its very essence [wesensmäßig]—as the valley belongs to
the mountain. (90)
In this way, what is at issue in the description of the descent is still truth as
unconcealment, but now the privative in un-concealment is emphasized
whereas it is ignored in the identification of truth with what is simply
manifest. 6
_________
6 As Heidegger states the point in the 1933/34 version of the course, to the essence of
truth belongs untruth (2001, 187). Truth is therefore a battle: “every standing in the truth is
confrontation [Auseinandersetzung], a fighting [Kämpfen]” (185). In this way Heidegger
88 FRANCISCO J. GONZALEZ

B. Truth as “Correctness” in the Cave

What is important in the present context is noting that in Heidegger’s in-
terpretation of the Cave analogy in the 1931-32 and 1933-34 courses the
conception of truth at issue in the description of both the ascent and the
descent is nothing but truth as unconcealment. What then of the transfor-
mation of truth from unconcealment to correctness? The reason why Hei-
degger turns to a reading of the Cave analogy is, in his own words, to fol-
low (nachgehen) this transformation (1988, 17). It is therefore surprising
that Heidegger’s reading does little or nothing to demonstrate this sup-
posed transformation. Only in Socrates’ claim that the freed prisoner,
when turned towards what is more, will see (
 (515d) does Hei-
degger find a conception of truth as correctness. Yet Heidegger insists
that, far from replacing or transforming truth as unconcealment, the cor-
rectness introduced here is made completely dependent on truth as uncon-
cealment. In the 1931/32 course, Heidegger interprets the use of the word
(
 as showing only how truth as correctness is grounded in truth
as unconcealment (1988, 34-35). In 1933/34, he elaborates:
Emergence of correctness in connection with unconcealment. The correct-
ness of seeing and regarding [Die Richtigkeit des Sehens und Besehens] is
grounded in the actual turning-towards and proximity of being [in der
jeweiligen Zuwendung und Nähe des Seins], in the manner and way in which
beings are manifest and unconcealed [offenbar und unverborgen]. Truth as
correctness is impossible without truth as unconcealment. (2001, 138)
This grounding is what Heidegger here takes Plato’s sentence to express
and not some transformation of the one conception of truth into the other.
One can and should, however, go even further than Heidegger here and
say that the word (
 is in the cited passage so grounded in uncon-
cealment that it appears to mean no more than “turned towards what is
more unconcealed.” In other words, the comparative here clearly cannot
mean “greater correspondence”: the word (
 describes not a
greater correspondence to the same objects but a turn to a completely dif-
ferent set of objects. The prisoner sees (
 not because he now sees
the shadows more accurately, but because he has completely turned away
from the shadows towards objects that are more real and more uncon-
cealed. Therefore, the comparative in (
 can only mean a relation
to a greater degree of being and unconcealment. But then one must ask if
there is a conception of truth as correctness here at all. One could interpret
_________
highlights the elements of struggle, pain, and even mortal danger to be found in Plato’s
analogy. That this account of the descent suppresses, or in the case of 1933/34, perverts its
political dimension is a problem that cannot be pursued here. See Gonzalez 2003.
 PLATO’S QUESTION OF TRUTH 89

“seeing (
” as meaning “seeing more clearly” as long as one
specifies that the source of the greater clarity is not to be found in some
property of seeing itself (better eyes, for example) but rather in the inher-
ently greater clarity of the objects to which seeing is now turned. To see
(
 is not to improve one’s vision of the same objects, but rather to
turn one’s vision to completely different objects that are more real and
more inherently manifest. While truth may here characterize seeing, the
truth of seeing is immediately dependent on, and derived from, the truth of
being and is as far as possible from expressing any kind of correctness or
correspondence.
Yet even if Heidegger is wrong to find a conception of truth as correct-
ness expressed in the word (
, the important point in the present
context is that even he does not see in the cited passage, in either the
course of 1931-32 or the course of 1933/34, a transformation of the es-
7
sence of truth. Even if truth is there understood as correctness, it is made
completely derivative of truth as unconcealment. But where, then, does
Heidegger attempt to demonstrate a transformation of the essence of truth
in Plato? In the 1931/32 course, after providing his interpretation of the
ascent, Heidegger does suggest that Plato does not understand unconceal-
ment radically (ursprünglich) enough because concealment (as distinct
from what is merely false or apparent) is not understood radically (ur-
sprünglich) enough (93). But he does not pursue this objection, claiming
that we must first attempt to understand Plato’s Idea of the Good and thus
turn to a different analogy: the Sun. It is on the basis of a certain interpre-
tation (and, as I argue elsewhere, misinterpretation) of that analogy that
Heidegger can argue, though only with some effort and a fair amount of
self-contradiction, that Plato did not think truth as unconcealment. Fur-
thermore, in the courses from the 1930’s Heidegger must turn to the
Theaetetus to show that Plato thought untruth as falsehood rather than as
concealment and therefore in opposition to truth understood as correct-
ness. 8

_________
7 A marginal notation to the text of WS 1931/32 interprets ( as follows: “(
(rectus - Recht): gerade-zu, ohne Umschweife, ohne Umwege, nicht über die Schatten, die
Sache selbst” (1988, 34). “Seeing more directly” is certainly closer to what Socrates means
than is “seeing more correctly.” However, even the interpretation of “seeing more directly”
is problematic, since the prisoners see the relevant objects for the first time; one cannot
strictly say that they already saw these objects indirectly through their shadows since they
did not, and still do not, recognize these shadows as shadows.
8 For a discussion of Heidegger’s reading of the Theaetetus, and of why this dialogue
eventually drops out of his account of Plato’s ‘doctrine of truth,’ see Gonzalez 2007.
90 FRANCISCO J. GONZALEZ

But instead of pursuing these rich and essential detours, I wish here to
stay with the Cave Analogy to see what happens to Heidegger’s interpreta-
tion of it in the 1940 essay: an essay that focuses explicitly only on this
analogy, omits entirely the interpretation of the Theaetetus, and only al-
ludes to the interpretation of the Sun analogy. It will be seen that Heideg-
ger can defend his thesis regarding the transformation in the essence of
truth only by suppressing much of what he himself, a decade earlier, found
in the Cave analogy. As for the argument that Plato did not experience
concealment originally enough and, specifically, did not understand it as
belonging to the essence of truth, that will be belied by Heidegger’s own
interpretation of the Myth of Er in 1942.

II. The 1940 Essay Plato’s Doctrine of Truth: from Unconcealment to

Bildung

With regard to the story of liberation recounted in the Cave analogy, Hei-
degger’s reading in the 1940 essay agrees with his earlier reading to the
extent of acknowledging that a conception of truth as unconcealment is at
work here. Thus, as in the earlier courses, he still interprets the descent
back into the Cave as showing that “truth originally means that which is
wrenched free of concealment [das einer Verborgenheit Abgerungene]”
(129). Heidegger even proceeds to suggest that the Cave analogy would be
impossible without an understanding of truth as un-concealment: “the un-
concealment related to what is concealed (distorted and hidden) [Ver-
borgenes (Verstelltes und Verhülltes), and only it, has an essential relation
to the image of a cave lying underneath the daylight. Where truth is of
another essence and is not unconcealment, or is not at least partially de-
termined by it, there a ‘Cave analogy’ has no hold on the imagination
[keinen Anhalt der Veranschaulichung]” (130). Here Heidegger is simply
repeating a central thesis of his earlier courses.
Yet a crucial departure takes place when Heidegger immediately quali-
fies the cited claim by asserting that despite the experience of truth as un-
concealment in the Cave analogy, “a different essence of truth pushes it-
self forward into the predominant position [in den Vorrang] instead of un-
concealment” (130). This is a much stronger assertion than anything found
in the earlier courses. There, as we have seen, though Heidegger claims to
see some indications of an emerging conception of truth as correctness, the
conception of truth as unconcealment remains predominant. Now Heideg-
ger sees the conception of truth as correctness as dominating the concep-
 PLATO’S QUESTION OF TRUTH 91

tion of truth as unconcealment, though this allows him still to acknowl-

edge that the latter remains to some extent present.
This general shift in Heidegger’s reading is facilitated by a specific
change in his interpretation that allows the sense of truth as correctness to
pervade and dominate the very history recounted in the Cave analogy in a
way that it did not before. This change concerns the notion of $, the
very notion which Socrates identifies as what the analogy is supposed to
illustrate. In the 1940 essay, Heidegger first characterizes $ in gen-
eral terms as “this re-accommodation and accommodation [Um- und
Eingewöhnung] of man’s essence to the realm assigned to him at any
given time” (122). But then he proceeds to suggest precisely the transla-
tion of $ which, in 1931/32 (1988, 114) and 1933/34 (2001, 207,
217), he had categorically rejected: “Bildung.” Given that on his reading
of the Cave analogy in the earlier courses it belongs to the essence of a
human being to exist within the unconcealment of beings as a whole, Hei-
degger there interpreted $ as explicitly taking a stand within this
unconcealment, which requires recognizing it as such, in a way that is de-
cisive for one’s essence (1988, 114-115). This is why Heidegger could
proceed to identify $ as thus interpreted with “philosophizing, un-
derstood as questioning one’s way through to being and unconcealment,
i.e., to that which still empowers these themselves” (115). Taking a stand
within unconcealment as such, a stand that must always be a questioning
comportment, is what Heidegger also called freedom. On this interpreta-
tion, then, $ is defined by a conception of truth as unconcealment;
$ is a certain way of existing in unconcealment.
In the 1940 essay, in contrast, the German word Heidegger claims to
come closest to what is named by $ (“am ehesten noch, wenngleich
nicht völlig, genügt”) is none other than “Bildung” (123). Here, of course,
he does not understand this word in its nineteenth century meaning, but
rather in a meaning derived from its etymology: Bildung is a bilden, a
forming, by means of an “anticipatory measuring-up to a paradigmatic
look [vorgreifenden Anmessung an einem maßgebenden Anblick], which
is therefore called the model [Vor-bild], ‘Bildung’ is at the same time an
impressing [Prägung] and a being-led by an image [ein Bild]” (123). Here
we can see that the difference between the earlier interpretation of $
and the present one goes far beyond the choice of words: in the earlier in-
terpretation, $ is a stand in unconcealment characterized by ques-
tioning. Now $ has become something much more narrow and much
less radical: it means being guided by a certain look or “picture” of the
world, measuring oneself up to this look or “picture” as to a paradigm. If
$ was earlier the process of liberation into ever greater unconceal-
92 FRANCISCO J. GONZALEZ

ment, it is now a turn to paradigmatic objects and conformity to those ob-

jects. Judged against the earlier identification of $ with freedom,
$ as ‘Bildung’ is the negation of such freedom, and this because it is
the absence of any genuine questioning. To be “cultivated” is not to ques-
tion, but only to measure up to a paradigm already given and accepted as
self-evident.
But why this radical change in Heidegger’s interpretation? Why this
transformation of $ from a questioning of being and unconcealment
into mere conformity to some ideal picture or look? The reason is not hard
to see: the current interpretation of $ allows truth as correspondence
and correctness to dominate the Cave analogy in its entirety. The ascent
which the analogy describes thereby becomes an ascent of increasingly
greater adequation to some paradigmatic looks, i.e., the Ideas outside the
Cave. Heidegger has transformed the $ which the Cave analogy is
meant to illustrate into nothing but the correspondence between a looking
(blicken) and a look (Anblick). That this is Heidegger’s strategy is made
clear on p. 124, where he suggests, with some air of mystery, that the
“Bildung” described in the analogy presupposes a fundamental transfor-
mation in the essence of truth.
But apart from Heidegger’s need to introduce truth as correctness into
the Cave analogy, what recommends his interpretation of $? Appar-
ently nothing in Plato’s text; on the contrary, Socrates’s description of the
ascent seems impossible to square with Heidegger’s reading. On p. 128
Heidegger describes $ as “the continual habituation in the fixing of
one’s look on the fixed limits of things that stand fixed in their look [die
stetige Eingewöhnung in das Festmachen des Blickes auf die festen Gren-
zen der in ihrem Aussehen feststehenden Dinge]” (128). Does this corre-
spond to the  (516a5) that takes place outside the Cave? What
Socrates describes is the freed prisoner’s gradual adjustment to the light so
that increasingly more and brighter objects become visible to him in this
light. As was the case also within the Cave, he “sees better” only in the
sense that truer objects gradually become visible to him. Furthermore,
each thing that becomes visible does so only as a step towards something
else becoming visible; the prisoner does not look at the reflections in the
water for their own sake but only on the way to seeing the originals. The
entire process of course culminates in seeing the sun itself as the source of
all light. But here too Socrates does not describe the prisoner as simply
staring at the sun but instead as “reasoning” (
C

, 516b9; also
517c1). So is there anywhere in the  where Socrates describes a
learning of how to fix one’s look fast onto the fixed limits of things stand-
ing fast in their look? Clearly not. The habituation has nothing to do with
 PLATO’S QUESTION OF TRUTH 93

learning how to make one’s vision correspond as accurately and correctly

as possible to fixed objects with fixed limits. There is only a becoming
visible of things previously not visible as one’s eyes adjust, not to the
things themselves, but to the light; there is no question here of seeing the
things more or less ‘correctly.’ This is not to say anything that Heidegger
did not himself see in the1930’s. In the 1931/32 course, he characterizes
the  Socrates describes as a “slow becoming-familiar [langsames
Vertrautwerden]—not so much with the things, but rather with the illumi-
nation [Leuchten] and the brightness [Helle] itself” (1988, 43); and as has
already been noted above, he accordingly characterizes the habituation
outside the Cave as a process of ever greater unconcealment, rather than as
some improvement in the accuracy and fixedness of one’s looking. 9
In conclusion, Heidegger’s reversal of his own earlier interpretation of
$ is not demanded by anything in the text, but only by what he has
already determined to be Plato’s “doctrine of truth.” Of course, if Heideg-
ger later in the essay could support his claim that a conception of truth as
correctness dominates Plato’s analogy, then this might in retrospect pro-
vide some support for his reading of $. But little support is in fact
provided by Heidegger’s later remarks. As pointed out above, in the earlier
courses the only indication of a conception of truth as correctness that
Heidegger detected in the text of the Cave analogy itself was the descrip-
tion of the prisoners as seeing (
 when freed from their chains and
turned around towards the fire (515d3-4). There, however, he rightly in-
sisted that a conception of truth as unconcealment is still dominant here,
since the prisoners see (
 only in being turned towards beings that
are more unconcealed. In sharp contrast, Heidegger’s reading of the same
passage in the 1940 essay, by quoting it out of context and only after as-
serting his central thesis, can maintain: “Everything depends on the
(, on the correctness of looking [Richtigkeit des Blickens]” (136).
Here truth as correctness has become the dominant and even exclusive
conception of truth expressed in the word (
 and with absolutely
10
no textual support.
_________
9 On p. 135 of the 1940 essay, Heidegger identifies $ with “making man free and
firm [frei und fest] for the clear constancy of the view of the essence [die klare Beständig-
keit des Wesensblickes].” This too is not only a major departure from the earlier identifica-
tion of human freedom with unconcealment, but is also untenable as an interpretation of the
text. What is described as taking place outside the Cave is not some clear constancy of a
look, but a fluid process of every greater unconcealment culminating in a process of reason-
ing.
10 Strangely, Hyland, despite his critique of Heidegger’s reading, grants Heidegger
much more than he should when he writes: “Without question, Plato does have his Socrates
94 FRANCISCO J. GONZALEZ

III. The 1942 Interpretation of  in the Myth of Er (Republic Book 10)

A. The Roman versus the Greek Conception of Truth

But what of the argument, made already in 1931/32, that Plato, by seeing
falsehood, rather than concealment, as the opposite of truth, could not help
but understand truth as correctness? As already suggested, this thesis is
undercut by Heidegger’s interpretation of the myth of Er. But here we can
add that it is also undercut by the extensive reflection on the nature of
&@$
 and  with which he prefaces his interpretation. I cannot do
justice here to Heidegger’s discussion in its entirety; instead I will draw
attention only to those points most relevant to the present context. Rela-
tively early in the course Heidegger raises the problem that the opposite of
 in Greek is not , but &@$
. Yet he immediately (Heideg-
ger, 1982, 39-40) proceeds to warn against drawing from this the hasty
conclusion that therefore truth could not have been understood by the
Greeks as unconcealment (Unverborgenheit). He argues that if the essence
of truth is for the Greeks in some way determined by  &@$
 as its op-
posite, “on the other hand, presumably  &@$
, the false, nevertheless
in its essence always remains a form of what is concealed [des Ver-
borgenen] and of concealing [des Verbergens]” (33). It is precisely this
dimension of concealment that Heidegger’s account of  &@$
 empha-
sizes. Taking the modern and Greek-derived word “pseudonym” as a clue
and arguing that this word does not mean “false name” but rather a name
that both conceals and reveals the author in a certain way (43-4), Heideg-
ger concludes that &@$
 means a concealing that is also a letting-appear.
The &@$
 belongs within the essence [Wesensbereich] of covering-up [des
Verdeckens], and thus to a form of concealing [einer Art des Verbergens].
The covering-up that pervades the &@$
 is, however, always at the same
time a revealing [Enthüllen] and showing [Zeigen] and bringing-to-
appearance [Zum-erscheinen-bringen]. (45)
This account of  &@$
 stresses its character as semblance (Scheinen),
i.e., as a concealing that is also an unconcealing and vice versa, in other
words, as a peculiar belonging-together of concealment and unconceal-
ment. Furthermore, Heidegger’s aim in the present course is to show pre-
cisely how  &@$
 belongs together with  understood as uncon-
cealment. From a passage in Homer Heidegger concludes that “N@$
 is
a dis-torting concealing [Verbergen], ‘hiding’ [‘Verhehlen’] in the narrow
_________
use the term ‘correctness’ (orthotes) several times in the cave analogy . . .” (2004, 61). Sev-
eral times? I find only one instance. Furthermore, as Heidegger himself points out in the
course, orthotes does not simply mean, if at all, “correctness.”
 PLATO’S QUESTION OF TRUTH 95

sense” (48) and sees this characterization as explaining how &@$

 can be
the opposite of . Furthermore, he uses a passage from Hesiod to
suggest that being &$ is grounded in being , and thus in un-
concealing (48). It seems, therefore, that, rather than, as in 1931-32, inter-
preting  &@$
 as uprooted from its original ground of concealment and
thus as uprooting  from the original experience of unconcealment,
Heidegger here stresses the extent to which  &@$
 still means con-
cealment and thus remains grounded in  as unconcealment.
There is a reason for this shift: the Romans. In contrast to what he
claimed in 1931/32, Heidegger now sees the translation of  &@$
 as
“the false” as a Roman interpretation that for the first time and definitively
uproots it from the original experience of concealment and semblance.
“False” comes from falsum, fallo, : bringing to a fall, rendering
unstable [wankend machen], toppling (57). In understanding untruth from
the perspective of making fall, the Romans are understanding it from the
perspective of imperium (58-60), of what stands upright and commands.
Untruth is thus experienced as a trick that trips up, undermines, makes fall.
But in this conception of “the false,” any relation to concealment has been
lost.
The Greek &@$
 as what hides [das Verhehlende] and as what from this
perspective is also ‘deceptive’ [‘Täuschende’] is now no longer experienced
and interpreted from out of concealing [Verbergen], but instead from out of
tricking [Hintergehen]. (61)
Heidegger nevertheless still maintains that the transformation in the es-
sence of  begins with Plato, but now adds that it took place “above
all [vor allem] through the thinking of Aristotle” (72). The location of the
transformation primarily in Aristotle seems new and surprising. Yet the
reason is that Heidegger now locates the transformation in the word
H"
, indeed even asserts that this word becomes “as it were the au-
thoritative [maßgebende] ‘representation’ of ” (73); and only in
Aristotle can one find even an implicit association of this word with the
11
conception of truth. Heidegger interprets H"
 as meaning “the un-
covering [entbergende] corresponding [Entsprechen] that asserts
[ausspricht] the unconcealed [das Unverborgene]” (72). In this way, and in
this way only, can he see the Greek and Roman conceptions of truth as
displaying a fundamental kinship despite the great difference already indi-
cated. For what he can now claim to be the same in both cases is an under-
_________
11 Ast’s lexicon (1908) lists only three occurrences in Plato of the word H"
, which
Ast translates as similitudo: Epinomis 990d, Republic 454c, and Theaetetus 176b. None of
these passages have anything to do with truth.
96 FRANCISCO J. GONZALEZ

standing of truth in terms of correspondence, even if in the case of the

Greeks “this correspondence [Ensprechung] still maintains itself and car-
ries itself out fully within the space of the essence [Wesensraum] of
 as unconcealment” (72), while in the case of the Romans it ceases
to do so. Specifically, Heidegger describes the kinship as follows:
The Greek H"
 as uncovering [entbergende] correspondence [Entspre-
chung] and the Roman rectitudo as measuring-oneself-against [Sichrichten
nach] . . . both have the character of an approximation [Angleichung] of the
assertion [Aussage] and thinking [des Denkens] to the state-of-affairs that lies
before and stays put [an den vorliegenden und feststehenden Sachverhalt].
Approximation [Angleichung] means adequatio. (73)
It is only by means of this claim that Heidegger can proceed to make the
conception of truth as adaequatio intellectus ad rem, the adequation of the
intellect to the thing, “metaphysics’ concept of truth” and thus a concept of
truth already present in the beginning of metaphysics with Plato. What is
lost in the transformation of H"
 into rectitudo is the essential realm
of uncovering [Entbergung] (73), but the “adequation” remains the same.
This thesis is merely asserted here and it quickly crumbles with the fol-
lowing reflections:
1) Of the apparently only three occurrences of the word H"
 in
Plato, not a single one has anything to do with truth. One of these occur-
rences is Plato’s use of the word in the Theaetetus to describe our relation
to god, where the word clearly does not mean “correspondence” but “as-
similation” and “becoming-like.” It expresses not a “matching” between
two things, but a transformation and appropriation. To claim therefore
(171) that Plato transforms the essence of truth into H"
 understood
as adaequatio seems preposterous.
2) Even in Aristotle there is no explicit identification of truth with
H"
. Bonitz (1961) finds only one occurrence of the word H"

in Aristotle (7 O 826b34) and in a context that has nothing to do
with truth. Of course, there is the famous passage of de Interpretatione
where Aristotle describes “the passions of the soul” (! " <
&<) as “likenesses of the things themselves” (H"
3" 
 ", 16a7-8). First, however, it is clear from the context that this
H"
 is not truth, but rather a prior condition of truth; truth itself is to
be found only in combination (-) and division ($). Sec-
ondly, Heidegger himself elsewhere, in a course delivered in 1925/26,
rightly insists that the H"
3" at issue in this passage have nothing to
do with “some kind of Angleichung of a state of the soul to a physical
thing—something which is absurd . . .” (1995, 167). And what Heidegger
is arguing in the context surrounding his reading of de Interpretatione is
 PLATO’S QUESTION OF TRUTH 97

precisely that “the Aristotelian concept of truth and in general the Greek
concept of truth are neither to be oriented towards a copying [Abbilden]
nor are to be understood at all in the sense of this kind of correspondence
[Übereinstimmung], but rather are to be oriented in comprehension to-
wards unconcealing and concealing [im Verständnis auf Entdecken und
Verdecken hierauf zu orientieren]” (162-3). In the face of this earlier char-
acterization of H"
 as a letting-be-seen and unconcealing that cannot
be understood as a copying and corresponding, 12 it is certainly hard to see
how and why Heidegger in 1942 understands Aristotelian H"
 as a
correspondence that corresponds by way of asserting.
3) Given the radical transformation which Heidegger himself sees as
separating the Roman from the Greek conception of truth, it is a priori
implausible that rectitudo would preserve the Greek H"
 largely un-
altered. Indeed, if H"
 still maintains itself completely within the
realm of unconcealment, then it must be, as Heidegger himself earlier
maintained, an unconcealing rather than some “adequation” between two
things. Truth becomes adequation only when the original experience of
unconcealment is lost. It is as if Heidegger, having shifted the major trans-
formation of the essence of truth to the Romans, can preserve his original
thesis regarding the transformation of the essence of truth in Plato and
Aristotle only by himself Latinizing the Greeks.
Heidegger in any case grants much more here to the Greek experience
of truth and untruth than he does elsewhere. Not only does he provide, as
we have seen, a much richer interpretation of &@$
 that stresses its char-
acter as concealment, but he also attributes to the Greeks the experience of
many other forms of concealment (92-92) which are left unnamed by
them, not through neglect, but because “they are so essential” (95). Yet the
Greeks do name, according to Heidegger, one type of concealing [Verber-
gung] that, unlike the &@$
, cannot be identified with distortion [Ver-
stellung] and that “is constantly present in Greek existence, but which in
its essence is not further considered, unless the word itself which names
_________
12 In a note Heidegger suggests that “H"
 is spoken from out of the perspective of
oQv: a letting-see approximates [gleicht sich an] in the only way that makes sense
in the case of such a comportment [in der Weise wie das einzig bei einem solchen Verhalten
Sinn hat]: in v ” (1995, n. 2, p. 167). H"
 is here a letting-see, an unconcealing of
what is unconcealed. Citing as an illustration of H"
 Aristotle’s claim at 19a33 that

 are true ( +) in a way similar to (H"
) the way in which things ( ") are
true, Heidegger translates/interprets: “
 (namely, the demonstrative [aufweisende] let-
ting-be-seen of beings) are unconcealing (entdeckend) in the same way beings are, insofar
as they are unconcealed [entdeckt]” (n. 1, p. 167). Obviously, the conception of truth Hei-
degger here sees at work in H"
 is truth as unconcealment.
98 FRANCISCO J. GONZALEZ

this concealing and its sphere already held for the Greeks sufficient revela-
tions” (95). This word is .
If the Greeks name , this does not mean that they say much about
it. Yet Heidegger defends the Greek silence about  as appropriate.
After all, would not speaking about this ultimate concealment betray it? If
the Greeks did not express the essential relations between  and
 in the way we must, this is because: “The Greeks are very silent
when we reflect on what is essential to them [ihr Wesenhaftes]. When,
however, they do say this, they do so in a way that is at the same time
characterized by silence [in einer zugleich verschweigenden Weise]”
(116). The Greek silence regarding  is thus not a sign that they did not
experience it in a profound and original way, but quite the opposite.

B. Saying  in the Myth of Er

It in any case turns out that the Greeks did not refrain entirely from think-
ing and saying . This saying, however, did not take the form of a sci-
entific treatise asserting propositions about : such a form of discourse
would be completely inappropriate to what is neither an object nor a being
standing in unconcealment, but rather withdrawal into concealment. The
only appropriate discourse would be one that unconceals in such a way as
also to conceal, as to allow what it unconceals its proper concealment; a
discourse, in other words, that in its saying maintains silence. This dis-
course is myth. According to Heidegger,
‘the mythical’—"@
-character [-hafte] is the unconcealing and concealing
[Entbergen und Verbergen] sheltered in the unconcealing-concealing word
[im entbergend-verbergenden Wort], as which unconcealing and concealing
the fundamental essence [das Grundwesen] of Being itself originally appears.
(104)
That  should come to word in a myth is therefore certainly not a limi-
tation betraying a primitive and unscientific way of thinking. Myth and
 go together, so that an interpretation of saying as mere assertion nec-
essarily renders the experience of  inaccessible behind a conception
of untruth as mere incorrectness.
But where does Heidegger find this myth of ? The answer should
shock: not in the early Greek poets, not in the Presocratics, but in Plato.
This is shocking for two related reasons. First, according to Heidegger we
have in Plato the transformation of the essence of truth and untruth from
unconcealment and concealment to correctness and incorrectness. This
transformation is a move away from an understanding of &@$
 as con-
cealment and a fortiori from the experience of that concealment more
 PLATO’S QUESTION OF TRUTH 99

original than &@$

 and named by the word . How very strange,
then, to find in Plato what Heidegger himself characterizes as “a thinking
of [ein Denken an], not only a thinking ‘about’ [ein Denken ‘über’] ”
(190). Secondly, if in Plato truth becomes correctness, this is at least partly
because, according to Heidegger, with Plato and Aristotle “the word be-
comes 
 in the sense of assertion” (113). How strange, then, to find in
13
Plato 
 in the sense of "@
!
How does Heidegger explain this “anomaly”?
People have already been puzzled about why in general ‘myths’ sometimes
crop up in Plato’s dialogues. The ground lies in the fact that Plato’s thinking
prepares itself for [sich anschickt] giving up the original thinking [das an-
fängliche Denken] in favor of what is later named “metaphysics”; that even
this beginning metaphysical thinking, however, must at the same time pre-
serve a recollection of the original thinking. Therefore, the saying [die Sage].
(145)
Oddly, then, the presence of myths in Plato’s dialogues is taken by Hei-
degger to be only a confirmation that a completely different kind of think-
ing is beginning here; after all, must not traces of the earlier thinking be
found in such a beginning? If, however, we are not convinced by Heideg-
ger’s demonstration elsewhere, as in the 1931/32 course, that we have in
Plato the beginning of the reduction of 
 to assertion and of 
to correctness, then Plato’s myths appear in a very different light. The
question is whether the myths represent an archaizing, anomalous and
marginal moment in Plato or whether they form an indispensable part of
Plato’s central and defining relation to 
 and truth. That the myths
generally do not appear at the beginning of a dialogue but at its conclusion
or climatic turning point, and that, far from being fragments of earlier
myths, they are extraordinarily elaborate and original, certainly suggest the
latter alternative.
In turning to the myth’s content, Heidegger devotes much time to inter-
preting the 
  $"
 to which Er is said to go after his death
(614b9-c1). What especially concerns him is the word $"
, in which
he finds expressed the extra-ordinary (das Un-geheure). The $"
 is
_________
13 And in the myth Heidegger finds an Ent-sprechung that differs significantly from the
adaequatio he otherwise seeks to find in Plato. In speaking of the way in which for the
Greeks myth (das Sagenhafte) must “ent-sprechen dem Gotthaften,” Heidegger writes:
“Dieses Ent-sprechen ist überhaupt das anfängliche Wesen aller Entsprechung (Homologie),
das Wort ‘Entsprechung’ wesentlich wörtlich genommen. Mit der Einsicht in diese
Entsprechung, in der ein Spruch, ein Wort, eine Sage dem Sein ent-spricht, d. h. es als das
Selbe in einem Gleichen sagend entbirgt, sind nun auch wir in den Stand gesetzt, die noch
ausstehende Antwort auf eine früher gestellte Frage zu geben” (169-70).
100 FRANCISCO J. GONZALEZ

the extra-ordinary not in the sense that it goes against the ordinary, but
rather in the sense that “it everywhere encloses what currently offers itself
as the ordinary [das jeweilig Geheure umgibt] and everywhere offers itself
in everything ordinary [in alles Geheure sich dargibt], without, however,
being the ordinary” (150). The extra-ordinary as thus understood is not
“exceptional,” un-natural or super-natural, but rather what is “most natu-
ral” in the sense of - (151). Indeed, it seems that on Heidegger’s read-
ing the 
  $"
 is - itself. If those beings that exist -
are “the ordinary,” then - is that extra-ordinary out of which the ordi-
nary emerges and within which it is enclosed. The effect of such a reading
is to make the place to which Er goes not some remote “beyond,” not
some “other world,” but rather the extra-ordinary and yet all-pervasive and
all-encompassing dimension of this world: the extraordinariness or uncan-
niness of being itself (-).
The $"
 
 is therefore also the place where the unconceal-
ing/concealing play of  especially comes to presence. Thus Hei-
degger observes:
We find it hard to get at this simple essence of the $"
 because we do
not experience the essence of . For the $"
, the self-showing and
indicating [Sichzeigenden, Weisenden], are what they are and are as they are
only in the essential realm [Wesensbereich] of unconcealing [Entbergung]
and of self-unconcealing Being itself [sich entbergenden Seins selbst]. (151)
It is in  as unconcealment that both myth, as the unconcealing-
concealing word (166), and the $"
, as the looking-in and self-
showing extra-ordinary, meet. T is the place for both. The
$"
 
, in short, is extra-ordinary Being in its self-emerging
(-) and self-unconcealing ( ).
It is therefore all the more extraordinary that , the forgetfulness that
forgets itself, the complete withdrawal into concealment, should have its
home in this $"
 
. Not only is  opposed to  but the
myth also makes clear its opposition to -. At 621a3ff. we are told by
Er that the plain of  is “empty of trees and of whatever else the earth
brings forth [ $$  7 9 < -].” U does not allow
anything to emerge (-) from the concealment and sheltering of the
earth; as a holding-back and withdrawing it is opposed to the self-
emergence named -. Thus Heidegger writes: “ appears as the
counter-essence [das Gegenwesen] to -” (176). But this does not
mean that  is the mere negation of - and Being; it is not nothing,
as is made clear by its being located within the $"
 
. As Hei-
degger asserts, “. . . The ‘away’ of what is withdrawn itself comes to pres-
ence [west an] in the prevailing of the withdrawal [im Wesen des
 PLATO’S QUESTION OF TRUTH 101

Entzugs]” (176). Rather than being nothing and nowhere, is not , pre-
cisely as the counter-essence to  and -, what first makes pos-
sible un-concealment and e-merging?
This is not to suggest, however, that  is simply ‘in’ the place of
, as if it were subordinate to and encompassed by . In in-
terpreting the myth’s reference to the “field of ,” Heidegger insists
that  is not in a place, but is itself place. The field or plain is not
something external to ; instead  “is the place-ly [Orthafte] and
the Where, so that the withdrawing concealment [entziehende Verber-
gung] no longer occurs somewhere in a field, but rather for its part unfolds
itself as the Where for what must belong there” (181). Does not indeed
withdrawing concealment shelter and “place” what emerges into uncon-
cealment? Are not, after all, the $"
 
 named T and the
place of  the same place in the myth or, if one prefers, two dimen-
sions of the same place?
Heidegger finds this belonging-together of  and  expressed
with special clarity at 621a7ff. where we are told that the human souls
must drink a certain measure of the water of the river that flows through
the plain of  before they can start on the new lives they have chosen.
The effect of drinking this water is to forget what has been (621b). The
myth does not explain why entering into a new life and thus a new mode of
unconcealment, i.e., a new relation to beings in their being, must be pre-
ceded by this oblivion. Heidegger offers the following explanation:
Every man who measures out upon the earth the journey pregnant with death
[die todesträchtige Fahrt] is on the earth and in the midst of beings in such a
way that on account of this drink a concealing and withdrawing of beings
holds sway [eine Verbergung und ein Entzug des Seienden waltet], so that
beings are only to the extent that at the same time and counter to this conceal-
ing and withdrawing [entgegen dieser Verbergung und dieser Entgängnis] an
unconcealment [Unverborgenheit] holds sway [waltet], in which the uncon-
cealed [das Unverborgene] remains capable of being held and is held [behalt-
bar und behalten]. (178)
If we must drink oblivion and concealment before entering upon life, then
the unconcealment that defines life, that unconcealment within which we
encounter and relate to beings, must occur against and be won from the
prior concealment. That we drink the water which induces  means, as
Heidegger suggests, that it “enters into the human being and determines
him from out of the interior of his essence [aus dem Innern seines Wesens
bestimmt]” (187). But in this case, - in our relation to Being can
occur and hold sway only in a constant struggle with X. Thus Heideg-
ger interprets the need for all human beings to drink a measure of the wa-
ter that flows through the plain of  as showing that “ belongs to
102 FRANCISCO J. GONZALEZ

the essence of ” and that “therefore un-concealment itself cannot

be simply the mere setting-aside [Beseitigung] of concealment” (183).
According to Heidegger in another important aspect of his interpreta-
tion, it is because our relation to the being of beings must be a constant
saving and preserving of unconcealment against concealment that Plato
characterized this relation as " (184). In making this identifica-
tion Heidegger resists a ‘psychological’ interpretation of " as the
psychical act of “remembering” something that has been forgotten. Here it
should be recalled that on Plato’s account all learning, all coming-to-know
the $ of something is ". Not only things that we happen to
have forgotten, but the being of anything whatsoever can be known only in
and through ". This means that every $ is concealed in 
as soon as we are born and therefore can be known only through being
retrieved and saved from this . What is at issue in " is not
the play of psychological states but the strife and belonging-together of
 and .
However, Heidegger proceeds to claim that with the transformation of
the essence of truth in Plato, i.e., the transformation of truth and untruth
into properties of human seeing and saying, there is necessarily a move in
the direction of a purely psychological interpretation of ":
Indeed, with Plato begins, at the same time as the transformation of the es-
sence of  into H"
, a transformation of , i.e., here of the
" that works against it. The event [Ereignis] of withdrawing con-
cealing transforms itself into the human comportment of forgetting [zum
menschlichen Verhalten des Vergessen]. At the same time, what stands
against  becomes a fetching-back-again [Wiederzurück-holen] through
man. (185)
As usual Heidegger gives no evidence for the transformation. He grants
only that what Plato says cannot be interpreted only from the perspective
of that towards which he begins the transformation. Again we have the
contrast between the transformation which Heidegger asserts is there and
what his own reading of the text shows: in this case, while Plato’s text on
Heidegger’s own reading shows the belonging-together of  and
 in Being itself and accordingly characterizes our relation to Being as
", we are asked to believe that the transformation of  and
 into mere “subjective states” begins with Plato.
Though Heidegger himself does not cite the account of " in the
Phaedrus, it provides a crucial contrast and supplement to what we are
told in the Myth of Er. There, in another extraordinary myth, Socrates
again states a condition for entering human life and explicitly maintains
that an understanding of learning as " necessarily results from
 PLATO’S QUESTION OF TRUTH 103

this condition. What is striking, however, is that the conditions stated in

the two myths appear completely opposed: while in the Myth of Er, what
is necessary for souls to enter a human life is drinking the water from the
plain of  (M  +
 8 +), what the myth in the
Phaedrus asserts to be necessary is seeing the truth: “For a soul that has
never seen the truth will never enter this [human] form” (
# ! Y 
"
 $
@ *   $ Y'  <", 249b5-6). This con-
trast is made all the more striking by the fact that what the souls are de-
scribed as visiting prior to their embodiment in the Phaedrus is not  <
U $
 but rather   $
 (248b6). How can these two
accounts be reconciled? It is precisely the notion of " that points
to the answer. In the Phaedrus, immediately after claiming that every soul
must have seen the truth before entering human form, Socrates adds that it
is therefore necessary ($+) to recollect ( ", 249c2). If a soul has
necessarily already seen the truth before taking on human form, then its
learning of the truth in this life can only take the form of “recollection.”
But there is another side to this that the Phaedrus does not mention but
that becomes the focus in the Myth of Er: if learning is ", this is
not only because we have seen the truth, but also because this truth has
immediately fallen into oblivion and concealment. The myths of the Re-
public and the Phaedrus recount two sides of the human destiny that nec-
essarily go together: we are equally under the power of  and ,
we emerge into this life from both the plain of  and the plain of
, and this is why our relation to being has the character of continually
freeing and saving what is unconcealed from concealment: ". We
are born in oblivion and in truth. What thus characterizes human existence
is the simultaneous concealment and unconcealment of beings. Indeed,
reading the myths of the Phaedrus and the Republic together perhaps
shows that  < U $
 and   $
, while visualized
as two distinct places, one in the place beyond the heavens and one in Ha-
des, nevertheless essentially belong together. As we have seen, Heideg-
ger’s interpretation of  < U $
 as found in the $"


, effectively turns it into   $
.
What needs to be stressed here is that Plato characterizes both 
and  as places to which the human soul goes and from which it
comes. What could be further from a transformation of  and 
into mere properties of human speaking and saying? In addition, a com-
parison of the myth in the Phaedrus with the Myth of Er only further con-
firms what Heidegger’s interpretation finds in the latter on its own: that
104 FRANCISCO J. GONZALEZ

Plato thought  and  together. 14 Furthermore, the characteriza-

tion of our relation to Being as " is for Plato demanded by the
inextricably intertwined events of concealment and unconcealment that
define human existence as such. Finally, it needs to be stressed that
", as the way in which we strive after the being that was both
unconcealed and concealed to us, is not some merely mental capacity or
faculty, but a way of existing. According to the account of the Phaedrus,
" occurs in falling in love: a love that completely overwhelms
15
and transforms one’s existence. What could be further from a transforma-
tion of " into a purely subjective state of remembering? 16
Heidegger’s interpretation of other details of the myth that cannot be
considered here confirms this overall interpretation. Specifically, Heideg-
ger’s interpretation of the name of the river that runs through the plain of
, i.e., T", “Careless,” argues that this carelessness is an entrust-
ing to concealment that, rather than being opposed to caring for uncon-
cealment, is inseparable from it (177-8). Also, his interpretation of the
“proper measure” shows that what characterizes the philosopher’s
 and thus distinguishes him from those who lack  is not
the complete overcoming of carelessness and forgetfulness, not the over-
coming of concealment in a clear and distinct intuition and comprehension
of beings, but rather the right measure of carelessness and forgetfulness,
i.e., the granting of the right measure of concealment to beings (178-9).
On Heidegger’s own reading, in other words,  is not the correct-
ness of an intuition or of a body of propositions, but instead a way of ex-
_________
14 Even Szaif, who otherwise resists the etymological interpretation of , ac-
knowledges that Plato, in the contrast between the two places named in the two myths and
in the reference to recollection in both cases, may be making an etymological connection,
which his fellow Greeks would have heard as such, between  and  (1998, n.
143, p. 181).
15 For how this connection between anamnesis and eros solves the problems that have
traditionally plagued the interpretation of anamnesis, see my “How is the Truth of Beings in
the Soul?: Interpreting Anamnesis in Plato,” Elenchos: forthcoming.
16 Heidegger draws attention to another way in which the Myth of Er illustrates the idea
of recollection and thus the belonging-together of concealment and unconcealment. At the
end of the myth, Socrates comments on how this "@
, the "@
 of , was “saved”
(%3) from , since Er was prevented from drinking the water of the river. Socrates
then adds that this saved "@
 is precisely what can save us (621b). Er’s story is of course
itself a “recollection,” but peculiarly it is a recollection of . Such a remembering of
forgetfulness, such an unconcealment of concealment, can “save” us, not by eliminating
, but by, in Socrates’ words, enabling us to pass through in the proper manner the river
flowing through the place of  ( < U 
" = $/" . . .” (621b). On
Heidegger’s reading, this proper passing-through is what Socrates earlier describes as drink-
ing the proper measure.
 PLATO’S QUESTION OF TRUTH 105

isting that in caring for beings to be un-concealed also entrusts them to

their concealment, as a constant recollecting of what has necessarily fallen
into oblivion.
What, then, has become of the supposed transformation of truth and un-
truth into a correctness and an incorrectness that both exclude each other
and are properties of thinking and speaking? Where is there so much as a
hint of such a transformation in that concealing saying of concealment that
is the Myth of Er? Heidegger sees this “transformation,” and thus the loss
of the fundamental experience of , in the central position the myth
accords to human existence.
The necessity of  and its essential relation to  as its preceding
ground is according to the "@
 interpreted from the perspective of the ori-
gin of man’s essence and his destiny. This focus [betonte Blick] on man is al-
ready the sign that the fundamental position of thinking within Greek exis-
tence is undergoing a transformation. . . .  is no longer experienced
purely as an event [rein ereignishaft erfahren], but rather is thought from the
perspective of the behavior [Verhalten] of man in the sense of the later ‘for-
getting.’ (192)
Therefore, while truth and untruth may not be interpreted as correctness
and incorrectness in Plato’s myth, the emphasis this myth puts on our rela-
tion to truth and untruth already puts truth and untruth on the path towards
becoming mere properties of our seeing and speaking.
The first thing to note is that this account of the transformation truth un-
dergoes in Plato is quite different from, and much weaker than, the thesis
of Plato’s Doctrine of Truth. It allows that truth in Plato is still interpreted
as an unconcealment in which we dwell and to which concealment neces-
sarily belongs: the claim is only that there is an emphasis on the human
relation to truth that will eventually result in a reduction of truth to the
human, i.e., humanism. Furthermore, interpreted thus the transformation
obviously cannot be said to begin with Plato in any meaningful sense.
Heidegger himself draws attention to the fact that in Homer  and
)
 are named together (193), thereby hinting, without explicitly claim-
ing, that the transformation is already well underway with the poet. Also,
interpreted thus one could argue that the transformation is still taking place
in Heidegger’s own Being and Time: a text that seeks to interpret the
meaning of being by way of interpreting the being of that being for whom
being is an issue. But for these reasons we must become increasingly sus-
picious of Heidegger’s thesis that at some point (when?) truth underwent a
transformation, a “catastrophe” in the literal sense, that set it on the road to
becoming (inevitably? why?) mere “correctness.” One must ask if Hei-
degger, rather than uncovering a historical transformation, is not instead
106 FRANCISCO J. GONZALEZ

pointing to an inescapable human limitation or, put more positively, hu-

man “measure” in the sense described in the Myth of Er. How can we not
think  and  “together with” our existing and speaking? And
why must doing so result eventually in a reduction of truth to correctness
and of  to forgetting?

IV. The End of Philosophy and the End of Truth

These questions and objections might in part explain why Heidegger, in

“The End of Philosophy and the Task of Thinking,” written in 1964, ex-
plicitly retracts his earlier thesis of a transformation of the essence of truth
in the Greeks. While Plato is not explicitly mentioned here, Heidegger’s
retraction clearly undermines the thesis of the 1940 essay. However, if we
look at why Heidegger retracts this thesis, we see that this retraction is not
so much a retraction of his interpretation of Plato as it is a more extreme
17
version of this interpretation. For if Heidegger now believes that truth did
not become correctness in Plato, this is only because he now believes that
Plato, and the Greeks generally, though naming , thought, and
even experienced, only the correctness of representing and saying (nur als
(
, als die Richtigkeit des Vorstellens und Aussagens erfahren
18
wurde; Heidegger 1969, 78).
This change, however, is only a further step, and indeed the final step,
on the path already taken from the courses of the 1930’s to the 1940 essay.
As has been seen, while the courses find in the Cave analogy a genuine
thinking of unconcealment along with correctness, where the former re-
mains predominant, the later essay essentially equates the truth thought by
Plato with correctness, acknowledging at most a residual experience of
unconcealment in the Cave analogy. What Heidegger does now is simply
_________
17 Dostal rightly observes that Heidegger does not in this essay “or anywhere else recant
his description of the traditional metaphysical notion of truth or his attribution of that notion
to Plato. Accordingly this late concession affects not so much his Plato interpretation as his
interpretation of the Pre-Socratics, and his attempt to find there the doctrine of truth as un-
concealment” (1992, 66-67). What is affected is the distinction Heidegger describes in the
following text from the 1938/39 Die Überwindung der Metaphysik: “Unterschied des
vorplatonischen Denkens und der Philosophie, die mit Plato beginnt:  zur (”
(1999, 36). Now even the “Pre-Platonists” belong to the “philosophy” which is at an end,
while “thinking” is a task that belongs entirely to the future.
18 Though Berti does not discuss this essay, his own conclusion is similar to
Heidegger’s here: “Sowohl bei Platon als auch bei Aristoteles ist die Wahrheit letztlich vor
allem Richtigkeit der Definition, die ihrerseits Kenntnis der Ursachen, d.h. Verstehen
(% ) ist” (1997, 105).
 PLATO’S QUESTION OF TRUTH 107

eliminate the remaining residue. Consequently, Heidegger will no longer

allow his own attempt to think unconcealment as Lichtung or Offenheit to
be named by the Greek term philosophy. With its confinement to a con-
ception of truth as correctness, philosophy is brought to an end. Philoso-
phy still has a long life ahead of it, but not as thinking: it can live on only
as the calculative reasoning of the technicized sciences (63-65). Plato’s
doctrine of truth, fully reduced now to correctness, has become cybernet-
ics.
Yet if, as I have argued above, the direction of Heidegger’s reading of
Plato was already misguided in the 1930’s, its culmination in the present
essay is a reductio ad absurdum. If what Heidegger in the 1937-8 course
Grundfragen der Philosophie referred to as “the fact that, as soon as it
became a matter of bringing the essence of truth to knowledge, 
became H"
 (correctness)” (1992, 205), is not a fact at all, there he at
least acknowledged that the Greeks experienced truth as unconcealment,
indeed claimed this to be indisputable (unbestreitbar). Now even this ex-
19
perience is denied. How are we then to explain the experience Plato de-
scribes in the Cave analogy, an analogy which Heidegger himself earlier
claimed to be impossible without an understanding of truth as unconceal-
ment? How can the comparative 
, of which Heidegger made
so much earlier and which is used to describe objects, be interpreted as
_________
19 Though introducing the essay on the “End of Philosophy” as a major departure from
Heidegger’s earlier interpretation (211-214), Boutot in conclusion suggests that the rupture
may not be as great as at first appears since Heidegger never claimed that the Greeks explic-
itly thought truth as unconcealment (1987, 215). However, Heidegger did claim that they
genuinely experienced truth as unconcealment, while even that he now denies. Boutot ap-
peals to Beaufret’s assertion that “Heidegger n’a jamais dit que les Grecs auraient ressenti
(gefühlt), éprouvé (erfahren) ou pensé (gedacht) la vérité comme . Bien au con-
traire” (1984, p. 12, n. 5). But this claim is simply wrong. One could cite a number of
passages, but one suffices: “Daß die griechischen Denker die Unverborgenheit des Seienden
erfuhren, ist unbestreitbar” (Heidegger 1992, 204-5). And since Heidegger proceeds to say
that this experience of  was later lost (verloren) (205), he clearly is not speaking of
an experience of  as correctness, which is all he will grant the Greeks as ever having
had in the later essay. Even in Platos Lehre von der Wahrheit, Heidegger speaks of “der für
die Griechen selbstverständlichen Grunderfahrung der ” and “Das im Sinne der
 anfänglich griechisch gedachte Wesen der Wahrheit, die auf Verborgenes
(Verstelltes und Verhülltes) bezogene Unverborgenheit” (130). Sallis writes: “One might be
tempted, then, to conclude that unconcealment was simply not experienced among the
Greeks, and that it was not in any sense named in sayings such as that of Parmenides which
spoke of . But this is not Heidegger’s conclusion. Rather, the fact that 
meant correctness and not unconcealment signifies for him that unconcealment was experi-
enced by the Greeks only as (through, in the perspective of) correctness” (1995, 178). Even
so, the cited texts show even this to be a significant departure from his earlier position.
108 FRANCISCO J. GONZALEZ

meaning simply “more correct”? Heidegger does not return to these texts
in the present essay.
Why then does Heidegger take this final step towards a reduction of
Plato’s philosophy, and Greek philosophy in general, to a conception of
20
truth as correctness? Is the cause really, as is sometimes claimed, Paul
Friedländer’s critique of Heidegger’s translation of  as “un-
21
concealment”? The irony is that Friedländer at around the same time re-
tracted his critique of Heidegger as “unjustified” (unberechtigt, 242). All
he now maintained against Heidegger’s translation was that “perhaps”
(vielleicht)  was not originally (ursprünglich) a negative and that it
was not experienced as a negative to the same extent (nicht so durchaus) as
words such as -/ and - (234-5). He did continue to oppose
Heidegger’s thesis of a transformation of the essence of truth, but not by
asserting that the Greeks always thought and experienced  as cor-
rectness but, on the contrary, by insisting that they from the beginning
experienced it as having three distinct but related senses: 1) “the uncon-
cealing, disclosing correctness of saying and opining” [die unverbergende,
enthüllende Rightigkeit des Sagens und Meinens, 2) “the unconcealed,
undisclosed reality of what endures, of being [die unverborgene, un-
22
verhüllte Wirklichkeit des Bestehenden, Seienden”; 3) “the unforgetting,
23
non-deceiving correctness, truthfulness of man, of character” (236). Hei-
_________
20 Even Barnes, who categorically rejects an interpretation of  as “un-
concealment,” still insists on a “sens objectif” that cannot be reduced to the “sens proposi-
tionnel” of “correctness” (1990, 191-3). For the same insistence, see Hestir 2004. Hestir also
rejects Heidegger’s etymological interpretation (n. 7, p. 112) but without explaining why
and while granting that  “has various important connotations for Plato” (n. 7, 113).
Peperzak writes: “But Heidegger conceded too much when in The End of Philosophy (p. 77)
he wrote that alétheia had always had the meaning of correct (richtig) or reliable (zuverläs-
sig). He forgot the meaning that he himself had listed as the first meaning in his essay On
the Essence of Truth and that—as we shall see—is the most important one in the text and
the context of the parable of the cave: the Sachwahrheit or truth of being itself” (1997, 80).
21 As Courtine suggests when he claims that Heidegger here “se range entièrement à
l’analyse de Friedländer et prend acte de ses objections. Singulière concession que ne rend
pas simplement caduc l’essai de 1942, mais risque de ruiner des pans entiers de la lecture
heideggérienne de la philosophie grecque!” (1990, 153).
22 Though not entirely wrong, this is an inadequate interpretation of  in Plato
because, as p. 239 makes clear, it identifies this sense of  with the Idea.  is
not the “undisclosed reality” of the Idea, but what makes possible this undisclosed reality.
And yet at least Friedländer does not reduce Plato’s  to correctness.
23 In the 1958 lecture Hegel und die Griechen, Heidegger does not see the Homeric
texts Friedländer cites as at all counting against the interpretation of  as “Unver-
borgenheit” (1978b, 437); and Friedländer, at least in the final version of his remarks, would
agree.
 PLATO’S QUESTION OF TRUTH 109

degger can therefore be said to reject the thesis of a transformation of the

24
essence of truth for the opposite reason Friedländer does.
There must be a deeper reason for Heidegger’s retraction of his earlier
thesis. Is it not given by the title of the essay? For philosophy to come to
an end and a radically new type of thinking to begin, the “truth” which
Greek philosophy sought to understand must be distinguished sharply and
definitively from what “calls for thinking.” Only by identifying the truth
of Greek philosophy with correctness can Heidegger maintain that what he
seeks to think is something so different that it can no longer be called
“truth.” Is not Heidegger’s reversal then only his new and most extreme
strategy for what he has sought perhaps all along, but especially since his
reading of Nietzsche: to “twist free of” Plato once and for all? If the evi-
dence compels him to retract his thesis of a transformation of the essence
of truth, he cannot do so by agreeing with Friedländer that Plato, as well as
the Greeks before him, thought and experienced  both as correct-
ness and as unconcealment (see 240): this would entangle his own think-
ing in Plato’s and thus prevent the turn to a radical new beginning. 25
The argument here is not that Plato has an adequate (or inadequate)
“doctrine” of truth, but rather that he no more has such a doctrine than
26
Heidegger does. On the one hand, it is difficult to deny what Jan Szaif
characterizes as a principle result of his exhaustive study of Plato’s con-
cept of truth, i.e., that the words  and  occur in Plato pri-
marily in connection with assertion and judgment (17, 69). On the other
hand, what the present Auseinandersetzung with Heidegger’s interpreta-
tion has shown is that truth in Plato is also and always that unconcealment
which is the object, as well as the ground, of all our thinking and desiring
and that therefore even in relation to assertion or judgment it is not plausi-
bly identified with either correctness or correspondence. Thus even Szaif
is forced repeatedly to acknowledge the tendency in Plato to treat truth as

_________
24 Furthermore, even if a philological argument can still be made against the etymology
of  as -, i.e., as originally a negative, since when has Heidegger allowed a
philosophical interpretation to be dictated by a philological argument? In Hegel und die
Griechen, Heidegger asserts that etymology is not what is decisive for the interpretation of
 (1978b, 433, 436).
25 For some reflections on how an identification of Plato’s ‘unsaid doctrine’ with the all-
pervasiveness of concealment () problematizes Heidegger’s distinction between the
first beginning and another beginning, see Sallis 2006.
26 Therefore the most basic objection that can be made against Heidegger’s reading is
that Plato did not have a “doctrine”: see Hyland 2004, 55-56. Both Holz 1966 and Hyland
2004 (see 63) find, rather than a doctrine or theory, an exhibition of truth in the very drama
of the dialogues as a play of concealment and unconcealment.
110 FRANCISCO J. GONZALEZ

an object rather than as a property of assertion (71) or, as he also describes

it, “the tendency . . . not to assign directly to assertions or cognitions the
true or the truth that is known or asserted, but rather to locate the true or
the truth on the level of the real objects and their being at which our asser-
tions and cognitions are directed” (39; see also 69). 27 All this represents
not so much a doctrine of truth on Plato’s part as rather a resistance to the
reduction of truth to any doctrine. After all, is not any “doctrine of truth”
as such already a doctrine of truth as correctness because an attempt to
make a correct assertion about truth, because an ortho-doxy of truth?
Whether in the Cave analogy or the Sun analogy or the Myth of Er,
 transcends us as both the hidden place and the place beyond the
heavens to which we belong and in which we dwell, as that chiaroscuro
within which we and the beings we encounter not only see and are seen
but exist at all. As thus understood  also characterizes not only our
speaking and seeing but even our very existing. Truth for Plato is therefore
indeed both unconcealment and a characteristic of our speaking and see-
ing, but not in such a way that the latter sense of truth so suppresses and
transforms the former as to require an end of philosophy and a new begin-
28
ning. If Heidegger’s philosophy therefore makes an important contribu-
tion to our understanding of truth in Plato, it is also the case that truth in
Plato brings into question central assumptions of Heidegger’s philoso-
phy. 29 But this is only to say that the dialogue between Plato and Heideg-
_________
27 Szaif sees this tendency, which he comes to label the ontological-epistemological
conception of truth, as persisting in Plato’s ‘late work,’ including the Theaetetus and the
Sophist (1998, 514). However, in these two dialogues Szaif also sees emerging an
alternative conception of truth which he calls ‘the logical’: “wird jetzt Wahrheit eindeutig
und primär als Eigenschaft der Urteils- oder Aussagehandlung verstanden, die, gewisser-
maßen vor die Wahl gestellt, zu affirmieren oder negieren, dies in Übereinstimmung mit
dem vorgegebenen Sein oder Nichtsein zu tun versucht” (519). Szaif’s interpretation of the
Theaetetus and the Sophist, to which I have fundamental objections to make, cannot be
assessed here. What is important to note in the present context is that Szaif must acknowl-
edge that the ontological conception of truth persists in the “late” dialogues (522-4), specifi-
cally the Timaeus and the Philebus, and that there is no attempt there to mediate systemati-
cally between the two conceptions, so that “man im übrigen auch nicht gesichert davon
sprechen kann, daß bei ihm [Platon] der logische Wahrheitsbegriff jetzt eindeutig die Pri-
orität besitze” (528) and “man eigentlich nicht von dem Platonischen Wahrheitsbegriff spre-
chen . . . sollte” (530).
28 Hyland has begun making this point in 2004, 61-62. See especially his insistence that
“the very notion of unhiddenness is incoherent without correct looking. The two are not only
compatible but necessarily connected” (62). See also Margolis 2005, 138; and Hestir 2004,
109-150.
29 Boutot is only drawing out the ultimate implication of Heidegger’s reading when, to-
wards the end of his book on Heidegger and Plato, he writes, with an honesty and bluntness
 PLATO’S QUESTION OF TRUTH 111

ger on the essence of truth is one that neither Platonists nor Heideggerians
can afford to ignore. 30
SKIDMORE COLLEGE

_________
one can only admire, the following barbaric sentence: “Platon n’a rien à nous dire sur l’être
et sa vérité” (1987, 314).
30 This paper was written with the support of the National Endowment for the Humani-
ties and the Alexander von Humboldt Foundation.
 COMMENTARY ON GONZALEZ

GARY M. GURTLER, S. J.

ABSTRACT
By including a discussion of the divided line, Heidegger’s interpretation of
the simile of the sun and the allegory of the cave can actually be seen as part
of a wider consideration of Plato’s theory of knowledge. While Heidegger
sets the problem in terms of truth as unconcealment and correctness, analytic
philosophers describe it as open or complete. The contrast becomes even
more stark in traditional language as skepticism and dogmatism. What is
positive in Gonzalez’s account of Heidegger’s interpretation is his emphasis
on Heidegger’s attention to both logos and myth as constituting the proper
mode for doing philosophy. This is rooted in Plato’s recognition of the limits
of language and so of the human condition. The divided line adds support
from the unexpected quarter of mathematics, with Plato making a clear dis-
tinction between the reasoning that is characteristic of mathematical sciences
and the more intuitive method of dialectic.

There is a slight gap in Professor Gonzalez’s account of the Republic, the

razor thin gap of the divided line, which separates the simile of the sun
from the allegory of the cave. I bring it up as a way of testing Gonzalez’s
account of those two images and their interpretation by Heidegger. 1 The
divided line is one of the more elusive similes that Plato developed. It is
not clear whether it is horizontal or vertical, how long it is, and how it is to
be divided. It seems that the divided line is intended, nonetheless, to shed
light on the other two images in this central part of the Republic, providing
a dialectical and mathematical counter to those more poetic images. It is,
however, not these aspects that I want to examine, but the rapid ascent up

_________
1 The anonymous reviewer observed that by introducing the divided line I was ignoring
Gonzalez’s paper and its interpretation of Heidegger. I have attempted to redress this in the
text by making the connection more obvious, without disproportionately lengthening the
comment. Let me point out immediately that consideration of the divided line supports
Heidegger’s concerns about the relation between unconcealment and correctness from an
unexpected quarter. Even if tangential, there is a deep point of contact between the divided
line and the more poetic images that occupy the efforts of Gonzalez and Heidegger. Plato’s
attention to the indefinite and incommensurable in mathematics is expressed here in the
difficulty of bringing the division described to actuality. This is clear from a close look at
the text, but the power of the interpretative veil to keep scholars from seeing what is in a
text cannot be underestimated. This veil seems to account both for Heidegger’s move away
from his own earlier analyses of Platonic dialogues and for Balashov’s statement, 1994,
below that takes the divided line as indicating that knowledge can be completed once and
for all.
 COMMENTARY ON GONZALEZ 113

the line to the unhypothetical first principle and the light this sheds on
Plato’s account of knowledge.
At the behest of students in my survey of ancient philosophy, I searched
for an article about the divided line and found one by Yuri Balashov,
1994, about whether it is possible to divide a line on the basis of Plato’s
description. In brief, Balashov’s conclusion is that all proposed methods
of division are wanting. More interestingly for our purposes, he points out
some of the epistemological implications of the divided line, giving two
opposing views. One, as he puts it, “seems to presuppose the open and
never completed status of knowledge” (p. 15, citing J. P. Dreher, 1990).
This implies a parallel between the limit state of knowledge and the exact
division of the line, with neither ever coming to actuality. Balashov ar-
gues that such an open view of knowledge is alien to the Plato of the Re-
public, Books 6-7. His argument for this concerns the unhypothetical first
principle. “The process whereby a dialectician apprehends the basic non-
hypothetical principle of all knowledge can actually be completed (see,
e.g., 511b8, 511d4, 516b, 532b1)” (p. 15). This actually completed
knowledge is central to the problem of truth as Professor Gonzalez has
construed it from the Platonic text and from Heidegger’s struggle with it,
albeit in different language. 2 Completed knowledge refers, in Heidegger’s
terms, to truth as correctness, and open knowledge as that dialectical in-
terweaving of unconcealment and forgetfulness that Gonzalez explores in
Heidegger’s analysis of key texts in Plato’s Republic and defends against
Heidegger’s own distancing from it in his later writings. Balashov, on the
contrary, makes the case for Heidegger’s analysis all the more obvious,
since his completed knowledge looks suspiciously like taking truth exclu-
sively as propositional and taking propositional truth as exhaustively com-
prehensive. He in fact dismisses the alternative, an open knowledge that
can never be completed, as possibly appealing to some scientists today but
far from the intent of Plato’s Republic. Thus Balashov’s position, rather
than Plato’s, represents the object of Heidegger’s critique in his 1964 arti-

_________
2 My comment juxtaposes different approaches, the Heideggerian one central for Gon-
zalez and that of the analytic articles on the divided line that I use as a foil. The reviewer
saw this as indifference to Gonzalez’s paper by introducing different, and perhaps incom-
patible, ways of viewing Plato’s dialogues. My purpose is to show that the issue that Hei-
degger and Gonzalez are addressing is also being raised in other contexts, even in analytic
circles. My own approach, further, is neither analytic nor continental, but one that, I hope,
is open to other methods and language. When approaches that seem incompatible come to
similar conclusions about the meaning of a text, I find that an indication that incompatibili-
ties may be exaggerated. Generally I have found those of continental background to be
similarly open to different approaches.
114 GARY M. GURTLER, S. J.

cle, with the end of philosophy as thinking and its living on only as the
calculative reasoning of the technicized sciences (Gonzalez, p. 107).
I looked back to the four texts Balashov cited as evidence for his asser-
tion and found some rather interesting results. He cites first the descrip-
tion of moving beyond the method of geometry and other sciences to the
method of dialectic.
Learn also that the other division of the intelligible, I mean that which reason
itself touches by the power of doing dialectic, makes these hypotheses not
first principles but by being hypotheses, as steps to start from, in order to go
right to the unhypothetical first principle of everything ( 
 `

" "<" 
@ 

@ 
 " 
@

j # H 
 1 :

@ $  $", ! 
 

-"

# ! ! D 

,
k
 %/  7 H", q " 
@ 

 %7 * 
@
 * 3. Rep 6. 511b3-7).
If this unhypothetical first principle is like the sun in the other two images,
grasping or touching it does not imply actually completed knowledge, but
a principle beyond the line itself and its division between objects and the
ways they can be known, as Gonzalez phrases it. This first principle is,
thus, very interesting in this context, since it is described in the language
of dialectic, but refers to something that the other images also describe as
the source of all knowing and being.
Balashov’s second passage concludes the section where Plato compares
knowledge of the forms with reasoning about the objects of geometry and
other such sciences. We use the forms to trace back to the unhypothetical
first principle and then come back down to make conclusions about the
forms in relation to other forms, all the while keeping hold of that first
principle. On the contrary, the knowledge of geometry and the other sci-
ences is distinct, forced to use reasoning, making it less clear and its ob-
jects not really understood. “And you seem to me to call reasoning the
habit of geometers and other such experts, but not thinking, as reasoning is
something between opinion and thought ($
 $2 + "
 $
+
*  "  7 *  

- `' ’
# 
@, K
"'-  $'  7 
@ * $

=, Rep 6. 511d2-5). Thought
or 
@ is concerned here with all the forms above the hypotheses of
mathematics. Plato does not say that we grasp all of them, but that by
grasping any of these forms we ascend straightaway to the unhypothetical
first principle and then descend back to consider these forms derived from
the unhypothetical first principle in relation to one another. The
conclusion of this section is even clearer. After mentioning the four ways
of knowing, thinking, reason, opinion and imagination, Plato writes:
“Order them in a ratio: for them it is the case that as it shares in truth, so
those things it rules over share in clarity (7 '
 #! ! 
,
 COMMENTARY ON GONZALEZ 115

6 %’
k %  ",
5 @  4 "

", Rep 6. 511e2-4).
The other two passages, Republic 7. 516b and 532b1, occur within the
discussion of the allegory of the cave, with Balashov recognizing the
continuity between the two images. The first does not refer to the divided
line at all but to contemplating the sun in its own place, while the second
presents an explicit parallel between what dialect reaches at the end of the
intelligible and what the senses reach at the end of the visible. Balashov
takes these as indicating actually completed knowledge, but it is not clear
that this is the case. Seeing the sun in its own place or region is not
grasping it completely, but rather directly, as opposed to seeing the sun
reflected in mirrors or bodies of water. The grasp is clearer and the object
has more being, or in the words that Gonzalez uses, degrees of truth char-
acterize not statements but degrees of being. Nowhere does the text imply
that this seeing of the sun is complete. Similarly, sensation and noetic
grasp reach their respective ends, but the text actually implies that this
must be done constantly, constituting the journey of dialectic illustrated by
the ascent outside the cave and necessitating the descent back into it. 3
The description of knowledge as open or complete differs in
terminology from Heidegger’s discussion of truth in Plato as un-
concealment or correctness. The task is to see if there is a way of
mediating this difference. Let me begin with a brief look at the way Plato
describes language in the Republic. At the end of Book 2, language is
divided into real falsehood, hated by gods and men alike, and falsehood in
words, needing to be used like a drug. At the beginning of the allegory of
the cave in Book 7, the prisoners name the shadows on the wall and take
great pride in their accomplishments in doing so. In these two
descriptions, I take Plato as indicating our naive use of language, where
we take what we talk about, or rather the world we construct through
language, as if it were both true and real. If the cave is our prison, we
ourselves are the ones constructing it.
Truth as completeness or correctness is very much tied to this naive use
of language, and the central similes of Books 6 and 7 are designed to
indicate the limits of language and the paradoxical nature of the human
condition by describing truth in terms of sensing or thinking and their
proper objects. The movement within the cave and the ascent out of it
thus articulate our reflective awareness of this limitation inherent in
_________
3 In this instance, the divided line clarifies in terms of dialectic what the language of as-
cent and descent leave somewhat obscure in the allegory of the cave, the nature and neces-
sity of the descent, confirming the importance of the descent that Heidegger emphasizes in
his earlier writings on the Republic.
116 GARY M. GURTLER, S. J.

language. In place of the speaking of the prisoners, their constant chatter

about the shadows on the wall before them, Socrates describes those
ascending as actually seeing for the first time their own situation within
the cave in stunned silence or apparently foolish babbling. With the
necessary adjustments, the prisoners outside the cave are still using their
eyes to see the world around them with no mention of what they say.
Gonzalez emphasizes exactly this point in his comments about the
allegory of the cave and Heidegger’s more considered comments about it.
He is right to point out that such a view does not leave room for
Heidegger’s notion of correctness, based on taking the forms as some sort
of fixed images or patterns. Similarly, it does not leave much room for the
thesis of Balashov’s article, which takes knowledge as complete,
especially knowledge expressed in language.
Plato’s description of language as falsehood and as imprisoning is also
paradoxical. It is through language, after all, that we become reflective of
our situation and are able to look at it as if from the outside. Gonzalez and
Heidegger appreciate Plato’ use of myth or story to illustrate his mastery
in bringing us to reflection about the truth of our situation, which is
elusive and never quite loses its paradoxical character. In the allegory of
the cave, this truth is described in terms of the process the story describes,
about which, as Gonzalez points out, Heidegger is very attentive. The
discussion of the divided line indicates the same truth, but through the
medium of logos rather than myth. This use of logos may in fact help us
to see how Plato himself might argue against Heidegger’s later claim of a
transformation of the forms into fixed patterns, substituting correctness for
unconcealment, 4 and thus how Plato argues for an open truth that is not

_________
4 In this instance, I see a particular oversight in Gonzalez’s approach to Heidegger’s
misread of Plato. Gonzalez justly defends Plato from Heidegger’s later tracing of this trans-
formation to correctness as already present in the Platonic dialogues. Heidegger had also
attributed this transformation to the Roman conception of truth in terms of adequatio, espe-
cially in denoting truth as correspondence. Adequation is not, however, about correspon-
dence, as if the intellect is brought to correspond to the thing, but rather of the presence in
intellect of the thing, a presence which retains all the ambiguity Heidegger wants to find in
truth as unconcealment and which preserves being as the proper object of intellect. The
removal of presence from the equation is the result of modern thought, where the idea in the
mind has an independence that gives rise to the problem of correspondence. In medieval
thought, the concept, for example, always remains the means by which the object is known,
more or less clearly, and is not the idea as the object of thought in modern epistemology,
whether traceable to Descartes or Kant. While it is true that Gonzalez has articulated most
clearly the strange backtracking in Heidegger’s interpretation of Plato and some of the rea-
sons for it, I find this insufficient because the problem is also symptomatic of a more gen-
eral misreading of the history of philosophy and cannot be restricted to Platonic texts alone.
 COMMENTARY ON GONZALEZ 117

complete, at least relative to us, as Heiddegger alludes in citing Aristotle’s

Metaphysics.
When Plato discusses the intelligible section of the divided line, he
deliberately leaves out anything that we might count as fixed. Instead we
have instead hypotheses, the ungrounded hypotheses of mathematics and
the forms as hypotheses that lead rather quickly to the unhypothetical first
principle. This is indeed a modest version of the theory of forms, but one
that Plato never really abandons. For all his emphasis on the forms as the
ground of our knowledge, precisely in its approach to truth, very few
forms are ever really discussed, and this passage reminds us that when
they are discussed, they are rather fluid among themselves and retain the
hypothetical character they have here. Hypothesizing the world of forms
is dialectical in the divided line, but Plato makes clear that dialectic is
described not in the language of mathematical reasoning but that of
intellectual insight. In the allegory of the cave, moreover, this dialectical
description is supplemented by the poetic imagery of the story where
intellectual insight is made explicitly parallel to visual perception. Both
modes of discourse indicate the tentative nature of our grasp of truth, its
open character, or its combination of concealing and unconcealing at once,
as Heidegger puts it.
Finally, I constantly point out to my students that truth in Plato is always
contextual. This is why he writes dialogues, where truths emerge from
particular discussions by certain individuals at a certain time, limited by
their assumptions and biases. This puts Plato on a razor’s edge, with
readers pushing him off, toward skepticism or dogmatism, as suits their
whims. In Book 10 of the Republic, we find Plato’s reply to such
strategems in his talk about the artisans and poets as being at second and
third remove from real being. It is another reminder of the mediating
nature of language, as at second or third remove from its object. This is
always true in our discussion of anything, even an event such as this
lecture tonight. We are here experiencing it, but as soon as the event is
over, our account can be no more than at second remove, with derivative
accounts at third remove at best. Plato here recognizes the limits of the
human condition, capable always of grasping truth, but not of nailing it
down once and for all.
BOSTON COLLEGE
 GONZALEZ/GURTLER BIBLIOGRAPHY

Ast, F. 1908. Lexicon Platonicum. Berlin.

Balashov, Yuri. 1994. Should Plato’s Line Be Divided in the Mean and Extreme
Ratio. Ancient Philosophy 14: 283-95.
Barnes, J. 1990. Heidegger spéléologue. Revue de Métaphysique et de Morale 95: 173-
195.
Beaufret, J. 1984. Parménide: Le Poème . 2d ed. Paris.
Berti, E. 1997. Heideggers Auseinandersetzung mit dem Platonisch-Aristotelischen
Wahrheitsverständnis. In Richter (ed.): 89-105.
________. 2005. Heidegger and the Platonic Conception of Truth.. In Partenie and
Rockmore (ed.): 96-107
Bonitz, H. 1961. Index Aristotelicus. Vol. 5 of Aristotelis Opera. Berlin.
Boutot, A. 1987. Heidegger et Platon: le problème du nihilisme. Paris.
Courtine, J.-F. 1990. Heidegger et la phénoménologie. Paris.
Dostal, R. 1992. Beyond Being: Heidegger's Plato. In Macann (ed.): 61-89.
Dreher, J. P. 1990. The Driving Ratio in Plato’s Divided Line. Ancient Philosophy 10:
159-72.
Friedländer, P. 1964. Platon. Band I. 3rd ed. Berlin.
Gonzalez. F. 2003. Heidegger’s 1933 Misappropriation of Plato’s Republic.
‹
/X": quaderni di filosofia 3: 39-80.
________. 2007. Heidegger on a Few Pages of Plato’s Theaetetus. Epoché 11, no. 2:
371-392.
Heidegger, M.1969. Das Ende der Philosophie und die Aufgabe des Denkens. In Zur
Sache des Denkens. Tübingen: 61-80.
________.1978. Wegmarken, 2nd ed. Frankfurt am Main.
________.1978a. Platons Lehre von der Wahrheit. In Wegmarken: 201-236.
________. 1982. Parmenides. Gesamtausgabe 54. Frankfurt am Main.
________. 1978b. Hegel und die Griechen. In Wegmarken: 421-438.
________. 1988. Vom Wesen der Wahrheit: Zu Platons Höhlengleichnis und Theätet.
Gesamtausgabe 34. Frankfurt am Main.
________. 1992. Grundfragen der Philosophie. Gesamtausgabe 45. 2d. ed. Frankfurt
am Main.
________. 1995. Logik: Die Frage nach der Wahrheit. Gesamtausgabe 21. 2nd Ed.
Frankfurt am Main.
________. 1999. Metaphysik und Nihilismus. Gesamtausgabe 67. Frankfurt am Main.
________. 2001. Sein und Wahrheit. Gesamtausgabe 36/37. Frankfurt am Main.
________. 2002. The Essence of Truth: On Plato’s Cave Allegory and Theaetetus.
Sadler. T. (tr.) London.
Hestir, B. 2004. Plato on the Split Personality of Ontological Alétheia. Apeiron 37, no.
2: 109-150.
Hyland, D. 2004. Questioning Platonism: Continental Interpretations of Plato.
Albany.
Hyland, D., and J. P. Manoussakis (ed.) 2006. Heidegger and the Greeks.
Bloomington, IN.
Macann, C. (ed.) 1992. Martin Heidegger: Critical Assessments. Vol. 2. New York.
Margolis, J. Heidegger on Truth and Being. In Partenie and Rockmore (ed.): 121-139.
Partenie, C., and T. Rockmore (ed.). 2005. Heidegger and Plato: Toward Dialogue.
Evanston, IL.
 GONZALEZ/GURTLER BIBLIOGRAPHY 119
Peperzak, A. 1997. Did Heidegger Understand Plato’s Idea of Truth? In Platonic
Transformations: with and after Hegel, Heidegger, and Levinas, 57-111. Lanham,
Maryland.
Richter, E. (ed.) 1997. Die Frage nach der Wahrheit. Frankfurt am Main.
Sallis, J. 1995. Delimitations: Phenomenology and the End of Metaphysics. 2d ed.
Bloomington.
________. 2006. Plato’s Other Beginning. In Hyland and Manoussakis (ed.): 177-190.
Szaif, J. 1998. Platons Begriff der Wahrheit. 3rd ed. München.
Wolz, H. 1966. Plato’s Doctrine of Truth: Orthotes or Aletheia? Philosophy and
Phenomenological Research, vol. 27, no. 2: 157-182.
 COLLOQUIUM 5

PLATO’S ANTI-HEDONISM

MATTHEW EVANS

ABSTRACT
Nearly all contemporary philosophers reject the ancient hedonist view that an
agent has a reason to commit an action just in case that action is pleasing to
that agent. But few (if any) do so on the ground that an action’s being pleas-
ing to an agent gives that agent no reason whatsoever to commit that action.
Instead they opt for a compromise position according to which the rational
authority of one’s own pleasures is unassailable, but not unlimited. My aim
in this paper is to show that a frequently overlooked argument from Plato’s
Philebus poses a powerful challenge to this widely held compromise position.
If my interpretation of this argument is correct, then Plato’s case against he-
donism is more radical, more subtle, and ultimately more appealing than any-
one has yet realized.

It often seems obvious to us that our pleasures can justify our actions. If I
ask you why you’re reading right now instead of dancing, and if your an-
swer is that reading, unlike dancing, is just something you like to do, then
(all else equal) your answer seems perfectly sufficient. To demand that
you specify some further end you have in enjoying yourself would seem
unreasonable if not bizarre. As Elizabeth Anscombe observes, “‘It’s pleas-
ant’ is an adequate answer to ‘What’s the good of it?’ or ‘What do you
want that for?’ I.e. the chain of ‘Why’s’ comes to an end with this answer”
(Anscombe 1957/1999, 77). And from this observation alone it is natural
to infer—as Plato’s Academic colleague Eudoxus apparently does—that
your taking pleasure in something gives you a special, non-derivative rea-
son to pursue that thing. Consider the following argument, attributed by
Aristotle to Eudoxus in book X of the Nicomachean Ethics:
[W]hat is most choiceworthy is what we choose neither because of, nor for
the sake of, anything else. And it is agreed that pleasure is a thing of this sort,
since no one asks anyone ‘For the sake of what are you pleased?’, and this
implies that pleasure is choiceworthy in itself. 1 (
#$ ! %M 

` Y$ K > *
= Œ* * 4$
) (EN 1172b19-23)

_________
1 This translation modifies that of Rowe in Broadie and Rowe 2002.
122 MATTHEW EVANS

If Aristotle’s report here is accurate, then Eudoxus makes essentially the

same observation that Anscombe does, and infers on that basis that pleas-
ure is, for each of us, something worth pursuing as an end in itself. 2 He
infers, in other words, that pleasure is worth pursuing by each of us be-
cause of how it is in itself, regardless of the relations in which it stands to
other things, and regardless of the antecedents or consequences of pursu-
ing it; it is something that, because of its own nature, makes it worth pur-
suing by each of us. And this is to say, in effect, that each of us has a spe-
cial, non-derivative reason to do what we enjoy doing. 3 According to Eu-
doxus, then, our practice of accounting for our actions shows that some
reasons—call them hedonic reasons—can warrant our actions just be-
cause, and just insofar as, these actions please us. Let’s call this view

Liberalism

An agent has a non-derivative reason to perform a certain action if and

because that agent’s performing that action will be pleasing to that
agent.

Eudoxus is a liberal, to be sure, but not only a liberal. Like other ancient
Greek hedonists, he is not content to claim that we have hedonic reasons
to do things; he wants also to claim that these hedonic reasons are the only

_________
2 Apparently Eudoxus also makes at least one of the following three additional infer-
ences: (a) that pleasure is worth pursuing only as an end in itself, (b) that pleasure is the
only thing worth pursuing as an end in itself, or (c) that pleasure is the only thing worth
pursuing only as end in itself. In what follows I will be ignoring these more questionable
inferences.
3 The concepts in play here, as I understand them, are strictly interconnected: for any x
and any agent S, x is worth pursuing as an end in itself by S if and only if there is a non-
derivative reason for S to pursue x; and there is a non-derivative reason for S to pursue x if
and only if how x is in itself is what makes x worth pursuing by S. In other words, an end in
itself—again, as I understand the concept—is something that has final value because it has
(a certain kind of) intrinsic value: it has final value in that it is worth pursuing as an end, not
merely as a means; and it has intrinsic value in that what makes it worth pursuing as an end
is how it is in itself, and not merely how it stands in relation to something else. For some
useful recent work on the distinction between final value and intrinsic value, see Korsgaard
1983, 169-95; Kagan 1998, 277-97; and Langton 2007, 157-85. Thanks to an anonymous
referee for demanding more precision from me on this point.
 PLATO’S ANTI-HEDONISM 123

non-derivative reasons we ever have to do anything. 4 Nuances aside, this

more radical claim can be called

Hedonism

An agent has a non-derivative reason to perform a certain action if,

only if, and because that agent’s performing that action will be pleas-
ing to that agent.

Hedonism is widely rejected by contemporary philosophers, and rightly

so. All that’s required to refute it is a sound argument to the effect that
someone has a non-derivative, non-hedonic reason to do something. So if
you have a reason to help some people just because they need help, or just
because you want to help them, or just because everyone would be happier
if you helped them, or just because they are your parents, or just because
they are rational beings, … then Hedonism is false. As a result, few (if
any) ethical theorists still take Hedonism seriously as a comprehensive
account of what we have most reason to do. 5 Nor should they. Hedonism
is an extreme and implausible view—so implausible, in fact, that its rejec-
tion is not controversial enough to be of much philosophical interest to us
at this point. For better or worse, we are all anti-hedonists now.
But we need not, for this reason alone, be anti-liberals. We can claim,
without contradicting ourselves, that our non-derivative reasons for action
include both the hedonic and the non-hedonic. Indeed, the prevailing view
today seems to be a compromise position of just this sort, according to
which some—but not all—of our non-derivative practical reasons are he-
donic. An especially prominent defender of this view is Thomas Nagel,
whose account of the value of physical pleasure and pain is an attractive
combination of Liberalism and Anti-Hedonism: “I am not an ethical he-
donist,” he writes, “but I think pleasure and pain are very important….
Almost everyone takes the avoidance of his own pain and the promotion
_________
4 The other ancient hedonists I have in mind here are the Cyrenaics and the Epicureans.
For a comprehensive critical overview of these two schools, of Eudoxus, and of Greek he-
donism in general, see Gosling and Taylor 1982.
5 Two especially influential negative appraisals of Hedonism can be found in Broad
1930, 227-39, and Brandt 1959, 295-329. A qualified exception is Fred Feldman, who
staunchly defends the view that one’s own pleasures are solely responsible for making one’s
own life good. But even Feldman is ready to admit that not all of one’s non-derivative rea-
sons for action are fixed by facts about what makes one’s own life good. For details see
Feldman 2004, 7-12.
124 MATTHEW EVANS

of his own pleasure as subjective reasons for action in a fairly simple way:
they are not backed up by any further reasons” 6 (Nagel 1986, 157). In Na-
gel’s view, then, Liberalism is true and Hedonism is false: our pleasures
can justify our actions, but so can lots of other things. Views of this sort
have prevailed among recent and contemporary ethical philosophers, pre-
sumably because they offer a plausible, low-cost way to accommodate
pleasure’s obvious deliberative salience. 7 The consensus, in effect, is that
Eudoxus gets things exactly half right: the rational authority our pleasures
have over us is unassailable, but not unlimited.
My aim here is to show that a frequently overlooked argument from
Plato’s Philebus has the potential to disrupt this consensus. I call it the
Aiming Argument, and I claim that it runs from 53c4 to 55a11. If my in-
terpretation is correct, then the dialectical role of this argument is to per-
suade Protarchus—the dialogue’s appointed defender of Hedonism—that
no pleasure is ever worth pursuing as an end in itself, and consequently
that we never have any hedonic reason to do anything. To support this
strikingly anti-hedonist and anti-liberal conclusion, Socrates first claims
that every pleasure is necessarily “for the sake of” (`) something other
than itself. He then claims that if a thing is necessarily “for the sake of”
something other than itself, then it is not worth pursuing as an end in itself.
And from these two premises he (validly) infers that no pleasure is worth
pursuing as an end in itself.
What makes this argument exciting, rather than merely odd, is that it
aims to discredit hedonists like Eudoxus by attacking the most plausible
and popular component of their view. As we have seen, most philosophers
today are inclined to agree with Eudoxus that none of our pleasures is “for
the sake of” anything else, and that this is why Liberalism is true; 8 but
_________
6 In this particular passage Nagel explicitly limits his discussion to “sensory pleasures”
such as those of “food, drink, sleep, sex, warmth, and ease.” But his overall point seems to
be that, from a subjective standpoint, your taking pleasure in something constitutes an
agent-relative reason for you to pursue that thing. For a broadly similar assessment of the
value of pleasure and pain, see Goldstein 1989.
7 See, for example, Sidgwick 1884/2001, 162-95; Ross 1930/1988, 134-41; Sumner,
1996, 81-112; and Kagan 1998, 25-48. Two noteworthy contemporary heretics are Elijah
Millgram and Christine Korsgaard, both of whom hold—very roughly, and on very different
grounds—that taking pleasure is a way of perceiving reasons for action, not acquiring them.
See Millgram 1993, 394-415, and Korsgaard 1996, 145-55.
8 As Verity Harte points out, this particular formulation is too strong. There’s an impor-
tant difference between (I) the claim that every pleasure is worth pursuing as an end in it-
self, and (II) the claim that every pleasure is worth pursuing only as an end in itself. (See
above, note 2.) What I meant to say here is not that every liberal is committed to (II), but
that every liberal is committed to (I).
 PLATO’S ANTI-HEDONISM 125

Socrates seems to be arguing here that every one of our pleasures is “for
the sake of” something else, and that this is why Liberalism is false. So if
Plato’s Aiming Argument is sound, then he will have established not only
that Hedonism is false, but also that Hedonism is false because Liberalism
is false. 9 This is what makes Plato’s Anti-Hedonism interesting.
But is the Aiming Argument sound? In what follows I will suggest, and
tentatively argue, that it is. First I will try to determine more precisely
what the phrase “for the sake of” (`) is supposed to mean in this con-
text. Clues are available in and around the argument itself, but in my view
they are cryptic and need to be supplemented with further evidence, pref-
erably from other parts of the Philebus. As I hope to show, this further
evidence yields two distinct but closely related ways of understanding
how, for Plato, pleasures are “for the sake of” things other than them-
selves: first, (some) pleasures are processes through which animals are
restored to their optimal equilibrium states (31b-36c); and second, (some)
pleasures are—like beliefs—attitudes with objective conditions of correct-
ness (36c-41a). I will argue that although either interpretation makes good
sense of Plato’s reasoning, the second gives his argument real philosophi-
cal force. In particular, it yields a powerful objection to the widely held
view that one’s pleasures can justify one’s actions. If this reading is right,
then Plato’s attack on Hedonism is more radical, more subtle, and ulti-
mately more appealing than anyone has yet realized.

II

As an anti-hedonist dialogue, the Philebus reaches its crescendo with the

Aiming Argument. For this is the first argument that converts Protarchus
to the view that no pleasure is worth pursuing as an end in itself. 10 Since
_________
9 In what follows I will be operating on the assumption that most of the substantive
claims and inferences voiced by Socrates in the dialogues can be safely attributed to Plato.
For an extended defense of this assumption, see Kraut 1992.
10 It is a matter of controversy whether this is the claim that the argument is designed to
support. I’m persuaded by Hackforth, who argues—on the basis of 54c10-d2—that the con-
clusion of the Aiming Argument “clearly means not merely that pleasure is not the good, the
sole good or even the chief good, but that it is not a good; in fact that ‘good’ cannot be
predicated of any pleasure” (Hackforth 1972, 106). Though some later commentators have
doubted Hackforth’s reading, their objections strike me as unconvincing. Gosling and Tay-
lor, for example, claim that the argument is meant to apply only to bodily pleasures and
(maybe) a proper subset of psychological pleasures, since Socrates never explicitly says that
all psychological pleasures have the relevant suspect feature (Gosling and Taylor 1982, 152-
54). But this leaves it quite mysterious why Socrates never explicitly says that only these
126 MATTHEW EVANS

Protarchus cannot concede this point and still defend anything even re-
motely resembling Hedonism, his unambiguous assent to the argument’s
conclusion marks both his own final defeat and, for this reason, the struc-
tural peak of the dialogue’s anti-hedonist program. But the argument ap-
pears rather suddenly in the text, and the very suddenness of its appear-
ance has led some commentators to claim that it is at best an afterthought,
and at worst a half-hearted parting shot. 11 This reading is uncharitable, in
my view, because there is a clear rationale for the argument’s being right
where it is. It begins just after Socrates’ extended account of the pure
pleasures (50e5-53c2), at which point his standing suggestion is that these
pleasures, unlike the impure ones, might be worth cultivating after all
(53b8-c2). So the subsequent appearance of the Aiming Argument makes
good sense if its purpose is to modify or refine this standing suggestion.
And on my reading, at least, that is precisely its purpose: to establish that
all pleasures—even the pure ones—have a suspect feature that makes
them unworthy of a certain kind of pursuit. Therefore, on my reading, the
Aiming Argument is both appropriately placed and philosophically sub-
stantial. It is no mere appendix.
Its first step is an initially obscure proposal concerning the nature of
pleasure. I will call this proposal the Genesis Theory. Socrates introduces
the theory as follows:
Have we not heard about pleasure that it is in every case a becoming, and that
there is no being at all of pleasure? ( 4$
< . . . K 7  %,

# $2
# )   4$
<;) For certain elegant minds (
"&
7 . . .
) attempt to reveal this account to us, and we should be grateful to
them. 12 , 13 (53c4-7)

_________
pleasures have the suspect feature, and why he implicitly says, just after completing the
argument, that he and his opponents have been assuming all along that pleasure is “in the
psyche” [% &:] (55b1-5). Gabriela Carone, on the other hand, thinks that the lesson of
the Aiming Argument is that pleasure cannot be “the only” good (Carone 2000, 257-83).
But she offers no textual evidence in support of her reading and does not address Hack-
forth’s argument.
11 The leading advocates of this view are Hackforth and Gosling, both of whom suggest
that the argument should be read as a provisional and hastily constructed appendix (Hack-
forth 1972, 105) (Gosling 1975, 220-2).
12 Many scholars have correctly pointed out that Socrates does not take credit for the
Genesis Theory. Instead he attributes it to an unnamed group of “elegant minds” (53c4-7; cf.
54d4-e2). Some have taken this to indicate that Plato, by distancing Socrates from the theory
in this way, is signaling that he himself does not endorse it (Hackforth 1972, 105-6) (Carone
2000, 265). On this view, Socrates implicitly treats the “elegant minds” as he treats the
“clever scientists” [$
 7 -] at 44b9—that is, as merely temporary allies in a
common struggle against Hedonism. But this cannot be right. Though it is true that Socrates
never explicitly endorses the view of the “elegant minds,” he takes great care to warn Pro-
 PLATO’S ANTI-HEDONISM 127

Protarchus finds the Genesis Theory puzzling at first (53c8) and so should
we. For it is nearly impossible to see from this extremely compressed ac-
count alone what the distinction means, how its terms are related to each
other, and why it is relevant to the analysis of pleasure in the first place.
After acknowledging a lack of clarity in its initial formulation (53c9-d1),
Socrates tries to illuminate the distinction with four examples, each of
which is supposed to apply to the two original terms:
(1) A being is something “itself by itself” (# > ) while a be-
coming “aims in every case at something else” ( 7 %"

0
). (53d3-4)
(2) Beings are “in every case most venerable” ("
 7) while
becomings are not. (53d6-7)
(3) Beings are to becomings as “good-looking and upstanding boys”
($ 
 ! 7 !) are to “their manly lovers” (%!
$
 #). (53d9-10)
(4) Becomings are “in every case for the sake of some one of the be-
ings” (` 
   )> ) while beings are “at each time
that for the sake of which what comes to be for the sake of some-
thing in every case comes to be” (
j  ‘
   `
 "
 7  ). (53e4-7)
Taken together, these four examples yield an apparently consistent view
about the relation between becomings and beings. Though elliptical, (1)
suggests that beings are self-contained or self-sufficient in a way that be-
comings are not, since the latter, unlike the former, constitutively “aim at”
things other than themselves. And while (2) suggests that what is aimed at
belongs to a higher order of value than what aims, (3) implies that this
higher order of value explains their being aimed at in the first place. The

_________
tarchus not to believe the “clever scientists” when they claim that every pleasure is merely
an escape from pain (44c1-5). Moreover, he explicitly accuses the “clever scientists” of
analyzing pleasure “without skill” [
# L] (44c6) and then says quite clearly both that
and why his own view differs from theirs (44d2-5). Socrates does nothing like this with the
“elegant minds.” He voices no disagreement with their theory and twice expresses strong
gratitude to them for publishing it (53c7; cf. 54d6). That Socrates does not claim to be the
theory’s author is itself neither here nor there. For Plato frequently distances Socrates from
the source of doctrines that Socrates—and Plato himself—clearly endorse. In the Philebus
itself Socrates refuses to take credit for the “Promethean method” of collection and division
(16c1-6) and the “dream solution” to the conflict between pleasure and intelligence (20b6-
7). Surely this should not be taken as a sign that Plato does not endorse either one. For simi-
larly respectful deferrals of authority in other dialogues, see Meno 81a-b, Symposium 201d,
and Phaedrus 244a.
13 My translations of passages from the Philebus will be modifications of Frede’s in
Frede 1993.
128 MATTHEW EVANS

“good-looking and upstanding boys” are the targets of “manly love” pre-
sumably because they have a certain value to which these lovers are at-
tracted. In (4) Socrates introduces the “for the sake of” relation to capture
the sense in which becomings “aim at” beings.
At this point Protarchus gets impatient—perhaps because he thinks that
Socrates is being unnecessarily opaque—and suggests a gloss on (4) that
exploits an analogy between becoming and producing. As Protarchus sees
it, the issue is “whether shipbuilding is for the sake of ships or ships for
the sake of shipbuilding” (54b2-4). Socrates concurs, saying “that’s just
what I mean” (  
@> #) (54b5). 14 He then offers Protarchus a
summary account of his overall position, which seems to combine (1)-(4)
with the analogy between becoming and producing:
I say that all ingredients, all tools, and all materials are in every case provided
for the sake of becoming, that every becoming in every case comes to be for
the sake of some particular being (‘ $2  0 0
#
 ‘ `  ), and that the whole of becoming comes to be
for the sake of the whole of being. (54c1-4)
Here Socrates expands on Protarchus’ shipbuilding analogy by specifying
three different types of thing, each of which has a fixed place in a certain
hierarchy of value. 15 His claim seems to be that some things are, because
of their own nature, somehow subordinate or superior to other things.
Consider shipbuilding for example. According to Socrates, I take it, the
ingredients, tools, and materials needed to build a ship—wood, metal, saw,
level, and so on—are somehow subordinate to the producing of a ship, and
the producing of a ship is somehow subordinate to the ship to be produced.
These two subordination relations strike me as intuitive. If you aim to
build a ship, then you can achieve your aim only if you have the right in-
gredients, tools, and materials to build a ship, and only if you work with
these ingredients, tools, and materials in such a way that you will, if suffi-
ciently unimpeded, produce a ship. Therefore the work you are doing pro-
vides standards of evaluation for the ingredients, tools, and materials you
will be working with, and the (intended) product of your work provides
standards of evaluation for the work you are doing. In this sense at least,
_________
14 This tells against Thomas Tuozzo’s claim that “Socrates never explicitly endorses the
view that pleasure is a ‘coming-into-being’ in the way that shipbuilding is” (Tuozzo 1996,
503). But Tuozzo is right to point out that the analogy to shipbuilding need not be taken as a
substantive claim about the nature of all becomings. Surely neither Protarchus nor Socrates
means to say that every becoming is a producing. At most the analogy suggests the con-
verse: every producing is a becoming.
15 Useful discussions of the metaphysical motivations and implications of this passage
can be found in Shiner 1974, 49-105, and Benitez 1989, 103-5.
 PLATO’S ANTI-HEDONISM 129

it’s clear how materials are subordinate to producing, and how producing
is subordinate to product. In each case the superior thing provides deter-
minate standards of evaluation for the subordinate thing. This is an espe-
cially salient feature of the relation between producing and product, since
producing is essentially subordinate to product. After all, part of what it is
to produce something is to do work that is subject to standards of evalua-
tion fixed by the nature of the (intended) product. Part of what it is to pro-
duce a ship, for example, is to do work that is properly criticizable in light
of what it is to be a ship. Every producing is, in this sense, essentially
regulated by some product. Since Socrates wants to say that becoming is
subordinate to being just as producing is subordinate to product, he seems
to be implying here that every becoming is essentially regulated by some
being.
Before proceeding any further with this line of thought, however, let’s
pause for a moment to ask ourselves how this distinction between becom-
ing and being is supposed to help us understand what pleasure is. If my
interpretation is right so far, then Socrates wants to say that every pleasure
is essentially subject to standards of evaluation that are fixed by the nature
of some being. On first hearing, however, this claim seems unmotivated
and opaque. What sense does it make to say that every pleasure is a be-
coming, or that every pleasure is subject to an external standard of evalua-
tion, or that every pleasure is essentially regulated by some being? And
why would anyone want to say these things at all?
The best place to look for answers to these questions, in my view, is not
the immediate text and context of the Aiming Argument, since additional
clues are sparse there. I suggest instead that we look to some earlier sec-
tions of the dialogue, where Socrates defends a number of substantial and
relevant claims about the nature and value of pleasure. Two claims in par-
ticular strike me as helpful, given our current concerns.
The first claim emerges during Socrates’ initial attempt to theorize about
bodily pleasures and pains. Complications aside, his view here is that
(some) pleasures are processes by which animals restore themselves to
their optimal equilibrium states (31b-36c). 16 I will call this the Equilib-
rium Theory. It is easy to see how the Equilibrium Theory and the Genesis
Theory might be related. Indeed, on the plausible assumption that every
restoring is also a becoming, the Equilibrium Theory implies the Genesis
Theory. Also noteworthy here is Socrates’ claim that the equilibrium states
of animals are, unlike the processes by which they are restored, “beings”;
_________
16 For more detailed interpretation, discussion, and analysis of this view, see Gosling
and Taylor 1982, 129-42; Frede 1992; Tuozzo 1996; and Evans 2007.
130 MATTHEW EVANS

he even says that pleasure is, for the animals who undergo it, “the path to
their being” (* $>  * 
# H$) (32b3). Given the way that
Socrates spells it out, then, the Equilibrium Theory seems to imply that
every restoration is a becoming, that every equilibrium state is a being,
and that every restoration is “for the sake of” some equilibrium state.
The examples he gives here are illuminating. He says that hunger is “a
dissolution (-) and a pain” (31e6), and that eating (when hungry) is “a
replenishment () and a pleasure” (31e8); similarly, thirst is “a
destruction (
) and a pain” and drinking (when thirsty) is a “replen-
ishing” (
@) and a pleasure (31e10-32a1). Following Socrates’
lead, we could try to think of these various states and processes in terms of
fulfillment. On this approach, an animal’s equilibrium states are ways of
being fulfilled, and eating and drinking are ways of becoming fulfilled or
aiming at being fulfilled. This would make decent sense of the view that
every pleasure is a becoming “for the sake of” some being. Moreover,
since there are obviously better and worse ways to seek fulfillment, espe-
cially over a lifetime, we might well think that part of what it is to become
fulfilled (or to aim at being fulfilled) is to be subject to standards of
evaluation fixed by the state of being fulfilled. And this in turn would
make decent sense of Socrates’ view that every pleasure is essentially
regulated by some being. All things considered, then, the Equilibrium
Theory gives us a potentially useful way to think about the meaning and
motivation of the Genesis Theory.
But I believe there is another way as well. What I have in mind is the
line of thought that emerges during Socrates’ famous attempt to persuade
Protarchus that some pleasures are false (36c-41a). Here Socrates draws a
crucially important distinction between attitudes and objects of attitudes:
S: Believing is something in us, right? ()  
-  $
'C 4"+)
P: Yes.
S: And being pleased? (Y$)
P: Yes.
S: And that which is believed is also something, right? (7 "* 7 
$
'C" % )
P: Of course.
S: And that by which the one who is pleased is pleased? (7   ’ 
4$"
 Y$)
P: Certainly. (37a1-10)
The apparent purpose of this exchange is to establish that (some) pleasures
are, like beliefs, attitudes taken toward objects. Just as it is appropriate to
 PLATO’S ANTI-HEDONISM 131

ask of a given attitude of belief what is believed, it is appropriate to ask of

a given attitude of pleasure what pleases. 17 And to answer these questions
is to specify the objects of these attitudes. As Socrates goes on to point
out, this distinction between attitude and object seems to open up the pos-
sibility that some attitudes are incorrect:
S: So, if what is believed is mistaken, should we agree that the belief making
the mistake at that time is incorrect, and is believing incorrectly? (0 $ 
“""
  $
'C"
 ”, * $'  “"
 
#
(* H"



#$> ( $
'C
)
P: How could we not?
S; But what if we observe that some pain or pleasure is making a mistake
about that by which it is pained or pleased? (• = - –  4$
* 7
 %> ’ + — 
#
 “"
 %
") Shall we then call
it “correct” or “good” or any other fine names?
P: That would be impossible, if in fact a pleasure makes a mistake. (37e1-9)
What Socrates wants Protarchus to see here, I take it, is that attitudes of
pleasure are on a par with attitudes of belief insofar as they can be as-
sessed as correct or incorrect on the basis of a distinct and conceptually
prior assessment of their objects. In the case of belief, at least, Socrates’
view is clear: an attitude of belief is correct just in case its object—that
which is believed—is true (38b-d). In the case of pleasure, however, his
view is not so clear. But the analogy he draws between belief and pleasure
strongly suggests that, as far as he is concerned, attitudes of pleasure have
the same sort of correctness conditions that attitudes of belief do. So even
if we ignore worries about how the correctness conditions of a pleasure are
fixed by its object, we can understand what Socrates is getting at. In his
view, certain attitudes—including those of belief, pleasure, and pain—are
subject to conditions of correctness that are determined by the character of
their objects.
Following Socrates’ analogy, then, we could try to think of attitudes and
their objects in terms of objective correctness. On this approach, the con-
ditions of correctness for attitudes are ways of being correct, and the tak-
ing (or having) of a given attitude is a way of becoming correct or (better)
a way of aiming at being correct. This would make adequate sense of the
view that every pleasure is a becoming “for the sake of” some being, espe-
cially if the “for the sake of” relation is understood as a type of aiming

_________
17 As Victor Caston has argued, it is essential to attitudes of this kind that they have the
property now known as intentionality (Caston 1993, 213-60) (Caston 2001, 23-48).
132 MATTHEW EVANS

relation. 18 And it would make obvious and excellent sense of the view that
every pleasure is subject to standards of evaluation fixed by some being.
So Socrates’ talk of attitudes and correctness here, like his earlier talk of
restorations and fulfillment, offers us a potentially helpful way to under-
stand the significance of the Genesis Theory.
Since both of these approaches to the theory are supported by the text, I
think we should feel free to draw on them if we find ourselves puzzled
about the role the theory is supposed to play in the Aiming Argument. As
we shall see, both approaches illuminate the argument in interesting ways.

III

Let me begin this section by offering a concise reconstruction of the Aim-

ing Argument, as I understand it:

(1) The Genesis Theory

Every pleasure is a becoming that comes to be for the sake of
some being. No pleasure is a being. (53c4-5; 54c6-7)

(2) The Value Requirement

Anything that comes to be for the sake of something else does not
belong “in the class of the good” [% : 
@ 
@ "
;]. 19 But
that for the sake of which something comes to be does belong in
the class of the good. (54c9-11; 54d1-2)

(3) Therefore, no pleasure belongs in the class of the good. (54d6-7;

54e1-2)

By endorsing the conclusion of this apparently valid argument, Socrates is

denying not just that every pleasure is good, but also that any pleasure is
good. 20 Taken at face value, however, this conclusion is both extreme and
_________
18 When describing the attitude of belief in other dialogues, Plato often uses the lan-
guage of aiming. See especially Theaetetus 193e-194a and Cratylus 420b-c. An illuminating
treatment of these and similar passages can be found in Caston 1993, 226-30.
19 Plato’s use of "
+ (allotment, portion, part, rank) here suggests that he is concerned
not so much with the one good itself as with the many things that, taken as a whole, belong
to the good or have a share in the good. This is why I prefer “the class of the good” to “the
good.”
20 Some commentators resist this. They claim that Socrates, in his exposition of the
Aiming Argument, uses language implying that he himself takes the Genesis Theory to be a
 PLATO’S ANTI-HEDONISM 133

puzzling. For it is hard to understand at first why Socrates thinks he is en-

titled to the Value Requirement, given his antecedent commitment to the
Genesis Theory. If we want to say that every becoming comes to be for the
sake of some other thing, and if we want to say that this other thing is
good, then presumably we need to say that every becoming has value of
some kind—even if this kind of value is somehow inferior to, or derivative
of, the kind of value the other thing has. Perhaps shipbuilding is not as
good as a ship; but if a ship is good, then so is shipbuilding. Perhaps be-
coming fulfilled is not as good as being fulfilled; but if being fulfilled is
good, then so is becoming fulfilled. And so on. As these examples indi-
cate, it would be utterly unreasonable for Socrates, given his own com-
mitments, to claim that becomings have no value of any kind. He has to
admit that becomings have value, even if only of an instrumental, subsidi-
ary, or otherwise derivative kind. So why does he deny here that becom-
ings are any good at all?
The most charitable way to interpret the Value Requirement, I think, is
to assume that by “the class of the good” Socrates really means the class
of ends in themselves. This would jibe nicely with an earlier discussion in
the Philebus, where Socrates claims that “the class of the good” (*
 
@ "
+) is “complete [and] sufficient,” such that “anyone who
recognizes it aims at and pursues it, wishing to get hold of it and possess it
for his very own, and cares for nothing else except what is accomplished
along with goods” (20d1-10). This passage strongly suggests that “the
class of the good” contains all and only those things that are worth pursu-
ing as ends in themselves. 21 For these are the things worth caring about
not because they are accomplished or acquired along with other (and bet-
ter) things, but just because they are what they are. (In what follows I will
call these things perfect goods.)

_________
mere hypothesis, and not really something to be believed. As these commentators point out,
Socrates twice refers to the theory with the conditional phrase “if in fact pleasure is a be-
coming” [4$
  .  %] (54c6; 54d1). Hackforth and Carone claim that the
use of . here is a sign that Plato—because he is skeptical of the Genesis Theory—insists
only on the validity of the Aiming Argument, not on its soundness (Hackforth 1972, 105-6)
(Carone 2000, 205). But this reads too much into the phrasing. The use of . in the con-
text of a deductive argument can often function merely as a reminder of the argument’s
logical structure. Socrates might just as well be emphasizing the role of the Genesis Theory
as a premise in the overall argument. (See Aristotle, De Caelo I 24 for a good example of
this.) His language here yields no good independent reason to suspect that Plato is withhold-
ing his assent from any of the argument’s premises.
21 For more detailed discussion of the standards of “completeness” and “sufficiency”
see Bury 1897/1973, 211-14; Cooper 2004, 270-8; and Richardson Lear 2005, 53-9.
134 MATTHEW EVANS

So, if “the class of the good” has more or less the same meaning across
the dialogue, then the Value Requirement is not all that jarring. It amounts
to the potentially substantive claim that no becoming is a perfect good
because every becoming is “for the sake of” something else. On this read-
ing, the conclusion of the argument—though still radical—is not quite as
radical as it might have seemed at first. In claiming that no pleasure is
good, Socrates is not claiming that no pleasure is any good at all. Indeed,
as the Genesis Theory suggests, this is pretty nearly the opposite of what
he thinks. He is claiming only that no pleasure is a perfect good.
Although this interpretation of the Value Requirement saves Socrates
from incoherence, it also saddles him with an apparently dubious premise.
If the Value Requirement is intended to be a substantive claim, rather than
a merely stipulative one, we need some independent reason to believe that
nothing worth pursuing as an end in itself could be “for the sake of” some-
thing else. For this claim seems easy to deny, at least on its face. A dedi-
cated shipbuilder might admit that shipbuilding is in every case “for the
sake of” a ship, but still insist that shipbuilding is worth pursuing as an end
in itself; a dedicated gourmand might admit that eating is in every case
“for the sake of” being fulfilled, but still insist that eating is worth pursu-
ing as an end in itself; and so on. If these responses are coherent—as they
seem to be—then they show how easy it is to challenge the Value Re-
quirement. Even on the assumption that the Genesis Theory is true, we can
apparently wonder whether—and if so, why—pleasures are not perfect
goods. So unless Socrates has a way to deal with this challenge, his Aim-
ing Argument will fail.

IV

I believe that Socrates does have a way to deal with this challenge. For he
gives what I take to be a further argument—I’ll call it the Supplementary
Argument—in support of the Value Requirement. His apparent strategy
here is to show that it is irrational to pursue any becoming of any kind as
an end in itself; and from this he infers, not unreasonably, that no becom-
ing of any kind is worth pursuing in that way. The Supplementary Argu-
ment begins with a nod to the author of the Genesis Theory, and then un-
folds as a kind of psychological diagnosis of those who pursue pleasure in
the wrong way:
S: So, just as we said at the beginning of this argument, we should be grateful
to the person who showed us, about pleasure, that it is a becoming, and that it
 PLATO’S ANTI-HEDONISM 135

has no being at all. And it is clear that he just laughs at those who claim that
pleasure is good.
P: Most definitely.
S: And this same person will also laugh on every occasion at those who find
an end in becomings (7  % +  

"
 ).
P: Why? And what sort of people do you mean?
S: I mean those who, when curing hunger or thirst or anything that a becom-
ing cures (9  %'M), are pleased on account of the becoming inso-
far as it is itself a pleasure (
 $! *  1 4$
<
˜
#<) and claim that they would not agree to live without thirsting and hun-
gering and experiencing all the effects that follow upon [thirsting and hunger-
ing].
P: They’re likely to, at any rate.
S: But wouldn’t we all say that destruction ( ) is the opposite of
becoming (D  )?
P: Necessarily.
S: So someone choosing this life would choose destruction and becoming, but
not that third life, in which there is neither being pleased nor being pained,
but only thinking in its purest possible form.
P: Well, Socrates, it seems that a great absurdity ( 
) follows if one holds
that pleasure is good. (54d4-55a11)
The target of Socrates’ attack here is the person who would reject every
pleasureless life available—even the painless life of purest thought—
simply because that life is pleasureless. Let’s call this person the pleasure-
lover. According to Socrates, the pleasure-lover’s decision is to be ex-
plained in part by the fact that he “finds an end in becomings.” But what is
it for a person to “find an end” in something, as Socrates uses this phrase?
Clearly part of it is for this person to take pleasure in that thing (54e3-6);
and another part, I propose, is for this person to pursue that thing as an
end in itself. This would explain why Socrates thinks the pleasure-lover
would reject the pleasureless life of pure thought, even though this life is
utterly free from destruction and pain. The pleasure-lover would reject this
life because it lacks things that, in his view at least, make a life worth liv-
ing.
The pleasure-lover’s mistake, on this reading, is to treat pleasures as
ends in themselves when deciding how to live. 22 But why does Socrates

_________
22 If Socrates is consistent throughout the Philebus, then he does not think that the
pleasure-lover’s mistake consists simply in rejecting the pleasureless life in favor of the
136 MATTHEW EVANS

think this is a mistake? His primary claim seems to be that treating pleas-
ures as ends in themselves is tantamount to treating becomings as ends in
themselves; and this entails placing value on destruction, he thinks, since
destruction is a precondition for becoming. He supports this claim with
some examples, each of which stems from the Equilibrium Theory as ap-
plied to the pleasures of eating and drinking. If you treat eating (when
hungry) and drinking (when thirsty) as things to be pursued as ends in
themselves, then—according to Socrates—consistency demands that you
treat hunger and thirst as things to be pursued; for you can eat (when hun-
gry) and drink (when thirsty) only if and only when you are hungry and
thirsty. To put the point more generally, those who pursue becoming ful-
filled as an end in itself must also, if they are rational, pursue not being
fulfilled. Since defect and destruction are the conditions of possibility for
becoming, and since you are rationally required to pursue the conditions of
possibility for what you pursue as an end in itself, you are rationally re-
quired to pursue defect and destruction if you pursue becoming as an end
in itself. This is why Socrates thinks that the pleasure-lover, insofar as he
is a lover of becoming, must also be a cultivator of defect and destruction.
But if this line of thought alone is supposed to establish that the lover of
becoming is irrational, then so far it seems incomplete. For it gives no
grounds for thinking that the lover of becoming cannot meet all the rele-
vant standards of rationality. Even if the lover of becoming must pursue
defect and destruction in order to pursue what he loves, there is nothing
obviously inconsistent about this. The lover of becoming is in trouble, I
take it, only if he cannot consistently justify his pursuit of both becoming
and destruction. But is it clear that he cannot? Obviously he is committed
to the pursuit of opposites, but he is not obviously committed to the pur-
suit of each opposite in the same way. For although he pursues becoming
as an end, he does not pursue destruction as an end; he pursues destruction
only as a condition of possibility for pursuing something else. If we as-
sume that destruction is indeed something that makes becoming possible,
and if the lover of becoming pursues destruction only as something that
makes becoming possible, then it’s not clear that the lover of becoming is
irrational in any way.
Many commentators suggest that the crux of Socrates’ criticism here is
not that the pleasure-lover is committed to pursuing opposites, but that he
_________
mixed life. For Socrates elsewhere (and often) agrees with Protarchus that the pleasureless
life would not be worthy of choice for any human being, especially when the alternative is a
life that is moderately pleasant and moderately thoughtful (21d-22b; 60e-61a; and 66e-67b).
Since the pleasure-lover’s alleged mistake here cannot be his life-choice, it must rather be
the grounds on which he makes this life-choice.
 PLATO’S ANTI-HEDONISM 137

is committed to living an unsatisfiable, purposeless, or otherwise patho-

logical life. 23 On an interpretation of this sort, Socrates is denouncing
those who seek destruction as a means to becoming because they are
thereby committing themselves to a life of endless or meaningless or dis-
organized effort. In describing such a life, and in explaining why Socrates
finds it rationally unacceptable, Dorothea Frede alludes to the Greek figure
most closely associated in today’s popular consciousness with existential
absurdity: “people who choose a life of pleasure for its own sake,” she
writes, “condemn themselves to Sisyphean labors, to creating ever new
needs to have something to fulfill” (Frede 1993, lvi). Like Sisyphus, the
lover of becoming is driven to cultivate destruction and so (according to
Frede) cannot ever achieve satisfaction. From this it is apparently sup-
posed to follow that there is something substandard or erroneous or irra-
tional about the lover of becoming and his life.
But it remains to be seen whether this charge really exposes anything
more than the prejudice of the prosecutor. Taken at face value, Frede’s
criticism—when put in the mouth of Socrates—is patently question-
begging. For there are two ways in which the pleasure-lover (understood
as a lover of becoming) might be thought to lead an unsatisfiable life. If
being satisfied requires being free of desire, then of course the pleasure-
lover is leading an unsatisfiable life. Yet it is difficult to see why the
pleasure-lover would or should accept this construal of what it is to be
satisfied, since it rules out in advance the pattern of evaluation that the
pleasure-lover is proposing. In response to Frede’s challenge he could ei-
ther assert (without penalty) that he has no interest in being satisfied, or
deny that being satisfied requires being free of desire. A suitably neutral
construal of what it is to be satisfied would hold that being satisfied is a
matter of achieving or acquiring perfect goods. But then Frede (and Socra-
tes) need an independent argument to the effect that the pleasure-lover is
not already doing this. If Sisyphus believes that the only thing worth pur-
suing as an end in itself is the activity of pushing rocks up hills, then our
continuing insistence that his life is unsatisfiable would simply beg the
question against him. We need to argue, on independent grounds, that the
activity of pushing rocks up hills is not a perfect good. And Frede has not
shown that Socrates ever provides such an argument.
Is there a way to understand Socrates’ remarks such that they expose
genuine irrationality in the lover of becoming, and do not beg the question
against him? One possibility, I think, is to hear Socrates as saying not just
that the pleasure-lover is a Sisyphean figure, but also that the pleasure-
_________
23 See for example Frede 1993, lvi; Carone 2000, 268; and Russell 2005, 197-99.
138 MATTHEW EVANS

lover is rationally required to make contradictory commitments. Accord-

ing to this alternative interpretation, the Supplementary Argument is de-
signed to establish that if you commit yourself to the view that becoming
is a perfect good, then you thereby also commit yourself to the view that
becoming is not a perfect good. The lover of becoming, on this view, must
claim that the thing at which becoming essentially aims is both to be pur-
sued and not to be pursued. To see how this view might be fleshed out,
consider Socrates’ own preferred example of drinking. As the Equilibrium
Theory predicts, drinking is essentially a matter of becoming fulfilled, so
part of what it is to pursue drinking as an end in itself is to be committed
to the aim of being fulfilled. As we have seen, however, another part of
what it is to pursue drinking as an end in itself is to be committed to the
aim of not being fulfilled. So, if we assume that one is rationally commit-
ted to the aim of being fulfilled only if one is not committed to the aim of
not being fulfilled, we can derive an apparent contradiction from the com-
mitments of the lover of becoming. 24
If this interpretation is on target, then the Supplementary Argument can
be expressed more simply as follows:
For any F,
(1) If becoming F is a perfect good, then being F is to be pursued.
(2) If becoming F is a perfect good, then being F is not to be pursued.
(3) There is no G such that being G is both to be pursued and not to be
pursued.
(4) So becoming F is not a perfect good. (1, 2, 3)
This reading, unlike Frede’s, gives Socrates some substantive grounds for
claiming that the lover of becoming is irrational. For if this reading is cor-
rect, then the problem with the lover of becoming is not that he must pur-
sue destruction in order to pursue becoming, nor that he must lead a life of
endless and fruitless striving, but that he must claim that one and the same
thing is both to be pursued and not to be pursued. 25 If Sisyphus comes to

_________
24 This assumption strikes me, but not everyone, as obviously true. Here’s a brief argu-
ment for it: If S both aims at being F and aims at not being F, then S has contradictory aims.
But if S has contradictory aims, then S is irrational. So, if S is rational, then S aims at being
F only if S does not aim at not being F. Thanks to Sam Rickless for pressing me on this
point.
25 Perhaps the most obvious objection to this argument is that there are many properties,
such as being full, that qualify as counter-examples to premise (3). On a view of this sort,
being full is to be pursued when one is empty, but is not to be pursued when one is not
empty. So, if it is permissible to qualify the conditions under which something is to be pur-
sued, then obviously one and the same thing can be both to be pursued and not to be pur-
sued. One could resist this objection by claiming that the status of a given thing as to be
 PLATO’S ANTI-HEDONISM 139

love rolling his boulder, for example, then to that extent he must both
value and not value the boulder’s being at the top of the hill. Since the
lover of becoming must for this reason make contradictory commitments,
and since it is irrational to make contradictory commitments, the lover of
becoming—and indeed, anyone who rejects the Value Requirement—must
be irrational. But then the Value Requirement must be true.
Is this third way of reading the Supplementary Argument any more ef-
fective than the first two? There’s a standing general worry that, no matter
how the argument is interpreted, its conclusion is just too incredible to be
believed. For if the Value Requirement is true, then—by Socrates’ own
lights—no essentially regulated things are perfect goods. Yet there seem to
be lots of essentially regulated things that are (at least potentially) perfect
goods. Think here not only of eating and drinking, but also of writing,
learning, communicating, helping, competing, and caring. Each of these
activities is essentially a matter of aiming at something, and yet all of them
would appear to be perfect goods. Here I think Socrates must bite the bul-
let and deny that they really are. He must insist that, when we are pursuing
these activities rationally, we pursue them not as ends in themselves, but
as contributors to the pursuit of ends in themselves. But of course many of
us would find this position unconvincing.
Maybe a shift in focus would help. Up to this point we have been trying
to understand the Supplementary Argument as applied to activities like
drinking, but we have not been trying to understand it as applied to atti-
tudes like believing. As we have seen, however, there is good reason to
suppose that Socrates thinks of believing—no less than drinking—as in
every case “for the sake of” some being. So he presumably also thinks that
the Supplementary Argument explains why believing is not a perfect good.
And this is where things get interesting. For it seems much more dubious
to us that believing is a perfect good than that drinking is, and the reason
why it seems more dubious is in fact captured quite nicely by the Supple-
mentary Argument.
Remember that, according to Socrates, part of what it is to believe
something is to be subject to a standard of evaluation that is met if and
only if what is believed is true. Since meeting this standard is, for each
attitude of belief, being correct, part of what it is to be an attitude of belief
_________
pursued is not relative to persons or times. On a view of this sort—which strikes me as
Platonic in spirit if not in letter—a particular person is warranted in pursuing a given thing
at a particular time only if, and only because, that thing is absolutely or objectively worth
pursuing. More needs to be said about this, of course, but enough has been said already to
show that the objection is not devastating. Thanks to an anonymous referee for raising this
as a problem.
140 MATTHEW EVANS

is to aim at being correct. So those who pursue believing as an end in itself

are to that extent committed to believing only things that are true. But
those who pursue believing as an end in itself are also committed to not
being committed to believing things that are true. For if believing is worth
pursuing as an end in itself, then the value of believing does not depend on
the truth of what is believed. It must not matter to the lover of belief that
what he believes is not true, because if it did matter to him, then he would
not be committed to the pursuit of believing—that is, the attitude of belief
simpliciter—as an end in itself. Therefore, if believing is a perfect good,
then believing only true things is both to be pursued and not to be pursued.
We have a contradiction.
This strikes me as an interesting and possibly correct explanation of why
it makes no sense to pursue belief as an end in itself. If you were looking
for advice about how to comport yourself as a believer, and if someone
told you that you should pursue believing as an end in itself, you would
rightly regard that person as either malevolent or crazy. Surely the best
way for you to believe things is not for you to believe things indiscrimi-
nately, promiscuously, and intensely. But why not? According to Socrates,
the answer lies in the very nature of the believing attitude: the truth of
what is believed is both necessary and sufficient for the correctness of the
believing attitude, and each believing attitude essentially aims at being
correct. 26 So the Supplementary Argument, as applied to the case of belief
at least, provides a powerful explanation for why it is irrational to treat
belief itself as a perfect epistemological good.
Similar considerations might show just as well that, and why, it is irra-
tional to treat pleasure itself as a perfect ethical good. If we follow Socra-
tes’ lead and assume that pleasures are attitudes with objective correctness
conditions, then the same line of reasoning would apply. To see why,
imagine for a moment that the analogy between pleasure and belief is to be
spelled out as follows: just as an attitude of belief is correct if and only if
what is believed is true, an attitude of pleasure is correct if and only if
what pleases is good. 27 If this is right, then part of what it is to take pleas-
_________
26 It is important to notice here that the correctness conditions for an attitude are not
equivalent to the conditions under which an agent ought to adopt that attitude. Sometimes
the only rational thing for you to believe is something that is, unbeknownst to you, false. In
that case, you ought to adopt an attitude of belief that is incorrect. Likewise, it is possible for
you to be irrationally persuaded to believe something true. In that case, you ought not to
adopt an attitude of belief that is correct.
27 In an unfinished paper I argue, on the basis of evidence from the very passage we are
considering now, that this is in fact Plato’s view. On my reading, Socrates criticizes the
pleasure-lover for taking pleasure in being pleased, because it is this error that accounts for
 PLATO’S ANTI-HEDONISM 141

ure in something is to be subject to a standard of evaluation that is met if

and only if what pleases is good. Since meeting this standard is, for each
attitude of pleasure, being correct, part of what it is to be an attitude of
pleasure is to aim at being correct. So those who pursue pleasure as an end
in itself are to that extent committed to taking pleasure only in things that
are good. But those who pursue pleasure as an end in itself are also com-
mitted to not being committed to taking pleasure in things that are good.
For if pleasure is worth pursuing as an end in itself, then the value of
pleasure does not depend on the goodness of what pleases. It must not
matter to the pleasure-lover that what he takes pleasure in is not good, be-
cause if it did matter to him, then he would not be committed to the pursuit
of pleasure—that is, the attitude of pleasure simpliciter—as an end in it-
self. Therefore, if taking pleasure is a perfect good, then taking pleasure
only in good things is both to be pursued and not to be pursued. We have
another contradiction.
Interpreted in this way, Socrates’ Supplementary Argument applies
straightforwardly to any family of attitudes with objective correctness
conditions. Since the Genesis Theory holds that pleasure is an attitude of
just this sort, the Supplementary Argument provides a more intuitive ex-
planation for why every pleasure—understood now in accordance with the
Genesis Theory—falls under the Value Requirement and so cannot be a
perfect good. If this is what Socrates means when he concludes that no
pleasure is good, then there is a very clear sense in which the Aiming Ar-
gument succeeds in casting the legitimacy of all hedonic reasons into
doubt. For it shows quite effectively why someone who accepts that there
are such reasons is guilty of confusing the positive attitudes we take to-
ward things with the things toward which we take such attitudes. As Soc-
rates points out, this confusion stems from two connected intellectual fail-
ures: first, a failure to distinguish the attitude of pleasure from the object
of pleasure; and second, a failure to see that the value of every attitude of
pleasure is contingent upon the value of its object. If we can manage to
avoid committing these errors, then we will not be tempted to think that
there are any such attitudes worth having in their own right. We will make
neither the epistemological mistake of claiming that belief is a perfect
good, nor the ethical mistake of claiming that pleasure is a perfect good. 28
_________
the pleasure-lover’s mistaken belief that some pleasures are perfect goods. In his view, sec-
ond-order pleasure carries a commitment to the perfect goodness of pleasure. And if this
view is generalized to cover all pleasures, not just second-order ones, it entails that every
pleasure carries a commitment to the perfect goodness of its object.
28 Compare Elijah Millgram’s recent critique of Hedonism: “Practical reasoning tends to
take one from a position of lesser pleasure to a position of greater pleasure…. Some phi-
142 MATTHEW EVANS

To be sure, this interpretation of the Supplementary Argument draws on

resources from the Philebus that outstrip what is available in the text of the
argument itself. As I have tried to suggest, however, these resources are
worth exploiting if only because the text of the argument is cryptic and
compressed. In order to make sense of the reasoning behind what Socrates
says here, some interpolation is required. And my reading, unlike some,
credits Socrates with an argument that is neither question-begging nor phi-
losophically weak. On the contrary, it credits him with an argument that is
both substantive and plausible—assuming, of course, that the Genesis
Theory is true, and that the analogy between pleasure and belief is apt.
Moreover, it shows how the Supplementary Argument reinforces the Aim-
ing Argument’s most vulnerable premise. (According to that premise, re-
member, no becoming is worth pursuing as an end in itself.) Without some
further argument in hand, we have no good reason to think that this prem-
ise—which I have been calling the Value Requirement—is true. But if my
interpretation of the passage as a whole is correct, then the Supplementary
Argument provides that missing reason: no becoming could be a perfect
good because, if it were, then the being at which it aims would have to be
both worth pursuing and not worth pursuing. So my interpretation not only
attributes to Socrates an interesting pair of arguments, but also provides a
way to see how the second argument supports the first.

I would like to conclude now with a very brief discussion of three different
ways in which Plato’s overall account, as I have interpreted it, might be
resisted. Though I suspect that Plato has a decent chance of overcoming all
three, I do not expect to vindicate his view here. I wish only to show that
the Aiming Argument, if charitably interpreted, is not as easy to resist as
first impressions might suggest.
_________
losophers have noticed this tendency, and concluded that pleasure is one’s sole and neces-
sary goal. In this they could not be more mistaken. Hedonists err in roughly the way that
someone who thinks that the goal of enquiry is to maximize conviction might err. Normally,
one’s enquiries tend to take one from a position of lesser conviction to a position of greater
conviction…. However, … one’s goal is not conviction: one’s goal is truth. Conviction is
epistemically important as a guide to truth, but conviction per se is not the object of my
efforts” (Millgram 1993, 397-401). Obviously Millgram’s account is akin to the one I am
attributing to Plato, but there are some important differences. One worth mentioning here is
that Millgram thinks of pleasure as a feeling (like conviction) rather than an attitude (like
belief). He is sensitive to pleasure’s qualitative or phenomenal character in a way that Plato,
in my view, is not.
 PLATO’S ANTI-HEDONISM 143

One immediate objection to Plato’s overall account is that it cannot ac-

commodate Eudoxus’ (and Anscombe’s) observation concerning the ap-
parent rational authority of pleasure. If Plato is right in thinking that pleas-
ure is not to be pursued as an end in itself, then he needs to explain why no
one is inclined to ask anyone what his end is in being pleased. Eudoxus
explains this by claiming that pleasure is, for each of us, a perfect good.
But if this explanation is wrong, as Plato clearly thinks it is, then what
explanation can he offer as an alternative? To answer this question, let’s
assume for the moment that pleasures aim at being correct, and that pleas-
ures succeed in being correct just in case their objects are perfect goods. I
think we can, on the basis of this assumption, reconstruct an interesting
response on Plato’s behalf. For if this assumption is true, then pleasures
have a significant guiding role in our deliberative economy: their function,
in effect, is to pick out things worth pursuing as ends in themselves. This
suggests a subtly different explanation not only for the truth of Eudoxus’
observation, but also for the invalidity of his inference. Eudoxus might be
right to think that, when you assert that you are pursuing something just
because it pleases you, you are asserting that there is a non-derivative rea-
son for you to pursue it. But Eudoxus is wrong to think that this assertion
commits you to the existence of hedonic reasons. For there are (at least)
two ways to understand what is being asserted here. You are making either
the stronger claim (A) that your enjoyment of this thing gives you a non-
derivative reason to pursue it, or the weaker claim (B) that your enjoyment
of this thing gives you a reason to think that there is a non-derivative rea-
son for you to pursue it. According to Plato, (A) is false but (B) is true:
your taking pleasure in something is a (fallible, subjective) sign that the
thing you are taking pleasure in is a perfect good. This suggests a reason-
able alternative explanation for why no one is inclined to ask you what
your end is in enjoying something. For your saying that you enjoy some-
thing is equivalent, in effect, to your saying that you take that thing to be
an end in itself.
A second objection to Plato’s account is that the Genesis Theory, even
in its most sophisticated form, is false. Many philosophers are inclined to
think that pleasures are not attitudes but sensations, or phenomenal mental
states, or phenomenal features of mental states. On this view, we pursue
pleasure as an end in itself because it is, in itself, a feeling we just like to
have. If this view is true, then the Genesis Theory is false. But this view is
not obviously true. To see why, consider the distinction between an
agent’s attitudinal and sensory pleasures: for any agent S, let S’s attitudi-
nal pleasures be S’s intense and unreflective attitudes of liking, and let S’s
sensory pleasures be phenomenal states that S intensely and unreflectively
144 MATTHEW EVANS

likes to be in. 29 Since there are significant discrepancies between the likes
and dislikes of various agents, and since some of these likes and dislikes
will no doubt take phenomenal states as their objects, the distinction be-
tween attitudinal and sensory pleasures is well-motivated. Once we accept
this distinction, however, we find ourselves inclined to say that pleasures
are really attitudes, not phenomenal states. For if the distinction is genu-
ine, then sensory pleasures are pleasures only if, and only because, they
are the objects of attitudinal pleasures. But in light of the previously men-
tioned discrepancy in likes and dislikes, it seems to be a contingent matter
whether or not any given phenomenal state is the object of any given
agent’s attitudinal pleasure. And in that case it seems reasonable to indi-
viduate pleasures by their attitudinal features rather than their phenomenal
features. 30 Whether this argument is conclusive or not, it is certainly
strong enough to protect the Genesis Theory from a curt dismissal.
The last objection I want to consider here is a reply to Socrates’ charge
that liberals like Eudoxus are stuck having to claim that attitudes of pleas-
ure are worth pursuing as ends in themselves. It seems to me that liberals
would be justified in claiming that this is not an implication of their view.
For there is an important distinction to be drawn between two different
liberal positions: the first holds that attitudes of pleasure are worth pursu-
ing as ends in themselves; the second holds that objects of pleasure are
worth pursuing as ends in themselves just because they are objects of
pleasure. The Value Requirement seems to rule out the first view, but it
seems not to rule out the second. Moreover, the second view strikes many
of us as intuitive: we often think that some things are worth pursuing as
ends in themselves just because we enjoy doing or having them. One ob-
vious problem with this view is that pleasures understood as attitudes often
do appear to have objective conditions of correctness. Few of us would be
inclined to say, for example, that one’s taking pleasure in acts of cruelty or
torture is sufficient to make these acts perfect goods. On the contrary,
cases such as these incline us to think that there are facts about the objects
of pleasure that make the pleasures taken in these objects more or less cor-
rect or incorrect. But if that is so, then attitudes of pleasure are not so
much value-makers as value-trackers. Your taking pleasure in something
may give you a reason to think that the thing you are taking pleasure in is
a perfect good; but your taking pleasure in that thing does not make it a
perfect good. To put the point in terms familiar from the Euthyphro, it is
_________
29 For more on this distinction, see Feldman 2004, 55-90.
30 For further discussion of this point, and a slightly different formulation of this argu-
ment, see Evans 2007, 90-3.
 PLATO’S ANTI-HEDONISM 145

not because we enjoy these things that they are good; it is because they are
good that we (ideally) enjoy them. The claim that pleasures make their
objects good is, for Plato, no less incoherent than the claim that beliefs
make their objects true.
The dispute does not end here, of course. Liberals have plenty of re-
sources with which to carry on their resistance to Plato’s attack. I myself
suspect that Plato can ultimately overcome this resistance, but I cannot
hope to support my suspicion in any further detail here. I can only hope to
have brought Plato’s view back into play as a fierce competitor in the long
battle over the ethical status of pleasure. 31
NEW YORK UNIVERSITY

_________
31 Many thanks to John Cleary, Don Garrett, Jim Hankinson, Liz Harman, Phillip Mit-
sis, Michael Pakaluk, Sam Rickless, Nishi Shah, David Sosa, Sharon Street, Christie Tho-
mas, and Steve White for critical comments and advice. Thanks also to an anonymous refe-
ree, whose disagreements were especially helpful. Special thanks to Verity Harte for all of
that and patience.
 COMMENTARY ON EVANS

VERITY HARTE

ABSTRACT
The Reply is focused on the two central interpretative and philosophical
claims of Evans’s paper: that the ‘Aiming Argument’ of Philebus 53c4-54d3
has as its target ‘Liberalism’; and that it constitutes a powerful argument
against this view. With regard to the latter, it is argued that the supplementary
material that, it is agreed, must be added to the Aiming Argument, if it is to
succeed against Liberalism, must be understood as a robust claim about the
origin of value. Further, the ‘Equilibrium Theory’ of pleasure of Philebus
31b-36c, no less than the discussion of false pleasure, may be understood in
such a way as to support such a claim, and, when understood in this way, the
claim can indeed be seen to cover all the dialogue’s examples of pleasures,
including the so-called ‘true pleasures’ of Philebus 50e-55c.

Professor Evans’s excellent paper is a welcome boon to those of us con-

cerned to champion the structural coherence and philosophical value of
Plato’s Philebus against its sometime critics. And it poses a challenge,
albeit a welcome and interesting one, for the dutiful respondent. For it de-
fends a conclusion I applaud and with which I have much sympathy: that
what Evans calls ‘the Aiming Argument,’ at Phil. 53c4-54d3, together
with a supplementary argument identified as running from 54d4-55a11, far
from being an untidy and poorly connected appendix to the discussion of
pleasure, constitute its well-placed conclusion. Evans argues that this ar-
gument offers a philosophically powerful challenge, not only to the Eu-
doxan-style hedonism that is the dialogue’s immediate target, 1 but also to
a weaker Liberal position, which has many modern adherents. Further,
Evans proposes, although he does not claim to argue fully here, that this
Platonic challenge is not just valid, but sound.
The duty of a respondent is, nonetheless, clear. I will focus my questions
on the most central and striking of the paper’s interpretative and philoso-
phical claims: that the Aiming Argument has as its target the view identi-
fied as Liberalism, and that it constitutes a powerful argument against this
view.

_________
1 Evans does not claim that Eudoxus is the explicit target of the Aiming Argument.
However, if the ‘for the sake of’ language in EN X.2 stems from Eudoxus, not Aristotle, the
linguistic relation does make it tempting to put this Eudoxan argument and the Philebus
argument into some, direct relation.
 COMMENTARY ON EVANS 147

The first question I shall consider is: what is the explicit conclusion of
the argument at Phil. 53c4-54d3? Evans characterizes it as ‘the first argu-
ment that converts Protarchus to the view that no pleasure is worth pursu-
ing as an end in itself’ (p.125). But I think there is reason to pause to
evaluate both this way of characterizing its conclusion and the claim that it
is the first to convert Protarchus to it. The conclusion of the argument that
is drawn explicitly, at 54d1-2, is that pleasure, on the basis of its being a
becoming or genesis, cannot be placed ‘in the lot or rank of the good’ (*

@ 
@ "
+, d2), on the basis of which it is agreed that one who
holds this view of pleasure would find the view that pleasure is (the or a)
good ridiculous (54d6-7; and cf. 55a9-11). The position ridiculed—that
pleasure is (the or a) good—complete with its in-built unclarity is, of
course, precisely the position that Protarchus inherited from Philebus at
the start of the dialogue and of which he is the appointed champion (see
11b4-6, with, e.g., 13b7). This thesis faced its first, and, it appeared by
Protarchus’ lights, already successful challenge, in the Choice of Lives
Argument, at Phil. 20c8-22c4, prompting the reformulation of the dia-
logue’s competition between pleasure and reason as a contest for second
prize. (For this thesis as target for this earlier argument, see, in particular,
20b6-c2 and 22c2. For Protarchus’ apparent acquiescence, see 21e3-4 and
22b9. Even Philebus does not challenge the defeat of this thesis directly,
content to observe that Socrates’ thesis is in the same boat, 22c3-4. 2 ) The
argument, then, does not present itself as being the first to draw its explicit
conclusion (nor to persuade Protarchus of it). But it is, we might agree, the
first to do so on the basis of a claim about the nature and normative status
of pleasure.
The connection between the Aiming Argument and the Choice of Lives
Argument does not go unnoticed by Evans. And the connection is rein-
forced by a fact that he notes, that this earlier argument was the first to use
the striking phrase ‘the lot or rank of the good’ (at 20d1), repeated in the
conclusion of the Aiming Argument at 54d2. It is to an interpretation of
the meaning of this striking phrase that Evans appeals in explaining how
an argument whose explicit target is the strong and admittedly implausible
hedonist thesis of Philebus may be characterised as an argument against
the weaker and considerably more plausible thesis of Liberalism. Evans
argues that, on a correct understanding, the claim that pleasure is not in the
lot or rank of the good amounts to the claim that pleasure is not a final
good, that is, that it is not among things worth pursuing as ends in them-

_________
2 Given the nuance with which Socrates initially stated his thesis, as a comparative
claim (11b6-c2), this observation seems not to be true.
148 VERITY HARTE

selves. And this counts against Liberalism, understood as the view that
pleasure is, not the only, but one among things that are (at least defeasibly)
worth pursuing as ends in themselves.
This seems to me an interesting and fruitful suggestion as to how best to
understand the, apparently shared, conclusion of both the Choice of Lives
and the Aiming arguments; as I think he would agree, it is a conclusion in
need of more interpretative support than Evans was in a position to give it
here. Taking this as given, my second question concerns the success of the
argument, being so characterized, as a strike against Liberalism.
The argument, as interpreted, combines two claims that do not of neces-
sity go together: first, the claim that every pleasure is a becoming or gene-
sis, understood as the claim that every pleasure is chosen for the sake of
something else; and, second, the claim that no pleasure is a being, under-
stood as the claim that no pleasure is pursued as an end in itself. The first
claim does not necessitate the second, providing it is possible for some
thing to be chosen both for its own sake and pursued for the sake of some-
thing else. But this does indeed seem to be possible. Aristotle allows that
things may be chosen both for their own sake and for the sake of some-
thing else (in particular for the sake of eudaimonia); indeed he includes
pleasure among things of this sort (EN I.7, 1097b3-4). 3
An argument against Eudoxus can safely ignore this option, since Eu-
doxus is presented as denying that any pleasure is chosen for the sake of
something else. An argument against Liberalism, however, clearly cannot
so freely ignore it. For Liberalism states only that pleasure is (one among)
things that are worth pursuing as ends in themselves. And this leaves open
whether pleasure is, in addition, chosen for the sake of something else.
The distinction between Eudoxus and Liberalism to which I here appeal
is, we may note, somewhat different from the distinction between Hedon-
ism and Liberalism that is drawn in the paper. The distinction between
Hedonism and Liberalism turns on whether pleasure is the only thing pur-
sued as an end in itself, or whether there might be other such ends as well.
Of course, it follows from the Hedonist picture that pleasure, when pur-
sued as an end in itself, could not, in addition, be pursued for the sake of
anything else, unless for some further pleasure.

_________
3 It is, of course, a complicated question how to understand the relation between the
likes of pleasure and happiness, in Aristotle’s view or in general. See, for recent discussion
of this question, Richardson Lear 2004.
 COMMENTARY ON EVANS 149

In general, this distinction between Eudoxus and Liberalism seems clear

in the paper. 4 Notice, however, that the weaker position comes from the
positive thesis with which Aristotle reports Eudoxus as having concluded:
‘that pleasure is choiceworthy in itself.’ The stronger position comes from
the negative thesis with which he reports Eudoxus as having begun: that
pleasure is the sort of thing that ‘we choose neither because of, nor for the
sake of, anything else.’ And, whereas Liberalism is generally framed in
the positive, there is, at least, one place in which the modern Liberalist
consensus is framed as though taking the stronger, negative view (p. 124
‘... most philosophers today’).
Read in this stronger, negative way, I find the intuition upon which Lib-
eralism would be based to be less than persuasive. According to the
Anscombe intuition from which we begin, once having declared myself to
be enjoying myself reading, to ask for ‘some further end [I] have in enjoy-
ing [myself reading] would seem unreasonable if not bizarre’ (p. 121). But
what does this mean? If it means that the question to ask myself is what
further end I have, not in reading, which I enjoy, but in enjoying reading,
then it does seem not only unreasonable, but also bizarre. But this, I think,
is because of what it reveals, not about the ends that govern our choices,
but about the arena for choice. Only in rather unusual circumstances is
enjoying something a thing one actively chooses to do or not; hence en-
joyment as such is not an activity whose rationale it makes sense to ask
someone about. If, on the other hand, the question to ask is what further
end I have in reading, which I enjoy, or what further end I have in enjoya-
bly reading, the question seems neither unreasonable, nor bizarre. I may
choose to read, which I enjoy, both for its own sake and because half an
hour’s enjoyable reading helps me sleep at night. In this case, the
Anscombe intuition persuades, I think, only because the pragmatic context
of the question and answer suggest that the occasion envisaged is rather
one in which I have as a matter of fact chosen the activity I enjoy just for
its own sake, so the question ‘why do you do it?’ has been asked and an-
swered. This view that pleasurable activity can, but need not, be chosen
just for its own sake bring us back to the weaker Liberalist reading.
If Liberalism does leave open the possibility of something being pur-
sued both for itself and for the sake of something else, then the Aiming
Argument does not succeed as an argument against it, taken by itself. That
this is Evans’s understanding of Liberalism is supported by the fact that it
would appear to be precisely this failure that the supplementary argument
_________
4 See, in particular, Evans’s n. 2, where a variety of possible stronger positions are at-
tributed to Eudoxus, but distinguished from the weaker position about to be identified as
Liberalism.
150 VERITY HARTE

is intended to address (see p. 134 especially). The strategy is to show that,

in identifying pleasure as a becoming, the argument has the strength to
show that it would be in some way incoherent to pursue such a thing as an
end in itself. Notice that, while the supplementary argument comes in the
lines immediately following the Aiming Argument, its force comes from
the interpretations of the thesis that pleasure is a becoming that have been
taken to ground the Aiming Argument itself. I wonder whether these are
everywhere interpreted strongly enough to do the work.
Evans takes the thesis that pleasure is a becoming to be correctly under-
stood as the view that, as a becoming, pleasure ‘is essentially regulated by
some being’ (p. 129, emphasis author’s); that is, ‘pleasure is essentially
subject to standards of evaluation that are fixed by the nature of some be-
ing’ (p. 129). In identifying support for this picture of pleasure, Evans
draws on two aspects of the dialogue’s preceding discussion of pleasure:
(i) the Equilibrium Theory, according to which ‘pleasures are processes by
which animals restore themselves to their optimal equilibrium states (31b-
36c)’ (p. 129) and (ii) Socrates’ defence of the view that some pleasures
are false, from which it emerges that pleasures, like beliefs, have objective
standards of correctness (pp. 131-132). In the end, it is this second view
that Evans takes to produce a successful reading of the supplement needed
for the Aiming Argument. My question is: what makes it work on this oc-
casion?
What does the work, I think, is a view about the origin of value. 5 Thus,
in the parallel case of belief, Evans argues, ‘if believing is worth pursuing
as an end in itself, then the value of believing does not depend on the truth
of what is believed’ (p. 140). But this, he persuasively reasons, is precisely
what shows that it does not make sense to pursue belief as an end in itself.
The value of belief in fact depends on something external to belief itself,
what makes it true. By the same token, he proposes (and elsewhere, it is

_________
5 As a referee points out, this strategy raises a question—both about the original argu-
ment and about the modifications to it that I propose—as to whether claims about the origin
of value (in effect, about its intrinsic or extrinsic character) can be legitimately used to
ground conclusions about finality (about whether a thing is worthy of pursuit for its own
sake), for the relation between final and intrinsic goodness is not straightforward. (See the,
now classic, discussion of this distinction in Korsgaard 1983.) Licence to make intrinsic
goodness a condition on finality is granted by Evans’s explicit connecting of them (cf. n. 3);
hence, I follow him here. The legitimacy of so connecting them—as a matter of fact or as a
matter of the interpretation of Plato—is a substantive question which there is not space to
pursue here. We may note, however, that considerations arising from dissatisfaction with the
treatment of goodness in Plato’s Republic led Bernard Williams to propose that final goods
be taken to be a species of intrinsic goods, though not to propose this as an interpretation of
Plato’s own, later view (Williams 2006: 135).
 COMMENTARY ON EVANS 151

noted, argues), the Philebus offers a picture of pleasure in which the value
of pleasure originates in something external to it, the goodness of what
pleases (p.140, with n. 27).
This is a conclusion with which I have considerable sympathy, both in
general and as a reading of the dialogue’s discussion of false pleasure. 6
And it is one, suitably strong reading of what is involved in the view of
pleasure as something essentially regulated. But it is not a reading one
finds consistently throughout the paper. And this leads to three, I believe,
unnecessary problems or missed opportunities.
First, Evans proposes that Socrates must deny that any activities that can
be characterized as being ‘essentially regulated’, activities including ‘writ-
ing, learning, communicating, helping, competing, and caring’ (p. 139),
can be rationally pursued as ends in themselves. But, while it is true that
these are activities that are governed by norms that must be satisfied in
order for the activity to be successfully enacted, I don’t see that it follows
from this that they are activities whose value is sourced in something ex-
ternal to themselves. On the stronger reading of essential regulation, only
an external source of value would make this a bullet Socrates must bite. 7
Second, since the stronger reading emerges clearly only from the second
of the passages drawn on as support for the dialogue’s picture of pleasure,
the discussion of false pleasure, it is left uncertain how the Equilibrium
Theory fits in. But the Equilibrium Theory, too, can I think be read in a
way more obviously aimed at the stronger story. Consider the four exam-
ples offered when Socrates elaborates the Equilibrium Theory:
Hunger, presumably, is dissolution and pain?—Yes.—Whereas eating, the
filling occurring in return, is pleasure?—Yes.—And thirst, in turn, is destruc-
tion and pain, whereas the process that once again fills with liquid what has
been dried out is pleasure. And the separating and dissolving that occurs con-
trary to nature, the afflictions of stifling heat, are pain, whereas the restoring
and chilling in accordance with nature is pleasure.—Certainly.—And the co-
agulating of liquid occurring contrary to the nature of an animal as a result of
frost is pain, whereas the path of their returning and separating in accordance
with nature is pleasure. (31e6-32a8)
Notice that, in the last two pairs of pain-pleasure examples—the pain of
overheating and the pleasure of being cooled; and the pain of freezing and
the pleasure of thawing out again—Socrates gives us examples in which a
process of the same type—heating or cooling—is offered both as an ex-

_________
6 See Harte 2004.
7 It may be that, for some of these activities, a case could be made for the stronger read-
ing; my point is that this is the case that would need to be made, and, being made, might
ease the bullet’s digestion.
152 VERITY HARTE

ample of something painful and as an example of something pleasant. 8

This choice of examples is not, I think, accidental. The examples are cho-
sen to demonstrate that heating or cooling aren’t pleasant—from the he-
donist’s point of view, aren’t valuable—just in virtue of themselves; cool-
ing, for example, is pleasant and valuable just when it is the process of
being cooled when one is hotter than one ought to be, than it is good for
one to be. The value of heating or cooling—and of the pleasures that they
sometimes are—depends on the value of the harmonious constitution that
they restore. 9 In this way, Socrates makes a first move (as yet no more
than this) towards his diagnosis of pleasure as something whose value de-
pends on something external to itself. 10
This second point was made in the interests of aligning, rather than al-
ternating, the two sources of support from the dialogue’s preceding discus-
sion of pleasure on which Evans has drawn. My third and final point asks
whether, on the stronger reading, we might bring into view yet a third
source of support and close what might otherwise seem to be a gap in the
paper. Recall that Evans’s account of the rationale for the occurrence of
the Aiming Argument just when and where it does is ‘to establish that all
pleasures—even the pure ones—have a suspect feature that makes them
unworthy of a certain kind of pursuit’ (p. 126, emphasis author’s). These
pure pleasures, discussion of which immediately precedes the Aiming Ar-
gument, include such pleasures as the pleasure of the sound of a pure tone,
of the visual experience of a pure colour or shape; the pleasures of learn-
ing (when not preceded by a hunger to learn).
It has yet to be shown how the Aiming Argument might be thought to
apply to these pure pleasures. Further, with the exception of learning, it is
not immediately obvious what it would mean to describe these experiences
as ones that are regulated, essentially or otherwise. But it easier to see how
they might be described as experiences whose value depends on some-
thing external. True pleasures are appreciations of a beautiful object, ap-
preciations because of the beauty of the object appreciated. As such, true
pleasures are comparable to the only example that Socrates, as opposed to
_________
8 It is offered as such, we may notice, in partial defiance of the phenomenology of the
experiences in view; the onset of frostbite is, as I understand it, typically characterized by a
loss of sensation, not by pain; and the process of restoration to normal body temperature and
return of sensation would characteristically be painful, not pleasant.
9 This framework is supported by the preceding general discussion of the constitution of
mixtures.
10 Here, the pleasures in view are pleasures in the sense of that in which pleasure is
taken. That the value of that in which pleasure is taken should determine the value of the
pleasure taken therein is, in my view, among the issues in question in the subsequent discus-
sion of false pleasure (see Harte 2004).
 COMMENTARY ON EVANS 153

Protarchus, provides in the course of presenting the Aiming Argument, the

relation between lover and beloved (53d9-10).
My remarks under this heading are intended to encourage Evans’s ar-
gument further in a direction in which he is already inclined to take it; in
the service of the view that the Philebus’ discussion of pleasure forms a
coherent, carefully structured, progression, which culminates in the Aim-
ing Argument. In regarding the discussion this way, it helps to view the
preceding stages, not as an attempt to provide an exhaustive examination
of pleasure, but as putting in place the materials for pleasure’s eventual
diagnosis. Evans’s paper is a welcome contribution to the understanding of
the philosophical power and significance of this diagnosis. 11
YALE UNIVERSITY

_________
11 Thanks to the BACAP consortium for the invitation to participate in this occasion and
to Michael Pakaluk for his kindness and generosity as host. Particular thanks to Matt Evans
for his stimulating paper and for helpful discussion both during and after the BACAP ses-
sion; and to an anonymous referee for some very useful criticism to which I fear I have not
been able to do full justice within the constraints of the format here.
 EVANS/HARTE BIBLIOGRAPHY

Anscombe, G. E. M. 1957/1999. Intention. Ithaca.

Benitez, E. 1989. Forms in Plato’s Philebus. Assen.
Brandt, R. 1959. Ethical Theory. Englewood Cliffs.
Broad, C. D. 1930. Five Types of Ethical Theory. London.
Broadie, S. and C. Rowe. (tr.) 2002. Aristotle, Nicomachean Ethics. Oxford.
Bury, R. 1897/1973. The Philebus of Plato. New York.
Carone, G. 2000. Hedonism and the Pleasureless Life. Phronesis 45: 257-83.
Caston, V. 1993. Towards a History of the Problem of Intentionality Among the
Greeks. Proceedings of the Boston Area Colloquium in Ancient Philosophy 9: 213-
60.
_______. 2001. Connecting Traditions. In Perler (ed.): 23-48.
Cooper, J. 2004. Knowledge, Nature, and the Good. Princeton.
Evans, M. 2007. Plato and the Meaning of Pain. Apeiron 40: 71-93.
Feldman, F. 2004. Pleasure and the Good Life. Oxford.
Frede, D. 1992. Pleasure and Pain in the Philebus. In Kraut (ed.): 425-63.
_______. 1993. Plato, Philebus. Indianapolis.
Goldstein, I. 1989. Pleasure and Pain. Philosophy and Phenomenological Research 50:
255-76.
Gosling, J. C. B. 1975. Plato, Philebus. Oxford.
Gosling, J. C. B. and C. C. W. Taylor. 1982. The Greeks on Pleasure. Oxford.
Hackforth, R. 1972. Plato’s Philebus. Cambridge.
Harte, V. 2004. The Philebus on pleasure: the good, the bad and the false. Proceedings
of the Aristotelian Society II, 2003-4: 111-128.
Kagan, S. 1998. Rethinking Intrinsic Value. Journal of Ethics 2: 277-97.
_______. 1998. Normative Ethics. Boulder.
Korsgaard, C. 1983. Two Distinctions in Goodness. The Philosophical Review 92: 169-
95.
_______. 1996. The Sources of Normativity. Cambridge.
Kraut, R. 1992. Introduction to the Study of Plato. In Kraut (ed.): 1-50.
Kraut, R. (ed.) 1992. The Cambridge Companion to Plato. Cambridge.
Langton, R. 2007. Objective and Unconditioned Value. The Philosophical Review 116:
157-85.
Millgram, E. 1993. Pleasure in Practical Reasoning. The Monist 76: 394-415.
Nagel, T. 1986. The View From Nowhere. Oxford.
Perler, D. (ed.) 2001. Ancient and Medieval Theories of Intentionality. Leiden.
Richardson Lear, G. 2005. Happy Lives and the Highest Good. Princeton.
Ross, W. D. 1930/1988. The Right and the Good. Indianapolis.
Russell, D. 2005. Plato on Pleasure and the Good Life. Oxford.
Shiner, R. 1974. Knowledge and Reality in Plato’s Philebus. Assen.
Sidgwick, H. 1884/2001. Methods of Ethics. Boston.
Sumner, W. 1996. Welfare, Happiness, and Ethics. Oxford.
Tuozzo, T. 1996. The General Account of Pleasure in Plato’s Philebus. Journal of the
History of Philosophy 34: 495-513.
Williams, B. 2006. Plato’s Construction of Intrinsic Goodness. Reprinted in The Sense
of the Past, ed. Myles Burnyeat, Princeton: 118-37. [Originally published in
Perspectives on Greek Philosophy: S.V. Keeling Memorial Lectures in Ancient
Philosophy 1991-2002, ed. R.W. Sharples, London: 1-18.]
 COLLOQUIUM 6
THE GOOD IS BENEFIT:
ON THE STOIC DEFINITION OF THE GOOD

KATJA MARIA VOGT

ABSTRACT
“Only virtue is good” is a key thesis of Stoic ethics, and it is perceived as
wildly implausible both by modern commentators and ancient critics. The
Stoics themselves initiate the reception of their philosophy as paradoxical.
But, at the same time, they claim that their theories are in agreement with
preconceptions, i.e., with those notions that we acquire early in life as part of
the natural development of reason. According to the Stoics, our preconception
of the good is that the good benefits. That is how the Stoics define the good—
“the good is benefit.” As they claim, their ethical theory is in agreement with
preconceptions because it fully captures this idea. This paper explores the re-
lationship between the preconception, the definition, and the conception of
the good, and offers an analysis of how the ideal agent comes to acquire the
conception of the good.

The Stoics describe the key moment of ideal development, the moment in
which the agent forms the conception of the good and understands that
only virtue is good, as an instantaneous transition, from foolishness to
wisdom. 1 Scholars have scrutinized this developmental story; it belongs to
the most famous aspects of Stoic ethics. 2 In this paper, I shall discuss what
I take to be an important, and underappreciated, component of Stoic
thought about this development: the definition of the good as benefit. This
definition, as I hope to show, is integral to the Stoics’ account of how the
ideal agent acquires the conception of the good, and central to the claim
that their ethics, as paradoxical as it may seem, is in agreement with our
preconceptions.
Like the ancient critics of the Stoics, contemporary scholars have a hard
time finding Stoic ethics plausible: the instant transition to virtue, and the
momentous recognition that everything one used to consider good—things
_________
1 Throughout this paper, I shall be concerned with the early Stoics. I shall not attempt to
discriminate between their views (even though it seems likely that many of the ideas I dis-
cuss have been formulated by Chrysippus).
2 I shall engage in particular with two recent papers: Michael Frede pursues the question
of how the Stoics can present a process of coming to acquire the conception of the good as
natural (2001). Brad Inwood discusses the Stoic emphasis on experience within the theory
of concept-formation, and asks how we can understand the acquisition of the conception of
the good in this context (2005).
156 KATJA MARIA VOGT

which according to the Stoics have value, like health, wealth, strength,
beauty, life, etc.—really are not good, because only virtue is good. We are
tempted to say that what is most prominent about the Stoic theory is, in the
words of the ancient discussion, how paradoxical it is. And surely, this is
how the Stoics themselves see their theories. They proudly advertise them
as paradoxical in a quite literal sense: as against opinion. 3
But Stoic methodology should give us pause. According to Chrysippus,
the Stoic theories are amazing. 4 But this does not mean that they aim to
revise our most fundamental assumptions. Stoic theories are in agreement
with our preconceptions, and that means, very roughly speaking, in
agreement with what, in some way, we have thought all along, merely in
virtue of having reason. It is a key aspect of Stoic methodology and epis-
temology that preconceptions are a criterion of truth. 5
But how can the Stoics meet this criterion? Presumably, preconceptions
are what we have before (ideally) turning into wise persons (‘sages’). Prior
to this transition, we call health, wealth, life, beauty, etc., good, i.e., we
seem to be deeply confused about the good, or, loosely speaking, entirely
‘on the wrong track.’ So how can our preconceptions, which must some-
how play a role in the way we—as fools—think about the good, be a crite-
rion for the theory of the good? Is it not the case that once we come to un-
derstand what is good and bad, we are adopting a wholly new outlook, one
that is really not in agreement with our previous perspective? What I hope
to show is that the Stoic theory of the good—and by implication, Stoic
ethics quite generally—in fact aims to meet the criterion of being in agree-
ment with the preconception of the good, and that the content of this pre-
conception is captured in the definition of the good as benefit.
The Stoic definition of the good has received less scholarly attention
than other aspects of Stoic thought about the good. To some extent, I sus-
pect, this is because we are still the victims of a tradition which the Stoics

_________
3 For Zeno, see Gnomologion Monac. 196 (Gnomol. Vatic. Ed. Sternb. 295) (= SVF
1.281), for Cleanthes see Arrianus, Epict. Diss. IV 1,173 (= SVF 1.619); see also Cicero,
Paradoxa Stoicorum 4.
4 In Chrysippus’ words, it is due to the exceeding greatness and beauty of the Stoic
teachings that they seem like fiction and not on the level of humans and human nature (Plu-
tarch, On Stoic Self-Contradictions 1041F).
5 According to DL 7.54, Chrysippus is “at variance with himself” when he says that pre-
conceptions and sense-perception are criteria. The variance, I think, must refer to the fact
that the Stoics are well known for emphasizing that cognitive impressions are the criterion
of truth. As I hope will emerge in the course of this paper, the thesis is not in disagreement
with Stoic epistemology. With respect to my discussion of Stoic preconceptions and defini-
tions, I am much indebted to Brittain (2005).
 ON THE STOIC DEFINITION OF THE GOOD 157

themselves initiated, a tradition which emphasizes the paradoxical side of

Stoic ethics in a one-sided way. 6 Further, the evidence on those Stoic theo-
ries which shall be central to my discussion—first and foremost, the theo-
ries of preconception and definition—is very sparse. In fact, it is a compli-
cated question whether the Stoics actually thought that there is a precon-
ception of the good. 7 While it may seem that this should obviously be the
case, trying to find an explicit statement to this effect proves difficult. 8
_________
6 Further, the most general difficulty in the study of early Stoic ethics—that we only
have fragmentary evidence—almost inevitably leads us to later authors, such as Seneca and
Cicero. Seneca and Cicero, each in their own way, seem to me to minimize Stoic emphasis
on the good as benefit. Seneca complains that those who think that the good is the useful are
those who consider wealth, horses, wine, and shoes as good. They take a cheap view of the
good (Letter 120, 2). Cicero, when he engages with Panaetius’ views, seems from the outset
of De officiis to be interested in the idea that the honorable and the useful might provide us
with two distinct types of considerations (1.9-10). It is further interesting to note that Cicero
does not seem to like the early Stoic definition of the good as benefit. He writes: “The defi-
nitions offered [by the Stoics] do differ from each other, but only very slightly; for all that,
they are getting at the same point. I agree with Diogenes who defined good as that which is
perfect in its nature. He followed this up by defining the beneficial (let us use this term for
™") as a motion or condition which is in accord with what is perfect in its nature.”
(De fin. 3.33; tr. Inwood 2005, 273-4).
7 Epictetus clearly thinks that there is a preconception of the good (Discourses 1.22.1-3
= LS 40S). He also discusses the issue of how possessing evaluative preconceptions is one
thing, and applying them to particular instances is another; according to Epictetus, it is in
the application of these preconceptions that opinion comes to be added (2.11.1-8; cf. Long’s
translation and discussion of the passage in 2002, 79-82). But we cannot take it for granted
that Epictetus is in agreement on this point with the early Stoics. M. Frede argues that a
preconception of the good is central to the Stoic theory of motivation; however, he also
mentions that Cicero’s report of early development does not talk about such a notion (2001,
75 and 78). On the Stoics’ conception of reason and evaluative notions, cf. Cooper 2004.
8 Sources on the Stoic theory of the good mostly relate to three topics: (i) The thesis that
only virtue is good, (ii) the Stoic definition of the good as benefit; (iii) the ideal develop-
ment of an agent, who, at the moment of becoming wise, acquires the conception of the
good. Stobeaus relates a variety of distinctions which at least partly do not seem to be on the
same, principal level of the Stoic theory as (i), (ii), and (iii). Some of them seem to elucidate
the minutiae of how virtue is good; e.g., virtuous moods like joy are classified as ‘goods in
process,’ and virtuous dispositions like undisturbed stability as ‘goods in state’ (on this
distinction, and similar matters, cf. Stobaeus 2.731-13; 2.58,5-15; 2.70,21-71,4). Others
seem to take up technical terms which are familiar from other philosophers, and relate them
to the Stoic theory; cf. a distinction between goods that are  and others that are

 (2.71,15-72,6). Cf. also the distinction between goods of the soul, external goods,
and those which are neither this nor that—this distinction is evidently inspired by similar
distinctions in other schools, and does not amount to much in Stoic ethics (SE, PH 3.181).
On such terminological anachronisms (a term I borrow from Sedley), cf. the observations
which Sedley (1983) makes on some other aspects of the presentation of Stoic ethics in
Stobaeus.
158 KATJA MARIA VOGT

Accordingly, after introducing a core piece of testimony in Section 1, I

shall devote Sections 2 and 3 of this paper to a detailed discussion of the
evidence on this issue, and the various considerations which make it less
straightforward than it might seem. As I shall argue in Section 4, we need
to turn to the definition of the good in order to see, via the relationship
which the Stoics think holds between preconceptions and definitions, how
they think about the preconception of the good. Its content is that the good
benefits. If this is correct, the Stoics seem to be able to claim that their
ethics is in agreement with the preconception of the good. In Sections 5
and 6, I then turn to two objections which might be raised against my in-
terpretation.
I should add a brief remark on terminology. It has become customary to
refer to that concept of the good which is integral to knowledge of the
good as the ‘conception’ of the good, and I shall abide by this convention.
Another way in which we might refer to this concept is by calling it the
‘scientific concept’—that concept of the good that the sage, who has a
scientific understanding of things, has. My question can thus be rephrased
as asking how the preconception of the good relates to the scientific con-
cept of the good, and, along the way, to the notions of the good which
progressors might have.

I. The Good is Benefit or Not Other Than Benefit

Under the heading ‘On what is good and bad and indifferent,’ Sextus says
that according to Epicurus we cannot investigate anything without a pre-
conception, and he continues as follows:
[…] Well then, the Stoics, holding on to ‘common conceptions’ (so to speak),
define the good in this way: “Good is benefit (™) or not other than
benefit,” by ‘benefit’ meaning virtue and virtuous action, and by ‘not other
than benefit’ the virtuous human being and the friend. For virtue, which is a
disposition of the commanding-faculty, and virtuous action, which is an ac-
tivity in accordance with virtue, are, precisely, benefit; while the virtuous
human being and the friend, also being themselves among the good things,
could not be said to be either benefit or other than benefit, for the following
reason. Parts, the sons of the Stoics say, are neither the same as wholes nor
are they different from wholes; for example, the hand is not the same as the
whole human being (for the hand is not a whole human being), nor is it other
than the whole (for it is together with the hand that the whole human being is
conceived as a human being). Since, then, virtue is a part of the virtuous man
and of the friend, and parts are neither the same as wholes nor other than
wholes, the virtuous human being and the friend have been called ‘not other
 ON THE STOIC DEFINITION OF THE GOOD 159

than benefit.’ So that every good is encompassed by the definition, whether it

is benefit or not other than benefit. 9 (M11.21-24) 10
The complete definition of the good as benefit is ‘the good is benefit or
not other than benefit.’ Virtue is good, and it is benefit straightforwardly.
But persons (as a whole) do not exhibit virtue; the souls of persons can be
virtuous. The wise man and his friend—and that means, quite generally,
wise persons—are good insofar as virtue is a part of them. The distinction
between ‘benefit’ and ‘not other than benefit’ thus cautiously observes the
difference between the wise person and her virtue. 11
Sextus says that this definition is in agreement with common concep-
tions; but I shall rephrase his claims in terms of preconceptions. 12 The
most straightforward way of reading the passage is to think that, according
to Sextus, the Stoics’ definition of the good is in agreement with the pre-
conception of the good. Sextus thus implies, first, that according to the
Stoics there is a preconception of the good. He further claims that the
definition ‘the good is benefit or not other than benefit’ is in agreement
with this preconception. This seems to indicate that he assumes there is a
certain kind of relationship between a preconception and a definition: the
definition captures the content of the preconception. But it seems implau-
sible that ‘the good is benefit or not other than benefit’—a rather complex
phrase—is the content of a naturally acquired preconception. The claim
that this definition is in agreement with the preconception, however, does
not commit us to thinking that ‘the good is benefit or not other than bene-
fit’ is the content of the preconception. We might think that, for the defini-
tion to be in agreement with the preconception, it would be enough if the
content of the preconception was that the good benefits. As we can gather
from a passage in Aristotle’s Topics (VI.9, 147a34), it is a commonplace
_________
9 The text continues with a distinction between three senses in which ‘good’ is used,
which may appear to add evidence to how the Stoics distinguish between definitions and so-
called ‘delineations’ (SE, M 11.25-27). As Brittain convincingly shows, this is a misleading
impression (2005, 197-199). M 11.25-27 and related passages in Stobaeus and Diogenes
Laertius are discussed by Reesor (1983).
10 Tr. based on Bett, LS, and Bury, with changes; cf. the shorter version in PH 3.169-
171.
11 Stephen Menn alerted me to the fact that my original way of putting things implied
that the soul, rather than the virtue of the sage, is the ‘part’ which the Stoics discuss here.
12 This is not to say that there are no important differences between preconceptions and
common conceptions. I tend to think that the two notions have been conflated in later antiq-
uity, which explains why I am not attributing much importance to Sextus’s precise choice of
terminology here. Cf. Obbink (1992); Jackson-McCabe (2004, esp. 324-325); and Brittain
(2005). With respect to the Stoics’ own use of these terms, it seems that common concep-
tions are a sub-set of conceptions; cf. Brittain (2005, 177-179).
160 KATJA MARIA VOGT

in ancient discussions to say that the good benefits. This must be what
Sextus says the Stoic definition is in agreement with, and as I shall argue,
this is a plausible picture.

II. Preconceptions

The Stoics’ developmental account of how we ideally achieve virtue is

difficult to understand. A particularly vexing, but as yet little discussed,
question is this: which notion of the good does the agent have before un-
dergoing the transition to virtue, thereby acquiring the conception of the
good? 13 It is not surprising that this question has not received much schol-
arly attention. The Stoics carefully avoid all mention of the good before
getting to the grand finale of the story, the agent’s arrival at the conception
of the good. In other contexts, they do not tire of telling us that the fool
considers things like health and wealth good, rather than recognizing that
only virtue is good. But within the developmental story, there is no men-
tion of these earlier notions of the good. As we can see in Cicero, the agent
progresses from (i) early impulses for self-preservation and affiliation, to
(ii) an increasingly consistent selecting and deselecting of things of value
and disvalue, namely health, wealth, illness, poverty, etc., to (iii) the rec-
ognition that only consistency is genuinely good (De finibus 3.20-22). 14
Neither in stage (i) nor in stage (ii) is the good mentioned. The Stoics do
not portray progressors as selecting what they mistakenly consider good;
rather, they portray them as selecting things of value or natural things.
Would it not seem that the progressor selects what she considers good?
The Stoics steer away from this way of putting things in the context of the
developmental account. Perhaps they are not burdening the already diffi-
cult theory of ideal development with yet another intricate issue: the issue

_________
13 Jackson-McCabe insists on keeping two questions apart: how the conception of the
good is acquired (according to Seneca, Letter 120 and Cicero, De Fin. 3.20-22, by analogy),
and how human beings come to have a preconception of the good (2004, 336 f). More
strictly speaking, I think we should distinguish three questions: (i) how one ideally acquires
the conception of the good when turning into a sage, which, in my view, is the topic of
Cicero, De fin. 3.20-22; (ii) how one might come up with the notion of virtue being the only
good without, at this point, turning into a sage, which, in my view, is the topic of Seneca’s
Letter 120; (iii) what it means to have a preconception of the good.
14 This passage is discussed in detail in Frede (2001).
 ON THE STOIC DEFINITION OF THE GOOD 161

of keeping separate the preconception of the good from the various views
about the good which progressors come to hold. 15
In order to understand this in more detail, we need to consider some as-
pects of the Stoic theory of preconceptions. According to the Stoics, hu-
man beings acquire preconceptions (
&) in the first years of their
life. We are not born with reason; the commanding-part of the soul is like
an empty sheet of paper, ready for writing upon. Preconceptions are that
which is first ‘inscribed’ on it, and the first method of inscription is sense
perception. When we perceive something white, we have a memory of
‘white’ once the perception has departed; many memories of the same
kind constitute experience; in this way, we form the preconception of
white (Aetius 4.11.1-4 = SVF 2.83 = LS 39E). Coming to possess precon-
ceptions makes us rational. Once human beings have acquired preconcep-
tions, they begin to think. 16 More particularly, this means that we come to
have thoughts with content, and can refer to things, for example, as white.
Once we have reason, we will almost certainly undergo further cognitive
development. This involves acquiring new concepts, as well as refining
the concepts we already possess as preconceptions. Further, we shall form
opinions, such as ‘this flower is white.’ Note that it is a difficult question
what opinions do to our preconceptions. An opinion such as ‘this flower
is white’ most probably does not affect them. But what if we acquired the
opinion that flowers are made of plastic? To what extent would this opin-
ion affect our ability to apply the preconception of flowers? 17 Preconcep-
tions are acquired naturally, and nature guarantees their truth; they thus are
criteria of truth. But it might seem that we can only use a preconception as
our criterion if, first, it is not obscured by false opinions, and second, if we
know its content.

_________
15 My commentator Stephen Menn argues that there really is no need to mention any
misguided assumptions about the good because the account is an account of ideal develop-
ment. Since I do not engage in a detailed study of the passage here, I shall leave the follow-
ing open: I am not sure whether the Stoics think that the ideal development is ideal to the
extent of not involving a phase in which the agent is grown-up (thus having reason) and not
yet a sage; if so, she would never be a fool. Even if the Stoics hold such a case to be possi-
ble, I think that the ‘less ideal’ scenario which I discuss also needs explanation: that an
agent arrives at wisdom, including the conception of the good, but prior to this was a fool.
16 Cf. Brittain (2005) and M. Frede (1994). See also M. Frede (1996), Sandbach (1971),
and Doty (1976). For a recent discussion of the ontology of Stoic concepts, cf. Caston
(1999).
17 Brittain discusses whether ‘foolish inquirers’ (i.e., inquirers who are not only fools
insofar as they are not sages, but insofar as they do not make the right kinds of efforts to
gain knowledge) lose or ‘substract an element’ from the preconception (2005, 181).
162 KATJA MARIA VOGT

Among the scarce evidence on preconceptions, passages which discuss

the preconception of the gods are comparatively prominent. Epicurus
thinks the gods are not invested in the successes and misfortunes of our
daily lives. The Stoics attack this view by referring to our preconception of
the gods as provident, beneficial, and caring. Epicurus’ theology, they
claim, is not in agreement with our preconceptions. Drawing on these
texts, Bonhöffer argued that, according to the Stoics, there are only ethical
and theological preconceptions. 18 In recent decades, on the basis of a
much-improved understanding of Stoic logic and epistemology, scholars
no longer hold the view that preconceptions are limited to evaluative and
normative notions. On the contrary, the theory of preconceptions now
seems to scholars to be closely connected to perception or experience, and
thus more plausibly concerned with preconceptions that have descriptive
content (witness the example ‘white’). 19
Thus it might seem that, rather than limiting preconceptions to evalua-
tive and normative notions, the Stoics may actually find it hard to account
for such preconceptions. However, the Stoics do not seem to conceive of
this as a problem. 20 Think of the preconception of the gods as provident
and caring. Clearly, this cannot in any straightforward sense arise from the
same kind of perception that a preconception of ‘white’ arises from, and it
may actually involve other operations than sense-perception (see below).
We should thus not think that the mere fact that ‘good’ is an evaluative
notion means that there can be no preconception of the good.

III. Concepts

In the absence of direct evidence for such a preconception, I suggest turn-

ing to sources on the various ways in which human beings refine their
concepts or come to acquire new ones. From preconceptions we move on
to more advanced concepts. For the latter, being taught and making an
effort play a role in concept-formation (Aetius 4.11.3). This criterion of
being taught and making an effort is obviously somewhat fluid. It is not
essential to the Stoics’ theory to be able to say whether, for example, hu-
_________
18 Bonhöffer (1968, 187-222); Cf. Obbink (1992, 197), who calls into question whether
we are actually dealing with a primarily theological debate, rather than an epistemological
debate about criteria of truth.
19 Sandbach (1971) already disputed Bonhöffer’s assessment. See Inwood (2005, 270)
for a discussion of the appropriately broad sense of experience we have to assume here.
20 I think that Seneca presents this point well in Letter 120. Cf. Inwood (2005) for a de-
tailed discussion.
 ON THE STOIC DEFINITION OF THE GOOD 163

man beings will acquire a preconception of snow. Clearly, this will depend
on where a child grows up. It is entirely possible to acquire reason without
ever having encountered snow. Other preconceptions—and the preconcep-
tion of the good might be one of them—are likely to be shared by all hu-
man beings who have reason.
According to the Stoics, we acquire concepts in a variety of ways: by di-
rect experience, resemblance, analogy, transposition, composition, opposi-
tion, and privation; the notions of the just and the good are acquired by
nature (DL 7.52-53). The sources do not make it clear whether this list, or
some part of it, refers exclusively either to preconceptions or to more ad-
vanced concepts; quite likely, at least some of these operations are rele-
vant to the formation of both. For our purposes, the question whether we
are here concerned with preconceptions or more advanced concepts is
most pressing with respect to the good. Should we think that the scientific
concepts of the just and the good are acquired naturally? Or are the pre-
conceptions of the just and the good acquired naturally? Both are entirely
possible and I tend to think that both are, for the Stoics, true. Growing up
will, on the Stoics’ views about one’s early impulses and affiliation, natu-
rally give rise to ethical notions. But it is also an important Stoic idea that,
if a human being lives fully naturally (i.e., lives a life in agreement with
nature), this will lead to the formation of the scientific concept of the
good. 21 While I think that both are true, it is conceivable that the passage
refers only to this second point. The passage thus does not constitute in-
disputable evidence for the claim that the early Stoics thought that there is
a preconception of the good.
It now seems that there are, most basically, two ways of arriving at a
concept which is more advanced than a preconception. A first possibility
is that the advanced concept of X ‘grows out of’ the preconception of X.
This is what I shall call Refinement. However, through many of the
aforementioned operations, we can also acquire concepts to which no one
preconception corresponds. In these cases, there is no direct line from a
_________
21 Jackson-McCabe (2004) takes DL 7.52-3 to refer exclusively to the acquisition of
preconceptions. In brief, this is his argument: The passage as a whole must refer to precon-
ceptions since the notions of the good and the just are said to be acquired naturally; and this
is the mark of preconceptions. However, the Stoics can speak of a ‘natural’ development in
more than one context. Even though, as Seneca says in Letter 120, nature can only give us
the seeds of the conception of the good, coming to acquire it is, for the Stoics, a natural
process (cf. M. Frede 2001). Jackson-McCabe in fact needs to admit that, for the list in DL
7.52-3 to make sense, ‘natural’ here must refer, as he says, to more than the naturalness
which characterizes the acquisition of all preconceptions. This ‘more,’ on Jackson-
McCabe’s account, is that ethical preconceptions are inborn (2004, 339).
164 KATJA MARIA VOGT

preconception of X to the concept of X. This is what I shall call Novelty.

If there is a preconception of the good, arriving at the scientific concept of
the good should be a case of Refinement. If there is no preconception of
the good, it would have to be a case of Novelty. In the case of Refinement,
the preconception of X can serve as a criterion for the theory of X. In the
case of Novelty, there is no preconception of X which the theory of X
could be in agreement with; the theory, in this case, must be in agreement
with other preconceptions.
Let us first consider Refinement. Suppose someone has a preconception
of ‘human being,’ and now asks herself the question what human beings
actually are. Let us assume she makes some progress. She comes to think
that human beings acquire reason in such-and-such a way, act in such-and-
such a way, etc., i.e., she acquires various opinions about human beings.
But as it turns out, one needs to understand nature quite generally in order
to understand fully what human beings are: a certain part of the large liv-
ing being that the universe is. Thus, when she finally comes to know what
human beings are, she becomes a sage. We might now say that she has a
scientific concept of what human beings are. She set out with a preconcep-
tion, moved through a gradual process of concept-refinement and acquisi-
tion of opinions, and finally arrived at the scientific concept of a human
being and at knowledge of what human beings are.
To complicate things, let us assume that some of the views the agent
comes to hold do not only fall short of knowledge by being opinions (thus
being changeable, etc.). Rather, suppose the agent assents to incognitive
impressions. Would she, in this case, ‘lose’ her preconception of ‘human
being’? Suppose she acquired the opinion that it is natural for human be-
ings to grieve when someone they love dies. In this case, it does not seem
to me that she would lose her preconception. She might still come to see
why this is misguided, and eventually turn into a sage. But suppose she
were to acquire the opinion that human beings can fly. In this case, she
might lose her preconception. An agent who is prepared to adopt the view
that human beings can fly is unlikely to ever turn into a sage; she is deeply
confused. Perhaps the Stoics would say that by virtue of being rational, we
actually would not accept a theory of human beings that tells us that hu-
man beings can fly.
Given the emphasis the Stoics place on how radical a shift in perspec-
tive takes place when we acquire the conception of the good, acquiring
this conception might not seem to be a case of Refinement. Rather, it
might seem to be a case of Novelty. Consider the Stoic example of the
notion of the center of the earth (DL 7.53). Before we arrive at the concept
of the center of the earth, we do not have a preconception of ‘center of the
 ON THE STOIC DEFINITION OF THE GOOD 165

earth.’ Rather, we have preconceptions of other things, like ‘center,’

‘sphere,’ ‘earth.’ Now we learn that the earth is a sphere, and thus we can
come up with the notion of a center of the earth. We shall also come up
with the opinion that there is a center of the earth. If, however, we come to
understand nature fully, and to have a scientific concept of the center of
the earth (supposing for the moment that the earth really is a sphere), the
view that there is a center of the earth will be a piece of knowledge. In this
case, the scientific concept refers to something that, as long as the agent
only had preconceptions, she was not able to refer to at all.
So is this perhaps how we arrive at the concept of the good? Do the Sto-
ics refrain from mentioning the preconception of the good in their devel-
opmental account because they want to say that there is no preconception
of the good? Do they claim that, starting from preconceptions of what is
appropriate, ‘belongs to us,’ is valuable, etc., we are somehow able to sum
up these ideas and transform them into one more complex notion, the con-
cept of the good? 22 Chrysippus says that his theory of good and bad things
is “most in harmony with life and connects best with the inborn precon-
ceptions” (Plutarch, On Stoic self-contradictions 1041E = SVF 3.69 = LS
60B). 23 Note that Chrysippus does not say that his theory of the good and
bad is most in harmony with the preconception of the good. Is it plausible
that he means to imply that there is no preconception of the good?
For at least two reasons, I do not think so. First, it seems entirely im-
plausible to assume that the process of acquiring preconceptions, which is
a process of growing ‘into’ the world, would not lead us to form the pre-
conception of the good. ‘Good’ is a concept that pervades everyday inter-
actions to such an extent that it seems difficult to see how a child might
grow up without acquiring it. And ‘good’ is actually one of the predicates

_________
22 In Chapter 4 of my book Law, Reason, and the Cosmic City (2007), I discuss these
and other evaluative and normative preconceptions that are relevant to the Stoics’ account of
appropriate action. (By ‘what belongs to us’ I refer to what is oikeion.)
23 Sandbach argues that )"
 need not be translated as ‘inborn’ (1971, 48). While I
find it artificial to refrain from translating )"
 as ‘inborn,’ I think we should assume
that Chrysippus here uses this term in a loose sense, according to which it does not say that
human beings are born with reason (but rather with the early impulses that steer the natural
development of acquiring preconceptions and thus reason). Jackson-McCabe argues that
Sandbach’s hesitations are ungrounded. On his account, ethical preconceptions are innate
because they are given to us like the first impulses. However, as it turns out, this still com-
mits him to a less than strict reading of )"
: “The human individual is not born with
ethical conceptions per se, only with an innate predisposition to form these concepts owing
to oikeiôsis” (2004, 340).
166 KATJA MARIA VOGT

that are mentioned in the context of our preconception of the gods. 24 It is

not clear how ‘good’ could belong to our preconception of the gods, with-
out there also being a preconception of good. It thus seems that acquiring
the conception of the good must be a case of Refinement, rather than Nov-
elty.

IV. The Task of Definition

But even if we assume that, according to the Stoics, human beings have a
preconception of the good, this preconception can only function as a crite-
rion if we know its content. Where can we turn for evidence on the content
of this preconception, if the Stoics do not discuss it? I suggest that we turn
to the Stoics’ views on definition. Unfortunately, these views are even less
well preserved than Stoic thought on preconceptions. A number of ideas
are transmitted, and I shall, in this paper, not attempt to cover all of them,
or explain how they make up one consistent theory. 25
According to Chrysippus, a definition (9
) expounds the peculiar
characteristic (.$
) of the definiendum (DL 7.60). 26 Presumably then, to
be ‘benefit or not other than benefit’ is the peculiar characteristic of the
good. But according to another well-known passage, ‘to benefit’ is the
peculiar characteristic of the good (DL 7.103).
Perhaps the shorter definition is merely shorthand for the longer one.
But it might also be that the Stoics ascribe several functions to defini-
tions. 27 As we saw, a definition names the peculiar characteristic of the
definiendum. It might do so, it seems, in various ways. In a technical defi-
nition—such as ‘the good is benefit or not other than benefit’—it might do
so in a way which satisfies the standards of scientific knowledge, taking
_________
24 Note that it is not clear whether the Stoics want to claim that we have a preconception
of the gods as good, caring, beneficial, where these characteristics are all on the same level.
Perhaps our preconception of the gods is that they are good, and that our preconception of
the good is that the good benefits. By putting this together, we arrive at the preconception of
the gods as beneficial (for testimony which suggests this train of thought cf. Clement, The
teacher I.1.63.1-2 = SVF 2.1116, part = LS 60I).
25 For a detailed interpretation of Stoic thought on definitions, see Brittain (2005).
26 See also Scholia on Dionysius Thrax, 107,5-7 = SVF 2.226 = LS 32A.
27 Brittain discusses whether we should think that the Stoics distinguish between pre-
liminary definitions, which serve the purposes of spelling out the content of preconceptions,
and what he calls ‘real’ definitions, which then are the technical definitions that the sage
arrives at. The evidence, however, does not confirm this view. It seems quite possible that
Chrysippus’ thesis, according to which definition names the peculiar characteristic, does not
only refer to the fully formulated, technical definitions (2005, 186-197).
 ON THE STOIC DEFINITION OF THE GOOD 167

complicated theoretical considerations into account. But the sources also

report another task of definition: definitions capture the content of precon-
ceptions. Galen reports that, according to the Stoics, a definition (9
) “is
that which by a brief reminder brings us to a conception of the things un-
derlying words” (Medical definitions 199.348,17-349,4 = LS 32D; cf. SE,
PH 2.212). 28 If this is what definition—or some kind of Stoic definition—
does, then we have reason to assume that the content of the preconception
of the good is that the good benefits. The simpler definition of the good,
that it benefits, would capture the content of the preconception. And by
doing so, it would help us enormously. It would tell us what our theory of
the good needs to be in agreement with.
Preconceptions are not available to us as explicit bundles of claims (or
as one explicit claim). In some cases, this may not be too problematic. For
example, we might think that someone who begins to investigate the na-
ture of plants can dismiss a theory according to which plants are made of
plastic without first making the content of her preconception explicitly
available to herself. But things might seem more complicated in the case
of the good. Here, our misguided views seem to shape our perspective so
deeply that, as long as we have not made the content of our preconception
explicit, we might not be able to accurately dismiss any theory of the
good. Once definition has done its job of elucidating our preconception,
we can see a very important point: that it is not part of the content of the
preconception of the good that things like health and wealth are good. As
inquirers, we can thus give up our attempt to formulate a compelling ethi-
cal theory which ascribes some kind of goodness to these things. And we
might think that a diligent inquirer could mistakenly have considered that
her task: to explain how the way in which virtue and wisdom are good fits
with the way in which things like health and wealth are good (this would
be a mistake because health and wealth are not good; but it would be a
plausible research project for anyone who has not yet clarified for herself
the content of her preconception of the good, because health and wealth
typically matter to human beings). With the help of the definition of the
good as benefit, the inquirer can build her theory, using her preconception
as a criterion. Acquiring the conception of the good is thus a case of Re-

_________
28 Cicero focuses on this aspect of Stoic definitions, and praises it: “However much we
attack this school, as Carneades used to, I’m afraid that they may be the only real philoso-
phers. For which of those definitions [the Stoic definitions of courage] does not uncover the
tangled conception of courage which lies buried in us all?” (Tusc. Disp. 4.53 = LS 32H, my
emphasis).
168 KATJA MARIA VOGT

finement (however, one that is particularly demanding, since our opinions

tend to obscure our preconception of the good).

V. Benefit and Good Use

For the remainder of this paper, I shall pursue two objections which might
be raised against this interpretation. First, one might think that my inter-
pretation makes a rather narrow, formal point.
Against my interpretation, one might suggest that the Stoics’ notion of
benefit is a much greater resource to them, a resource which indeed makes
Stoic ethics somewhat less paradoxical. And one might argue that, in some
sense, this is what is indicated at the beginning of this paper: that, by pay-
ing close attention to the way in which preconceptions are criteria of truth,
we can better understand the way in which Stoic ethics, in spite of the
prominent paradoxes, has a ‘plausible’ side. One might suggest that the
notion of the good as benefit is, in a key piece of testimony (DL 7.103),
associated with the notion of good use. And accordingly one might think
that the Stoic story goes something like this. When we grow up and come
to select and reject things of value, we see them as beneficial (or not). To
select and reject them correctly is to ‘use’ them correctly; if we do so, they
actually benefit us. As we progress toward virtue, and eventually turn into
virtuous agents, we come to see that only virtue benefits simpliciter. But
virtue is the wise use of valuable things. And thus there is a sense in which
valuable things benefit—used wisely, they benefit.
But this picture cannot be correct. Just as much as only virtue is good,
only virtue benefits. According to the Stoics, valuable things do not bene-
fit, no matter how perfectly well they are selected (or used). It is thus mis-
leading to think of benefit as some kind of ‘link’ between things of value
and the good—as if both valuable and good things could benefit, if only
the valuable things are dealt with correctly. The Stoic notion of ‘benefit’ is
no less ambitious than the Stoic notion of ‘good.’ 29
Let me turn to the famous passage which associates the definition of the
good as benefit with the notion of ‘good use.’
(1) For just as heating, not chilling, is the peculiar characteristic of what is
hot, so too benefiting, not harming, is the peculiar characteristic of what is

_________
29 We might think that the famous passage in Plato’s Republic that identifies the good
and the useful is an apt inspiration here. It is in the context of discussing God’s goodness
that Socrates explains that whatever is good is useful (Rep. II, 379b).
 ON THE STOIC DEFINITION OF THE GOOD 169

good. 30 (2) But wealth and health no more do benefit than they harm. There-
fore wealth and health are not something good. (3) Furthermore they say: that
which can be used well and badly is not something good. But wealth and
health can be used well and badly. Therefore wealth and health are not some-
thing good. (DL 7.103; tr. LS; numbers K.V.)
Does this passage not commit us to placing the Stoics’ notion of benefit,
first and foremost, within a theory of good use? My answer to this ques-
tion is a qualified ‘no.’ Sections (2) and (3) of the text offer arguments for
the Stoic thesis that things like health and wealth, not being good, do not
benefit. However, (2) and (3) are quite difficult to interpret, and in both
cases, as I shall suggest, we should not entirely take the report at face
value.
Section (2) rephrases the claim that health does not benefit (let us take
this as a sample claim of the relevant kind) as the claim that health no
more benefits than harms. ‘No more’ (ou mallon) is a notion that origi-
nates with Democritus and becomes an important technical term in Scepti-
cism. A Sceptic may say that honey is no more sweet than bitter. In saying
so, he reports that for him, honey sometimes seems sweet and sometimes
seems bitter. He is, to use another Sceptical metaphor, stuck between the
pull that each appearance has on him, and thus will not assent to either of
them. The Stoics claim that health neither benefits nor harms, and we
might think that, strictly speaking, this is not the claim that it no more
benefits than harms. The subtle difference is this: the Stoics emphasize the
point that health does not at all benefit or harm, rather than the point that,
since it sometimes seems to benefit and sometimes seems to harm, we had
better suspend judgment on whether, in reality, it is a beneficial or a harm-
ful thing. Of course, ‘no more’ might be used in a less than strictly techni-
cal, Sceptic sense. 31 For the Stoic doctrine to be presented with the help of
this expression, it would seem to me that we should envisage a context in
which someone aims to explain the Stoic thesis, using a term which reso-
nates with those who participate in Hellenistic debates, but goes beyond
the Stoics’ own terminological repertoire.

_________
30 The Stoic argument, as it is presented here, is strikingly similar to an Epicurean ar-
gument. According to Epicurus, one perceives that pleasure is the good in the same way in
which one perceives that honey is sweet, fire hot, and snow white (De Fin. I.30).
31 On technical and non-technical uses of the expression, cf. Bett (2000, 30-32). In the
course of explaining Socrates’ reasoning in Plato’s Euthydemus 278e-281e, Long arrives at
the conclusion that, at this point in the dialogue, “it seems that health in general is no more
good than it is bad.” Long here uses ‘no more’ as an illuminating paraphrase, rather than a
technical term, and we might think that the Stoics could have used it as such (1996, 27).
170 KATJA MARIA VOGT

Section (3) ascribes a train of thought to the Stoics which makes use of
another notion that resonates with the participants of Hellenistic debates:
the idea that the things which are conventionally considered goods can be
used well and badly. Section (3) tends to remind scholars of Platonic ar-
guments to the effect that things which can be used well or badly are not in
themselves good (cf. Euthydemus 280e-f and Meno 87e-88a). 32 In a num-
ber of dialogues, Plato has Socrates discuss how things like wealth or
health need to be used correctly, or with wisdom, rather than incorrectly,
or foolishly (with ignorance). The precise interpretation of these passages
is controversial, and we cannot enter into it here. What is important for our
purposes, and perhaps uncontroversial, is this: (a) Things like health and
wealth can be used well and badly. (b) While they are not good in them-
selves, they actually can be or become good, or, they actually can benefit,
namely if they are used correctly (wisely, etc.).
DL 7.103 attributes (a) to the Stoics, but not (b). So much seems correct.
Even though the passage associates the Stoic notion of benefit with the
notion of good use, it does not do so in the sense discussed above—the
text does not suggest that, according to the Stoics, good use of indifferents
makes them beneficial. Another aspect of the report seems more problem-
atic. (a) is presented as the Stoics’ reason for saying that things like health
and wealth are not good. This further claim should in my view be taken
with some caution. 33 The Stoics would certainly agree that a rich person
can use her wealth in better or worse ways. But this is not their principal
way of explaining why wealth is a mere preferred indifferent, rather than
good. The Stoics have their own—and actually, as I suggest, quite differ-
ent—story to tell, a story which perhaps might be presented as their way
of developing further and spelling out the Socratic notion of good use.
According to the Stoics’ account, preferred indifferents are ‘in agreement
with nature’; things like health and wealth have value because of the way
in which they contribute to the natural life of a human being. And this
value is a ‘mere value’ ( '), rather than the value of real goodness, be-
cause health and wealth do not contribute to an agent’s happiness. The
_________
32 Cf. Long and Sedley’s commentary in LS Vol. 2, 350. Francesca Alesse argues that
the use of ‘<’ in DL 103 is in line with Euthydemus 280d4-6. From this passage, she
writes, ‘derives’ the Stoic notion of  (2000, 326-7). It is important to keep in mind that,
for it to be true that a passage in Plato was influential among the Stoics, we by no means
need to assume that the Stoics did not develop the ideas they found further. For an approach
of this kind, which focuses on Stoic engagement with and further development of certain
Socratic views, cf. Striker (1996) and Long (1996).
33 The same reasoning is mentioned in Plutarch, On Stoic Self-Contradictions 1048C
and Sextus Empiricus, M 11.61. See also DL 7.104.
 ON THE STOIC DEFINITION OF THE GOOD 171

wise person selects and rejects valuable things perfectly, and it is this per-
fect selection which makes her life go well, or, be a ‘happy life.’ Whether
she gets any of the valuable things that her selection and rejection is con-
cerned with is irrelevant to her happiness.34 This, I suggest, is at least the
most prominent argument which the Stoics make for why health and
wealth are not good.
The ideal way of dealing with things of value is, by the Stoics, described
as selecting and rejecting them perfectly. We might paraphrase this and
say that, if one does so, one uses these things well. Or, we might think that
perfect selection and rejection is the Stoics’ way of spelling out what good
use is. If this is so, then it is not surprising that the sources mention the
Socratic idea of good use. At the same time, the Stoics’ way of explaining
this idea might be quite distinctively their own. 35
But if, on the suggested interpretation, the Stoic thesis that only virtue
and virtuous action benefit is as paradoxical as the thesis that only virtue
and virtuous action are good, then it does look like I have overstated
things in my introduction, saying that we tend to focus in a one-sided way
on the paradoxical side of Stoic ethics. While Stoic ethics is no less para-
doxical for the definition of the good, I would still want to resist this con-
clusion. The fact that Stoic ethics seems paradoxical to us shows how con-
fused we are. There is a point to the Stoics’ claim that really, once we un-
derstand things, Stoic philosophy turns out to be fully in agreement with
the very assumptions which make us rational. If we were less confused, it
would no longer seem extreme to us; rather, Stoic philosophy would end
up looking like the most obvious theory. That naturally acquired notions
are criteria of truth is, as it were, the promise that, if we only make enough
progress, things will ‘fall into place,’ and the Stoic theories will appear
entirely natural to us. This is not an obvious implication of the Stoics
thinking that their theories are true; for one might also hold that a view is
true and yet recognize that it is (thoroughly) revisionist. What I am sug-
gesting is that the Stoics are, in effect, claiming that their theories only
seem revisionist. If we were actually to become wise, we would see how,
quite to the contrary, it is the everyday opinions of fools which in fact ‘re-
vise’ the notions that nature supplies us with.

_________
34 Cf. the example of the archer in Cicero, De fin. 3.22.
35 In his discussion of Euthydemus 278e-281e, Long notes the difference between using
things like health and wealth, and selecting them. However, Long then employs a notion,
‘controlled by wisdom,’ which makes the difference between both conceptions less visi-
ble—both ‘wise selection’ and ‘wise use’ could be described as ‘wise control’ (1996, 28-
29). Cf. M. Frede (2001, 89).
172 KATJA MARIA VOGT

VI. A Typical Case of Refinement

Let me turn to the second objection which might be raised against my in-
terpretation. Perhaps I have emphazised the continuity between the pre-
conception and the conception of the good too much. Granted, according
to both the preconception and the conception the good benefits. But do the
sources not suggest that, nevertheless, coming to call virtue ‘good’ is like
using ‘good’ for the first time with a proper understanding of what the
term actually means? Is the scientific concept of the good thus not more of
a new concept than I have presented it as being?
According to Cicero, we acquire the conception of the good by analogy
(collatio rationis). 36 The mind ‘climbs up’ by means of analogy from the
things which are in accordance with nature (i.e., things of value, like
health or wealth). However, at the point when we actually acquire the con-
cept of the good, something other than analogy is in play. We perceive the
good, and name it ‘good’ not by comparison with other things, but by its
own specific power. 37 When we eat honey for the first time, we perceive
its sweetness, but not by comparison with other sweet things that we have
tasted earlier. Rather, it is the honey’s specific sweetness which we per-
ceive and which makes us call it ‘sweet.’ Similarly, we ‘perceive’ the ac-
tual good—virtue—as good through the way in which it, independently
from other things that we formerly regarded as good, appears to us
(Cicero, De fin. 3.33-4 = SVF 3.72 = LS 60D).
What interests us now is the moment of acquiring the conception of the
good. According to Cicero, it does have an element of acquiring a whole
new concept. When calling virtue good, we do not compare it to things
which are less good, or good in a qualified sense. Only virtue is good, and
when we for the first time actually experience virtue, this ‘hits us.’ The
fact that we have the preconception of the good as benefit does not mean
that we are actually acquainted with anything good or anything beneficial.

_________
36 Seneca too ascribes an important role to analogy (Ep. 120.4). I am not discussing Let-
ter 120, since I take Seneca to be concerned with a different issue from the one that Cicero
discusses. As Inwood emphasizes, the letter is written with the assumption that we (not
being sages) already have the notion of the good (2005, 284).
37 In Cicero’s account, this latter aspect sounds as if it involved a perception of the
good. However, we need to be careful to not confuse this idea with a Platonic notion of
‘seeing the Form of the Good.’ Plutarch preserves a quote from Chyrsippus on how good
and bad things are perceptible: what we perceive is the condition of someone’s soul; we can
perceive whether someone is wise or courageous, etc. (On Stoic self-contradictions 1042E-F
= LS 60R = SVF 3.85).
 ON THE STOIC DEFINITION OF THE GOOD 173

This might seem to be a peculiar feature of the preconception of the

good. How can there be a relationship of Refinement between the precon-
ception of X and the scientific concept of X, when acquiring the scientific
concept makes us realize that we used to be deeply misguided about what
X is? I am suggesting that there can be such a relation. Let us first con-
sider other evaluative notions. When we, for example, learn the notion
‘caring’ through interaction with our family, and thus acquire a preconcep-
tion of ‘caring,’ this does not involve encountering a single instance of
caringness. In the end, only wise persons and gods are caring, and even
very good parents are not (unless they are wise). 38 Acquiring the concept
‘caring’ thus is like acquiring the concept ‘good.’ Even though there is a
preconception of caring, which is the criterion of truth for the scientific
concept of caring, we really do not understand what it is to be caring until
we have the scientific concept.
But what about other concepts, say, concepts of natural entities? Recall
my example of the concept ‘human being.’ Again, we see the same kind of
change. Full knowledge of what human beings are can only be achieved as
part of the overall achievement of wisdom. When one understands nature
as a whole, one’s concept of a human being will change quite drastically
(and this would be true for all natural beings). Rather than consider them
as natural beings with certain features, one would come to see them as
certain kinds of parts of a large living being.
As I suggest, every case of Refinement which eventually arrives at a
scientific concept that is, in some loose sense, basic to our way of seeing
the world—i.e., at a concept which is part of the overall knowledge of the
sage—will involve this kind of drastic change. A full understanding of
nature involves seeing things really quite differently from how any lesser
grasp makes them appear. The good is no exception; the same is true for
all evaluative and normative concepts, as well as all concepts of natural
entities. Nevertheless, these are cases in which the preconception of X
functions as a criterion of truth for the conception of X.
The relationship between the preconception of the good and the concept
of the good is thus, on my interpretation, a case of Refinement. The con-
ception of the good relates back to the preconception of the good, rather
than to some set of other preconceptions. The preconception of the good,
rather than another set of preconceptions, captures what it is that the the-
ory must, most immediately, be in agreement with. As in other cases of
Refinement, the progress that must be made in order to arrive at the scien-
_________
38 I am grateful to John Cooper for discussing these issues with me; the example came
up in conversation.
174 KATJA MARIA VOGT

tific conception of the good is very significant, and involves deep changes
in our ways of thinking about things. In order to begin our inquiry, we
need an explicit statement of the content of our preconception of the good,
and this means a certain kind of definition. The definition of the good as
benefit tells us what we naturally take the good to be, and accordingly tells
us that our theory of the good must meet the criterion of portraying the
good as benefit. As radical as the transition to wisdom may be, once we
get there, our theory of the good will no longer strike us as revisionist.
Rather, it will seem like the most natural theory to hold.
COLUMBIA UNIVERSITY
 COMMENTARY ON VOGT

STEPHEN MENN

ABSTRACT
I argue that the Stoics do not use their definition of the good as what benefits
to explain how the sage acquires the concept of goodness. Rather, the ideal
progressor begins with tendencies to select things according to nature, with-
out believing that these things are good or that they benefit; then, by reflect-
ing on her own harmonious selection of these things, she arrives at the con-
cept of goodness. However, the definition of the good as what benefits is im-
portant for the Stoics in refuting the false beliefs of non-sages about the good,
and thus helping those whose moral development has miscarried to return to
the right path. In particular, the definition of the good as what benefits can be
used to argue that what can be used wrongly is not good; I defend the attribu-
tion of this argument to the Stoics.

Katja Vogt’s paper is not so much about the Stoic definition of the good as
about ways the Stoics used that definition to motivate and justify their of-
ten paradoxical conclusions about what things count as good. I’ll start with
some cautions and clarifications about how exactly the Stoic definition, or
definitions, of the good should be formulated, and then turn to some larger
issues about the role of the concept of benefit in our coming to knowledge
of the good, and about the concept of “use” and of right and wrong uses of
a thing.
In the version of her paper presented at Dartmouth, Katja spoke of “the
good is benefit” as the Stoic definition of the good, but it is not, strictly
speaking, the or even a Stoic definition. Different Stoics offered different
definitions of the good, probably not meant to be incompatible; thus
Cicero cites Diogenes of Babylon as defining the good as “id quod esset
naturâ absolutum,” perhaps  #
2 ! - (De finibus III,x,33).
“The good is benefit” is not presented in our sources as a definition in this
way. We are told that “the good is benefit or not other than benefit” (Sex-
tus Empiricus Adversus mathematicos XI,22), and this is described as a
definition, but a disjunction is on the face of it an odd way to capture a
basic concept like goodness, and Sextus also tells us that this disjunction is
coextensive with a non-disjunctive definition, “what is capable of benefit-
ing” (XI,27; he also notes stricter senses of good, XI,25-6, but this formula
covers all cases). This last is a natural definition, since to benefit is the
.$
 of the good (Diogenes Laertius VII,103), and a definition is “the giv-
ing of an .$
” (Chrysippus quoted by the scholia on Dionysius Thrax,
SVF II,226). And this definition, short and lacking in detail, gives us a
176 STEPHEN MENN

starting-point that can be filled out into more precise accounts of what
things are good, such as “id quod esset naturâ absolutum,” or “benefit or
not other than benefit,” which is not so much a definition as a division of
the kinds of things that fall under the definition. The reason why such a
division is needed is that not everything that benefits is a benefit: “by
‘benefit’ they mean virtue and good action, and by ‘not other than benefit’
the good person and the friend” (Sextus XI,22). Things other than virtue
are not benefits, and indeed they are not beneficial, since they benefit no
more than they harm, since benefit results when they are conjoined with
virtue and are thus used virtuously and wisely, while harm results when
they are conjoined with vice and are thus used viciously and foolishly. But
a whole consisting of, say, an oil-scraper and virtue, whenever it exists
(that is, whenever the oil-scraper and virtue occur together), is never used
badly: so this whole is a beneficial thing, and falls under the definition of
good. This whole is not a benefit, but the Stoics say that it is not other than
benefit, since virtue is a benefit and the Stoics say that a whole is not other
than its parts; and this is how they can say that all good things are either
benefit or not other than benefit. Sextus gives as examples “the good per-
son and the friend”: these examples work because the good person and the
friend (who must also be a good person) are virtuous, and the Stoics think
that qualities such as virtue are bodies and bodily parts of the qualified
things; so the good person and the friend are not other than virtue, and are
thus not other than benefit.
But how does the definition of the good as what benefits help to explain
how we can come to a conception of the good? (Katja says that when in a
Stoic context we ask how we come to “a conception of the good,” we are
using this phrase as shorthand for “a scientific concept of the good”: we
might have other, non-scientific concepts of the good beforehand, and in-
deed we must if, as Katja claims, we acquire the scientific concept of the
good by “refinement” rather than “novelty,” but she takes the sources, no-
tably Cicero in De finibus III, to be asking how we come to the final, cor-
rected, scientific concept of the good. This seems peculiar to me: I think
Cicero is just asking how we come to a concept of the good.) As Katja
notes, when Cicero gives (in the name of the Stoics) an extended psycho-
logical account of how we acquire the concept of the good, he makes no
mention whatever of this definition of the good, or of any concept of bene-
fit prior to our acquiring the concept of the good. And I doubt that this is
due to sloppiness or incompleteness on Cicero’s part: he or his sources
seem to me to be showing excellent judgment. It is difficult for us to form
a concept of the good, since none of the things we are ordinarily familiar
with are good, and so Cicero wants to explain how this concept-
acquisition is possible; but since it is also true that none of the things we
 COMMENTARY ON VOGT 177

are ordinarily familiar with benefit, it seems no easier for us to acquire the
concept of benefit, and, if so, the concept of benefit will not be any help in
our acquiring the concept of good. (It is not just that it is no easier for us to
point to instances of benefit than of good, but also that our concept of
benefit will be no clearer than our concept of good.)
That doesn’t mean, however, that the definition of the good as what
benefits is of no use in Stoic ethics. One very common use of definitions
in Stoic philosophy is for what Rachel Barney has called a “strict sense
project.” That is: I want to argue that none of the things that we ordinarily
encounter and that we ordinarily take to be F are really F; only some quite
rare and special objects are F in the strict sense (for instance, while we
might think that we and many other people are sane and free, it turns out
that only the sage is sane and free, and all other adult humans are slaves
and mad). In order to convince you of this paradoxical conclusion, I need
to start from some premiss about F which you will accept. So I formulate a
definition of F, articulating the ordinary concept of F in a way that we can
agree on: then I argue that the various necessary conditions for being F,
spelled out in this definition, are not in fact satisfied by the ordinary ob-
jects ordinarily taken to be F, but only by some quite special class of ob-
jects. A definition of F is useful in this particular way, not when the inter-
locutor does not have a concept of F, but when he has a concept of F, cor-
rect as far as it goes, which he misapplies; the definition helps to generate
an elenchus which will purge him of his false beliefs about what things are
F. It is clear that the Stoic definition of the good as what benefits is indeed
used in this elenctic way, to refute the false beliefs that health and wealth
and so on are good: if they were good, then since it is the .$
 of the good
to benefit, they would always benefit and never harm, but in fact they
benefit when they are used virtuously and harm when they are used vi-
ciously, and so they no more benefit than harm, and so they are not good.
However, Cicero in De finibus III, and the parallel extant Stoic accounts
of how we come to acquire the concept of good, do not make this argu-
ment; and there is no reason why they should, since it does not help to
explain how we acquire the concept of good. As Katja says, “within the
developmental story, there is no mention of these earlier notions of the
good” on which we would think that health or wealth are good; the word
“good” is first mentioned in the story when we get to the developing per-
son’s recognition that only consistency is genuinely good. 1 Before this
point, the developing person’s motivations are described in terms of im-
pulses to appropriate action [<
, officium] and impulses to select the

_________
1 Vogt, 160.
178 STEPHEN MENN

things that are according to nature, with no mention of the person’s beliefs
(either true or false) about what things are good. Katja asks “Would it not
seem that the progressor selects what she considers good? The Stoics steer
away from this way of putting things in the context of the developmental
account. Perhaps they are not burdening the already difficult theory of
ideal development with yet another intricate issue: the issue of keeping
separate the preconception of the good from the various views about the
good which progressors come to hold.” 2 I think the Stoics have a more
serious principled reason for not saying here that the progressor selects
what she considers good, not just a reason of economy of exposition. This
is an account of ideal development, explaining how a human being would
ideally develop, if taught by nature (that is, by Zeus) and not corrupted by
the false opinions current in society, from the first impulses which babies
are argued to have toward whatever preserves or is in accord with their
constitution, to the kind of action and understanding that are characteristic
of normal healthy adult human beings. It turns out that normal healthy
adult human beings are very rare for the Stoics—perhaps one or two since
the last cosmic conflagration—and therefore the normal healthy process of
development that results in them must also be very rare; this is still the
norm that nature aims at, and anything else is a deviation. And, in the ideal
process of development that nature aims at, there is no stage at which we
select health and so on because we falsely think they are good. If we did
that, we would be fools, slaves, and mad (indeed utterly foolish, slavish,
and mad, since foolishness, slavery and madness do not come in degrees);
this is not something that happens to us along the ideal path of develop-
ment, and if it did, we would not be able to extricate ourselves from it by
the gentle progressive methods Cicero describes.
Of course, since Cicero is describing how we would come to be sages,
he should not assume that we are sages to begin with; but the Stoics do not
hold that every human being must be either a sage or a fool. Rather, they
hold that every adult human being must be either a sage or a fool; and they
are trying to describe an ideal natural course of development on which we
would go from being children to being sages without ever becoming fools.
This can happen if we start from our natural and non-erroneous impulses
to select things that are according to nature without believing that they are
good, and develop those impulses into a consistent selection of what is
according to nature: we can then reflect on the harmony and consistency
of our actions, discover that that is of incommensurably greater value than
the things that are merely according to nature, and so on. Obviously this

_________
2 Vogt, 160-161.
 COMMENTARY ON VOGT 179

would not work if our actions were not in fact harmonious and consistent,
which they would not be if they were mad foolish actions accompanied by
the passions which arise from believing that things outside our control are
good or bad for us. For those of us whose development miscarries, so that
we go from being children to being fools rather than sages, the subsequent
development to wisdom will not be so smooth: while we will have some
sound cognitions and impulses which can be a basis for subsequent cogni-
tive development, we will also have false beliefs which contradict these
and get in the way of our building upon them. In particular, we will have
false beliefs about the good, taken from our parents and others (for a Ro-
man aristocrat it starts with your wet-nurse, Tusculans III,i,2) or reached
by our own mistaken reasoning, which will disrupt our rational valuation
of the things according to nature and make it hard for us to develop a new
concept of the good by reflection on the rationality in our valuation of the
things according to nature, still harder to recognize that only that rational-
ity and its consequences are good. So we will need to supplement the posi-
tive education described by Cicero with a negative education, consisting in
elenchus of all the false beliefs, especially about the good, that are ob-
structing our positive education; only the person who is never a fool can
dispense with this.
Is the ideal natural development toward a concept of the good what
Katja calls “refinement” or “novelty”? It looks more like novelty. We
come to the concept [notio = )
, De finibus III,vi,21] of the good by
reflection on the order and harmony of appropriate actions or of the things
according to nature which are their objects, not on anything which we
have already either rightly or wrongly taken to be good. If we have had a
pre-scientific concept of good, it has been entirely latent, never applied
either rightly or wrongly to any particular thing. On the other hand, at the
time when the progressor acquires the scientific concept of the good, she
must already be in such a condition that, when she perceives the consis-
tency of her actions, she is able to recognize it, perhaps not precisely as
good—she certainly does not need to know the word “good”—but at any
rate as having an attraction incommensurably greater than the things that
are merely according to nature. This is not just being able to recognize the
truth of propositions such as “I ought to prefer the consistency of my ac-
tions to any particular thing according to nature that I might attain through
my actions,” but being able to recognize the distinctive quality which the
consistency of my actions possesses (like the special taste of honey, De
finibus III,x,34), from which the truth of such propositions follows. Cicero
says that the progressor is able to recognize this, but he does not make it
especially clear how. Perhaps this already implicit recognitional ability
can be described as a pre-existing concept of the good. How is this ability
180 STEPHEN MENN

to be explained? Katja says that for Cicero we acquire the conception of

the good by analogy (she cites the phrase collatio rationis from De finibus
III,x,33) from things according to nature, such as health. Actually I am not
sure this is what Cicero thinks—he contrasts collatio rationis with simili-
tudo, which suggests that it should be translated “inference” rather than
“analogy,” but unfortunately he does not say what the inference is—but it
is a reasonable proposal and Seneca develops it in Letter 120, even bor-
rowing the Greek word 
 into Latin for the occasion. We form
concepts of the health and strength of the mind [animus] by analogy with
the health and strength of the body; we witness actions which seem to pro-
ceed from such mental health and strength, although in fact they do not,
and we imagine them as they claim to be, ignoring their defects. But even
if we can arrive in this way at concepts of mental health and strength, how
do we know that they should always be preferred to bodily health and
strength?
Well, if the ideal progressor attains the concept of the good at the same
time that she attains virtue—or even a moment later, by reflection on the
virtue which she has attained first—then perhaps she just “tastes” virtue,
like tasting honey, and once you’ve tasted it, everything else just fades in
comparison. But those of us whose development went wrong before we
tasted virtue, and who are now trying to find our way back to the path, will
need a further elenctic argument to refute the belief that the health and
strength of the body are good in such a way that they can be weighed
against the health and strength of the mind, and to persuade us to refer the
goodness that we have falsely attributed to the bodily conditions to the
mental conditions instead. (If this elenchus is to be a refutation of false
beliefs about the good, there must be at least some rudimentary concept of
the good, or of benefit, that we share with the sage, thus “refinement”
rather than “novelty”; but this shared concept might be just “what should
be preferred to everything else,” or perhaps what should be pursued with a
distinctive kind of vehemence.) The most obvious argument is that the
soul uses the body, and through it also external things, either well or ill,
and that a soul in the right condition will use even a disordered body well,
while a soul in the wrong condition will use even a well-ordered body
badly; consequently we should care more about the condition of our soul
than about the condition of our body or our external possessions. The ar-
gument comes from Plato (versions at Meno 87d2-89a4, Euthydemus
280b5-282a7, Clitophon 407e3-408bc4, and cp. Apology 30a7-b4), and is
multiply attested for the Stoics, in the form of an argument that whatever
can be used both well and badly is neither good nor bad: there is Diogenes
Laertius VII,103-4, which Katja discusses, and also Plutarch De Stoicorum
 COMMENTARY ON VOGT 181

repugnantiis 1048c-d and Sextus Empiricus Adversus mathematicos

XI,59-61, which she mentions in a footnote.
I am not sure what Katja’s attitude is toward this argument. In the ver-
sion of her paper presented at Dartmouth, she seemed to deny that the ar-
gument was Stoic at all. She does not quite say that in the present version,
but she does say that “for the Stoic doctrine [sc. that health, etc. neither
benefit nor harm] to be presented with the help of this expression [sc. ‘
#
"M
,’ used in Diogenes Laertius VII,103-4], it would seem to me that
we should envisage a context in which someone aims to explain the Stoic
thesis, using a term which resonates with those who participate in Helle-
nistic debates, but goes beyond the Stoics’ own terminological reper-
toire”: 3 this implies that the argument Diogenes attributes to the Stoics is
not in fact Stoic at least in the formulation Diogenes gives. (She may not
intend this conclusion to apply also to the arguments in Plutarch and Sex-
tus, which do not use the phrase “
# "M
”.) But given the triple attesta-
tion, and no evidence against that I can see, it seems very hard to doubt
that at least the common core of the arguments in Diogenes and Plutarch
and Sextus, arguing that whatever can be used both well and badly is nei-
ther good nor bad, is Stoic, and, I would think, Old Stoic (compare also
Epictetus Discourses I,xx,14-15, attributing to Zeno the thesis that “the
essence of the good is the use of impressions as one ought”). Indeed, I
don’t see real ground for doubting even that the formulation Diogenes
gives, with “
# "M
,” is Stoic. Katja says, rightly, that the Stoics claim
that health neither benefits nor harms, and that this is importantly different
from the Pyrrhonists’ recommendation that we suspend judgment on
whether health is beneficial or harmful, which is what the Pyrrhonists
mean by their use of “
# "M
” according to Sextus Outlines of Pyr-
rhonism I,188-91. But a Stoic could perfectly well use the phrase without
meaning by it what Sextus means by it: it was not a Pyrrhonist trademark,
since it goes back (as Katja notes) to Democritus, and Sextus is at some
pains to distinguish the special Pyrrhonist sense of the phrase from its
meanings in other schools, which he is afraid that his readers will attribute
to the Pyrrhonists as well. (Indeed, the Pyrrhonists may not be consistent
on this: Sextus in Adversus mathematicos XI sometimes seems to be argu-
ing with full [negative] dogmatic conviction that nothing is either good or
bad by nature).
I certainly agree with Katja that the Stoics are not trying to “link” the
value of things according to nature with that of genuinely good things by
means of the concepts of use and benefit, by saying that things according

_________
3 Vogt, 169.
182 STEPHEN MENN

to nature benefit us when they are used wisely. It is true that health and
wealth can be used wisely and well, and that benefit then results (we can
hold off for a minute on whether it would be the health and wealth or
something else that would benefit us); but it is equally true for the Stoics
that disease and poverty can be used wisely and well, and that benefit will
equally result, and so these considerations establish no closer link between
health and goodness than between disease and goodness. I am not sure that
this is as much of a difference from Plato as Katja suggests. Plato does not
usually speak of using disease or poverty, but Euthydemus 281d6-e1 says
of health and wealth and so on that “whenever folly leads [4 <] them,
they are greater evils than their contraries, by so much as they are more
capable of serving what leads them when it is bad; but when prudence and
wisdom lead them, they are greater goods, and in themselves neither of
these [i.e., neither health, wealth, etc., nor their contraries] are worth any-
thing.” While the emphasis here is on health and wealth and so on, it does
seem that disease and poverty can also “serve” or be “led”—and this
seems synonymous with being “used”—although they cannot be used as
efficaciously as health and wealth either for good or for evil, and so health
and wealth can become both greater goods and greater evils than disease
and poverty. The Stoics agree that having wealth or political power can
give you greater scope for living well if you are wise, but they deny that
you will actually live better, and this does seem to be a disagreement with
Plato. Is there also a disagreement when Plato says that things that are in
themselves neither good nor bad can become good or bad when conjoined
with wisdom or folly, and so can benefit or harm? As Plato puts it, “the
neither-good-nor-bad are the things which participate sometimes in the
good, sometimes in the bad, and sometimes in neither, like sitting and
walking and running and sailing, and again like stones and sticks and other
things of this kind” (Gorgias 467e6-468a3), and, given his general under-
standing of participation, this is exactly what we would expect him to say.
But the Stoics, as I noted at the beginning, think that qualities are (not
separate Forms but) bodies which combine into wholes with the things that
come to possess them; and so the Stoics can say, not that health, when it is
conjoined with virtue, benefits and becomes good, but rather that when
health is conjoined with virtue, there comes to be a whole composed out of
health and virtue, and that whole (or, equally, a whole composed out of
disease and virtue) benefits and is good.
Katja suggests that the argument that health and wealth can be used well
or badly is not the Stoics’ main argument for their conclusion that health
and wealth are not good. But what other argument do they have? Katja
says,
 COMMENTARY ON VOGT 183

according to the Stoics’ account, preferred indifferents are “in agreement

with nature”; things like health and wealth have value because of the way in
which they contribute to the natural life of a human being. And this value is a
“mere value” ( '), rather than the value of real goodness, because health
and wealth do not contribute to an agent’s happiness. The wise person selects
and rejects valuable things perfectly, and it is this perfect selection which
makes her life go well, or, be a “happy life.” Whether she gets any of the
valuable things that her selection and rejection is concerned with is irrelevant
to her happiness. This, I suggest, is at least the most prominent argument
which the Stoics make for why health and wealth are not good. 4
These are certainly all things that the Stoics believe, but how are they an
argument that health and wealth are not good? How do I know that
whether I actually get the natural things I select is irrelevant to my happi-
ness? The only way I can see to fill out the argument is to say that health
can be used badly, and therefore is not sufficient for happiness, and that
disease can be used well, so that health is also not necessary for happiness.
Katja suggests that the Stoics’ concept of the right selection of external
things is their replacement for (and perhaps their interpretation of) the So-
cratic concept of the right use of external things, but it seems to me that
the concepts are different, and that the Stoics keep both of them without
confusing them. They use the Socratic argument to show that health and
wealth are not good, and that only virtue (or things that include or presup-
pose virtue) can be good, where virtue gives the necessary and sufficient
condition for using things well. But then they have to face the same chal-
lenge which Socrates also faces in the Euthydemus as a result of this ar-
gument: what is this virtue? Apparently a kind of knowledge (since any-
thing can be used badly if we lack the knowledge of how it should be
used), but knowledge of what, and what sort of )
 does it produce for
us? If we say that it is knowledge of the good, then it is a knowledge
which is knowledge only of itself, and even if such a knowledge is possi-
ble, it seems empty and not desirable; and if the only good )
 it pro-
duces is itself (say, by teaching the same knowledge to someone else),
once again, that seems sterile and not obviously desirable. The orthodox
Stoic answer (by contrast to the heterodoxies of Aristo and Herillus) was
to say that virtue involves knowledge of what is according to nature, and
that one )
 it produces is the consistent selection of what is according
to nature. And this will also be important in Cicero’s story of how the
ideal progressor grasps the concept of the good in the order and harmony
of her acts of selection of the things according to nature: if virtue did not
have the practical aspect of the consistent selection of things according to

_________
4 Vogt, 170-171.
184 STEPHEN MENN

nature, there would be no way to build up to it, and no harmony to grasp at

the end of the process. Thus the concept of right use and the concept of
right selection fill different logical needs in the theory, and both right use
and right selection will be needed for a happy life. If I am a genuinely vir-
tuous person, I will not merely recognize the goodness of virtue and the
badness of vice and the indifference of everything else, but will also rec-
ognize which indifferents are in accordance with nature, and I will select
them, which means that I will get them if nothing external obstructs. But,
of course, I will not always get the things I select: I may select health and
still receive disease. And, if disease is what Zeus gives me, it is my task to
use it well, even if I did not select it and would not have selected it; the
virtuous person will be able to use anything well, and as long as I use well
whatever Zeus gives me, I will be just as happy as if I had gotten and used
well what I selected.
MCGILL UNIVERSITY
 VOGT BIBLIOGRAPHY

Note: Abbreviations (shown in brackets) are used throughout to cite the following
works: Diogenes Laertius, Lives of the Eminent Philosophers: DL; Sextus Empiricus,
Outlines of Scepticism: SE, PH, Against the Mathematicians: SE, M; A. A. Long and
D. N. Sedley, The Hellenistic Philosophers: LS; Johannes von Arnim, Stoicorum
Veterum Fragmenta: SVF.

Primary Texts
Rackham, H. (tr.) 1942. Cicero. Rhetorica, De fato, Paradoxa Stoicorum, De partitione
oratoria. Loeb Classical Library. Cambridge.
Rackham, H. (tr.) 1931. Cicero. De finibus bonorum et malorum. Loeb Classical
Library. Cambridge.
Miller, W. (tr.) 1913. Cicero. De officiis. Text with English translation by W. Miller.
Loeb Classical Library. Cambridge.
Hicks, R. D. (tr.) 1991. Diogenes Laertius. Lives of the Eminent Philosophers [DL].
Vols. 1 and 2. Loeb Classical Library. Cambridge.
Cooper, J. M. (ed.) 1997. Plato. Complete Works. Indianapolis.
Cherniss, H. (tr.) 1976. Plutarch. On Stoic Self-Contradictions and On Common
Conceptions. In Moralia, vol. 13, pt. 2. Loeb Classical Library. Cambridge.
Inwood, B. (tr. with an introduction and commentary) 2007. Seneca. Selected
Philosophical Letters. Oxford.
Bury, R. G., (tr.) 1933-49. Sextus Empiricus. Vol. 1 >PH 1–3@ ; vol. 2 >M 7, 8@; vol. 3
>M 9–11@; vol. 4 >M 1–6@. Loeb Classical Library. Cambridge.
Bett, R., (tr. with an introduction) 2000 (19971). Sextus Empiricus: Against the
Ethicists. Oxford.
Wachsmuth, C. (ed.) 1884. Ioannis Stobaei Anthologii. Books 1 and 2. Berlin. O.
Hense (ed.) 1894-1909. Books 3 and 4. Berlin.

Collections of Fragments
Long, A. A., and Sedley, D. N. 1992. The Hellenistic Philosophers. Vol. 1.
Translations of the Principal Sources, with a Philosophical Commentary. Vol. 2.
Greek and Latin Texts with Notes and Bibliography. Cambridge [LS].
Von Arnim, J. 1903-24. Stoicorum Veterum Fragmenta. Vols. 1–4. Leipzig >SVF@.

Secondary Texts
Alesse, F. 2000. La Stoa e la Tradizione Socratica. Naples.
Bett, F. 2000. Pyrrho, his Antecedents, and his Legacy. Oxford.
Bonhöffer, A. 1890. Epiktet und die Stoa: Untersuchungen zur stoischen Philosophie.
repr. Stuttgart-Bad Cannstatt 1968.
Boudouris, K. J. (ed.) 1994. Hellenistic Philosophy. Vol. 2. Athens.
Brittain, Ch. 2005. Common sense: concepts, definition and meaning in and out of the
Stoa. In Frede, D. and Inwood, B. (eds.) Language and Learning. Philosophy of
Language in the Hellenistic Age. Proceedings of the Ninth Symposium
Hellenisticum. Cambridge: 164-209.
Caston, V. 1999. Something and Nothing: The Stoics on Concepts and Universals.
Oxford Studies in Ancient Philosophy 17: 145-213.
Cooper, J. 2004. Stoic Autonomy. In Cooper: Knowledge, Nature, and the Good.
Princeton: 204-244.
 VOGT BIBLIOGRAPHY 186
Doty, R. 1976. Ennoêmata, Prolêpseis, and Common Notions. The Southwestern
Journal of Philosophy, Vol. VII, no. 3: 143-148.
Fortenbaugh, W. W. (ed.) 1983. On Stoic and Peripatetic Ethics: The Work of Arius
Didymus. New Brunswick and London.
Frede, M. 2001. On the Stoic Conception of the Good. In Ierodiakonou, K. (ed.):
Topics in Stoic Philosophy. Oxford: 71-94.
_______. 1994. The Stoic Conception of Reason. In Boudouris (ed.): 50-61.
_______. 1996. Introduction. In Frede, M. and Striker, G. (eds.) Rationality in Greek
Thought. Oxford: 1-28.
Inwood, B. 2005. Getting to Goodness. In Inwood: Reading Seneca. Stoic Philosophy
at Rome. Oxford: 271-301.
Jackson-McCabe, M. 2004. The Stoic Theory of Implanted Preconceptions. Phronesis
49/4: 323-347.
Long, A. A. 2002. Epictetus: A Stoic and Socratic Guide to Life. Oxford.
_______. 1996. Socrates in Hellenistic philosophy. In Long: Stoic Studies. Cambridge:
1-34.
Obbink, D. 1992. What all men believe—must be true. Oxford Studies in Ancient
Philosophy 10: 193-231.
Reesor, M. E. 1983. On the Stoic Goods in Stobaeus, Eclogae 2. In Fortenbaugh (ed.):
75-84.
Sandbach, F. H. 1971. Ennoia and Prolepsis in the Stoic Theory of Knowledge. In
Long (ed.) Problems in Stoicism, London: 22-37.
Sedley, D. 1983. Comments on Professor Reesor’s Paper. In Fortenbaugh (ed.): 85-6.
Striker, G. 1996. Plato’s Socrates and the Stoics. In Striker: Essays on Hellenistic
Epistemology and Ethics. Cambridge. 316-324.
Vogt, Katja Maria. 2007. Law, Reason, and the Cosmic City: Political Philosophy in
the Early Stoa. Oxford.
 COLLOQUIUM 7
ON NAMES AND CONCEPTS:
MYTHICAL AND LOGICAL THINKING IN PLATO’S SYMPOSIUM

GÜNTER FIGAL

ABSTRACT
This essay is a discussion of the relation between myth and logos in which the
rationality of myth is stressed without reducing it to a prehistory of logos.
Rather, myth and logos are regarded as complementary: Myth is an address-
ing speech about something which is regarded as an overwhelming power,
and logos is the attempt to elucidate the very meaning of what has been
mythically addressed. The discussion refers to Plato’s Symposium. It concen-
trates on a phrase articulated by Socrates (Symposium 199d) in which the
word eros is doubled for the purpose of indicating its mythical and logical
sense. According to the concluding reflections, the relation of myth and logos
can be conceived as a hermeneutical one. Both are related to each other as
text and interpretation.

The genesis of philosophy has often been identified with the step from
myth to logos. Poetry, oral and anonymous as well as written and artificial,
was replaced by thought and investigation. The world was no longer in-
habited by powers which mostly were imagined as human-like or animal-
like figures. Instead of this the world became calculable and transparent,
accessible in its structure and order.
Although—or even because—this view has become so popular, 1 it has
often been doubted. Max Horkheimer and Theodor W. Adorno have stated
in their Dialectic of Enlightenment that the distinction between myth and
logos cannot be made strictly if myth has to be regarded as the prehistory
of logos. If logos has a prehistory the rationality of logos must also be pre-
formed in myth, and, as Horkheimer and Adorno suspect, the irrational
elements of myth may therefore also be embedded in logos. In a compara-
ble manner but with a different intention, Hans Blumenberg has stressed
the rationality of myth. Myth, according to Blumenberg is one of the ef-

_________
1 This—at least academic—popularity was supported by Wilhelm Nestle’s book Vom
Mythos zum Logos (1940 Stuttgart).
188 GÜNTER FIGAL

fective forms of logos itself (Blumenberg 1979, 34). 2 In myth, the distance
from the world which is essential for logos is already gained and estab-
lished. There is no essential difference between articulating the world by
giving names to powers and appearances and conceiving the world by de-
termining its plurality and, thereby, finding an intelligible order.
The objections against a rigid distinction between myth and logos are, at
least to some degree, plausible. The step from myth to logos could not
have been made if myth and logos were completely different from each
other. But nevertheless there is a difference, and this difference has to be
maintained. 3 Otherwise the two words, “myth” and “logos,” would in
principle have the same meaning. And, what is more important, otherwise
there would be no difference between poetry and philosophy at all. Of
course, one could say, as Nietzsche does in The Birth of Tragedy, that this
difference is a result of misunderstanding. What is called philosophy
would, according to Nietzsche, be only a derived form of poetry which as
such had not been understood in the philosophical tradition beginning with
Socrates and thus with Plato. But if this were the case, no critical or con-
ceptual knowledge of poetry would ever be possible—a knowledge we
prove to have already when we find more or less convincing answers to
the question of what poetry is. Although there are reflections on poetry in
poetry itself, without this question the nature or essence of poetical lan-
guage would remain concealed. The question however cannot emerge
from an illusion. Every question as to what something essentially is indi-
cates that this thing has lost its former self-evidence. Thus the question as
such indicates a point of view outside the subject put into question. The
loss of self-evidence is always a fact which cannot be doubted.
So the difference between myth and logos has to be maintained. But on
the other hand it cannot be maintained strictly—as if each were completely
different from the other. If they have something in common and, at the
same time, are different, one could understand and possibly explain why
they do not exclude each other. Then, the change from myth to logos
would not be an absolute one—as if myth had been completely replaced
by logos or had been completely transformed into it. There would be a
change, but what had been changed would at the same time remain. It
_________
2 Cf. also Blumenberg 1979, 18: “Der Mythos selbst ist ein Stück hochkarätiger Arbeit
des Logos.”
3 As to this conviction, I agree with Ernst Cassirer; see his Philosophie der sym-
bolischen Formen. Zweiter Teil: Das mythische Denken (1925) in Gesammelte Werke, vol.
12. Hamburg 2002, esp. pages 28-29. But the concept of myth that will be developed in the
following differs from Cassirer’s insofar as myth is not regarded as the unity of language
and world.
 ON NAMES AND CONCEPTS 189

would still be there, but because of the change it could be seen in a new
light and could play another role.
The complex relation between myth and logos as indicated here is ar-
ticulated paradigmatically in Plato’s dialogues. The critical revision of
myth and poetry in the second, the third, and the tenth book of the Repub-
lic leave no doubt that Plato programmatically goes beyond mythical
thinking and speaking. The Platonic program is a radical one least of all
because according to it myth has to be understood in a rather broad sense.
Not only Homer and Hesiod are critically examined, but also Parmenides,
Heraclitus, and Empedocles. According to a statement of the Eleatic
stranger in the Sophist, they—and also others who cannot so easily be
identified—did not philosophize, but each of them appears to have told a
myth, as if we, the readers of their writings, were children. 4
But on the other hand, Plato himself told and even invented myths, and
because of this it is very unlikely that he thought of himself as of one of
those criticized in the Sophist. It is not very likely that he thought of him-
self as one who did not take care whether others were able to follow his
words or not (Soph. 243a). Plato’s myths, like logoi, then have to be ra-
tional; they must not be narratives just taken for granted. But if the distinc-
tion between myth and logos is valid, they can nevertheless not be logoi
which are only disguised in a mythical appearance. They must not be re-
duced to logos or be regarded as a mere illustration of logically acquired
insights 5 which always can be articulated plainly as such.
But how is the rationality of myths to be understood? The question can
only be answered by orienting oneself to logos. The rationality of logos is
beyond any doubt. The rationality of logos is evident, and as such it has
always been the measure for the eventual rationality of myth. Because lo-
gos is different from myth, its rationality may, as such a measure, even
become especially clear when examined in contrast to myth. This contrast,
on the other hand, may be especially intelligible when it is not simply es-
tablished but enacted in speech. What is to be experienced, then, is ration-
ality in the making.

_________
4 Sophist 242 c: š@
  `
  "
 $ +  K
= 4"+.
5 This is the suggestion of Theo Kobusch, Die Wiederkehr des Mythos. In Janke, M. and
Schäfer, M. (eds.) 2002. Platon als Mythologe. Darmstadt: 44-57.
190 GÜNTER FIGAL

II

A situation of this kind is dramatically realized in the Symposium. Accord-

ing to the entertainment program of the banquet, after Phaidros, Pausanias,
Eryximachos, Aristophanes, and Agathon have praised Eros in speeches,
Socrates takes the floor by asserting that in trying to speak like the others
he could only make a fool of himself. As he says, he had thought that one
should say the truth about everything which is to be praised. The truth
should be given, and then one can choose the most beautiful aspect of it
and express it in speech in the most decent and seemly way (Symp. 198d).
Like the Eleatic stranger, Socrates could also have said that his predeces-
sors in speech had only told myths, as if the listeners were children.
But as to the dramatic situation of the Symposium this would not have
been very revealing. All who participated in the banquet and had agreed to
contribute a speech had explicitly intended nothing else but a myth. They
had agreed to praise Eros because he, unlike other gods, had never been
praised by the poets (Symp. 198d). Eros is taken to be a god, a god of an
outstanding divine character, a god of exceptional degree. 6
This generally held assumption that a god has to be praised is rejected
when Socrates, beginning to examine Agathon, asks whether Eros is of
such a kind as to be eros of something or of nothing— %


@

k
 8 
 H › ), –
#$ (Symp. 199d). The
phrase is decisive. It indicates the Socratic position in nuce insofar as it
draws a sharp distinction between mythical and logical thinking—a dis-
tinction that normally is not drawn in the dialogue, not even in Diotima’s
famous speech on Eros in which also the philosophical importance of Eros
is revealed. The phrase is a caesura, an interruption, a kind of landmark
established in the undifferentiated thinking and speaking which dominates
the dialogue. As such a landmark the phrase is especially remarkable be-
cause Socrates does not simply take another point of view—as if the
mythical speeches had been completely forgotten. Socrates doubles the
word “eros.” In the original Greek text this must have been even clearer
and at the same time more puzzling because the word was written twice in
the same way: ) ). So the ambivalence of the one word was more
conspicuous. The word, firstly, is a name, for a god or, as is revealed later
on in the dialogue, for a $". At the same time it is the Greek term for
love.

_________
6 Symp. 177a: 
-J  7 

-J D.
 ON NAMES AND CONCEPTS 191

As to the name, 7 one should be careful and not simply understand it as

the result of a personification. In the speeches of the Symposium, Eros is
mostly not imagined as a person—only Agathon speaks of him in this
way, and consequently also of his beauty and virtue (Symp. 196b). In all
other instances the name expresses an experience of a certain intensity
which could not be adequately expressed merely by designating it. Desig-
nation, as it were, does not do justice to the intensity of the experience, to
the disquieting character of the experience, and to the fact that what is ex-
perienced appears as an overwhelming power. On the other hand, the
name “Eros” is not a proper name like “Socrates” or “Aristophanes.”
Rather it is a linguistic expression which, although it could be used for
designation, has a name-like character. It is a word which can be under-
stood as a term, but can also be used as a name, and then it has the sense
of a name.
How the name-like character of the word is to be understood might be
indicated by the fact that in the Symposium eros is never discussed by be-
ing distinguished from something else. Such a discussion would have been
possible, e.g., by distinguishing between love and hate. But a distinction
like this presupposes a general view on human life and on the different
affections which belong to it. With designation of that kind eros would be
regarded as one aspect of human life among others. On the other hand
naming is something different. Names call something or someone as such.
Even if something or someone may have the same name as something or
someone else—like Socrates and the young man Socrates in the States-
man—a name refers to something or someone as being unique and origi-
nal. Therefore calling someone by his or her name is not only a reference
but even more an address. To address someone means: to take her or him
as being there independently, but as being at the same time in close rela-
tion to oneself. Also the word “eros,” understood as a name, is an address.
In this case something is addressed which normally is experienced and
which in experience is all too close to oneself. In this case there is no dif-
ference between the experiencing person and that which is experienced so
that the latter cannot be understood and described but only expressed.
When addressed by its name, what is experienced does not loose its close-
ness, but at the same time is shifted into some distance. It has thereby not
become an object but is rather like a person to encounter—close, but now
explicitly different from oneself.

_________
7 The importance of names for mythical thinking has been stressed by Hermann Usener
in Götternamen. Versuch einer Lehre von der religiösen Begriffsbildung (1896 Bonn).
192 GÜNTER FIGAL

This closeness does not get lost with Socrates’ doubling of eros. What
happens is, on the contrary, an intensification. The doubling lays a stress
on what is addressed. At the same time, however, another difference is
established, namely the difference between addressing and being ad-
dressed. Eros and eros—this is to be understood as if something were mir-
rored. Eros is reflected by eros, and thus attention is drawn to eros itself.
This reflection is the origin of conceptual thinking. It is not its beginning
at a certain time in the history of mankind but rather the situation in which
conceptual thinking emerges—and does so anew in every situation of that
kind. The name has already drawn attention to something. Eros, called by
his name as Eros, is not just love as it may happen to someone. Rather it is
love as such. In calling love by name, however, love has not been con-
ceived as such. It could not have been conceived because it would have
been addressed, as if it were someone to encounter. In addressing, atten-
tion is demanded by the addressed; without any reflection, attention goes
straight to the addressed. But now, by the doubling of the name, attention
is drawn back. Addressing has been interrupted. Doubling, saying a word
twice, not only lays a stress on the word but also shows hesitation. The
word itself commands attention. One knows the word, and one knows
what it means. Nevertheless the question emerges what it precisely means.
One wants to know how its meaning as such can be conceived.
Because Socrates wants to know the meaning of eros he asks: of what is
the Eros eros? Socrates makes the intention of his question clear already at
the beginning. He explains his question by saying that he does not want to
know whether Eros would be love of a mother or a father. This would be a
ridiculous question—in part because the speeches up to this point had not
been devoted to that kind of love. The question would be ridiculous be-
cause it would miss the problem. Socrates does not want to know some-
thing about a person named Eros and his factual relations to other persons.
Rather the question aims at the name Eros itself. The name is taken in the
same way as the word “father” or “mother.” Both words can also be used
to address someone. But in such an address one not only turns towards a
person but also expresses a relation. This relation however is essential for
the word’s meaning: Someone is a father or a mother only as the father or
mother of someone, namely of a son or a daughter. The meaning of “fa-
ther” and “mother” consists in a relation. In a comparable way this holds
true also for the meaning of Eros (as a name), which has thus turned into
the meaning of eros (as a concept).
The task of conceiving eros would then be to say more precisely which
particular relation is essential for the meaning of eros. But for the moment
I will not consider further how Socrates performs this task. It is more im-
 ON NAMES AND CONCEPTS 193

portant to stress that Socrates with his question follows his general convic-
tion, as stated in the Phaedo, that the meaning of words is also the mean-
ing of the things expressed by the words. Therefore concentrating on the
words (
) is not really a flight. To examine the things in words is to
examine them in images. But someone who examines them in deeds or, as
one may also say, in reality, in principle does the same (Phaedo 100a). In
contrast to reality, which is generally not regarded as an image, words are
obviously images and can therefore become the subject matter of reflec-
tion. So the meaning of things is accessible in words. In words, this mean-
ing may become transparent.
How this is to be understood becomes clear already with Socrates’ ex-
ample. One has to investigate and to make clear how something—as well
as the word corresponding to it—is essentially combined with something
else, and, correspondingly, with which other words a word belongs essen-
tially together. Thus in the case of eros, the particular relation which eros
essentially is, has to be clarified; also, one has to find out what the essen-
tial correlate of eros is. The concept of eros would consist then of a defi-
nite sequence of determinations of this kind.
According to Plato there seems to be no general rule of how a concep-
tual system of determinations is to be found and established. The most
elaborate discussion of the problem in the Philebus (16c-18e) is only valid
for something which essentially appears in different forms so that its unity
can only be conceived as that of a definite number of essential possibili-
ties. Such are phonemes, e.g., vowels, or musical sounds in distinction
from noises. One can only say what a vowel is by determining how many
vowels there are. By the determination of their definite number, all other
vowel-like linguistically relevant sounds are at the same time recognizable
as variations, which may be caused by different dialects or by the individ-
ual pronunciation of different speakers. The definite number of determina-
tions which defines something like a vowel is situated between the one
and the infinite (Phil. 16e). It defines something as a system of determina-
tions which essentially belong together, so that there never is “the” vowel
in general but only a particular one which is definitely distinguished from
all others.
The way in which the essence of Eros is defined in the Symposium dif-
fers remarkably from this procedure. Although Eros too is defined by a
definite number of determinations, eros is no essential plurality. Eros is
one, unique, and by the name “Eros” one addresses an overwhelming
power. Because of its essential plurality, a vowel could never be named
and addressed by a name. Vowels are no mythical figures. They can ap-
pear only in a myth like the one told by Socrates in the Philebus about the
194 GÜNTER FIGAL

invention of the art of writing ( "") by the Egyptian god Theuth
(17b-18e).
Eros, on the contrary, seems necessarily to be addressed in mythical
speeches. In the Symposium the conceptual determination of Eros is so
subtlety intertwined with mythical elements that it seems not to be com-
plete when developed without them. After Socrates has stated that the rela-
tion of Eros to something must be understood as his being in need of this
very something (200a), Socrates shifts from his dialogue with Agathon to
a report on Diotima’s speech by whom, earlier in his life, he had been
taught everything essential about Eros. Diotima characterizes Eros as in-
between ("'-) and then explains this in-between as the in-between of
human beings and gods (202a, 202d-e). Eros is essentially hermeneutical;
the relation essential for the meaning of eros is to be understood as media-
tion in sacrifice, prayer, and divine orders. Concerning the gods of whom
Diotima speaks, nothing could be said without myth. It is for this reason
that Diotima explains in a mythical way her determination of Eros as in-
between, telling the story of his parents, of Poros, whose name literally
means “way,” and of Poverty () (203a-203e). The two essential com-
ponents of longing for something and being able to reach it by mediation
are addressed here in the same mythical way as Eros himself and thus they
are also understood as powers. Even myth itself is the result of the erotic
relation to that of which one is in need. According to Diotima the in-
between has mainly to be understood as longing for immortality, and this
is not least of all realized in poetry. The poets, like Homer and Hesiod, are
supposed to have children who are more beautiful and more immortal than
those of normal human beings, children by whom they themselves gain
immortality (209c-d).
But none of these elements and aspects of Diotima’s speech can be re-
garded as a sufficient reason for the inevitability of myth. Such a reason is
provided only by the fact that the essential correlate of eros is the beauti-
ful. What Eros is longing for is to bring something forth upon the beautiful
(Symp. 206c), and this entails that the beautiful is essentially accessible in
eros. The mediation which eros is has in consequence to be understood as
the mediation of the beautiful.
Beauty however is essentially not accessible in conceptual thinking or
cognition. According to Diotima beauty itself appears neither as 
 nor
as %" (Symp. 211a). Beauty as characterized in the Phaedrus, is ap-
pearance itself in its most extreme intensification. The beautiful is
%
 (Phdr. 250d), i.e., the most evident of all, evidence without
any concealment. Beauty understood in this way cannot be made accessi-
ble by conceptual thinking and cognition because every thinking and rec-
 ON NAMES AND CONCEPTS 195

ognizing is made possible by it. Thinking and recognizing, every kind of

knowledge, can take place because of the openness of something that can
be thought, recognized, or known as something which is open to reference.
But this openness as such is always already given. Thinking and recogniz-
ing, 
 and %", are always too late.
But the openness or evidence of beauty can be addressed. It can be
called by a name, and thereby it can be confirmed—like the presence of a
person who is already there which is confirmed by addressing the person.
Eros, as characterized by Diotima, can then be regarded as the relation in
which such an addressing is possible and in which it is enacted. As being
already there in every naming eros himself has also to be named—“eros”
must become “Eros.”

III

These considerations lead back to the initial question concerning the rela-
tion between myth and logos and thus also to the question of how to un-
derstand the rationality of myth. According to the Symposium conceptual
thinking follows myth—not in the sense of a chronological order but inso-
far as it is enacted in a realm which has already been mythically articu-
lated. This holds true at least for those topics which, because of their very
nature, have to be addressed. In a slightly different way this has already
been noticed by Aristotle in his reflections on the initial experience of phi-
losophy. Philosophy has its origin in wonder (Metaph. œ 2, 982b 12), and
therefore a lover of myths ("
) is in some way a philosopher
(Metaph. œ 2, 982b 118). Myths are composed of wonderful things 8 be-
cause, as may be added, they are articulations of wonder. According to
Diotima, beauty is wondrous (Symp. 210e). As the most extreme intensifi-
cation of evidence, beauty is the most wondrous of all.
Naming and addressing constitute the most appropriate way to articulate
this wonder. In wonder something affects us; the wonderful is an interrup-
tion of everyday life and thus also a caesura set into self-confident knowl-
edge and action. The affection does not demand a reaction as if it were a
part of the every day life. In wonder, one is not concerned with one’s own
safety, and one is also not inclined to get possession of anything. Wonder
does not affect us directly like fear or attraction. What is wondrous re-
mains at a distance yet without becoming indifferent. It places a demand

_________
8 Metaph. œ 2 982b 19: H ! "@
 -  % ".
196 GÜNTER FIGAL

on us, and this demand is understood and answered in addressing, in call-

ing by name.
But the address is, in this case, not the beginning of a dialogue. What is
wondrous does not, as such, respond. It does not turn toward us but re-
mains aloof. Because of this essential aloofness the wondrous can only be
addressed in an impersonal way. Thus it is a speech about someone or
something. Often it is a narrative, but it can also take the form of praise as
in the speeches on Eros given in the Symposium.
Speeches like these are mythical. They are addresses as well as speeches
about someone or something, and thus retain the aloofness of wonder. In
this aloofness they are nevertheless affected by the wonderful, dominated
by enthusiasm (Apology 22c) in different modes and in different degrees
of intensity as these can be experienced in every speech of the Symposium.
This enthusiasm is not necessarily an affect of the speaker but it is a con-
stitutive element of the speech. The speech is as if it were vibrating with it.
As an address, the speech is affected by its very subject. It is an expression
of this affection and thus—although taking place in the aloofness of won-
der—it is without aloofness. The aloofness is not realized as such. It is
there only as a condition which renders the mythical address possible.
This becomes different as soon as the meaning of a mythical speech is
put into question. With this question, the distance from the subject of
speech is realized; what is sought in the question is something that must be
described and determined. Of course, there have already been elements of
description and determination in the address. Because mythical addresses
do not constitute a dialogue with the addressee they can only be realized in
speaking about someone or something. But with the question about the
meaning of a mythical speech description and determination have become
dominant. Thus the subject matter of the speech counts as subject matter.
The perspective of the question characterized above is that of examining
something by concentrating on how it is articulated in language. It is the
perspective which Socrates calls the flight into the 
. It can also be
called the perspective of interpretation. Interpretation is the attempt to
elucidate a speech or something written by making present its meaning.
Meaning in this case is not so much the intention of the speaker or author
but rather the order of the subject matter, given as a web of linguistically
articulated determinations. This order is the structure of the subject matter
as such. It is 
, the intelligibility of something as being revealed in
language, but also in lines and colors, in sounds. The Latin word for 

understood in this sense is textus. According to the rhetorical tradition,
textus, literally “web,” is the order of a speech and thus also the order of
its subject matter.
 ON NAMES AND CONCEPTS 197

Interpretation is the attempt to reveal the articulated structure or order

by doubling the articulation. But interpretation is nevertheless different
from reproduction or repetition. Its goal is to match the meaning by saying
something differently or even by saying something different. But interpre-
tation is not only a variation, such as may occur in conversation when
something that has not been understood is articulated differently. Rather
interpretation aims at the meaning as such. In giving an interpretation one
takes distance from a speech or from something written, intending its lo-
gos, the logos inherent in it, attempting to reveal it. Even if interpretation
is not exegetical but only a recitation or a performance it is only successful
in presenting the logos or text.
Thus one can say that interpretation is logical. It is an articulation of
logical thinking. Interpretation is logos because it refers to the logos inher-
ent in something. The logos referred to is the necessary condition for the
logical character of interpretation. But on the other hand the logos inherent
in something has to be revealed—presented in interpretation. Interpreta-
tion as logical thinking is the ratio cognoscendi of logos, whereas its ratio
essendi is given by the correlate of interpretation and by interpretation
itself. It becomes explicit in the doubling which interpretation as such is.
The rationality of myth can be understood in this way. It is inherent in
myth—otherwise myth could not be interpreted philosophically, i.e. in
logical thinking. But it has to be interpreted to become explicit in its logi-
cal structure. Such an interpretation can of course be a critical one—it can
be enacted by distinguishing between aspects of a speech or of something
written which do not fit the structure of meaning or which cannot be main-
tained in attempting to clarify a particular meaning. In this critical sense
Socrates in the Republic gives an interpretation of the poetical myths about
the gods—doubling myth by “theology” (Rep. II, 379 a). But theology in
this sense does not replace myth. In general, interpretation cannot replace
its correlates, and therefore myths as addresses cannot be replaced. They
may only be doubled, and thereby understood as being accessible for in-
terpretation. Every kind of speech or writing, once discovered as being a
correlate of interpretation, is understood from a logical point of view. Thus
a myth or poem may be understood logically and still be a myth or a
poem. There is something like a step from myth to logos, but with this step
nothing gets lost. The step does cross the border of mythical thinking but it
does not lead out of the region of myth once and for all. The border can be
crossed in both directions. There is no transformation which would be
198 GÜNTER FIGAL

Aufhebung in Hegel’s sense 9 —as if something would be negated and con-

served on another level, like poetry being conserved only academically in
departments of literature. Even if reflected and conceived in a philosophi-
cal way, poetry can be read as poetry. It does not lose its poetical charac-
ter; the experience of this character is still possible. The same holds true
for myth. Conceiving does not replace addressing. Eros will ever remain
something to be addressed.
ALBERT-LUDWIGS-UNIVERSITÄT FREIBURG

_________
9 This is the critical assumption in Dennis Schmidt’s elucidating commentary to this pa-
per.
 COMMENTARY ON FIGAL

DENNIS J. SCHMIDT

ABSTRACT
Tracing the contours of Günter Figal’s analysis of the character and conse-
quences of the emergence of philosophy as this emergence is presented in
Plato’s Symposium, this paper also takes some issue with Figal’s tendency to
grasp this emergence too much as a philosopher, that is, as still too committed
to the logic of the concept and its transparency. Figal argues that, in the end,
nothing is lost in the move into logos, into the language of philosophy, from
the language of mythos that precedes it. This paper highlights some of Figal’s
very interesting and innovative remarks on this issue in Plato, but concludes
by questioning the legitimacy of his final claim that “nothing is lost” in the
move into logos.

Beginnings are difficult to identify. Real beginnings, the origins of some-

thing new, mark a transformation, a shift, and so one might think that it
would be easy to say “here and now, in this way, something new has come
into the world.” But it is not easy at all. Hegel recognized that begin-
nings, even the most radical, are never pure, never fully absolved of what
precedes them and he understood this enigma of the beginning as the real
beginning of the dialectic, and so turned the question of the beginning into
the beginning of his own system. Even as the negation of a past, every
new beginning still stands in some relation with what it surpasses, sup-
plants, or suppresses. It is also the case that such negation of the past
needs to be taken as a reminder that every beginning is accompanied by an
ending.
This difficulty of identifying beginnings is never more acute, never
more difficult, than in the case of speaking of the beginning of philosophy.
Indeed, today we seem more comfortable identifying the “end” of philoso-
phy than we do speaking about its beginning. From the start, philosophers
have struggled with speaking of the beginning of philosophy itself. One is
not unjustified in suggesting that this question is one of the very first for
this self-reflexive project: once the elements of philosophy began to
emerge—concepts, logic, and the self-consciousness of the processes of
thought generally—those elements and the possibilities that emerge with
them became the guiding questions for reflection. Plato, who devotes
much effort to this question, defines it in terms of the movement out of a
manner of speaking according to “myth” and into the speech of “logos.”
Plato was well aware of the high stakes of this question: setting philoso-
phy apart from other forms of speech—but in such a way as to demon-
200 DENNIS J. SCHMIDT

strate its capacity to grasp the truth of those other forms—was crucial if he
was to legitimate the claim that the language of philosophy, the logos, is
justified in judging every form of speech.
Günter Figal has renewed this question of the beginning of philosophy
as it is elaborated by Plato and he is fully aware of its complexities and of
its stakes. He is also aware that such a beginning is never able to be un-
derstood as a chronological matter, but that it must be understood from out
of the “situation in which conceptual thinking emerges—and does so anew
in every situation of that kind.” 1 The beginning of philosophy is thus pe-
culiar in yet another sense: it is never finished beginning but remains es-
sentially wedded to its own emergence, to the birth of conceptuality and
the new perspective of thinking that language makes possible.
Figal makes clear that myth can never be reduced to logos and that the
difference between these forms of speech must be maintained; nonethe-
less, he knows as well that there must be a common ground holding them
in a relation which lets a transition from myth into logos be possible. 2
Nonetheless, the question that guides him betrays a curious prejudice
about the nature of intelligibility and truth. This happens when he asks:
“But how is the rationality [emphasis mine] of myths to be understood?
The question can only be answered by orienting oneself to logos.” 3 Say-
ing that this is a question that “can only be answered by orienting oneself
to logos” presumes a privilege and priority of logos over myth. Such is, of
course, the assumption that governs philosophy, but it is this assumption
that is called into question by the “end” of philosophy. The “end” of phi-
losophy is not its cessation or disappearance. Quite the contrary, the end
of philosophy marks its opening to other forms of speaking and thinking
that had, from its first beginning, been marginalized or even suppressed.
Among these other forms are the poetic practices about which Günter Fi-
gal speaks. One can, perhaps, see already my final dispute with Figal: I
will suggest that, in the end, he remains just a bit too much of a philoso-
pher, just a bit too willing to believe that nothing is lost to the logos and
rationality. But before elaborating upon that final point, I would want to
highlight two aspects of Figal’s analysis of these matters: first, his discus-
sion of “rationality in the making” in which he finds an account for both
the transition from myth to logos and for the birth of the conceptual lan-

_________
1 Figal, 192.
2 Here the assumption seems to be very much like the one that Aristotle asserts when he
writes that “Interaction between two factors requires a prior community of natures between
those factors.” De Anima, 429b25.
3 Figal, 189.
 COMMENTARY ON FIGAL 201

guage of philosophy; second, his remarks about beauty and the openness
that lets thinking be possible.
He begins this analysis by finding a parallel situation “dramatically real-
ized in the Symposium.” In Socrates’ speech we find this transition mir-
rored. It is a mirroring that begins with the doubling of the word “eros”
“when Socrates, beginning to examine Agathon, asks whether Eros is of
such a kind as to be eros of something or of nothing— %


@

k
 8 
 H › ), —
#$ [Symposium 199d].” 4 I
will not trace Figal’s very interesting discussion of the specific character
of the ambivalence of this doubling. That discussion is important, but
quite nuanced in the distinction drawn between addressing and being ad-
dressed. What must be pointed out though is the upshot of this doubling,
namely, that in this intensification, this concentration of what is said in
speaking of eros, “attention is drawn to eros itself.” Eros as such becomes
a question and, as Figal points out, “this reflection is the origin of concep-
tual thinking.” But the concept does not emerge all at once, finished, and
unproblematic. It emerges first as a question: “Doubling, saying a word
twice, not only lays stress on the word but also shows hesitation. The
words itself commands attention . . . . the question emerges what it pre-
cisely means. One wants to know how its meaning as such can be con-
ceived.” 5 What emerges is a new relation to the word, one in which the
meaning of the word steps forth as such and as a question. Self-
consciousness begins at this point as does the project of philosophy itself.
One might say that when the transition out of myth begins at that moment
the logos begins to be visible as such. But as the doubling of the name
named in myth, this transition does not entail the simple rejection of what
is said there, but rather marks its sublation. According to Figal, this is the
point at which the rationality of myth itself becomes visible. The name is
en route to becoming an idea.
The second point that I want to emphasize in Figal’s paper concerns
something which is only fully evident in thinking through the special char-
acter of eros itself—here one sees that Figal’s choice of the Symposium to
discuss this question about the myth and logos is not at all accidental.
Here one comes to understand the necessity, even the primacy, of myth for
logos.
Figal reminds us that “the essential correlate of eros is the beautiful.”
What eros longs for “is to bring something forth upon the beautiful.
Beauty however is essentially not accessible in conceptual thinking or

_________
4 Figal, 190.
5 Figal, 192.
202 DENNIS J. SCHMIDT

cognition. According to Diotima it does not appear as logos.” 6 This is

key point and one about which I find myself in some disagreement with
Figal. Like Kant, who argues in the third Critique that beauty needs to be
understood first by its apartness from and resistance to the concept,
Diotima sets beauty outside of the reach of conceptual reason. Logos fails
to grasp beauty because beauty is “appearance itself in its outermost inten-
sification.” Moving away from appearance as such, logos will always
miss the self-evidence of the beautiful which is firmly, untranslateably,
lodged in appearance itself. Here Figal’s point is especially well taken
(and in this, his reading of Plato pushes beyond what we learn about the
relation of the beautiful to the concept from Kant): “Beauty understood in
this way cannot be made accessible by conceptual thinking and cognition
because every thinking and recognizing is made possible by it. Thinking
and recognizing, every kind of knowledge, can take place because of the
openness of something which can be thought. But this openness as such is
always already given. Thinking and recognizing, logos and episteme, are
always too late” [emphasis added]. 7 He might have added that beauty is
the preeminent manner in which this openness is given. The loss of the
self-evidence of what appears, that is the rise of the question and the move
into logos, signals as well the withdrawal of the beautiful. This with-
drawal of the beautiful belongs to the same movement that describes the
movement out of myth. There is a retreat in this movement, not simply an
overcoming. My concern with Figal’s analysis is that he might underesti-
mate the force and consequences of this retreat: something is lost that can-
not be recovered here—or so I would argue.
Figal’s way of charting the move from myth to logos, the movement
into philosophy, does not oversimplify this relation as is so often the case.
If it is the case that the doubling of what is said in myth leads to a new and
reflected form of language, opening up the possibility of logos, it is also
the case that logos needs to be understood “as it is enacted in a realm
which has already been mythically articulated.” Philosophy is nourished
by this movement and so retains some trace of the conditions of its own
emergence.
But Figal’s chief concern is to understand the character and the conse-
quences of this move into the reflected language of the concept. His con-
clusion comments upon the most enduring consequence of this move to
examine “something by concentrating on how it is articulated in lan-
guage.” Figal refers to this new perspective of thought as “the perspective

_________
6 Figal, 194.
7 Figal, 194-195.
 COMMENTARY ON FIGAL 203

of interpretation.” The movement into logos is equally a move that re-

quires interpretation, that is “the attempt to elucidate a speech or some-
thing written by making present its meaning . . . . [which] is given as a
web of linguistically articulated determinations.” 8 The need for interpreta-
tion, the inexhaustibility of this move into philosophy, is able to be seen
from its incipient moment. From the outset, philosophy has a hermeneuti-
cal character.
Figal’s analysis of this move, this birth into philosophy, adds much that
is new to a longstanding debate. He recognizes the subtleties and delicacy
of this movement from myth to logos, neither flattening each into the
other, nor simply suggesting that logos trumps myth. His treatment of this
theme, not just in Plato’s Symposium, is original and compelling. How-
ever, I suggested at the outset of my remarks that I would conclude by
arguing that Figal remains a bit too much of a philosopher, a bit too caught
by its assumptions and so he might miss what it is that will continue to
elude philosophy, just as beauty will elude the concept. Figal’s final sen-
tence makes the point that I want to call into question. He writes: “Every
kind of speech or writing, once discovered as being a correlate of interpre-
tation, is understood from a logical point of view. Thus a myth or a poem
may be understood logically and still be a myth or a poem. There is some-
thing like a step from myth to logos, but with this step nothing gets lost”
[emphasis added]. 9 In coming to this Hegelian sounding conclusion, I
worry that the limits of philosophy, the limits of conceptual reason, that
Günter Figal has so carefully exposed, might be forgotten. The birth of
philosophy, the movement into the concept and “the logical point of
view,” is not the move into transparency. Conceptual reason, as Kant so
powerfully demonstrated in the Critique of Judgment, might find much
beyond its reach and much that resists any translation into the language of
the concept. It might well be that the language of philosophy must under-
stand itself as but one form of language in a dialogue where translation,
however necessary, is exceedingly difficult and much is inevitably lost.
THE PENNSYLVANIA STATE UNIVERSITY

_________
8 Figal, 196.
9 Figal, 197.
 FIGAL/SCHMIDT BIBLIOGRAPHY

Blumenberg, H. 1979. Arbeit am Mythos. Frankfurt am Main.

Cassirer, E. 2002 Philosophie der symbolischen Formen. Zweiter Teil: Das mythische
Denken. Gesammelte Werke, vol. 12. Hamburg.
Nestle, W. 1940. Vom Mythos zum Logos.
Kobusch, Theo. 2002. Die Wiederkehr des Mythos. In Janke, M. and Schäfer, M. (eds.)
Platon als Mythologe. Darmstadt: 44-57.
Usener, H. 1896. Götternamen. Versuch einer Lehre von der religiösen
Begriffsbildung. Bonn.
 COLLOQUIUM 8

YET ANOTHER WAY TO READ THE REPUBLIC?

ALASDAIR MACINTYRE

ABSTRACT
It is argued that a variety of attempts by commentators to identify the under-
lying doctrines of the Republic have all failed and that the nature of the diffi-
culties that confront anyone who attempts to identify them has too often been
misunderstood. Plato’s principal intention in the Republic was to present his
readers not with a set of doctrines, but with an aporia and to challenge those
readers to think their own way through it and out of it. Accounts of the char-
acter of Glaucôn, Adeimantus, and other members of Socrates’ audience, of
the exposition of the diagram of the Line, of the justifications of justice, and
of Plato’s jokes are advanced in support of this interpretation. And it is fi-
nally suggested that Plato directs his readers’ attention to one possible way of
resolving the central aporia.

In this paper I argue that in the Republic Plato’s primary concern is to pre-
sent us not with a theory, but with an aporia. It is an aporia that Plato
confronted as a result of developing a number of independent lines of
thought about Forms. The outcome of these developments had been a
characterization of Forms such that it had become puzzling what kind of
educational transformation a mind must undergo, if it was to become ca-
pable of apprehending Forms. The Republic is often interpreted as if its
primary purpose was to present and to defend a set of theses both about
the nature of Forms and about the nature of that educational transforma-
tion. And commonly those who interpret the Republic in this way con-
clude that Plato was in a variety of ways wrong-headed and mistaken, the
evidence for this being drawn from what are taken to be the unsatisfactory
arguments that Socrates advances to his interlocutors.
Against this I want to suggest that Plato’s views and arguments are not
to be identified with those put into the mouth of Socrates, that Plato was
well aware of at least some of the incoherences and confusions in those
views and arguments, and that his presentation of them is an invitation to
his readers—and doubtless originally to his hearers—to join him in think-
ing through the aporia with which he is engaged. I do not expect those
who know the text of the Republic best, the philosophers and scholars who
have commented on that text, to find this interpretation compelling, in part
because the assumptions that I bring to the text are in important ways at
odds with theirs. I therefore begin by identifying and putting in question
some of those assumptions.
206 ALASDAIR MACINTYRE

I. The Project: Putting the Commentators to the Question

The literature on the Republic is so vast that it seems foolhardy to add to

it. It seems that everything worth saying must have been said and indeed,
were we to read our way through the commentators, ancient, medieval and
modern, as a prologue to writing about the Republic, death would inter-
vene before we began to write. Yet this Great Too Much itself poses an
initial question: Why did it not inhibit more writers in the latter part of the
twentieth century and at the beginning of the twentyfirst?
Consider the genre of the one-volume or long essay introduction to the
Republic, taking note only of what is in English. Such introductions rely
on translations and the first of them were sequels to two nineteenth cen-
tury translations, that by Davies and Vaughan, published in 1852, and that
by Jowett, published in 1871. The earliest commentaries were both by
pupils of Jowett, Bernard Bosanquet and R. L. Nettleship, in 1895 and
1901. Bosanquet used Davies and Vaughan, Nettleship generally followed
Jowett, although making his own translations when he quoted. There is
then a gap of fifty years until N. R. Murphy’s The Interpretation of Plato’s
Republic in 1951. But after that no decade lacks its commentary: R. C.
Cross and A. D. Woozley in 1964, Leo Rauch’s Monarch notes in 1965, a
volume of essays edited by Alexander Sesonske in 1966, Alan Bloom’s
translation and commentary in 1968, Nicholas P. White in 1979, Julia An-
nas and C. D. C. Reeve in the nineteen eighties, Nickolas Pappas, Sean P.
Sayers and Daryl H. Rice in the nineteen nineties, Harry Eyres in 2001,
Basil Mitchell and J. R. Lucas as coauthors in 2003, Stanley Rosen in
2005, the Blackwell Guide to Plato’s Republic in 2006, and last, but not
least, the Cliffs Notes, of which there are two sets, a shorter one by Mary
Ellen Cross in her Greek Classics in 2001, preceded by a longer one by
Thomas Thornburg, in 2000. This makes a total of eighteen one volume
commentaries in English, all of them of some value and some outstanding,
in just over one hundred years. Whence this astonishing flow?
Part of the answer is: the market. Introductory philosophy teachers of-
ten use the Republic and the production of English translations for them
matches that of commentaries: two in the eighteenth century, four in the
nineteenth, at least fourteen in the twentieth, one already in the twentyfirst.
Commentaries, like translations, sell. And publishers believe that com-
mentaries can mediate between the Great Too Much of Platonic scholar-
ship and the basic needs of unsophisticated students and of their teachers,
who may themselves be philosophically sophisticated, but are rarely Plato
scholars. Yet on the face of it the Republic is not an obscure text, but
strikingly accessible. Certainly students do need to be guided through its
 YET ANOTHER WAY TO READ THE REPUBLIC? 207

arguments by a teacher. And it is as ideal teacher, as role model, that the

best of such commentators present themselves. But in fact they always do
something more than guide students through the text. All to greater or
lesser degree supplement the text, making themselves Plato’s co-authors.
How do they do this?
Consider the treatment of some disputed issues on which philosophi-
cally acute scholars have held opposing views. What is the relationship
between the Form of the Good (509 d-e), and that “something that is not
hypothetical,” that 

, at which reasoning arrives when it grasps
the , the first principle, of everything (511 b-c)? Are they the same or
different? Just how close is the relationship between the diagram of the
Line and the story of the Cave? Are they related as a landscape is to the
story of a journey through and beyond it? Or are they to some degree in-
dependent of each other, each with its own distinctive point? It is true that
Socrates remarks of the Cave and the Sun at 517b that “this whole im-
age…must be fitted together with what we said before” and identifies
some correspondences between the Line and the Cave. But how are we to
spell this out further? Here scholars divide.
Some have followed Richard Robinson in holding that “on the whole the
evidence seems sufficient” for identifying the Form of the Good with that


 that we arrive at when we reach the  of our mathematical
enquiries. 1 But Kenneth Sayre has argued that “formidable difficulties
appear when this assumption is examined critically.” 2 This disagreement
is related to others. Is the 

 which is the terminus of a dialecti-
cal enquiry into the mathematical sciences a proposition or a set of propo-
sitions which can then function as a premise or premises? Or is it an 
of quite a different sort? To these questions too rival answers have been
advanced.
The authors of one-volume commentaries uniformly take it that it is
their duty either to decide between such competing scholarly positions or
else to declare that there is no way to decide between them. (As have done
other writers on the Republic—including me. 3 ) In so doing they com-
monly make two assumptions that I want to question. The first is that in
_________
1 Richard Robinson, Plato’s Earlier Dialectic, London: Clarendon Press, 1953, chapter
X, section 4.
2 Kenneth Sayre, Plato’s Literary Garden, Notre Dame, IN: University of Notre Dame
Press, 1995, chapter 6, section 2.
3 See A. MacIntyre, “The Form of the Good, Tradition, and Enquiry” in Value and Un-
derstanding, ed. R. Gaita, London: Routledge, 1990, and A Short History of Ethics, second
edition, London: Routledge, 1998. The interpretation of the Republic in chapter V of Whose
Justice? Which Rationality?, London: Duckworth, 1988, is less unsatisfactory.
208 ALASDAIR MACINTYRE

the Republic the character Socrates is a mouthpiece for Plato’s views. A

second is that, when we are presented by Plato with passages susceptible
of rival interpretations, then either there is a correct interpretation, one
that, had Plato been confronted by it, he would have been able to recog-
nize as what he intended, or else the text is hopelessly opaque.
This second assumption can be reformulated, so that it omits any refer-
ence to Plato’s intended meaning. What makes one particular interpreta-
tion correct and its rivals mistaken, it may be thought, is that only that
interpretation is sufficiently coherent with what is said elsewhere in the
text or in other texts by Plato. The correct interpretation of a disputed pas-
sage, on this formulation supplies not what Plato had in mind, but what he
ought to have had in mind. Both formulations imply that Plato uninten-
tionally left the Republic unfinished and unclear and it is now up to us to
clarify and complete it, so becoming his co-authors.
It is just this that I deny and against either formulation of the commenta-
tors’ second assumption I argue as follows. With Plato’s as with other
ancient philosophical texts, where we have no reason to believe that the
text is corrupt, the identification of two or more incompatible meanings
that a passage can bear, may be evidence that Plato had not distinguished
those two meanings, but it is also possible that what we confront is not a
failure by Plato to distinguish, but rather a case of Plato posing alternative
possibilities to his readers. If in our interpretative strategy we are guided
by this thought, we will be ascribing to Plato a certain attitude to those
readers. So let me begin by characterizing this attitude.
The text of the Republic from Book V onwards provides evidence that
Plato had been pursuing more than one line of enquiry, that the relation-
ship between those lines of enquiry was not yet clear to him, and that he
was well aware of this. So the considerations that directed him towards
the 

, considerations that find their expression primarily in the
diagram of the Line, were not the same as those that had directed him to-
wards the Form of the Good, considerations that find their expression in
the story of the Cave. What we do not find in the Republic is any account
of how these two lines of thought are “to be fitted together.” Yet this is
not just an omission. Plato intends his readers—or some of them—to dis-
cover an aporia. He is not inviting us to decipher an unclear text, so as to
discover which views he held or should have held, but so as to do philoso-
phy for ourselves.
Put baldly like this, what I have suggested will not seem compelling.
Let me try to make it more plausible by identifying what I take to be
Plato’s intended relationship to the readers of the Republic, a relationship
 YET ANOTHER WAY TO READ THE REPUBLIC? 209

best defined by contrast with the relationship between Socrates and his
interlocutors and audience in the Republic.

II. Whom is Socrates Addressing?

About those interlocutors and that audience we, Plato’s readers, whether in
the 380’s or at any time since, know four things. The first is that some will
soon die violently at the hands of allies and friends of others who are pre-
sent. Polemarchus will be executed by order of the Thirty Tyrants, among
them Critias, uncle of Plato, Glaucôn and Adeimantus. Lysias will return
from exile with the invading democratic army in the civil war in which
Glaucôn may have died. And Socrates will be executed by the democrats.
The connection between $
- and death is more immediate than any
of the participants realize.
Secondly, those participants are lovers of spectacles and lovers of con-
versations. They have come down to the Peiraeus for a festival procession
in honor of a Thracian goddess, and are staying on for horse-racing with
torches and conversation with other young men (328a). Later they will
hear themselves characterized by Socrates as “in love with listening,”


, and “in love with sightseeing,” 

"
, lovers of sights
and sounds and therefore “unable to see and embrace the nature of the
beautiful itself” (476b), so that they cannot be philosophers. Even those
who have cultivated some , such as Thrasymachus and Lysias, will
find themselves as lovers of  lumped together with the lovers of
sights and sounds.
The context is important. Socrates has just asserted that “there can be
no happiness either public or private in any other city” (473e) and pref-
aced this by remarking that until philosophers become kings or kings phi-
losophers (473c) the just city will not be possible. By asserting that the
reign of philosophy is indispensable for the achievement of happiness Soc-
rates has appealed to a consideration that his hearers take very seriously.
For a third thing that we know about them is that they care about their own
happiness. About what happiness is, let alone about how it is to be
achieved, they are of course shown to be confused. But they—and Socra-
tes—agree that the evaluation of any way of life is a matter of whether it
does or does not make those who live it out happy. On this Adeimantus is
their spokesman and it is he who interrupts Glaucôn and Socrates to ask
how Socrates would defend himself against the accusation that the guardi-
ans will not be happy and that it will be their own fault that they are un-
210 ALASDAIR MACINTYRE

happy (417b-419a). Socrates defers his answer, but Plato has made it
clear that without an answer his enquiry will have failed.
Given that so long as they are lovers of spectacles and conversations
they are incapable of philosophy, and given that the reign of philosophy is
indispensable for happiness, Socrates’ interlocutors and hearers are in the
unhappiest of situations, something that they do not yet recognize. And
here a first contrast between us, Plato’s readers, and them, Socrates’ audi-
ence, appears. For Plato, by showing us what Socrates had so far failed to
show them, has posed the questions of how and to what extent we, like
them would have to be transformed in order to be able to engage in phi-
losophy and of what it would be to engage in philosophy. Another thing
that we know about Socrates’ interlocutors and audience is that, even if so
far incapable of philosophy, they are not wholly ignorant of it. When Soc-
rates first introduces the semitechnical notion of a Form in Book V (475e-
476b)—the word ‘8$
’ had already been used in its everyday prephi-
losophical sense in Book II (357c-358a)—he moves rapidly forward in a
way that seems to presuppose prior knowledge in his hearers and this is
confirmed in Book X, when Socrates says “As you know (my italics), we
customarily hypothesize a single form in connection with each of the
many things to which we apply the same name” (596a). (I have generally,
though not quite always used the Reeves-Grube translation in my quota-
tions.)
Moreover, when Socrates opens up the question of the nature of the
good, he says to his audience that “you have often heard it said that the
form of the good is the most important thing to learn about . . . . You knew
very well that I was going to say this…” (504e-505a). Then he adds that
“you certainly know that the majority believe that pleasure is the good,
while the more sophisticated believe that it is knowledge” (505b). So
those with whom Socrates converses have already encountered philoso-
phical discussion. But what kind of philosophical discussion? They may
after all have been present only at the kind of debate that Socrates de-
scribes as “clever and contentious, aiming only at achieving reputation and
strife, whether in the law-courts or in private discussions” (499a), the ef-
fect of which is to discredit philosophy with the many (500b). In which
case what has passed for philosophy with them will be one more barrier to
their understanding of the relationship of philosophy to happiness. So let
us ask how far Socrates may be able to take them from their initial condi-
tion.
 YET ANOTHER WAY TO READ THE REPUBLIC? 211

III. Who is Capable of Philosophical Enquiry?

The Republic is a drama of successive beginnings: first of all Socrates’

initial narrative in the opening sentences of Book I, then a second begin-
ning with the challenge of Glaucôn and Adeimantus in Book II, and a third
with the introduction of the philosopher as the central topic for discussion
three quarters of the way through Book V (474b). It is a familiar and also
a true thought that what both we and Socrates’ hearers principally learn
from Book I are the limitations and insufficiencies of elenchus in Socrates’
dealings with Thrasymachus. But Plato the author reveals sympathies
with Thrasymachus that Socrates the character does not share.
Consider the accusations that Thrasymachus levels at Socrates when he
first breaks into the conversation (336a-c). Thrasymachus is exasperated,
but so are we, Plato’s readers, and I think that Plato intended this effect.
Thrasymachus accuses Socrates of engaging in disputation out of “com-
petitiveness or love of honor” and it is, he suggests, a sign of this that Soc-
rates is prepared only to refute answers proposed by others, but not to pro-
pose answers himself. Plato through his portrayal of Thrasymachus has
posed a problem about Socrates. For, if Socrates proceeds to attempt only
to refute Thrasymachus, he will thereby have vindicated Thrasymachus.
And indeed for a time in the remainder of the Book I it seems that he may
be going to do just that.
The effect of Thrasymachus’ intervention is to connect two questions
whose relationship is going to be crucial in the rest of the Republic. To
what end is this whole enquiry directed? And by what type of desire is it
directed? C.D.C. Reeve has emphasized that, on the view expounded by
Socrates, what each of us is able to learn depends on the nature of the de-
sires that rule in our souls. 4 Different types of desire make us attentive to
and focussed upon different kinds of object with the result that what Reeve
calls the cognitive resources of different types of soul are partially deter-
mined by their dominant desires. And the clear-sighted Thrasymachus,
wrongheaded as he may be about the nature of justice and the nature of
ruling, is someone who attends to and focuses on whatever is relevant to
the competitive lover of victories and who therefore recognizes certain
truths, the acknowledgement of which is necessary if Plato’s purposes are
to be achieved.
The first such is that in every actually existing polis what is taken to be
just is what is prescribed by law and that those laws are made in what they
_________
4 C. D. C. Reeve, Philosopher Kings: The Argument of Plato’s Republic, Princeton:
Princeton University Press, 1988, 2.4 and thereafter.
212 ALASDAIR MACINTYRE

take to be their own interests by the ruling party. A second is that to be

unjust is to pursue what one takes to be one’s own good at the expense of
others, so that tyrants are the exemplars of injustice. Thrasymachus’ ini-
tial formulations are crude and in need of revision and among the com-
mentators Cross and Woozley devote admirable effort and ingenuity to
constructing an interpretation that rescues Thrasymachus from inconsis-
tency, just the kind of work that Plato intended his readers to undertake.
We should not then be surprised to discover in Books VIII and IX that
some of what was said by Thrasymachus in Book I is repeated by Socra-
tes. For in each type of polis, so Socrates asserts, laws are made in their
own interest by rulers and what passes for justice is administered on behalf
of rulers. It is in this context that we learn what it is to have a timocratic,
oligarchic, democratic or tyrannical soul. Each type is ruled by desires
whose modes of attention prevent it from directing its attention towards
the soul’s true good and so prevent it from understanding itself. But now
we have to remember that Socrates’ conversational partners in the Repub-
lic are all of the oligarchic or democratic party and as such will have de-
mocratic, oligarchic or at best timocratic souls, perhaps some mixture of
these. So they too will have desires that limit their cognitive resources and
their cognitive possibilities. What then will they make of Socrates’ teach-
ing?
Glaucôn and Adeimantus had defined their initial attitudes with refer-
ence both to Socrates and to Thrasymachus. Glaucôn provisionally identi-
fies himself with the core of the Thrasymachean position. Thrasymachus
had said that anyone with great powers outdoes others in satisfying his
desires (343e-344a). And Glaucôn says that anyone with the freedom to
do what he likes acts from a desire to get more and more. Thrasymachus
used the verb ‘
+,’ Glaucôn uses the noun ‘
'.’ But
Glaucôn differs from Thrasymachus in that he knows that he does not as
yet know whether what Thrasymachus asserted is in fact true or false. So
at this stage neither Thrasymachus nor Socrates have succeeded in secur-
ing the assent of either Glaucôn or Adeimantus, who want something more
and other than Thrasymachus had wanted, namely the truth.
What does that tell us about Glaucôn’s character? Reeve has an answer
that deserves serious consideration. Cephalus and Polemarchus he takes to
be money-lovers, Thrasymachus a lover of honor and victory, while
Glaucôn and Adeimantus he takes to be already wisdom-lovers. Reeve
interprets those later passages in which Socrates ascribes prior acquaint-
ance with the theory of forms to his hearers as referring only to Glaucôn
 YET ANOTHER WAY TO READ THE REPUBLIC? 213

and Adeimantus and infers that “they are philosophic men.” 5 My diffi-
culty with this is that, if Glaucôn and Adeimantus were already acquainted
with the theory of forms to the extent that would be necessary for them to
be “philosophic men,” they would not need the expositions of the Line and
the Cave and these could only be intended for the other members of Socra-
tes’ audience, just those who, on Reeve’s view, not being philosophic
men, would be unable to profit from these expositions. Moreover, at the
point at which Glaucôn and Adeimantus take over the dialogue with Soc-
rates from Thrasymachus they remain open to the possibility that Thrasy-
machus is in the right. But this means that they do not as yet understand
$
- from the standpoint of sophia. The standpoint of sophia is not
yet their standpoint. And this could only be so, if they were not yet lovers
of wisdom. So I conclude that Reeves is mistaken.
In Book VIII after Socrates had described a type of constitution that is a
mixture of oligarchy and timocracy, he asks “Who is the man who corre-
sponds to this constitution?” (548d). Adeimantus replies that “he would
be very like Glaucôn here, as far as the love of victory is concerned” and
Socrates agrees that Glaucôn is a lover of victory, while denying that
Glaucôn has other qualities of a man fitted to live under such a constitu-
tion. What this suggests is that Glaucôn, although not yet a wisdom-lover,
has the qualities necessary for philosophical enquiry, because he is stable,
courageous, graceful, a hard worker both physically and mentally, with a
good memory and a strong desire to learn (535a-c; see also 412b et seq.).
What he and Adeimantus still lack is the necessary education in dialectic.
Without it how far can either of them go?
They are of course, as we already noted, aware that elenchus will not
yield the kind of conclusion to which they aspire. And they accept from
Socrates the new method of enquiry that he pursues from Book II on-
wards. That method is never fully spelled out, but it amounts to this. Soc-
rates proceeds through a series of assertions and inferences, at each point
leaving it open to any participant in the conversation to voice an objection.
And the degree of confidence that he, they and we are entitled to have in
Socrates’ conclusions is a matter of the range and power of the objections
advanced and the quality of the responses to those objections. So refuta-
tions in the form of objections now function constructively, and so do
refutations of refutations in the form of responses to objections. But,
when the needed and relevant objections have not yet been advanced, let
alone debated, we can have at best no more than grounds for a tentative
and provisional acceptance of Socrates’ conclusions, as Plato makes clear
_________
5 Reeve 1988, 41.
214 ALASDAIR MACINTYRE

when he has Socrates say of one particular assertion that “in order to avoid
going through all these objections one by one and taking a long time to
prove them all untrue, let’s assume that our hypothesis is correct and carry
on. But we agree that if it should ever be shown to be incorrect, all the
consequences we’ve drawn from it will also be invalidated” (437a).
Socrates’ conversational partners are not troubled by this. But we,
Plato’s readers, should be. For we will have thought of a number of objec-
tions that badly need to be posed and answered—the commentaries pro-
vide a fine selection of these—objections ignored by both Socrates and
Glaucôn. So that the unconditional assent that Glaucôn gives to Socrates’
conclusions in the closing passages of Book IV could not possibly be justi-
fied. Plato had already warned both Glaucôn and us: “we will never
achieve an accurate grasp from the methods of argument that we are now
using, but for that there is a longer and greater road…” (435c-d).
We should in any case by now have been raising our own questions
about the ambiguous roles in which Glaucôn and Adeimantus have been
cast. At 382b Socrates had defined “what is truly called falsehood” as
“the ignorance in the soul of someone who has been lied to.” Real false-
hood, he goes on, is hated by both gods and humans. And he then asks
whether none the less falsehoods in words may not be useful against ene-
mies and to prevent harm and wrongdoing, answering that they are indeed
thus useful. But who may legitimately use them? Only the rulers of a just
city and no others. Rulers may lie to enemies or citizens for the sake of
the good of that city, but citizens may not lie to rulers and should be pun-
ished, if they do (389b-d).
Yet not too long afterwards Socrates proposes to tell his “noble lie”
(414b) to the rulers, as well as to the other citizens, according to which the
differences in their souls are due to the different metals in those souls
(415a-c). There are at least two problems here. The first concerns the
rulers who, as we learn later, are to be educated precisely in those disci-
plines that “lead us towards truth” (525b), so that their souls may be in-
formed by truth. Yet here Socrates is proposing the inculcation of an illu-
sion in just those same individuals. A second problem concerns the con-
trast between such rulers and Glaucôn, Adeimantus, and the others. The
latter are allowed to know about the rulers what the rulers are not to be
allowed to know about themselves. Their standpoint is in this respect su-
perior with regard to truth to the standpoint of the rulers. But the rulers
are philosophers and even if we ignore the fact the Glaucôn, Adeimantus
and the others are not or not yet philosophers, the question arises: how can
there be a standpoint higher than that of the philosopher?
 YET ANOTHER WAY TO READ THE REPUBLIC? 215

Commentators have devoted great energy and ingenuity in trying to res-

cue Plato’s Socrates from the charge of incoherence on these matters.
What I want to suggest is that Socrates is indeed prima facie incoherent,
but that Plato is not. What Plato is doing is intentionally portraying a par-
ticular kind of incoherence. What kind? To answer this question we need
first to consider the account of the Line (509d-511e).

IV. The Incoherence of the Line

There are of course a range of interpretative disagreements about the Line.

Consider the incompatible treatments of it by Reeve and Annas. Reeve
argues that Plato presents a complex ontology of forms, modes, qualities
and figures that “holds the key to a consistent interpretation of the Sun,
Line and Cave, and indeed of the Republic as a whole.” 6 Annas, by con-
trast contends that Plato uses the Line to provide both an analogy between
the visible and the intelligible, and “an ambitious classification of the dif-
ferent states a person may be in with regard to knowledge” and that “as
often happens with Plato, his eagerness to use analogy and images to illus-
trate a point leads him into intellectual unclarity.” 7 For “the line, uncom-
fortably, both distinguishes the visible and intelligible realms to compare
them, and also puts them on a continuous scale of epistemological
achievement.” 8 And Annas goes on to argue that in interpreting the Line
we are confronted with a choice between “two unsatisfactory alternatives,”
an interpretation that distinguishes the objects studied by mathematics
from Forms and a rival interpretation that identifies mathematical objects
with a subset of Forms. Annas identifies crucial difficulties in each view
and goes on to speak of “the insolubility of this problem.” 9
We are offered here not only a choice of interpretations, but also a
choice of Platos: either Reeve’s complex metaphysician, whose mode of
exposition had concealed his ontology, or Annas’s overambitious enquirer,
led astray by his too great love for the striking image. Which should we
choose? The answer is: if forced to make such a choice, then choose
Reeve, but, if possible, choose neither. For Plato himself invited us to call

_________
6 Reeve 1988, 53.
7 Julia Annas, An Introduction to Plato’s Republic, Oxford: Clarendon Press, 1981, p.
249.
8 Annas 1981, 249-250.
9 Annas 1981, 251-252.
216 ALASDAIR MACINTYRE

into question the surface meaning of the Line long before any commenta-
tor did so. Consider two of its important features.
The first is that in order to characterize adequately the objects that we
encounter at each stage in the ascent mapped by the Line, we must already
have ascended to the next higher stage. The shadows and reflections that
are the objects of our imaging are not adequately characterized until we
understand them as shadows of and reflections of, and what they are shad-
ows of and reflections of are physical objects, objects of perception. This
relationship of epistemological and ontological dependence also holds
between not only the mathematical objects that the mind encounters at the
next higher stage and the physical objects of perception, and again be-
tween the intelligible as such and the visible as such. The physical objects
of perception are not yet adequately characterized until we understand
them both as exemplifications of arithmetical and geometrical properties
and yet also as objects that, unlike those properties, present themselves as,
in Reeve’s phrase, cognitively unreliable, as having incompatible proper-
ties. And the visible as such is not yet adequately characterized until its
ontological and epistemological dependence on the intelligible is recog-
nized. That dependence is what makes it possible for us to judge truly, as
we often do, concerning images, physical objects and the mathematical
properties of the physical world. And it has the character that it has only
because mathematical objects are themselves adequately characterizable
only by reference to what lies one stage yet higher, the Forms. It follows
that I cannot understand anything whatsoever until I have apprehended the
Forms. But what does the Line tell me about the Forms? It is here that a
second feature of the Line is important.
Mathematical reasoning is limited in two ways. It cannot of itself as-
cend beyond hypotheses. None of its proofs are free of assumptions. And
it cannot dispense with diagrams. It cannot, that is, rid itself of depend-
ence upon physical representations of mathematical realities, which, just
because they are physical representations, are inadequate representations.
Mathematical reasoning cannot therefore yield a knowledge of Forms.
But, if no diagram-informed reasoning can yield a knowledge of Forms,
then the Line cannot yield such knowledge. For the Line is a diagram and
moreover a diagram that, on Socrates’ account of it, as we have just seen,
makes our understanding of each of the lower levels of apprehension de-
pendent on our apprehension of Forms. What the Line seems to demon-
strate then is that things cannot be as the Line says they are. The Line de-
constructs itself to the attentive reader.
The simile of the Sun and the Cave poses related problems of which
Glaucôn and Adeimantus also seem unaware, but which are inescapable
 YET ANOTHER WAY TO READ THE REPUBLIC? 217

for us. Those who have never left the Cave are unable to see things within
the Cave as they are, while those who have left the Cave for the light of
the Sun and then returned to it, allowing their eyes gradually to become
reaccustomed to the gloom are able to see how things are within the Cave.
But Glaucôn, Adeimantus and the others have never left the Cave. So the
most that they can learn from Socrates’ narrative is that their own judg-
ments as to the real and the unreal are not to be trusted and they do not
know as yet how to correct those judgments. So the comparison of the
Form of the Good to the Sun, both as a source of the intellectual light by
which what is is to be grasped and as itself the culminating object of intel-
lectual vision, once again leaves Glaucôn, Adeimantus and the others
without a standard by which to measure their own lack of intellectual ca-
pacity.

V. What Plato is Saying to Us

By this treatment of the Line, the Sun, and the Cave Plato puts in question
our ability to understand what he is saying. He may for a moment have
tempted us to regard ourselves as superior to Glaucôn and Adeimantus,
who not only do not know, but do not know that they do not know. For,
while we may not know, at least we now seem to know that we do not yet
know. After all, we have been able to identify flaws in the argumentative
method of Books II-IV, while Glaucôn, Adeimantus, and the others were
not. But this self-congratulatory mood is unlikely to persist, when we re-
alize that, if Plato’s account of Forms as standards is true, then we are in
the same predicament as Glaucôn and Adeimantus.
It is a twofold predicament. More generally, absent knowledge of the
Forms and of the Form of Good, not only do we not know what we hith-
erto took ourselves to know, but we do not know how to distinguish know-
ing from not knowing. And therefore, more particularly, we do not know
how to evaluate justifications of justice any more than do Glaucôn and
Adeimantus. And here we stumble upon the wry joke at the heart of the
Republic. In Book II Glaucôn and Adeimantus had challenged Socrates to
show them and us that justice is the kind of good to be valued for its own
sake and not only for any concomitant advantages that it may bring. Soc-
rates to meet this challenge had provided in Book IX not one, but three
justifications. Yet each of these justifications holds, if and only if the
Forms are as Socrates says they are.
218 ALASDAIR MACINTYRE

The first justification contrasts the happiness of the philosopher king de-
scribed in Book VII with the unhappiness of the tyrant first described in
Book VIII and then more fully in Book IX. The crucial difference be-
tween these is both in the objects of their desires and in their relationship
to their desires. The philosopher king is not a lover of ruling (521b), while
the tyrant desires passionately to have his own way with others, with gods
as well as with human beings (573c). The philosopher king desires noth-
ing that is not just and when he realizes that in the just polis, even if not in
his own (592a), justice requires that he rule, he will find his own happiness
in the happiness of the city (519e and 520d-e). So that this justification of
justice by reference to the happiness of the just can be understood as a jus-
tification only by those who have reconceived their happiness as only phi-
losophers can. But Glaucôn, Adeimantus and the others are not able to do
this. Socrates’ answer to their question ‘Can a justification be provided
for treating justice as a good to be valued for its own sake?’ is in effect
‘Yes, but not to you.’ And this is what Plato is also saying to us his read-
ers.
The same holds of the other two justifications presented in Book IX.
The philosopher alone can judge whether or not the life of the philosopher
with its love of wisdom and argument is superior to the lives of the honor-
lover and the profit-lover (582a-d). And, in order to make the distinction
between true pleasures and “pleasures that are mixed with pains, mere im-
ages and shadow-paintings of the true pleasures” (586b) on which the third
justification depends, it is necessary to have the kind of experience of the
life of reason and virtue (585e) that is restricted to philosophers. All that
Socrates has been able to provide for Glaucôn Adeimantus, and the others
is an account of the stages through which each of the three justificatory
arguments would have to proceed. But the arguments themselves cannot
have been made available to them, for that would require the kind of in-
sight that derives from an intellectual grasp of the Forms. So that those
who complain of flaws and defects in Plato’s statement of Socrates’ argu-
ments are missing the point. Annas complains that “Plato’s arguments
about the consequences of justice are uneven in quality; they are bedev-
illed rather than helped by the analogy with the city, and they are unclear
about pleasure.” 10 Yet such apparent defects are ineliminable from any
account that could be given either to Socrates’ interlocutors or to us.
Plato in the Republic is committed to a conception of Forms as objects
of knowledge and as constitutive of what is. Every particular is, to use
Aristotle’s terminology, a this-such and to understand what this or that
_________
10 Annas 1981, 314.
 YET ANOTHER WAY TO READ THE REPUBLIC? 219

particular this is is to understand it as informed by this particular such. It

is to grasp the unities that underlie the multiplicities of perceptual experi-
ence, to understand Many in terms of One. But the achievement of this
understanding is obstructed by two things, the nature of perceptual experi-
ence and the distortions of attention that derive from our disordered de-
sires. And so the theory of Forms—the word ‘theory’ carries no particular
weight—generates its own epistemological problem, a problem of circu-
larity. If I am to attain to a knowledge of Forms, then I must transform my
desires and transcend the limitations and incoherences of my perceptual
experience. But I can become someone who desires as one should desire
and who is no longer trapped by the distortions of perceptual experience,
only if I have already directed myself in the light afforded by the Form of
Good, supreme object of both knowledge and desire. So how am I to
break through to knowledge of Forms?
In the Republic, as we have already noticed, several types of answer to
this question are shown to be unsatisfactory, shown to us, Plato’s readers,
that is, but not to Glaucôn and Adeimantus. Consider the proposal that
philosopher kings should be produced by a childhood introduction to cal-
culation, geometry and the like, two or three years of physical training, an
education in the unified sciences from twenty to thirty, five years of dia-
lectic, and fifteen years of practical training, so that at the age of fifty they
may be “led to the goal and compelled to lift up the radiant light of their
souls to what itself provides light for everything” (540a).
Two aspects of this proposal need to be emphasized. The first is that it
throws no further light on the difficult question of the relationship of
mathematical understanding to the knowledge of the Forms. And this
question was never resolved in such a way as to produce general agree-
ment within the Academy. Speusippus abandoned belief in Forms, think-
ing that he had found the key to the structure of knowledge in mathemat-
ics, while Aristotle retained Plato’s distinction between Forms and mathe-
matical objects, but developed different accounts of both, and Plato con-
tinued to grapple with the problem. So that the educational proposal of
Book VII is an ‘as if’ proposal: suppose that we had already solved this set
of problems, then might the relevant education for rulers be something
such as this? But even as an ‘as if’ proposal, it is also a tongue in cheek
proposal, an elaborate joke.
The jokes in the Republic have received insufficient attention. They are
very good jokes, worthy of an author whose admiration for Aristophanes
is evident, so that the first example of someone with a truly philosophical
nature is a dog (376a-b), while those who enquire into harmonies are de-
scribed as tormenting and torturing their strings, so that the strings in turn
220 ALASDAIR MACINTYRE

engage in accusations and denials and boastings (531b), and Pythagorean

mathematics is parodied in the argument (587c-e) whose conclusion is that
the life of the philosopher king is 729 times more pleasant than that of the
tyrant. The same spirit that informs these jokes informs the conception of
education that Socrates articulates in Book VII. Socrates himself had
never had anything remotely like the education that he describes. And of
the several portrayals of Socrates that we have, that, for example, in the
Apology or that in other Platonic dialogues or that authored by Xenophon,
which Socrates is it who can be imagined as seriously intending such a
proposal? The answer is: none of them. It is only as an ironic proposal
that that conception can be credibly ascribed to any Socrates recognizable
as Socrates.
What I am suggesting is that Plato’s dramatic intention in this part of
Book VII, as earlier in the similes of the Line, the Sun, and the Cave, is to
bring out the apparent impossibility, if the theory of Forms is true, of com-
ing to learn that it is true through a series of arguments. Myles Burnyeat
has made the point succinctly: “What premises could the arguments start
from? If they are drawn from within our current stock of beliefs and as-
sumptions, they will not lead to the realm beyond. If they start from out-
side our present outlook, they will not persuade.” 11 But perhaps before
this was Burnyeat’s thought it was Plato’s.

VI. Towards the Resolution of the aporia

Plato had at one stage believed himself able to solve this problem by resort
to the notion of ". We become aware of the Forms through rec-
ollection. What difficulties had led him to reject this thesis we can only
guess. But in the Republic he is no longer prepared to appeal to it and so
presents us and himself with this massive aporia. Yet the Republic also
suggests that perhaps this aporia can be resolved.
Consider how when Plato cannot provide an argument, he often pro-
vides an image, characteristically an aesthetically striking image, and how
such images are always more than and sometimes other than surrogates for
arguments. Consider, for example, the image in Book IX of the being who
looks like a human being, but has inside him as well a multiform, many-
headed beast and a lion (558b-589a) or the image in Book X of the sea
god, Glaucus (611c-d), or the portrait of the man who engages unworthily
_________
11 Myles Burnyeat, British Academy Master-Mind Lecture, “Plato,” Proceedings of the
British Academy: 2000 Lectures and Memoirs, London, p. 13.
 YET ANOTHER WAY TO READ THE REPUBLIC? 221

in philosophy in Book VI (495d-e). Such images do at least part of the

work that argument might have been expected to do. But this, far from
rescuing Plato from his aporia, makes his predicament all the more severe.
For those images are poetic images and the question is at once raised: are
they the kind of poetry that has been or should be proscribed?
Plato’s answer is given in Book X. The censorship of dramatic poetry
in Book III had been justified without reference to the Forms. But it has
now become clear that it is only by the standards afforded by the Forms
that poetic images can be judged. And so Plato proceeds to an apparently
paradoxical judgment, one in which what is said in the earlier part of Book
X seems to be at odds with what is said in the later part. For in the earlier
part Socrates concludes that “all those engaged in poetic imitation, begin-
ning with Homer, imitate images of virtue and all the other things that they
write about and have no grasp of the truth” (600e), while in the latter part
Socrates recounts the myth of Er, a recounting which is a poetic and mi-
metic performance, and expresses the hope that his hearers will be per-
suaded by it.
It is of course possible to interpret the Republic in ways that obscure the
paradoxical character of Book X. Reeve says that “The truth about this
world is a matter of evidence and argument. The truth about the next
world is a matter of myth,” so exempting poetic representations of the
realm beyond death from the general strictures on poetic representation.12
But this seems a quite arbitrary attempt to rescue Plato. Annas by contrast
is not at all concerned to rescue Plato. She says of the first part of Book X
that Plato reduces the poet to “a mindless copier of appearances” and
speaks of Plato’s “passionate, hopelessly bad arguments.” 13 And with the
myth of Er she begins by commenting on its “vulgarity” and calls it “a
lame and messy ending.” 14
These responses to Book X should not surprise us. If Books II-IX were
what Reeve and Annas take them to be, a straightforward exposition of a
problem and of its proposed solution, in which the voice of the character,
Socrates, is the voice of the author, Plato, then Book X would indeed ap-
pear to present a problem, to be no more than a couple of extended after-
thoughts that Plato might have done better to omit. But on the reading that
I am suggesting Book X is integral to Plato’s enterprise. For it tells us that
we need urgently to appropriate the truth in the myth, while it has also
informed us that what is appropriated from myth is not appropriated as
_________
12 Reeve 1988, 263.
13 Annas 1981, 344.
14 Annas 1981, 353.
222 ALASDAIR MACINTYRE

truth. This is of course in one way no more than a restatement of the basic
aporia of the Republic. But at this late stage of the conversation Plato
envisages a possibility that is new to the Republic, the possibility of a phi-
losophical defence of poetry, of the philosopher as 

, someone
whose arguments show that poetry is “beneficial for regimes and for hu-
man life” (607d). It is mentioned only as a very bare possibility, but what
is striking is that it is mentioned at all. For consider the task of the phi-
losopher as 

.
It would be to show that poetic representations, including myths, can
lead us towards and not away from the truth, that what we can learn from
such representations would be vindicated and not discredited, if evaluated
from the standpoint of those with knowledge of the Form of the Good.
The message of 607a-e is that this has not yet been achieved, but the narra-
tion of the myth of Er, beginning at 614b, presupposes that it can be
achieved. For the recounting of the myth of Er makes no sense, unless it
presents truths to Socrates’ conversational partners which they, not being
philosophers, could not appropriate in any other way. But this is not the
only expression of Plato’s belief that perhaps, just perhaps the aporia of
the Republic can be resolved.
What is the mark of the philosopher who has attained knowledge of the
Form of the Good? It is that he is able to advance an account that distin-
guishes the Good from everything else, an account that he is able to de-
fend against all refutations (535b-d) from whatever point of view. Those
attempted refutations would be designed to identify the Good with some
pretender to its position, pleasure or knowledge or whatever. The philoso-
pher’s refutation of those attempted refutations would, like all examples of
elenchus, be an argument from his opponent’s premises, not from his own.
But he would be able to rely on some shared attitudes and presuppositions:
“Nobody is satisfied with obtaining what seems to be good rather than
what really is good . . . . Every soul pursues the Good and does everything
for its sake, divining it to be something, but being at a loss and not having
an adequate grasp of what it is…” (505d-e).
What this suggests is that the refutations of those with mistaken concep-
tions of the Good will convict them not just of inconsistency but of failing
to do justice to an incoherent, fragmentary, and inadequate grasp of the
Good that they already possess, yet a grasp exhibited in a directedness
towards the Good. Furthermore those refutations, by providing an increas-
ingly sophisticated account of what the Good is not, will impose con-
straints on any account of what the good is. So we are given an outline
sketch of how by dialectic someone might begin to move from a deeply
inadequate grasp of the Good towards a somewhat more adequate grasp.
 YET ANOTHER WAY TO READ THE REPUBLIC? 223

It is not unimportant, as the discussion of $ in Book VIII makes

clear, that this can only be achieved by those who have, by the cultivation
of the appropriate habits, turned themselves into souls who are not too
distracted from directedness towards the good by disorderly desires (518c-
e). But this transformation of habits can only have been effective because
there is already in each soul a striving towards the Good. And it is the
acknowledgment of this striving that suggests the possibility of a resolu-
tion of the central aporia of the Republic.
For those who are lovers of sights and sounds and crafts, of honor and of
victory, are none of them, so it turns out, lovers merely of these things.
They are also already, albeit implicitly and confusedly, lovers of the Good.
And so the gap between what Burnyeat called “our current stock of beliefs
and assumptions” and “the realm beyond” may not be quite as unbridge-
able as it had appeared to be. We must be careful not to exaggerate. How
that gap is to be bridged, if at all, remains in the Republic an unanswered
question. We, like Socrates’ interlocutors, remain unable so far to pro-
gress beyond images. We cannot as yet dispense with the Line, the Cave,
and the Sun, or the myth of Er. But, unlike Socrates’ interlocutors, we
have been made aware that the Republic provides not an answer, but a
question.
UNIVERSITY OF NOTRE DAME
 MACINTYRE BIBLIOGRAPHY

Annas, J. 1981. An Introduction to Plato’s Republic. Oxford.

Burnyeat, M. 2000. “Plato.” Proceedings of the British Academy: 2000 Lectures and
Memoirs, London.
MacIntyre, A. 1988. Whose Justice? Which Rationality? London.
_______. 1990. “The Form of the Good, Tradition, and Enquiry.” Value and
Understanding, ed. R. Gaita. London.
_______. 1998. A Short History of Ethics, London.
Reeve, C. D. C. 1988. Philosopher Kings: The Argument of Plato’s Republic.
Princeton.
Robinson, R. 1953. Plato’s Earlier Dialectic. London.
Sayre, K. 1995. Plato’s Literary Garden. Notre Dame.
 ABOUT OUR CONTRIBUTORS

John J. Cleary is Professor of Philosophy at Boston College and

Associate Professor of Philosophy at NUI Maynooth (Ireland). He
received his B.A. and M.A. from University College Dublin, and his Ph.D.
from Boston University. He was director of the Boston Area Colloquium
in Ancient Philosophy from 1984 to 1988, and is the founding general
editor of this series of Proceedings. He has published extensively on
ancient philosophy, including a monograph on Aristotle and Mathematics
(Leiden, 1995). Currently, he is writing a book on the role of paideia in
ancient political thought.

Matthew Evans is Assistant Professor of Philosophy at New York

University, where he has been teaching since 2004. He is a graduate of
the Joint Program in Ancient Philosophy at the University of Texas at
Austin. Since graduating from UT he has published a paper on Epicurean
friendship, as well as several papers on Plato’s moral psychology in the
Philebus. Currently he is writing a book on Plato’s metaphysics of mind
and value.

Günter Figal is professor of philosophy (chair) at the University of

Freiburg (Germany). He was educated at Heidelberg. His published work
is mainly on ancient Greek philosophy, Phenomenology, and Her-
meneutics. Among his books are: Martin Heidegger. Phänomenologie der
Freiheit (1988), Für eine Philosophie von Freiheit und Streit (1994;
English: For a Philosophy of Freedom and Strife, 1998), Sokrates (1995),
and most recently Gegenständlichkeit. Das Hermeneutische und die
Philosophie (2006). He is the editor of Internationales Jahrburch für
Hermeneutik.

Francisco J. Gonzalez is Associate Professor of Philosophy at Skidmore

College, where he has taught since 1991. He received his B.A. from
Northern Illinois University and both his M.A. and Ph.D. from the
University of Toronto. He is the author of Dialectic and Dialogue: Plato’s
Practice of Philosophical Inquiry (Northwesetern, 1998) and the editor of
The Third Way: New Directions in Platonic Studies (Rowman &
Littlefield, 1995). He has written articles on a wide range of topics within
226 ABOUT OUR CONTRIBUTORS

Platonic and Aristotelian Studies as well as on some of the major figures

of contemporary Continental Philosophy such as Heidegger, Gadamer,
Ricoeur, and Levinas. He has recently completed a book entitled Plato and
Heidegger: A Question of Dialogue.

Gary M. Gurtler, S.J., is Associate Professor of Philosophy at Boston

College. He was educated at St. John Fisher College (B.A.), at Fordham
University (M.A. and Ph.D.), and at the Weston Jesuit School of Theology
(M.Div.). He has published on ancient philosophy, with special attention
to Neoplatonism, including a book on Plotinus: The Experience of Unity
(1988). His most recent articles include “The Activity of Happiness in
Aristotle’s Ethics,” The Review of Metaphysics 56 (2003), 803-834,
“Plotinus: Matter and Otherness, On Matter (II 4[12]),” Epoché, 9 (2005),
197-214, and “Plotinus: Self and Consciousness,” History of Platonism:
Plato Redivivus (New Orleans, University Press of the South, 2005), pp.
113-29. Currently, he is continuing research on Plotinus’ reworking of
Platonic otherness as one of the key features of his retrieval of Plato’s
thought.

David Konstan is the John Rowe Workman Distinguished Professor of

Classics and the Humanistic Tradition and Professor of Comparative
Literature at Brown University. Among his recent books are Friendship in
the Classical World (Cambridge, 1997); Pity Transformed (Duckworth,
2001); a tranlsation of Aspasius’ Commentary on Aristotle’s Nicomachean
Ethics (Duckworth, 2006); and The Emotions of the Ancient Greeks
(Toronto, 2006).

Verity Harte is Professor of Philosophy and Classics at Yale University

and Honorary Research Professor in the Department of Philosophy at
King’s College London. She received her B.A., M.Phil., and Ph.D from
Cambridge University. She is the author of Plato on Parts and Wholes
(Oxford 2002) and of various articles in ancient philosophy.

Alasdair MacIntyre teaches philosophy at the University of Notre Dame.

He was previously Arts and Sciences Professor of Philosophy at Duke
University. Among his books are A Short History of Ethics (1966), After
Virtue (1981), Whose Justice? Which Rationality? (1988), and two
volumes of essays, The Tasks of Philosophy and Ethics and Politics
(2006). He is an Honorary Member of the Royal Irish Academy, a
Corresponding Fellow of the British Academy, a member of the American
 ABOUT OUR CONTRIBUTORS 227

Academy of Arts and Sciences, and a member of the American

Philosophical Society.

Stephen Menn is Associate Professor of Philosophy at McGill

University, and also teaches in the Classics Program there. He was
educated at Johns Hopkins University and the University of Chicago. He
is the author of Plato on God as Nous (Carbondale, 1995, now available
from St. Augustine’s Press), Descartes and Augustine (Cambridge, 1998,
revised paperback edition 2002), and The Aim and the Argument of
Aristotle’s Metaphysics (forthcoming from Oxford), as well as articles on
ancient and medieval philosophy and science. He is working on a
translation of Simplicius’ commentary on Aristotle’s Physics I,1-2 (with
Rachel Barney), an introduction to Aristotle’s Metaphysics, and a book on
al-Fârâbî and the history of the many senses of being.

Pierre-Marie Morel is Maître de conférences – Docteur habilité at

University Paris I – Panthéon-Sorbonne, where he teaches Ancient
Philosophy. He received his Ph.D. in 1994 from University Paris I. He has
written essays on Atomism, Aristotle and Plotinus. He has published
Démocrite et la Recherche des causes (1996), Atome et nécessité.
Démocrite, Epicure, Lucrèce (2000), Aristote. Une philosophie de
l’activité (2003), a French translation with notes of Aristotle’s Parva
naturalia (2000), and has edited, with Aldo Brancacci, Democritus:
Science, the Arts and the Care of the Soul (2007).

David Roochnik is a Professor of Philosophy at Boston University. His

Ph.D. is from the Pennsylvania State University. His most recent
publications include Beautiful City: The Dialectical Character of Plato’s
Republic (Cornell University Press), and Retrieving the Ancients: An
Introduction to Greek Philosophy (Blackwell).

Dennis J. Schmidt is Professor of Philosophy, Comparative Literature,

and German; he is also the editor of the SUNY Press “Series in
Continental Philosophy.” His book publications include: The Ubiquity of
the Finite: Hegel, Heidegger, and the Entitlements of Philosophy (MIT
Press, 1988; On Germans and Other Greeks: Tragedy and Ethical Life
(Indiana University Press, 2001); Lyrical and Ethical Subjects: Essays on
the Periphery of the Word, Freedom, and History (SUNY Press, 2005).
He is the translator of Ernst Bloch’s Natural Law and Human Dignity
(MIT Press, 1986), and the co-editor of Hermeneutische Wege: Hans-
Georg Gadamer zum Hundertsten (Mohr Verlag, 2000) and of The
228 ABOUT OUR CONTRIBUTORS

Difficulties of Ethical Life (Fordham University Press, 2007). He has

written extensively on a variety of topics in philosophy, literature, and art
with special interest in Contemporary European thought and Ancient
Greek thought.

Henry Teloh received his Ph.D. in philosophy from the University of

Wisconsin-Madison in 1972. Since that time he has taught Ancient
philosophy, social and political philosophy, philosophy of education, and
philosophy in literature at Vanderbilt University. He is as of 2007
Professor Emeritus of philosophy at Vanderbilt. Teloh has written two
books in Ancient philosophy: The Development of Plato’s Metaphysics
and Socratic Education in Plato’s Early Dialogues. He has recently
completed a book manuscript on rhetoric in the Gorgias and Phaedrus.

Katja Maria Vogt is Associate Professor of Philosophy at Columbia

University. She was educated in Munich, Germany, and has published
articles on ancient philosophy, Kantian ethics, and friendship. Most of her
work in ancient philosophy is on the Sceptics, the early Stoics, and
Seneca. She is the author of a book on Pyrrhonian Scepticism, Skepsis und
Lebenspraxis (Alber 1998), and a forthcoming book on early Stoic
political philosophy, Law, Reason, and the Cosmic City: Political
Philosophy in the Early Stoa (OUP 2007).

James L. Wood received his PhD in philosophy from Boston University

in 2006 and is currently Lecturer in the Core Curriculum program at BU.
He has written and presented numerous papers on Platonic and
Aristotelian philosophy, including publications in Ancient Philosophy and
Interpretation. His work focuses in particular on Plato’s Philebus and the
intersections of metaphysics and ethics in Platonic thought. He is
currently writing a paper on Heidegger’s interpretation of Platonic
metaphysics.

Raphael Woolf is Reader in Philosophy at King’s College London. He

was educated at the University of Cambridge and King’s College London.
He has translated Cicero’s De Finibus (‘On Moral Ends’) for the series
Cambridge Texts in the History of Philosophy, and published articles on
Plato, Aristotle and Hellenistic philosophy. Current interests include the
nature of epistemic authority in Plato, and pleasure and desire in Epicurus,
on which he is writing a chapter for the forthcoming Cambridge
Companion to Epicureanism.
 INDEX OF NAMES

Adeimantus, 205, 209, 211-214, 216- Carone, G., 126, 133, 137
219 Cassirer, E., 188
Adorno, T. W., 187 Caston, V., 131-132, 161
Agathon, 190-191, 194, 201 Cebes, 2, 6-7, 12, 15, 20
Alesse, F., 169 Cephalus, 212
Annas, J., 34. 206, 215, 218, 221 Chrysippus, 155-156, 165-166,
Anscombe, G. E. M., 121-122, 143, 175
149 Cicero, 27, 31-32, 36, 39, 41, 42-
Apollo, 20 43, 156-157, 160, 167, 170, 172,
Aristo, 183 175-180, 183
Aristophanes, 190-191, 219 Cleinias, 73
Aristotle, 10, 37-38, 85, 95-99, 117, Clement of Alexandria, 39, 42, 45
121-122, 133, 148-149, 160, 195, 200, Cooper, J., 133, 157
218-219 Courtine, J. -F., 108
Asmis, E., 26, 29-30, 45-47 Critias, 209
Ast, F., 96 Crito, 21
Cross, M. E., 206
Balashov, Y., 112-116 Cross, R. C., 206, 212
Barnes, J., 84-85, 108 Cushman, R., 57
Beaufret, J., 107
Benitez, E., 128 Davies, J. L., 206
Berti, E., 84, 107 De Lacy, P., 37, 51
Bett, F., 169 Diogenes of Babylon, 175
Bett, R., 159, 169 Diogenes Laertius, 26-28, 31-33,
Black, E., 58 39, 40-42, 46, 49-51, 54, 157,
Bloom, A., 206 159, 175, 180-181
Blumenberg, H., 187-188 Diogenes of Oinoanda, 32
Bonhöffer, A., 162 Dionysius Thrax, 175
Bonitz, H., 96 Diotima, 190, 194-195
Bosanquet, B., 206 Dorter, K., 6
Boutot, A., 107, 111 Dostal, R., 83, 106
Brandt, R., 123 Doty, R., 161
Brickhouse, T., 57, 60, 71-72 Dreher, J. P., 113
Brittain, Ch., 156, 159, 161, 166
Broad, C. D., 123 Echecrates, 2, 6-7
Broadie, S., 121 Empedocles, 189
Burnyeat, M., 220, 223 Epictetus, 157, 181
Bury, R., 133, 159 Epicurus, 25-32, 34-36, 38-42,
44-48, 49-54, 158, 161-162, 168
Callicles, 57-59, 63, 65, 67-69, 73-77, Eryximachos, 190
78-79, 81 Eudoxus, 121-123, 124, 143-144,
146, 148-149
230 INDEX OF NAMES

Evans, M., 129, 144 Jaeger, W., 57

Eyres, H., 206 Jowett, B., 206

Feldman, F., 123, 144 Kagan, S., 122, 124

Frede, D., 127, 129, 137-138 Kant, I., 44, 202-203
Frede, M., 155, 157, 160, 161, 163, Kaufman, C., 57-58
171 Keynes, J. M., 38
Friedländer, P., 108-109 Kobusch, T., 189
Konstan, D., 50
Gallop, D., 3 Korsgaard, C., 122, 124, 150
Gassendi, P., 31 Kraut, R., 125
Glaucôn, 205, 209, 211-214, 216-219
Glidden, D. K., 34, 39-40 Laches, 60, 76
Goldschmidt, V., 32, 36, 43-44 Langton, R., 122
Goldstein, I., 124 Lewis, T. J., 58
Gonzalez, F., 88, 90 Long, A., 26-27, 30-31, 34-35,
Gorgias, 57-58, 60-64, 66, 69, 73, 77, 38, 40, 46-47, 157, 169-171
78-81 Lucas, J. R., 206
Gosling, J. C. B., 123, 125-126, 129 Lynch, M., 10
Grube, G. M. A., 18, 210 Lysias, 209

Hackforth, R., 125-126, 133 MacIntyre, A., 207

Hall, R., 71 Manuwald, A., 33-34, 36-37, 41
Hammerstaedt, J., 34, 40-41 Margolis, J., 111
Harte, V., 124, 151-152 Meno, 21
Heal, J., 10 Millgram, E., 124, 142
Hegel, G. W. F., 198, 199, 203 Mitchell, B., 206
Heidegger, M., 83-111, 112-113, 115- Modrak, D., 32
117 Moline, J., 70
Heraclitus, 189 Morel, P. -M., 34, 36
Herillus, 183 Murphy, N. R., 206
Hesiod, 189, 194
Hestir, B., 108, 110 Nagel, T., 123-124
Homer, 189, 194 Nestle, W., 187
Horkheimer, M., 187 Nettleship, R. L., 206
Hyland, D., 94, 110-111 Nicias, 76
Nietzsche, F., 17, 22, 188
Inwood, B., 155, 157, 162, 172 Nussbaum, M., 57-58

Jackson-McCabe, M., 159, 160, 163, Obbink, D., 159, 162

165
 INDEX OF NAMES 231

Pappas, N., 206 Sextus Empiricus, 45-46, 158-

Parmenides, 14-15, 22, 108, 189 160, 170, 175-176, 181
Pater, W., 57 Shiner, R., 128
Pausanias, 190 Sidgwick, H., 124
Pepperzak, A., 108 Simmias, 1-7, 9, 12, 15, 17-19
Phaedo, 2, 6-7, 12 Socrates, 1-10, 12-15, 17-23, 38,
Phaidros, 190 57-77, 78-81, 85-86, 88-89, 91-
Philebus, 147 94, 103-104, 116, 124-142, 144,
Philodemus, 35, 37, 43 147, 150-152, 168-170, 183, 187-
Plato, 1, 11, 14-15, 17-18, 21-23, 25, 188, 190-194, 196-197, 201, 205,
38, 42, 45-46, 57, 59-60, 74, 78, 80- 207-218, 220-223
81, 83-84, 88-90, 92-97, 99-100, 102- Speusippus, 219
111, 112-117, 121, 124-127, 132-133, Spitzer, A., 58
139-140, 142-143, 145, 146, 150, 168- Stauffer, D., 58
170, 172, 180, 182, 187-189, 193, Stobaeus, 157-159
199-200, 202-203, 205-215, 217-222 Striker, G., 37-38, 170
Plutarch, 156, 165, 170, 172, 180-181 Szaif, J., 84-85, 104, 109-110
Polemarchus, 209, 212
Polus, 57-59, 64-67, 69, 73, 78-81 Talisse, R., 60, 77
Protarchus, 124-128, 130-131, 136, Tarrant, H., 58
147, 153 Teloh, H., 60, 76
Theuth, 194
Rauch, L., 206 Thompson, W. H., 57
Reesor, M. E., 159 Thornburg, T., 206
Reeve, C. D. C., 206, 210-213, 215- Thrasymachus, 209, 211-213
216, 221 Tuozzo, T., 128-129
Rendall, S., 57
Rice, D., 206 Usener, H., 191
Richardson Lear, G., 133, 148
Robinson, R., 207 Vaughan, D. J., 206
Roochnik, D., 57-58, 60-61, 71-76, 80 Velleius, 35
Rosen, S., 206
Ross, W. D., 124 Weiss, R., 60
Rowe, C., 121 White, N., 206
Russell, D., 137 Williams, B., 10, 150
Woolf, R., 7
Sallis, J., 108-109 Woozley, A. D., 206, 212
Sandbach, F. H., 161-162, 165
Santoro, M., 52 Zeller, E., 57
Sayers, S., 206 Zeno of Cytium, 156
Sayre, K., 207 Zeus, 178, 184
Scott, G. A., 60
Sedley, D., 27, 31, 35, 157-158, 169
Seneca, 157, 160, 162-163, 172, 180
Sesonske, A., 206