INQUIRY, FORMS, AND SUBSTANCES

PHILOSOPHICAL STUDIES SERIES

Founded by Wilfrid S. Sellars and Keith Lehrer

Editor
KEITH LEHRER, University of Arizona, Tucson

Associate Editor
STEWART COHEN, Arizona State University, Tempe

Board of Consulting Editors

LYNNE RUDDER BAKER, University of Massachusetts at Amherst
ALLAN GIBBARD, University of Michigan
DENISE MEYERSON, University of Cape Town
RONALD D. MILO, University of Arizona, Tucson
FRAN<;OIS RECANATI, Ecole Poly technique, Paris
STUART SILVERS, Clemson University
NICHOLAS D. SMITH, Michigan State University

VOLUME 62
 THOMAS A. BLACKSON
Department 0/ Philosophy, Temple University, Philadelphia

INQUIRY, FORMS,
AND SUBSTANCES
A STUDY IN PLATO'S
METAPHYSICS AND EPISTEMOLOGY

SPRINGER -SCIENCE+ BUSINESS MEDIA, B.V.

A C.I.P. Catalogue record for this book is available from the Library of Congress.

ISBN 978-94-010-4124-9 ISBN 978-94-011-0281-0 (eBook)

DOI 10.1007/978-94-011-0281-0

Printed on acid-free paper

All Rights Reserved

© 1995 Springer Science+Business Media Dordrecht
Original1y published by Kluwer Academic Publishers in 1995
Softcover reprint of the hardcover 1st edition 1995
No part of the material protected by this copyright notice may be reproduced or
utilized in any form or by any means, electronic or mechanical,
including photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner.
 For my mother, Louise Orlando (12/8/36-11/2/89).
Thirty-one years is not long enough to know a person like her.
 Acknowledgements

On early drafts of this book, I received helpful comments and

encouragement from R. E. Allen, Joe Margolis, Gary Matthews, Terry
Penner, Bill Prior, Alan Sidelle, and David Welker. An anonymous
referee provided me with a multitude of detailed and helpful comments
on the penultimate draft.
 CONTENTS

1. INTRODUCTION

i. Introductory remarks . . . . . . . . . . . . . . . . . . . .. 1
ii. Assumptions.......................... 3
Ill. Organization.......................... 4
IV. Socrates's influence . . . . . . . . . . . . . . . . . . . . . , 5
v. Plato's tribute to Socrates . . . . . . . . . . . . . . . . .. 7
VI. Philosophers against lovers of spectacles . . . . . . . . 11
vii. Herac1itean flux . . . . . . . . . . . . . . . . . . . . . . .. 15
viii. Forms and the four kinds of stuff . . . . . . . . . ... , 19
IX. Plato's place in history. . . . . . . . . . . . . . . . . . .. 22

2. PLATO'S TRIBUTE TO SOCRATES

i. Introductory remarks . . . . . . . . . . . . . . . . . . . .. 26
ii. Socrates's "What is F?" question . . . . . . . . . . . . . 26
iii. A new method of inquiry .... . . . . . . . . . . . . .. 30
iv. Against Vlastos . . . . . . . . . . . . . . . . . . . . . . . . 36
v. Against Annas . . . . . . . . . . . . . . . . . . . . . . . .. 41
vi. The Forms exist in a third realm . . . . . . . . . . . . . 45
vii. Forms and Recollection . . . . . . . . . . . . . . . . . .. 55
viii. Against Penner . . . . . . . . . . . . . . . . . . . . . . . . 57

3. PHILOSOPHERS AGAINST LOVERS OF SPECTACLES

I. Introductory remarks . . . . . . . . . . . . . . . . . . . .. 64
n. The Socratic challenge . . . . . . . . . . . . . . . . . . .. 64
iii. An argument from knowledge . . . . . . . . . . . . . . , 68
iv. The lover of spectacles . . . . . . . . . . . . . . . . . . . 75
v. Against Owen . . . . . . . . . . . . . . . . . . . . . . . . . 80
vi. Against Vlastos . . . . . . . . . . . . . . . . . . . . . . . . 88
vii. Against Irwin . . . . . . . . . . . . . . . . . . . . . . . . . 93

VII
Vlll Contents

4. HERACLITEAN FLUX

i. Introductory remarks . . . . . . . . . . . . . . . . . . . .. 99
ii. The Aristotelian explanation . . . . . . . . . . . . . . . . 99
iii. Careless assimilation . . . . . . . . . . . . . . . . . . . .. 105
iv. Heraclitus and Protagoras . . . . . . . . . . . . . . . . .. 106
v. Two mistakes . . . . . . . . . . . . . . . . . . . . . . . . . 116
vi. Forms are not "in" sensibles ... . . . . . . . . . . . .. 119
vii. Against traditional interpretations . . . . . . . . . . . .. 122

5. FORMS AND THE FOUR KINDS OF STUFF

i. Introductory remarks . . . . . . . . . . . . . . . . . . . .. 133

ii. Eternal Forms versus transitory sensibles . . . . . . .. 133
iii. Starting-points in ontology . . . . . . . . . . . . . . . . . 135
iv. Against Cherniss . . . . . . . . . . . . . . . . . . . . . . . 145
v. Against Cornford .. . . . . . . . . . . . . . . . . . . . .. 150
vi. Against other interpretations ................. 154
Vll. Against Frede . . . . . . . . . . . . . . . . . . . . . . . . . 160

6. CONCLUSION

i. Introductory remarks . . . . . . . . . . . . . . . . . . . .. 171

ii. The received wisdom . . . . . . . . . . . . . . . . . . . .. 171
iii. The development of Plato's thought . . . . . . . . . . . 173
iv. Aristotle on Plato . . . . . . . . . . . . . . . . . . . . . .. 175
v. Illuminating a perplexing passage . . . . . . . . . . . .. 177
vi. Plato's conception of reality . . . . . . . . . . . . . . . . 178

Notes .................................. 180

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 217
 1

INTRODUCTION

i. Introductory remarks

Plato, but not Socrates, concluded that the Forms are substances. 1
Whether the Forms are substances is not an issue that Socrates had in
mind. He did not deny it, but neither did he affirm it. If Socrates
were asked a series of questions designed to determine whether he
believed that the Forms are substances, he would admit that he had no
opinion about this philosophical issue. Unlike Plato, Socrates was not
a metaphysician. The same, of course, would not have always been
true of Plato. Unlike Socrates, he was a metaphysician. At some point
in his career, and at least by the time of the Phaedo and the Republic,
Plato did what Socrates never thought to do. Plato considered the
question and concluded that the Forms are substances.
Although this development occurred more than two thousand years
ago, time has not eclipsed its importance. It is one of the most seminal
events in the history of the philosophy. With his defense of Socrates's
method of intellectual inquiry, and the development of his Theory of
Forms, Plato caused a now familiar cluster of metaphysical and
epistemological issues to become central to philosophy.
Many Platonic scholars have tried to understand the origin of the
Theory of Forms, and although some of their attempted explanations
are intriguing, none has been compelling. None has provided a reading
of the dialogues that is true both to the corpus and to Plato's historical
context. Some scholars may try to dispute this assessment, but the lack
of a standard explanation in the text books tells the score. Nearly
every introductory text in the history of ancient philosophy records that
Plato concluded that the Forms are substances and starting-points, but
few, if any, offer any real explanation for this development. The
reason for this omission is not that the topic is too advanced. It is that
there is little agreement among scholars about why Plato concluded that
2 Introduction

the Forms are substances. Despite the many attempted explanations,

this scholarly issue remains unresolved. 2
My intention is to settle the debate. The Theory of Forms is the
theory that certain essences exist and that they exist as substances, i.e.,
ontological starting-points. I argue that each of these conjuncts
becomes part of the theory as Plato attempts to validate the leading
assumptions in the method of inquiry that Socrates uses in his search
for knowledge about matters involving justice, beauty, and other such
properties. Plato agreed with Socrates on the priority of definition in
the search for knowledge, and he decided that this epistemological fact
imposes significant and far reaching constraints on how reality can be.
Plato came to believe that reality must contain Forms to be the essences
of justice, beauty, etc., and because he was unsatisfied with the
ontologies that he associated with the sophists and the Ionian natural
scientists, he came to believe that these Forms must exist in reality as
substances and starting-points. Plato thought that the possibility of
knowledge entailed that reality could be no other way.
My explanation of the development of Plato's thought has several
virtues that make it more probable than its predecessors. It fits well
with Plato's place in history, particularly with his relation to the Ionian
natural scientists, Socrates, and the sophists. It makes Plato's thought
about the Forms be part of a philosophical and intellectual history that
begins in the pre-Socratic period, extends through Socrates and the
sophists, and continues into the twentieth century. My explanation
does not come to grief over the text. It fits well with the development
of Plato's thought from the early to the late dialogues. Unlike its
predecessors, my explanation provides for an illuminating and unified
reading of three passages that scholars have long recognized as keys to
the development of Plato's thought, but which have proved stubbornly
resistant to interpretation, both individually and as a group. They are
Phaedo 95e-102b, Republic 473d-480a, and Timaeus 48b-52d. These
passages stand as a veritable gauntlet that has been the ruin of previous
attempts to explain the origin of Plato's Theory of Forms.
 1. ii Assumptions 3

ii. Assumptions

I approach the evidence armed with three substantive assumptions. The

first, however, is not especially controversial. It is that Plato came to
his belief about Forms against a background of intellectual and political
events. Plato was a man. He lived in an historical context, and no
explanation that ignores this fact can be very illuminating.
The second assumption is more controversial. It is that Plato had
reasons for believing that the Forms are substances. He did not just
find himself with this belief. Nor did he come to have it in virtue of
some unconscious influence. Plato adopted his belief in response to a
philosophical problem. It was the conclusion of an argument.
The third assumption is still more controversial. It is that Plato's
reasons for this conclusion were not simpleminded mistakes. Twentieth
century philosophers might not accept his reasons, but the difference
of opinion would be substantive. Plato is one of the greatest
philosophers in the history of the subject. No explanation in terms of
trivial confusions is likely to be correct. 3
The second and third assumptions should be uncontroversial, but
they actually run contrary to much of the received opinion about Plato.
Both inside and outside the field of Platonic studies, many twentieth
century philosophers are content to suppose that Plato fell into his
belief in virtue of some simple confusion. This trend may have its
roots in Aristotle. He had little patience for the Theory of Forms. He
dismisses it as npf.TLUIlCiTa (Posterior AnaLytics 83a33), the sounds of
those who vocalize without expressing propositions.
The situation, unfortunately, has not improved with age. No
sympathy exists for the Theory of Forms among philosophers in the
analytic tradition that stems from Wittgenstein. These philosophers
suggest that Plato fell into his belief because he was held blindly in the
grip of certain mistaken views about grammar. According to this
explanation, Plato was misled by surface grammatical structure. He
failed to appreciate that abstract nouns are nominalized predicates.
This prevented him from understanding that formal truths about beauty,
justice, etc., are determined by linguistic practices, not by metaphysical
objects that exist in some special compartment of reality.
4 Introduction

I am convinced that both the Platonic corpus and its historical

context are inconsistent with this Wittgensteinian explanation. Indeed,
I am convinced that the corpus and its historical context are clearly
inconsistent with all such patronizing attempts to explain the
development of his thought, but instead of simply presenting the
evidence to prove this inconsistency, perhaps the only really effective
way to refute these explanations is to establish a better one in their
place. This is just what I intend to do. I think that more can be said
for Plato and his Theory of Forms than many scholars have been
willing to allow. If I am correct, much of the received wisdom about
Plato distorts both the development of his thought and his place in the
history of philosophy.

iii. Organization

My discussion proceeds according to the order dictated by the

traditional arrangement of the Platonic dialogues. I begin with an
analysis of the assumptions that inform Socrates's search for definitions
in the early dialogues, and I end with an interpretation of the passages
in the Timaeus in which Plato has his character discuss the Forms, the
Receptacle, and the four kinds of stuff. As I proceed, I defend my
readings against some of the alternative positions in the secondary
literature. Because this literature is so very extensive, I can only
defend my readings against what I take to be the most important and
influential alternative interpretations.
This restriction does little to ease my burden. My explanation of
the development of Plato's thought is contrary to the claims of many
distinguished scholars. The interpretations that I argue against belong
to many of the leaders in the field. They include Julia Annas, Harold
Cherniss, F. M. Cornford, Michael Frede, Norman Gulley, Terry
Irwin, G. E. L. Owen, Terry Penner, and Gregory Vlastos.
This first chapter constitutes an introduction to the most important
points that I try to establish in subsequent chapters. In section iv, I
discuss Socrates's influence and show why it is insufficient to explain
the genesis of Plato's belief that the Forms are substances. I present
the textual evidence in chapter 2. In sections v, vi, vii, and viii, I
 l.iii Organization 5

present my explanation of this development in Plato's thought. In

chapters 2, 3, 4, and 5, I develop this interpretation, present textual
evidence, and argue against alternative interpretations. I conclude in
section ix with a discussion of Plato's place in history and the
importance of his Theory of Forms in the history of philosophy.

iv. Socrates's influence

Socrates is the most important figure in Plato's historical context. This

fact is not in doubt, but there is much less agreement on the content of
Socrates's influence. I argue that his appeal to Forms in his search for
knowledge about matters involving beauty, justice, etc., is a crucial
part of the background against which Plato's thought develops.
Socrates's influence is evident from the way he figures as a
character in the Platonic dialogues. In the early definitional dialogues,
he appeals to the Forms to help account for the possibility of
knowledge with respect to certain questions that interest him and his
interlocutors. At the end of book I of the Republic, e.g., he suggests
that inquirers who seek knowledge about such matters must follow a
"method" (Jlf.8ooo~). He suggests, firstly, that they must know the
answer to the "What is justice?" question before they can come to
know the answers to the questions that prompt the inquiry, such as
whether a given "city" (1l'OA.L~) is just. Socrates suggests, secondly,
that there is an answer for him and his interlocutors to know because
there is a Form to be the essence of justice. 4
Plato agreed with Socrates. With respect to matters involving
justice, beauty, etc., Plato agreed that knowledge is not possible unless
definitional knowledge is possible. This principle informs the so-called
"priority of definition. ,,5 Plato also agreed that the Forms help make
definitional knowledge possible. This principle has not attracted
enough attention to garner a name, but I shall call it the "priority of
Form." Plato agreed that the Forms insure that there are definitional
truths to be possible objects of knowledge.
Socrates believed in Forms, but he was no metaphysician. 6 He
assumed that the Form the F itself is specified in answer to the "What
is F?" question, and he assumed that this definitional answer is part of
6 Introduction

the justification necessary for knowledge about matters involving

F-ness. As for whether the Forms themselves require explanation, or
are starting-points in ontology, he had no opinion and did not consider
the question. This philosophical issue was not one of his concerns. He
was not concerned with how Forms fit into a conception of reality that
accounts for the possibility of knowledge. Socrates sought knowledge
about matters involving justice, beauty, and the other properties that
figure prominently in the early dialogues, but he was simply not
concerned to argue about how Forms fit into reality. 7
Socrates's influence explains why Plato supposed that the Forms
exist, but it does not explain why he concluded that they exist as
substances. Socrates did not have a view on how certain properties
constitute essences. His retreat to Forms was in no important way
different from the ordinary retreat to definitional truths. If someone
were to retreat to such truths to explain why something with
such-and-such features counts, say, as beautiful, he would not
necessarily believe that these truths are ground in metaphysical objects
that exist in some special compartment of reality. Nor would he
necessarily believe that they are ground in the reality of the sensible
world and the ways that language users choose to use their words.
With respect to the Forms, Socrates never ventured beyond this
unsophisticated position. Plato did. By the time of the two great
middle dialogues, the Phaedo and the Republic, he had an ontological
theory about Forms. He believed, to borrow Quine's famous words,
that the Forms "have being independently of the mind," that "the mind
may discover them but cannot create them" ([1948], p. 14).8 Once
Plato came to believe that the Forms exist as substances, he had done
something much more than take over the use of Forms that he found
in the way that Socrates searched for knowledge.
In terms of Carnap's well known distinction between "internal" and
"external" questions ([1950]), Plato came to believe that the assertion
that there are Forms is an answer to an external question. The
grammatical resources in ordinary English provide for the
nominalization of predicates, and thus provide speakers with the means
to talk about the properties that predicates express. The grammatical
resources in ordinary English also provide for existential quantification,
 1. iv Socrates's influence 7

and thus provide speakers with the means to assert the existence of
what they talk about. To assert that the essence of justice exists,
therefore, is not necessarily to assert a proposition with metaphysical
import. The proposition can be a trivial consequence of nothing more
than an acceptable definition of justice and the grammatical resources
in ordinary English. When Plato concluded that the Forms exist as
substances, he did not suppose that they merely have this "internal"
existence. He accepted a proposition with radical metaphysical import.
He concluded that the Forms are "really real" (OIlTW<; all), that they are
part of a reality that is "external" to the sensible world and to the ways
that language users choose to use their words. Plato, but not Socrates,
believed that the Forms exist in a third realm.9

v. Plato's tribute to Socrates

I argue that part of the explanation for this development in Plato's

thought is that he understood Socrates's practice in his search for
knowledge as part of a reaction to the practices of the Ionian natural
scientists. Socrates's new way of searching for knowledge resulted in
a significantly changed conception of intellectual inquiry, and it helped
to redefine the practice of philosophy in the ancient world.
The Greek adjective 1>LMCJo1>o<; occurs rarely in the surviving
pre-Platonic texts.1O Perhaps the two best known occurrences are in
Herodotus's Histories 1.30.2 and in Pericles's Funeral Speech in
Thucydides's Histories 11.40, where they mean roughly "lover of
knowledge. "11 The intellectuals who could go by this description
were primarily those who sought the kind of "wisdom" (CJo1>ia) that the
ancients called 'inquiry into nature' (7rEPL'1>VCJ€W<; LCJTopiall)12 and that
can be called 'natural science' as long one keeps firmly in mind that the
interests of these pre-Socratic intellectuals were extremely broad and
included meaning, knowledge, and other such matters. 13
Anaxagoras was perhaps the most prominent such "philosopher"
residing in fifth century Athens, and Socrates may have been an early
member of Anaxagoras's school. Under the patronage of Pericles,
sometime after 480 BC, Anaxagoras established a school in Athens and
remained there until about 430 BC. This was the first such school in
8 Introduction

Athens, and Socrates may have been familiar with its teachings. The
character says in the Apology that the many who belittle him say that
"he busies himself studying things in the sky and below the earth"
(19b4-5),14 and Meletus, one of his accusers, charges him with beliefs
that Anaxagoras was known to hold (26dl-9).
Socrates, however, soon gave up this sort of inquiry, if he ever
engaged in it at all. As the early definitional dialogues clearly show,
and Aristotle reports in the Metaphysics, Socrates soon "was busying
himself with ethical matters, neglecting the world of nature as a whole,
seeking the universal in these ethical matters, and fixed thought for the
first time on definitions" (987b 1-4). Whether he ignored his
predecessors, or argued against them, Socrates's turn to the Forms in
his search for knowledge was the centerpiece of a revolution that
reshaped intellectual inquiry in the ancient world. 15
The most important evidence for understanding this revolution is
Socrates's intellectual autobiography in the Phaedo, but to appreciate
the full import of this passage, both for the development of Plato's
thought and for the place of Plato's thought in the history of
philosophy, one must depart from the orthodox interpretation. In the
autobiography, as I read this passage, Plato recounts the revolution and
pays tribute to Socrates for his contribution. Furthermore, and more
importantly for Platonic scholarship, Plato shows himself to conclude
that the Forms are substances. In his effort to entrench the shift in
perspective and to validate the new conception of inquiry, he marks the
natural scientists's method as "pseudo-philosophical" and part of a
conception of reality that is inconsistent with knowledge. On this
conception of reality, the Forms are not substances. The only
substances are sensible objects, such as portions of fire, water, and the
other stuffs that are traditional in Greek cosmology. Plato marks
certain assumptions in Socrates's method as "philosophical" and part of
a conception of reality that is consistent with knowledge. 16 This
conception is the Platonic conception of reality. The Forms are among
its starting-points. Plato supposes that if the Forms of justice, beauty,
etc., do not exist as substances, but either do not exist at all or must
supervene on more basic objects, then there are no grounds for the
 1. v Plato's tribute to Socrates 9

respective definitional truths that are necessary for the possibility of

knowledge about matters involving these properties.
The orthodox interpretation of Socrates's autobiography, at least
among contemporary English speaking historians in the analytic
tradition, is the one that Gregory Vlastos offers in his extremely
influential and now classic 1969 paper "Reasons and Causes in the
Phaedo." Although I am in agreement with many points in Vlastos's
paper, his interpretation is significantly different from the one that I
have outlined and believe is true. If Vlastos is correct, Plato discusses
the natural scientists in order to help his readers understand what
Aristotle was able to later express more clearly with his well known
Doctrine of the Four Causes. Vlastos argues that Plato maintains that
the Greek noun airLa and its cognates apply to different and
non-competing kinds of causes, including physical and logical causes,
and that these different kinds of causes are appropriate for different
kinds of explanation. According to Vlastos, Plato does not choose
Socrates and his method over the natural scientists and their method;
rather, like Aristotle after him, he tries to make room for both kinds
of intellectual inquiry. Vlastos argues that Plato realizes, and is trying
to help his readers to realize, that although Socrates and the natural
scientists both use forms of the noun ai.rLa to describe the objects of
their inquiry, they are concerned with different kinds of causes and that
each such kind is acceptable in its own domain.
The evidence is not completely straightforward, but it is clear
enough to show that Vlastos is mistaken. In trying to see Plato as
anticipating Aristotle's Doctrine of the Four Causes, Vlastos
misunderstands the development of Plato's thought. He removes Plato
from his rightful place in the history of the intellectual reaction to the
Ionian cosmological speculations and their novel conceptions of reality.
This, in turn, causes Vlastos to misread Socrates's autobiography and
to leave his readers with absolutely no understanding of Plato's reasons
for concluding that the Forms are substances.
Socrates may have argued against the natural scientists and
consciously turned against them, or he may have simply taken no notice
of them, but the character in the Phaedo supplies an argument for
rejecting the explanations that fall out the natural scientists's method of
10 Introduction

inquiry. With respect to matters involving justice, beauty, etc., he

supposes that if the explanations of the natural scientists were accepted
in the place of his explanations in terms of the definitional content of
Forms, knowledge would be impossible because there would be no
truths to know. The character supposes that if the natural scientists's
way of searching for knowledge were accepted as the philosophical,
i.e., "scientific" method of intellectual inquiry, then one would have to
accept a conception of reality according to which the essences of
justice, etc., either would not exist or would have their source in
something other than Forms that exist "themselves according to
themselves" (aVia mO' aUia), and he rejects both alternatives. On
matters involving justice, beauty, etc., Plato agreed with Socrates that
knowledge is possible only if the respective definitional knowledge is
possible. Further, in reflecting on how the essences fit into reality,
Plato came to believe that Socrates showed good philosophical sense
not to deny that they are starting-points. He came to believe that unless
the essences were starting-points there would be no definitional truths
to be possible objects of knowledge. Plato came to believe that
definitional truths about justice, beauty, etc., have their objective
ground in a third realm of Forms.
Contrary to Vlastos's interpretation, Socrates's autobiography is a
record of a clash between different methods of inquiry and their
corresponding conceptions of reality. Plato does not attempt to make
them live in peace. He does not suppose that the natural scientists's
explanations are unacceptable as logical explanations, but are acceptable
as physical explanations. He outright rejects the explanations that fall
out of their method of inquiry. He supposes that this method is part of
a conception of reality in which the essences do not exist as
starting-points, and he believes that in this case there can be no
knowledge about justice and the other properties that Socrates asks
about. Plato maintains that there can be such knowledge, he accepts
the leading assumptions in the method that Socrates uses to pursue this
knowledge, and he believes that these assumptions are inconsistent with
a conception of reality in which the essences do not exist as substances.
In trying to see Plato as anticipating Aristotle, Vlastos obscures these
important truths.
 1. vi Philosophers against lovers of spectacles 11

Vl. Philosophers against lovers of spectacles

I argue that in the Republic, the other great dialogue from the middle
period, Plato again shows himself to believe that the Forms are
substances. Indeed, he argues for this conclusion in his discussion of
the philosopher and the lover of spectacles in hook V of the Republic.
This argument is parallel to the earlier argument against the Ionian
natural scientists, but the opponents are now much more prominent in
fifth century Athens. As in the Phaedo, so also in the Republic, the
argument takes place against a background in which Socrates reacts to
his predecessors, but now the reaction is to opponents that Plato
characterizes as lovers of "opinion," or "fancy" (oo~a), as opposed
"wisdom" (aocf>ia). This characterization gives them a philosophical
position like the one that he associates with the sophists.
In book V, after he insists that only the philosopher is tit to rule,
Plato distinguishes the philosopher from his competitor, the lover of
opinions and spectacles, in terms of their respective beliefs about the
Forms. The philosopher accepts the ontology and explanation in terms
of the definitional content of Forms that Socrates defends in the
Phaedo. The lover of spectacles admits into his ontology the many just
things, beautiful things, etc., but he rejects the philosopher's way of
fitting the respective essences into reality. Instead of accepting the
Platonic claim that essences exist as substances, he insists that they
must find their place among "the many conventions (VOILLIW) of the
many." Plato is dead set against this conception of reality. He argues
that the lover of spectacles accepts an ontology that is inconsistent with
the possibility of knowledge about justice, beauty, and other such
matters. Plato believes that, if the lover of spectacles were right about
reality, then for the properties that Socrates uses his "What is F?"
question to ask about, there would be no fact of the matter about what
features a given thing must have to be F.
The lover of spectacles holds a metaphysical and epistemological
position similar to the one that Plato associates with the sophists. He
takes the sophists, especially the younger generation of sophists, to
advocate a conception of reality in which claims about justice, beauty,
etc., are part of a subjective and merely human perspective. Plato
12 Introduction

rejects this conception of reality because he believes that it wrongly

removes essences from objective reality and reduces them to nothing
more than the fancies of individuals and groups. He maintains that if
the many vop..Lp..a about justice, e.g., were accepted in place of the one
Form the just itself that exists "itself according to itself" (auTe) KaO'
aUTO), there would then be nothing that justice is, because in reality
justice would have no essence, and although individuals could fancy
themselves as speaking truths about matters involving justice, no
statement that predicates justice would have a truth value, and political
knowledge would therefore be impossible. 17
This reading should seem natural, perhaps even beyond dispute, but
many scholars actually favor a different interpretation. G. E. L. Owen
first put this interpretation forward in his 1957 paper "A Proof in the
II€pL Io€wv." He argues that Plato came to believe that the Forms exist
as both samples and standards. Plato came to this conclusion,
according to Owen, because he was impressed that disagreement over
whether, e.g., something is a finger, or a portion of fire, is resolved
by reference to "unambiguous samples" of fire and fingers, but he
found that the "world which contains unambiguous samples of fire and
fingers contains no comparable cases of goodness or similarity ... " (p.
109). On Owen's interpretation, the philosopher and the lover of
spectacles agree on how to resolve disagreement about matters
involving fingers, fire, etc., but they disagree on how to resolve
disagreement about matters involving the properties that Socrates uses
his "What is F?" question to ask about. The lover of spectacles is
content to try to make certain ordinary sensible objects of experience
function as "unambiguous Paradigms," but the philosopher insists that
only intelligible objects, not the ordinary sensible objects of experience,
can function as such paradigms (p. 109). He insists that if there were
only the ordinary sensible objects, then for the properties that Socrates
asks about, no "unambiguous Paradigms" would exist, and
consequently everything would be both F and not-F. To avoid this
result, and to allow for knowledge, the philosopher supposes that there
are Forms that exist as "unambiguous Paradigms."
Although perhaps Plato could argue in the way that Owen claims,
the text provides insufficient evidence to conclude that he actually does
 1. vi Philosophers against lovers of spectacles 13

argue this way. Owen is quite correct that the clash between the
philosopher and the lover of spectacles is a clash between different
conceptions of reality. On Owen's interpretation, as on mine, Plato is
concerned with the ontology that is necessary for the possibility of
knowledge about matters involving justice, beauty, etc., but Owen is
wrong to suppose that Plato accepted the very implausible proposition
that the Forms are special instances of justice, beauty, and the other
properties that Socrates asks about.
Owen argues that Plato accepts this implausible proposition
because he wants to extend a certain semantics for (Greek counterparts
of) the predicates 'is a finger,' 'is a portion of fire,' etc., to cover the
predicates that occur in Socrates's "What is F?" question. The middle
dialogues, however, reveal a very different proximate starting-point for
the Theory of Forms. Plato shows no interest in the semantics for
(Greek counterparts of) 'is a finger' and other such predicates. His
interest is in Socrates's search for detinitions and its place in the search
for knowledge. To validate the leading assumptions in Socrates's
method of inquiry, Plato struggles to incorporate Forms into reality in
a way that is consistent with the possibility of knowledge.
In principle, of course, Plato could come to believe that the best
way to complete this project is to make the Forms exist as
"unambiguous Paradigms," but given that he is sophisticated enough to
take on this philosophical problem, I doubt that he could really accept
the rather bizarre solution that Owen claims for him. Only extremely
clear and compelling evidence could show that Plato concluded that the
Forms are special instances of justice, beauty, etc., and neither in the
discussion in book V of the Republic, nor, for that matter, in any
discussion in any place in the entire Platonic corpus, is there any such
evidence. The Theory of Forms is not without problems, but Owen
saddles Plato with more problems than he deserves. l8
This unsympathetic approach to Plato is manifest in subsequent
endorsements of the main lines of Owen's interpretation. In his
important 1965 and 1966 papers "Degrees of Reality in Plato" and
"Metaphysical Paradox," although he gives him a different motivation,
Vlastos agrees that Plato came to believe in a third realm of Forms that
are both samples and standards. Vlastos holds that Plato "misconstrued
14 Introduction

universals as a higher grade of particulars" ([ 1965], p. 75). On his

interpretation, the lover of spectacles is content to try to extract
answers to the "What is F?" question from ordinary sensible objects of
experience that are F. Plato does not reject this procedure, but he
insists that it works only on intelligible objects. In locating the source
of definitions in sensible objects, the lover of spectacles leaves himself
no way to avoid supposing that everything is both F and not-F. The
philosopher does not have this problem because he grounds definitional
truths in intelligibles. He supposes that the answers to Socrates's
"What is F?" question are in Forms that exist in a third realm and that
are both samples and standards. Unlike the ordinary sensible objects
of experience, the Forms somehow wear their essences on their sleeves.
The separately existing soul can somehow see that the Form the F itself
is F, can see that it has other properties, and can conclude that these
properties are the essence of F-ness. When it is incarnated in the body,
if it is a philosopher and does not succumb to the body's many lusts
and desires, the soul can recollect what it previously knew.
In his variation on Owen's interpretation, Terry Irwin does not
maintain that Plato came to believe in a third realm of Forms that exist
as both samples and standards, but his interpretation is hardly generous.
He maintains that the Theory of Forms has its origin in a
straightforward conflation. On his interpretation, the lover of
spectacles is content to try to define justice, beauty, etc., in terms of
sensible properties. Plato supposes, according to Irwin, that lover of
spectacles makes a terrible mistake. He supposes that the lover of
spectacles must conclude that justice, beauty, etc., and their opposites
are reduced to the same sensible properties. He supposes, e.g., that the
lover of spectacles must conclude that beauty both is and is not a matter
of bright coloring. To avoid such unpalatable consequences, Plato has
the philosopher help himself to non-sensible properties to define justice,
beauty, and the other properties that Socrates asks about. Although this
debate between the philosopher and the lover of spectacles is not over
whether the Forms are substances, Irwin maintains that Plato
nevertheless invalidly jumps to this metaphysical conclusion. The
reason, according to Irwin, is that Plato conflates two very different
propositions. Irwin maintains that Plato does not appreciate the
 1. vi Philosophers against lovers of spectacles 15

difference between the proposition that justice, beauty, etc., are not
definable in terms of sensible properties and the proposition that they
exist as substances and separately from sensible objects.
Plato was probably wrong to conclude that the Forms are
substances, but I see no good reason to suppose that he made the
mistakes that Vlastos and Irwin claim for him. In book V of the
RepubLic, Plato struggles to incorporate essences into reality. The lover
of spectacles tries to understand claims about justice, beauty, etc., in
terms of nothing more than the "many conventions of the many." The
philosopher opposes the lover of spectacles. He insists that if the
"many conventions of the many" about F-ness were to replace the one
Form the F itself, then instead of the one objective definitional truth
that had its source in this Form, there would be many subjective
"opinions, or "fancies," that masquerade as definitional truths. He
insists that if, instead of the one definitional truth, there were these
many ersatz definitional truths, each specifying different properties as
the essence of F-ness, then everything would be both F and not-F, and
consequently knowledge would be impossible. He insists that the lover
of spectacles accepts a conception of reality that is inconsistent with the
possibility of political knowledge and expertise. To avoid this unhappy
state of affairs, the philosopher insists that the essences exist as
starting-points. Contrary to Vlastos's interpretation, Plato does not
show himself to misconstrue universals as a higher grade of particulars.
Nor is he guilty of the mistake that Irwin claims for him. Plato does
not jump to a metaphysical conclusion about the ontological status of
the Forms merely because justice, beauty, etc., are not definable in
terms of sensible properties.

vii. Heraclitean flux

Owen and the Oxford school of commentators have been influential in

Platonic scholarship in second half of the twentieth century, but
Aristotle has always been the dominate authority in the field. Given his
long tenure in Plato's Academy, his obvious brilliance, and his deep
and abiding interest in the philosophical issues that inform the Theory
of Forms, no serious attempt to understand the development of Plato's
16 Introduction

thought can afford to ignore Aristotle's explanation in the Metaphysics

at 987a32-b7, 1078b9-1079a4, and 1086a31-bll.
Aristotle's remarks are very familiar. Every serious student ofthe
history of philosophy knows that Aristotle says that Plato concluded
that the Forms are substances on the basis of considerations involving
definition, knowledge, and the Heraclitean "flux" of sensibles. The
problem is to understand what Aristotle means. His remarks have very
little straightforward content. They require substantial interpretation,
and any acceptable interpretation must be consistent with, and help to
shed light on, Plato's thought in the dialogues.
Many influential scholars have supposed, in solution to this
problem, that Aristotle's explanation should be taken to confirm the
view that the Theory of Forms has its origin primarily in Plato's
association with Cratylus and the extreme Heracliteans. On this
interpretation, Plato accepts from Socrates that knowledge is possible,
but his association with the extreme Heracliteans pushes him to the
conclusion that the Forms are substances. He agrees with them that the
sensibles are in radical flux. On the basis of this belief, together with
his belief that knowledge is possible, Plato concludes that the objects
of knowledge must be unchangeable Forms that exist in a third realm.
He concludes that the objects of knowledge cannot be the sensibles, for
these objects are caught up in radical flux.
This interpretation has many problems, but the worst is that it
wrongly subjects Plato to the twin evils of radical flux and skepticism
about the ordinary objects of experience. No where is there sufficient
evidence to warrant attributing these implausible beliefs to Plato. In
certain passages in the Phaedo, e.g., passages which many have taken
to provide the necessary evidence, Plato does draw a sharp contrast
between the Forms and the sensibles. As I read this contrast, however,
it does nothing to commit Plato to the paradoxical proposition that the
sensibles are in radical flux. Plato says that the Forms are eternal,
unchangeable, and intelligible. Plato says that the ordinary objects of
experience are transitory, changeable, and sensible. This does not
show that he believes that the sensibles are in radical flux.
Plato matches this sharp contrast between Forms and sensibles with
an equally sharp contrast between thinking and sensing. He says that
 1. vii Heraclitean flux 17

thinking about the Forms, as opposed to using the senses to look to

nature, is the most proper activity of the philosopher. Again, however,
at least as I understand this contrast, it is hardly sufficient evidence to
show that Plato was a skeptic about the sensibles. It does nothing to
show that he believed that there can be no knowledge about matters
involving the ordinary sensible objects of experience.
Some scholars claim that the Republic provides part of the
necessary evidence, but once again, the text is too weak to support the
position. Plato does argue that, in trying to understand essences in
terms of the many conventions of the many, the lover of spectacles
must accept a conception of reality according to which, instead of
political knowledge and expertise, only "opinion" and "fancy" are
possible. Plato's argument, however, does not depend on the premise
that the ordinary sensible objects of experience are in such radical flux
that they are not possible objects of knowledge. It depends on the
premise that an ontology in which conventions stand in the place of a
third realm of Forms is part of a conception of reality that is
inconsistent with the possibility of knowledge.
This way of understanding the premise fits with the central
argument in the Republic, which is Plato's most important argument in
the middle dialogues, and perhaps also in all of ancient philosophy.
Plato argues that only the philosopher is fit to rule, and one of his
premises is that only the philosopher accepts a conception of reality
according to which there can be political knowledge and expertise.
Plato is not a skeptic. He does not believe that no one can know
anything about the sensible world. This skepticism would be
diametrically opposed to the main point of the Republic. Contrary to
common belief, Plato does not abandon the worldly concerns of men.
Just the opposite is true. He is convinced, and tries to demonstrate,
that only the philosopher has a conception of reality according to which
such concerns fall within the scope of knowledge.
Irwin has tried to provide a better interpretation of Plato and
Aristotle, and his work has taken on the air of orthodoxy. On his
interpretation, Plato IS neither a skeptic nor an extreme Heraclitean
about the sensibles. The key to Irwin's more charitable interpretation
is his novel claim about the way that Plato understands the flux of
18 Introduction

sensibles. He claims that Plato takes the flux of sensibles to be the

reduction of opposites into sensible properties, not the succession of
opposites in sensible particulars. In this way, Irwin squares his
interpretation of Aristotle with his interpretation of Plato. Irwin,
recall, maintains that the Theory of Forms stems from a conflation. On
his interpretation, Plato argues that justice, beauty, etc., cannot be
defined in terms of sensible properties. In the terminology that Irwin
inherits from Owen, Plato claims that the Forms are "compressed" with
their opposites in the sensible properties. The compression of Forms
with their opposites in sensible properties has no import for how the
Forms exist, but Irwin claims that Plato believes otherwise. He claims
that Plato concludes that the Forms exist separately and as substances.
He makes this mistake, according to Irwin, because he conflates
sensible objects and sensible properties.
I am in complete agreement with Irwin's attempt to save Plato from
the twin evils of skepticism and extreme Heracliteanism, but this is
where my agreement ends. The Theory of Forms does not stem from
a simpleminded mistake. Irwin's interpretation of Aristotle's remarks
on the development of Plato's thought is only as good as his
interpretation of the Phaedo and the Republic, and much more can be
said for Plato in these great middle dialogues than Irwin would have his
readers believe. He frees Plato from the twin evils, but he still saddles
him with more problems than he deserves. Plato concluded that the
Forms are substances, not because he fell into confusion over sensibles,
but because he believed that no other way of fitting essences in reality
would validate the leading assumptions in Socrates's method of inquiry.
This is the import of both the Phaedo and the Republic.
Since Aristotle's explanation can be understood in a way that is
consistent with the Phaedo and the Republic, this is how it should be
understood. Aristotle's remarks are brief, but they are to the
philosophical point. He frames the debate not as a historian, but to
highlight what he takes to be the underlying mistake in Plato's
metaphysics. Plato and Aristotle agree that definitional truths are
necessary for the possibility of knowledge. Furthermore, although he
does not voice his agreement, Aristotle agrees with Plato's conclusion
against the lover of spectacles. He agrees that claims about justice,
 1. vii Heraclitean flux 19

beauty, etc., do not depend on the "fancies" of the many. The debate
is over whether the essences are ground in the reality of sensibles.
Aristotle insists that the essences are somehow "in" sensibles, that the
essences do not exist separately and as substances. In his sketch of the
accomplishments of his predecessors, he explains that not appreciating
this fact is the mistake that gives rise to the Theory of Forms.
Aristotle says that "< s > eparation is responsible for the difficulties
arising about < the Theory of> Ideas" (l 086b8).
On my interpretation, Aristotle's explanation both confirms and
helps to shed important light on Plato's reasons for concluding that the
Forms exist as substances. Socrates's autobiography in the Phaedo is
a difficult passage, but, at the very least, it shows that Plato rejected
the natural scientists's method of intellectual inquiry in the search for
knowledge. Because their method does not recognize the proper place
of Socrates's "What is F?" question, Plato believed that it carries a
conception of reality that eliminates the possibility of knowledge. Plato
leaves his readers to conclude that the natural scientists cannot provide
for the possibility of knowledge because they either deny that essences
exist or try to explain them in terms of more basic objects.
The way Aristotle highlights his debate with Plato helps to
illuminate this line of argument against the natural scientists.
Aristotle's remarks suggest that Plato thought that no ontology that
takes "sensibles" (aiu(hrr&) for starting-points is consistent with the
existence of "intelligibles" (V01J7&). Plato thought that because the
natural scientists try to make reality rest on certain sensible
starting-points, such as portions of water, air, etc., and because these
objects are "flowing," the natural scientists accept an ontology that can
provide no stable perch for the existence of essences.

viii. Forms and the four kinds of stuff

Not only does my interpretation make good sense of Aristotle's

remarks, it illuminates a set of extremely difficult and well known
passages in one of the late dialogues. The dialogue is the Timaeus. 19
Very early in the dialogue, because the inquiry concerns the natural
20 Introduction

world, Socrates relinquishes his lead to Timaeus. Unlike Socrates,

Timaeus is an expert in philosophy and cosmology.
Near the "works of necessity" (47e-69a), just before he introduces
the Receptacle into his ontology, Timaeus sets himself against the
Ionian natural scientists who suppose that the ontological starting-points
in reality are portions of fire, air, water, and earth. Plato's intent in
this and subsequent passages is much debated, but given Aristotle's
remarks on the origin ofthe Theory of Forms, and given Plato's earlier
arguments against the natural scientists and the lovers of spectacles, a
very natural and illuminating interpretation is possible.
Timaeus sets himself against these Ionians because he intends to
show that their ontology is inconsistent with the possibility of
knowledge. He supposes that only the Forms and other "intelligibles"
(II01JT(x) are fit to be starting-points in a conception of reality that is
consistent with knowledge. He insists that these objects are the
starting-points in formal explanations of "sensibles" (a'Ul(}1J7&), such as
portions of the stuffs of the four kinds. He argues that if, as is true for
the Ionian natural scientists, stuffs of the four kinds are the
starting-points in reality, then skepticism is unavoidable. He argues
that the Ionian natural scientists cannot provide definitional "accounts"
(M)'OL) for the terms 7rVP ('fire'), vowp (,water'), etc., that are
"trustworthy and firm" (49b5) enough to provide for the possibility of
knowledge about matters involving these four kinds of stuffs.
Timaeus states his argument briefly, but his main point seems to be
that the Ionian natural scientists must give unacceptable replies to the
"What is F?" question. He seems to suppose that the Ionian scientists
must point to sensible objects and utter sentences such as TOV70 7rVP
f.UrL ('This is fire') and TOU70 vowp EUrL ('This is water'), and he
rejects such responses as wholly inadequate ways to answer the
question. Theses responses are inadequate, according to Timaeus,
because the sensible objects of the respective demonstrative pointings
somehow do not have enough "firmness" (49d7), and somehow are not
"steadfast" (e3) enough, to provide accounts that are "trustworthy and
firm" enough to provide for the possibility of knowledge about matters
involving fire, water, and the other stuffs.
 1. viii Forms and the four kinds of stuff 21

The proper responses are of the form TOLOVTOJI uowp fan

('Such-and-such is water') and TOLOVTOJl 7rVP fan (,Such-and-such is
fire'). Responses must take this form, according to Timaeus, because
only such responses are consistent with the role of the Forms the fire
itself, the water itself, etc., to provide the information necessary to fill
the TOLOUTOJI ("such-and-such") schemata. He argues that only the
Forms are "steadfast" enough to provide accounts that are "trustworthy
and firm" enough to provide for the possibility of knowledge.
Plato and the Ionians agree that such knowledge is possible, but he
argues that they have completely failed to see the significance of this
possibility. He argues that the Ionians have failed to see that this
possibility imposes significant and far reaching restraints on how reality
can be. Plato argues that if, as is true in the Ionian cosmology, the
starting-points in reality are sensible objects, such as portions of fire,
water, etc., then the distinction between knowledge and true belief
would collapse, and fancy would be all that remains.
Unlike Socrates's autobiography in the Phaedo and the discussion
of the philosopher and the lover of spectacles in book V of the
Republic, there is no orthodox interpretation for these passages from
the Timaeus. They are notorious for their for impenetrability. All
scholars recognize their importance for understanding the Theory of
Forms, but despite repeated efforts, these passages have stubbornly
resisted analysis. In the secondary literature, a wide diversity of
interpretations have come to occupy an uneasy co-existence.
Although the passages in question are hardly straightforward, this
is not the only source oftheir resistance to analysis. Another important
and more easily eliminated source of the problems that plague many
commentators is in the background assumptions that they try to bring
to these passages. One particularly problematic such assumption stems
from what I have argued is a fundamentally mistaken interpretation of
Aristotle's explanation of the origin of the Theory of Forms. This
assumption is common to many interpretations. Cherniss and Gulley,
e.g., despite their other serious differences of opinion, both interpret
the Timaeus against the background assumption that Plato was
influenced by the extreme Heracliteans and came to believe that the
sensibles are always flowing in every way and that there can be no
22 Introduction

knowledge with respect to such objects. This is a mistake. Plato's

previous arguments against the natural scientists and the lovers of
spectacles provide the proper background assumptions for the Timaeus.
With these arguments in the background, the correct interpretation
comes into focus. Timaeus argues against a conception of reality in
which essences are not starting-points. As in the autobiography in the
Phaedo and in book V of the Republic, so also in this section of the
Timaeus, Plato shows himself to believe that such an ontology is
inconsistent with the possibility of knowledge.

ix. Plato's place in history

Plato's beliefthat the Forms are substances did not spring from a trivial
confusion. It was a response to a philosophical problem that he
inherited from Socrates. In his search for knowledge, in his insistence
that he and his interlocutors must first answer the "What is F?"
question before they can answer the other questions that concern them,
Socrates was part of an intellectual movement that struggled with issues
connected to the very perplexing third condition that is necessary for
knowledge, the so-called "justification" condition.
Socrates was not the only member of this movement. The
justification necessary for knowledge was already a topic of discussion
in the pre-Socratic period. The sixth century Milesian cosmologists
incorporated an optimism concerning the possibility of knowledge.
Thales and his "school" made bold assertions about the underlying
nature and behavior of the universe. These assertions undermined the
unreflective confidence in the orthodox conception of reality. This, in
turn, provoked a discussion of the methods for acquiring knowledge.
Some of this discussion is preserved in the fragments of Xenophanes,
Heraclitus, Parmenides, and other pre-Socratics.
In fifth century Athens, after Pericles's legal reforms made
oratorical skill necessary for political power, a different set of events
again provoked inquiry into the methods for acquiring knowledge.
With the spectacular rise of the suphists, and the ntw USt of rhetorical
skill to manipulate the will of political bodies, many confused the
power of the spoken word WIth the force of reason. Socrates was the
 l.ix Plato's place in history 23

exception. To expose and to combat this confusion, he insisted on the

method of discussion by question and answer. He tried to screen the
use of rhetoric and make persuasion come through reason.
These achievements are important, but they do not exhaust
Socrates's claim to philosophical immortality. In his search for
knowledge, he did not just turn his back on explanations according to
the natural scientists's method of inquiry. Nor did he just provide an
antidote to the sophists's tendency to contlate the power of the spoken
word with the force of reason. Socrates started an intellectual
revolution. He introduced a new method of inquiry. About matters
involving justice, beauty, and the other properties he used his "What
is F?" question to ask about, he insisted on the priority of definition.
He insisted that definitional truths are part of the justification necessary
for knowledge. He believed that, on such matters, knowledge is not
possible unless definitional knowledge is possible.
Socrates's belief about the structure of knowledge was new and
compelling, but it was also perplexing in ways that he did not
appreciate. His belief presupposes that on matters involving justice,
beauty, and other such properties, empirical truths depend on formal
truths. Empirical truths can seem straightforward enough, but formal
truths have always been deeply puzzling. Socrates's essences are not
easy to fit into reality, and this fact casts doubt on the coherence of his
conception of the structure of knowledge. If the essences are not to be
wholly subjective, they apparently must have their ground in something
more than just mere "convention," for otherwise, as Aristophanes says
in the Clouds, what passes for truth would be nothing more than the
whims and fancies of those who manage to "talk over the crowd"
(1422). What this something is, however, is not immediately evident.
Philosophers in the Aristotelian tradition have hoped to make sensible
objects do part of the necessary work, but just how they can do this
work has never been made completely clear.
Plato struggled with this problem. He accepted Socrates's assertion
about the structure of knowledge, and he became convinced that
essences must exist in reality as substances. Plato became convinced
that the alternative conceptions of reality, as he understood them, were
inconsistent with the possibility of knowledge. This is the import of
24 Introduction

Phaedo 95e-102b, Republic 473d-480a, and Timaeus 48b-52d. In these

passages, Plato argues against the conceptions of reality that he
associates with the Ionian natural scientists and the sophists. Plato
argues that if there were no third realm of Forms, then the distinction
between knowledge and true belief would collapse. He argues that
there would be no facts of the matter, and instead of knowledge and
expertise, only whim and fancy would be possible.
With Socrates as the catalyst, and against the background of the
Milesian cosmological speculations and the confusion over justification
among the sophists, Plato created a new conception of reality to
account for the possibility of human knowledge. Heraclitus,
Parmenides, and others had reacted to the Milesians. Socrates had
reacted to the sophists, but Plato's reaction stands apart as an
extraordinary and singular achievement. It resulted in the first great
world view in the history of philosophy. Plato created a conception of
reality, and the place of humans in it, that for a short time seemed to
validate the leading assumptions in the Socratic method of inquiry.
Plato supposed that knowledge is possible because there are Forms in
a third realm. He supposed that humans, if they are philosophers, can
acquire knowledge because before birth they were with these Forms,
and in life they can recollect what they once knew.
Few twentieth century philosophers accept Plato's specitic
metaphysical conclusions,20 and I have no intention of trying to
convince them otherwise, but I do believe in giving credit where it is
due. Plato went beyond Socrates and concluded that the Forms are
substances. This is true, but it is only one part of the story. The
remaining, and often untold, part of the story is that Socrates was idle
and Plato stepped past him and into the ring with a genuine
philosophical problem that Socrates only dimly recognized, and perhaps
did not recognize at all. To show that human knowledge is possible,
Plato recognized that he must present a conception of reality, explain
the place of humans within it, and show how such beings can acquire
knowledge. His conception is no longer compelling, but it is an
attempt to solve a genuine problem. It is not a pseudo-problem, born
in a simpleminded confusion, that some subsequent more sober-minded
philosopher could easily expose and quickly eliminate. Aristotle and
 l.ix Plato's place in history 25

subsequent philosophers have produced different conceptions, but no

one has yet produced a completely convincing conception. After all
these years, the philosophical problem remains unsolved.
 2

PLATO'S TRIBUTE TO SOCRATES

i. Introductory remarks

Socrates's thoughts about the proper way to search for knowledge are
a crucial part of the intellectual background against which Plato came
to believe that the Forms are substances. l Plato presents these
thoughts in the early definitional dialogues in the passages in which his
character presses the "What is F?" question in his search for
knowledge. Plato presents a history of these thoughts in the passage
in the Phaedo in which his character gives an autobiographical account
of his dispute with the natural scientists. 2 Plato takes Socrates's side
in this dispute, and Plato shows himself to conclude that the leading
assumptions in Socrates's method of intellectual inquiry are inconsistent
with a conception of reality in which the Forms are not substances.

ii. Socrates's "What is F?" question

The search for answers to the "What is F?" question is the most
prominent feature of the pre-Phaedo dialogues, and two aspects of this
search are striking. The first is Socrates's insistence on the "priority
of definition." The second has not attracted enough attention to garner
a name, but for reasons that shall soon be evident, I call it the "priority
of Form." With respect to matters involving justice, beauty, etc., the
priority of definition and the priority of Form constitute a central part
of Socrates's method of inquiry in his search for knowledge. 3
Every serious student of ancient philosophy knows that Socrates
seems to insist that he and his interlocutors must know the essence of
F-ness before they can answer the other questions that concern them.4
In the famous passage at the end of book I of the Republic, Socrates
seems to insist on this point with respect to justice:
 2.ii Socrates's "What is F?" question 27

I have not had a good banquet, but that was my fault, not
yours. I seemed to have behaved as gluttons do, snatching
at every dish that passes them and tasting it before they have
reasonably enjoyed the one before. So I. before finding the
answer to our first enquiry into the nature of justice, let that
go and turned to investigate whether it was vice and
ignorance or wisdom and virtue. Another argument came up
after, that injustice was more profitable than justice. and I
could not refrain from following this up and abandoning the
previous one so that the result of our discussion for me is
that I know nothing; for, when I do not know what justice
is, I shall hardly know whether it is a kind of virtue or not,
or whether the just man is unhappy or happy (354a13-c3).5

Socrates leaves his interlocutors to conclude that no one should try to

come to know whether justice is a virtue, or whether the just person is
happy, unless he first knows what justice is. Violating this injunction
is an intellectual vice akin to the moral vice of gluttony.
Although less well known, the Hippias Major' contains two
passages that provide equally good examples of Socrates's insistence on
the priority of definition. In this dialogue, Plato makes Socrates adopt
a humorous device to make his point. He makes him report the words
of a fictional questioner who plays the role of spoiler that Socrates
himself plays in other definitional dialogues:

Quite lately, my noble friend, when I was considering as

ugly some things in certain compositions, and praising others
as beautiful, somebody threw me into confusion by
interrogating me in a most offensive manner, rather to this
effect. You, Socrates, how do you know what things are
beautiful and what are ugly? Come now, tell me what
beauty is (286c5-d2).

I am called every kind of bad name by... that man who is

always cross-questioning me. He is a very close relative of
mine and lives in the same house, and when I go home and
he hears me utter these opinions he asks me whether I am
not ashamed of my audacity in talking about a beautiful way
of life, when questioning reveals that I do not even know
what beauty is (304dl-8).
28 Plato's tribute to Socrates

In these passages, the fictional questioner scolds Socrates for violating

the priority of definition. He says that Socrates should be "ashamed,"
that he is wrong to claim to know that certain things are beautiful when
questioning shows that he does not know what beauty is.
These passages from the Hippias Major and the Republic are not
the only passages in which Socrates insists on the priority of definition.
This principle informs the search for knowledge in nearly all the early
definitional dialogues.? Socrates relies on the priority of definition in
the search for knowledge as the reason for his relentless search for the
essences of beauty, justice, and the other such properties.
Once his interlocutors grant the priority of detinition, Socrates
presses the search for a definition in striking way. He takes great care
to focus attention on the "Form" that he supposes the "What is F?"
question is about, and he suggests that this Form somehow determines
the properties that make objects be F. He uses an instance of

(QI) What is that "by which" (instrumental dative)8

all F-things are F?

or a sentence relatively like it, e.g.,

What is that "because of which" (OLa) F-things are F?

What is the "cause" (atnop) of F-things being F?
or What is the "producer" (ll'OLOUP) of F-ness in F-things?

as a way of eliciting the preliminary response,

(Al) The F. (Or: The Form the F itself.~

After getting the preliminary (AI) response to the (QI) question,

Socrates asks the definitional question,

(Q2) What is the F?,

in hopes of getting an acceptable definitional answer. The Hippias

Major provides an especially good example of this practice:
 2.ii Socrates's "What is F?" question 29

He would say, Stranger from Elis, is it not by

(instrumental dative) justice that the just are just? Would
you answer, Hippias, as if he were asking the question?
I shall answer that it is by justice.
Then this, namely justice, is definitely something.
Certainly.
Again, it is by wisdom that the wise are wise, and by
goodness that all things are good?
U ndoubtedl y.
That is, by really existent things--one could scarcely say,
by things that have no real existence?
Quite so.
Then are not all beautiful things beautiful by beauty?
Yes, by beauty.
Which has real existence?
Yes, what else do you think?
Then tell me, stranger, he would say, what is this thing,
beauty (287c1-d3)?

In the guise of a fictional questioner, who no doubt is really Socrates

himself, Socrates insists that the Form the F itself makes F-things be
F. Once his interlocutor agrees, he uses his "What is F?" question to
press the search for a definition. 10
The priority of definition and the priority of Form are both
important for understanding why Plato came to believe that the Forms
are substances. Socrates may have accepted a strong and implausible
form of the priority of definition, but the definitional dialogues show
that he certainly accepted the weaker and much more plausible version
that if knowledge about matters involving F-ness is possible, then
definitional knowledge of F-ness must also be possible. Socrates
assumed that the Form the F itself determines what properties constitute
the essence of F-ness. When he asks the "What is F?" question, he
assumes that he is asking about the Form the F itself.
These assumptions are not beyond question, but they are reasonable
enough. With respect to matters involving aesthetic and moral
properties, the conditions necessary for the possibility of knowledge
probably do include a ground for the existence of formal truths. The
problem is not so much with the claim that there are formal truths and
that they have a foothold in reality. The problem is the ontological
30 Plato's tribute to Socrates

status of the Forms. Socrates invests them with an explanatory

function, but he is silent about how they exist. Like someone who
explains that certain properties make objects be beautiful because they
are part of the meaning of the predicate 'is beautiful,' Socrates explains
that certain properties make objects be beautiful because they are part
of the Form the beautiful itself. He does not say how the Form exists
in reality so that it can discharge this explanatory function. He just
says that Form the beautiful itself is that "by which" (instrumental
dative) beautiful things are beautiful.
This is an unstable combination. Socrates neither believed nor
denied that the Forms are substances, but the Hippias Major and other
early definitional dialogues clearly suggest that he invested the Forms
with an explanatory function that they cannot very well keep unless
they become substances. To prevent the explanation in terms of the
meaning of the predicate 'is beautiful' from giving way to a more
natural explanation in terms of sensible reality and the linguistic
practices of language users, meanings have to be relocated to a third
realm. If the Forms are to retain their explanatory roles, they too have
to be relocated to a third realm. They must become substances. l1

iii. A new method of inquiry

Plato recognized that Socrates's method constituted a major break from

previous practices, and although twentieth century philosophers do not
share Plato's enthusiasm, he applauded Socrates for breaking with
tradition and giving the philosopher a new method of inquiry. The
most important place where he helps his readers to understand the
significance of Socrates's break with tradition is in the autobiography
in the Phaedo. Plato makes his character give an account of his
intellectual history to showcase both the problems that Socrates pushed
to the center of philosophy and the reasons he may have had for
rejecting the way that his predecessors handled these problems.
The character explains that, although he now "in no way inclines
towards < this method of inquiry> " (97b6-7), as a young man he was
"remarkably keen on the kind of wisdom known as inquiry into nature"
(96a6-8). During this time in his youth, Socrates says that he accepted
explanations such as following:
 2.iii A new method of inquiry 31

< W > henever a large man standing beside a small one

appeared to be larger. <he is larger> by (instrumental
dative) a head; similarly with two horses (96d9-el).

< T > hat the ten was greater than the eight because (oL(i) of
the two in them, and that the two cubits were larger than the
one cubit because (o~a) of their exceeding the latter by half
(96e2-4).

When his interlocutor asks what he now believes about explanations

according to this method of inquiry, he responds that he rejects them:

I am far from supposing I know the explanation (d~Tia) for

any of these things, when I do not even accept from myself
that when you add one to one, it is either the one to which
the addition is made that comes to be two, or the one that
has been added and the one to which it has been added, that
have come to be two, because of the addition of one to the
other, because I wonder if, when they are away from each
other, this becomes a cause (d~ria) for their coming to be
two, namely to be placed together near each other. Nor if
you divided some one thing, am I still able to be persuaded
that this, in tum, the dividing, has become a cause (atria)
for their coming to be two. This is opposite to the previous
cause (aLTia) for becoming two. Then it was their being
brought close to each other and added, one to the other;
whereas now it is their being drawn apart, and separated
each from the other (96e6-97b3). < I reject> that method
of inquiry. Instead I rashly adopt a different method, a
jumble of my own, and in no way incline towards the other
(97b6-7).

He became disenchanted with his predecessors because, as the

definitional dialogues show, he wanted to know the essences of various
properties. If one presses the explanations of the natural scientists for
answers to such questions, he believes that unacceptable answers are
the result. These explanations provide answers such as

(1) Large things are large by heads.

32 Plato's tribute to Socrates

(2) Ten things are greater in number than eight

by possessing two things.
(3) Two things are two by being placed near
each other.

At some time in his youth, Socrates rejected these answers in his search
for knowledge and adopted a new "method" (p,f.(}ooo<;) to answer the
questions that prompt the request for definitions. This new method is
the same method that he follows in the early definitional dialogues:

< N > othing else makes (1I'OLew) it beautiful except that

beautiful itself, whether by its presence or communion or
whatever the manner and nature of the relation may be, as
I do not go so far as to affirm that, but only that it is by
(instrumental dative) the beautiful that all beautiful things are
beautiful (lO0d4-8).

You would beware of saying that when one is added to one,

the addition is the explanation (ai-ria) for their coming to be
two, or when one is divided, that division is the explanation.
You would shout loudly that you know no other way in
which each thing comes to be, except by participating in the
peculiar being of any given thing in which it does
participate. In these cases you own no other explanation
(atria) for their coming to be two, save participation in
twoness. Things which are going to be two must participate
in that, and whatever is going to be one must participate in
oneness (lOlb9-c7).

As he does in so many previous dialogues, Socrates insists on the (Al)

answer to the (Ql) question. In his search for knowledge, he insists on
this preliminary answer before he is willing to proceed to the (Q2)
question and the quest for a definition.
One reason he gives for rejecting the explanations of the natural
scientists is that they do not provide necessary and sufficient conditions
in response to the (Q2) question. He presses this objection in several
passages. In 96e6-b3, which I quoted, he gives example accounts in
which bringing two things close together and dividing a given object
into two objects are alleged to explain why two things are two. He
 2. iii A new method of inquiry 33

rejects these accounts as inadequate because he insists that "opposites"

cannot explain why something has a given property. This eliminates
accounts that do not provide necessary conditions. In 1OOe8-1 01 b2 and
101b4-7, he gives example accounts in which having a given feature is
alleged to explain "opposite" properties. He remarks:

< Y > ou would not allow anyone to say that one man was
larger than another by a head, and the smaller was smaller
by that same thing, but you would protest that you for your
part will say only that everything larger than something else
is larger by nothing but largeness, and because of this is
larger. because of the large, and that the smaller is smaller
by nothing but smallness, and because of this is small.
because of the small. You would be afraid, I imagine, of
meeting the following contradiction: if you say that someone
is larger and smaller by a head. then. first, the larger will be
larger and the smaller smaller by the same thing; and
secondly. the head, by which the larger man is larger, is
itself a small thing, and it is surely monstrous that anyone
should be large by something small.. ..

< Moreover, you would> be afraid to say that ten is greater

than eight by two, and that this is the explanation (air£a) for
its exceeding, rather than that it is by numerousness, and
because numerousness, or that two cubits are larger than one
cubit by half, rather than by largeness. Because, of course,
there would be the same fear.

Possessing a given pair of things, Socrates says, cannot explain why a

given group of things is greater in number than eight because no
property can explain "opposite" properties. This eliminates accounts
that do not provide sufficient conditions.
Socrates also gives another reason for rejecting the explanations
according to the method of inquiry of the natural scientists. This
reason is more serious, more interesting, and much more important for
Platonic scholarship than that the natural scientists provide accounts that
are open to counterexample. 12 In lOOc9-d8, which occurs just after
he and his interlocutors have agreed to accept only explanations in
terms of the Forms that exist "themselves according to themselves"
34 Plato's tribute to Socrates

(ctimx mO' aura), Socrates dismisses all explanations according to the

method of the natural scientists as outright non-starters. He makes this
point in no uncertain terms. Socrates announces that he no longer can
understand or recognize such explanations:

I no longer understand nor can I recognize those other wise

(CTocp6C;) explanations (a£rim), but if anyone says to me that
a given thing is beautiful because it has a blooming color, or
a shape, or something else like that, I dismiss those other
things--for they all trouble me--but in a plain, artless, and
possibly simpleminded way, I hold this close to myself:
nothing makes (7roLEW) it beautiful except that beautiful
itself... (lOOc9-d5), it is by (instrumental dative) the
beautiful that all beautiful things are beautiful (lOOd7-8).

Plato is not explicit about why Socrates rejects "those other wise
explanations" as non-starters in the search for knowledge, but his
reasons should not be difficult to grasp.
Plato believes that these explanations are unacceptable because they
are born in a methodology that is misguided with respect to the place
of the "What is F?" question in the search for knowledge. He rejects
explanations according to the natural scientists's method because he
believes that if one were to adopt it as the "philosophical," i.e.,
scientific, method of intellectual inquiry, then about matters involving
beauty, justice, etc., knowledge would be impossible. Such knowledge
would be impossible, he believes, because if explanations according to
their method of inquiry were to replace explanations in terms of the
definitional content of Forms, then one would have to accept a
conception of reality that makes this knowledge impossible. The
essences that were starting-points in the Platonic conception of reality
must either have explanations or not exist.
Plato does not say whether the scientists explain essences in terms
of more basic objects, but given Socrates's insistence in the
autobiography and earlier in the dialogue that thinking about the
Forms, as opposed to using the senses to look to nature, is the most
proper activity of the philosopher (9ge2-6, 65d-68b), he may believe
that they maintain that essences supervene on the reality of sensibles
and the linguistic practices of language users. This would trouble Plato
 2. iii A new method of inquiry 35

enough to make Socrates reject their method of inquiry as a non-starter

because it is part of a conception of reality that does not make essences
exist as substances and ontological starting-points. On this conception
of reality, no intelligible object exists as a substance. The only
substances are sensible objects, such as the portions of water, fire, etc.,
that are traditional in Ionian cosmology.
Alternatively, and perhaps more plausibly, instead of supposing that
the natural scientists have a view about how essences exist, Plato may
believe that they ignore the (Q 1) and (Q2) questions altogether. \3 He
may believe that the natural scientists make sensible objects be starting-
points in their conception of reality and are not interested in essences.
He may believe that they see no need to answer the definitional
question in the search for knowledge. This too would trouble Plato
enough to make Socrates reject explanations according to their method
of inquiry as outright non-starters. Plato believed that only Forms that
exist as substances, i.e., "themselves according to themselves" (cdlTCi
w.O' aU7a), can insure that there are definitional truths to be possible
objects of knowledge, and he firmly believed that unless there are such
truths, no one can know the answers to the other questions involving
beauty, justice, etc., that prompt the search for knowledge.
At the end of the autobiography, if the reader still has not
understood the Socratic revolution, Plato has Phaedo repeat the main
point near the end of the interlude after the autobiography, and before
he returns to the primary topic of the conversation: 14

As I recall, when these points had been granted to him, i.e.,

(epexegetical Kai) it was agreed that each of the Forms was
something and that by partaking in them the other things
took the names of the Forms, he next asked ... (102alO-b3).

Phaedo repeats a lesson that Socrates may have tried to impress on his
interlocutors on the last day of his life. 15 This is the very same lesson
that Plato is trying to impress on his readers. Socrates introduced a
new "method" (jJ.EOOOOC;) of inquiry, and he started a revolution among
the intellectuals of his day. He assumed, in his pursuit of truth, that
knowledge about matters involving justice. beauty, etc., presupposes
knowledge ofthe essences ofthese properties. Socrates turned his back
on the natural scientists, suggested that the essences necessary for
36 Plato's tribute to Socrates

knowledge have their source in Forms, and he left Plato to provide the
advertisement and to articulate the argument.
Socrates's autobiography contains the argument and is the
advertisement. In this passage, Plato's character supposes that
explanations in terms of the definitional content of Forms cannot be
eliminated in favor of explanations according to natural scientists's
method of inquiry. He supposes that this method is part of a
conception of reality that eliminates the possibility of definitional
knowledge. He supposes that this, in turn, eliminates the possibility of
knowledge with respect to the questions that prompt the inquiry. Plato
believed that knowledge is not possible unless essences are
starting-points in ontology, and the autobiography in the Phaedo is his
tribute to Socrates for turning philosophy to explanations in terms of
the definitional content of Forms and away from explanations according
to the method of the natural scientists.
Socrates's autobiography thus holds a distinguished place in the
ancient world that is importantly analogous to the place that Descartes's
Discourse on Method and Meditations hold in the modern world.
Descartes put forward a new method of intellectual inquiry, and he put
forward a new metaphysics to validate this method. Socrates and Plato
split this labor. Socrates put forward a new method of inquiry. He
insisted on the priority of definition and the priority of Form. Plato
put forward the metaphysics. He insisted that the Forms are substances
and ontological starting-points, i.e., that they exist in reality
"themselves according to themselves. "16

iv. Against Vlastos

In "Reasons and Causes in the Phaedo," Gregory Vlastos gives a now

classic and widely accepted interpretation Socrates's intellectual
autobiography. Vlastos insists that the importance of this passage can
"hardly be exaggerated," since "as much is to be learned from it about
Plato's metaphysics ... as from any other text of equal length in the
corpus" (p. 291). I agree with much of Vlastos's interpretation, and
am in full agreement with his claim that the autobiography passage is
extremely important for Platonic scholarship, but I deny that its main
lesson is the one that he claims for it.
 2.iv Against Vlastos 37

Because Socrates says that before he became interested in Forms,

he was interested in the kind of "wisdom" that contemporaries call
"natural science," but known in his day as "inquiry about nature,"
Vlastos tries to see Plato as anticipating Aristotle's Doctrine of the Four
Causes and as using a technique that has become part of the
stock-in-trade of analytic philosophers. He argues that Plato tries to
solve philosophical puzzles concerning cause and explanation by
exposing unapparent senses of the Greek noun airia. Vlastos writes:

All four puzzles in 96B8-E4 will yield to the same treatment

on the hypothesis that all of them crop up because in this
benighted phase of his philosophical evolution Socrates was
confusing physical aLrtm with logical ones ... (pp. 313-314).

I say "hypothesis," for certainly there is nothing whatever in

the wording of these six lines which states or implies that
their puzzles arise because physical factors are being
confused with logical ones. For this hypothesis I claim no
more than indirect verification from the context (n. 61 on p.
313).

Had he availed irimself, as Aristotle was to do, of the

expository device of philosophical lexicography, this
achievement would have been more perspicuous ....
However, we should not be put off by the fact that at no
point does he say in the style of his great pupil and critic,
"o:irio: has many different senses." There are other ways of
exhibiting distinctions, and one way of doing so is to use
them. This, I have argued, is what Plato does in our
passage ... (pp. 324-235).

Vlastos maintains that Plato realizes, and is trying to help his readers
to realize, that homonymy can be insidious and that the word aiTia is
homonymous in this way. Because in his youth Socrates did not notice
this fact, he fell into confusion. He failed to notice that the word atria
and its cognates are used for different kinds of causes, including logical
and physical causes,17 and that different kinds of causes are
appropriate for different kinds of explanations. IS By the time of the
Phaedo, on the last day of his life, Socrates has realized his mistake,
38 Plato's tribute to Socrates

and he explains this mistake to his interlocutors so that they need not
fall into the same confusion. On Vlastos's interpretation, Plato does in
a very rough way what Aristotle later does in a much more polished
and sophisticated way in the Metaphysics in his discussion of his
predecessors's views on cause and explanation.
Anyone in search of antecedents for Aristotle's Doctrine of Four
Causes might believe that the Phaedo is fertile ground, since Aristotle
refers it in his discussion of causes (Metaphysics 991b3), but Vlastos
has drastically overstated Plato's achievement. He is correct that
Plato's character is interested in what Aristotle might describe as
"formal explanation." He is also correct that this new interest is part
of a "conscious abandonment" of a previous "line of thought" and that
this turn away from the methodology of the natural scientists is "one of
the great turning-points in European natural philosophy" (p. 297).
Vlastos, however, is fundamentally wrong to try to see Plato as
anticipating the Doctrine of the Four Causes.
Aristotle is traditionally thought to be the first philosopher to
maintain that homonymy is not confined to situations where it is readily
detectable, as, e.g., in the case where 'light' is said of dark feathers,
but also may appear in contexts where its presence is not obvious, as,
perhaps, in the case where 'exist' is said of minds and bodies. Not
only does he seem to discuss homonymy in several places, but his most
characteristic views appear to involve claims of homonymy. The
Doctrine of the Four Causes is a good example. It is commonly
interpreted in part as the claim that the word aiTia is homonymous in
a way which escaped the notice of earlier philosophers.
On Vlastos's interpretation, Plato, not Aristotle, was the first to
insist that homonymy can be insidious, but I see no good reason to
accept this assertion. Nowhere in Socrates's autobiography, nor, for
that matter, in any passage in the corpus, does Plato say of any word
that it is homonymous in an insidious way. Vlastos claims only
"indirect verification from the context" for his "hypothesis." I do not
see that the "context" provides any verification, indirect or otherwise,
and even it were to provide some small amount of indirect verification,
this would not be enough evidence to support such a strong assertion.
It would not be enough to show that Plato anticipated Aristotle on the
ways in which words can be homonymous. Plato made important and
 2.iv Against Vlastos 39

lasting contributions to the philosophy of language, but this is not one

of them. On this point, I do not believe that there is sufficient evidence
to accept Vlastos's interpretation.
Nor do I agree with Vlastos that, by having Socrates reject the
explanations of the natural scientists, and insisting on explanations in
terms of the definitional content of Forms, Plato is trying to explain to
his readers that the word atria can be used for both physical and
logical explanations. If Plato were trying to help his readers grasp this
point, he certainly decided to proceed in an unhelpful way. Socrates
does not say that the accounts of the natural scientists are acceptable as
physical explanations, but not as logical explanations. He simply
rejects these explanations as inadequate.
In Vlastos's defense, one might try to argue that Plato intentionally
leaves his readers to conclude that the word ahia can be used for both
physical and logical explanations and that both kinds of explanation are
acceptable in their separate domains. The textual evidence, however,
is clearly against this interpretation. Both before and after the
autobiography, Plato has opportunities to explain that the word a'tria
is used for many kinds of cause and explanation, but he always fails to
capitalize on these opportunities. In the Hippias Major, e.g., instead
of drawing distinctions between kinds of causes, Socrates is twice
content to make Hippias agree simply that causes are producers:

And the producer is nothing other than the cause?

That is so, Socrates (296e8-9)

Look at it this way. The cause was seen to be the

producer, was it not?
Certainly (297a4-5).

Since the Hippias Major pre-dates the Phaedo, Vlastos can insist that
this passage does not undermine his interpretation. In the Philebus,
however, which post-dates the Phaedo, Plato again makes the very
same point:
40 Plato's tribute to Socrates

There is no difference except in name between a cause and

the nature producing, but the producer and the cause quite
properly would be called one?
Quite properly (26e6-8).

These remarks are not what one should expect if, as Vlastos claims,
Plato really believed that the word airta and its cognates are used for
many kinds of cause and explanation.
Furthermore, in the search for definitions, a search that in the
dialogues Plato takes up both before and after the Phaedo, only once
does he use the noun atria or the adjective ciiTLo~ in connection with
definition. This occurs in the Hippias Major at 29ge4. 19 If Vlastos
were correct, then, once again, this is not at all what one should
expect. If Plato really believed that the word aLria and its cognates
could be properly used in connection with the search for definitions,
one should expect to find him using these words in this way.
Since Plato does not intentionally use the words for both physical
and logical causes, does not make an effort to distinguish these uses of
the words, and does not say that the words have these uses, one cannot
very easily avoid the conclusion that Vlastos's interpretation is
incorrect. Plato simply does not show himself to believe that the word
atria can be used for multiple and non-competing kinds of explanation,
including both physical and logical explanation.
Not only is Vlastos's interpretation incorrect, but in trying to see
Plato as anticipating Aristotle's Doctrine of Four Causes, he has
obscured the development of Plato's thought. He has, in fact,
mischaracterized Plato's most important reason for discussing the
natural scientists. Socrates turned away from what Plato describes as
"the kind of wisdom known as inquiry into nature." Whether Socrates
simply ignored his predecessors or actually argued for his break with
tradition is difficult to determine, but in the autobiography in the
Phaedo, Plato makes his character supply an argument.
In the early definitional dialogues, as part of a new method of
intellectual inquiry, Socrates puts forward the suggestion that
knowledge has a structure. On matters involving beauty, justice, etc.,
he maintains that knowledge is not possible unless knowledge of the
essences of these properties is possible. He furthermore suggests that
these essences somehow have their source in Forms.
 2.iv Against Vlastos 41

In the autobiography in the Phaedo, the character defends this

suggestion by arguing against an ontology and conception of reality in
which essences are not starting-points. He does not try to show his
interlocutors that the word aLTia can be used for multiple kinds of
explanation and that each is acceptable in its place. Nor does he try to
show them that the word aiTia has multiple, but non-apparent senses.
Contrary to Vlastos' s interpretation, Plato does not try to make room
for explanations according to the method of inquiry of the natural
scientists. He maintains, rather, that explanations in terms of the
definitional content of Forms cannot be replaced by adopting
explanations according to the method of the natural scientists. Plato
supposes that such explanations are part of a conception of reality in
which essences do not exist as substances. He supposes that this, in
turn, eliminates the possibility of knowledge about matters involving
justice, beauty, and the other properties that interested Socrates and
figured in the affairs of fifth century Athens.
Vlastos's interpretation is mistaken. Socrates's autobiography is
not a poor attempt to express Aristotle's Doctrine of Four Causes.

v. Against Annas

Julia Annas also rejects Vlastos's assertion that in the Phaedo in

Socrates's autobiography Plato shows himself to anticipate Aristotle's
Doctrine of the Four Causes ([1982], p. 325). Indeed, according to
Annas, instead of anticipating the Doctrine of Four Causes, Plato is
actually badly confused in a way that Aristotle could have easily
avoided with his views on homonymy:

Plato sees himself as concerned throughout with a single

topic: airia or explanation. He is criticizing others'
attempts at it and producing an improved account of it
himself, assuming that there is a single "it" throughout. And
what he began from was certainly a concern with causal
explanation (p. 318).

The Phaedo passage is thus a classic case of what Aristotle

regards as confusion arising from failure to notice that a
philosophically important term is being used as though it had
42 Plato's tribute to Socrates

a single sense, whereas in fact it is crucially ambiguous ....

Plato has failed to see that he is confusedly treating very
different kinds of explanation. A grasp of Aristotle's point
in Physics 11.3 would have enabled Plato to transform his
confused discussion into an unconfused discussion of three
different kinds of explanation; but Plato shows no sign of
any such grasp (p. 325).

Annas maintains that Plato conflated the different senses of the word
ai.iia, or perhaps confuses the different kinds of aLTim,20 that
Aristotle later distinguishes.
Annas argues that Plato's conflation caused him to make a category
mistake. On her interpretation, he believed that causation is a transfer
of properties, that something, x, causes a thing, y, to be F by somehow
giving, or transferring, some F-ness to y:

Plato's demand on explanation is one which we no longer

find compelling; it applies convincingly only to a few simple
cases where the explanandum is a quality that can be
transmitted from one thing to another (p. 317).

Forms are explanations. then, not because Plato always

wants to find a use for Forms, but because only they do
justice to the intuitive demand that something's being F
cannot be explained by something which is the opposite of
F. but can only be explained by citing something which is F.
and indeed a better example of F than the thing being
explained ~p. 317; emphasis in original).

Annas supposes that Plato makes his character reject the accounts of the
natural scientists because their explanations are not consistent with the
so-called "intuitive demand" that causation is a transfer of properties.
He accepts accounts in terms of Forms because they are consistent with
this demand. Thus, if Annas's interpretation is correct, Plato took a
bad idea about cause and effect and applied it in formal explanation:

We are led to expect that he will < use the transference

idea> to offer an improved account of causal explanation,
 2. v Against Annas 43

but. .. what he offers is a very limited and idiosyncratic kind

of formal explanation (p. 318).

According to Annas, Socrates's autobiography is not a record of an

achievement. It is, rather, the record of an embarrassment. If Annas's
interpretation is correct, Plato's tribute contains little more than
confusion about explanation and causation.
I agree that Plato did not distinguish either different kinds of cause
and explanation or different senses of the word atria, but I deny that
he is confused in a way that Aristotle could clear up with his Doctrine
of the Four Causes. I also deny that Socrates turned away from the
natural scientists for the confused reasons that Annas claims. The
autobiography is a difficult text for modern readers, but Plato is not
guilty of the confusion that she claims. Plato's primary concern is to
help his readers realize that Socrates recognized that there is a question
which is first in the search for knowledge, and that answers to this
question do not figure in explanations according to the method of the
Ionian scientists. This hardly amounts to an interest in kinds of cause
and explanation, or in senses of the word atTia, but neither does it
force Plato to conflate causal and formal explanations.
Just what method of inquiry the natural scientists employ is not
very clear,2l but despite Annas's suggestion, not all their explanations
are clearly causal explanations. Aristotle, e.g., reports that Anaxagoras
and his followers gave explanations of the following sort:

Things, they say, appear different from one another and are
said to be one thing rather than another because there is a
numerical preponderance of that among the innumerable
constituents in the mixture, for nothing is purely and wholly
white or black or sweet, or bone or flesh, but that which it
contains most, this is taken to be the nature of the thing
(Physics 187b3-7).22

The nature of this Anaxagorean style of explanation is not evident, but

it does not look too much like a causal explanation. It appears to be
a logical explanation in terms of "constituents" for why things are "said
to be one thing rather than another."
44 Plato's tribute to Socrates

Furthermore, even if the natural scientists were trying to give

causal explanations, Plato need not be guilty of conflating causal and
formal explanations. He is trying to show his readers that Socrates
recognized that, as philosophers, i.e., as inquirers in the search of
knowledge, the scientists should be interested in his "What is F?"
question and that explanations according to their method of intellectual
inquiry are part of a conception of reality that does not contain an
objective ground for answers to this important question.
There need be no conflation in Plato's mind on this point. With
respect to matters involving the properties that Socrates asks about,
Plato accepted Socrates's "method" (/Lf(}OOOr; = /LETa + ooor;) of
inquiry. He thought that if, in the place of Socrates's method, the
natural scientists's method were accepted as the correct method of
intellectual inquiry, then one would have to accept a conception of
reality according to which knowledge about these matters would be
impossible because the respective definitional knowledge would be
impossible. The Phaedo contains Plato's tribute to Socrates for his part
in this revolution. Plato explains that Socrates's dogged insistence on
the priority of definitional knowledge constituted an important break
from the natural scientist's way of searching for knowledge. Socrates
decided to follow a new "way" (boor;) of inquiry "in the pursuit of"
(/LETa) knowledge, and Plato applauded him for his decision.
Although their interpretations manifest this problem in different
ways, because they fail to understand Plato's reasons for discussing
Socrates's break with the method of inquiry that the natural scientists
use in their search for knowledge, Vlastos and Annas obscure Plato's
reasons for concluding that the Forms are substances. Vlastos makes
Plato assert that explanations according to the method of the natural
scientists and according to the method of the philosophers are somehow
two non-competing kinds of explanation, but Plato actually makes no
such assertion. On the contrary, he makes his character argue for his
explanations in terms of the definitional content of Forms by asserting
that explanations according to the method of the natural scientists do
not provide knowledge about matters involving justice, beauty, and
other such properties. Annas seems to recognize that Plato's character
argues for explanations in terms of the Forms over explanations
according to the method of inquiry of the natural scientists, but she
mislocates the source of his complaint against these scientists. She
 2.v Against Annas 45

makes Plato assert that the accounts of the natural scientists are
inadequate formal explanations because they fail to conform with a
certain conception of the nature of what Aristotelians would call
"efficient causation." Again, however, Plato makes no such assertion.
His complaint against the method of inquiry of the natural scientists
does not really concern causation. Socrates's examples do suggest
constraints on acceptable explanations, but his interest in these
constraints does not stem from a concern with the nature of efficient
causation. It stems, rather, from a concern with the possibility of
knowledge about matters involving the properties that he asks about.
Plato believes that the natural scientists's method of inquiry is part of
a conception of reality that is inconsistent with the possibility of such
knowledge. In their interpretations of the Phaedo, neither Vlastos nor
Annas recognize this important point.

vi. The Forms exist in a third realm

The way that Plato developed Socrates's use of Forms has made trouble
for him among philosophers as different in outlook and separated in
time as Aristotle and Wittgenstein. Although Wittgenstein and the
twentieth century analytic philosophers are interested in theories of the
truth-conditions for certain statements, they insist that the only adequate
such theories are theories in terms of sensible reality and the linguistic
practices of language users. They insist, to make Michael Dumrnett
their spokesman, that "uses" determine senses and that senses determine
truth-conditions:

The solution offered by analytical philosophy is that what

constitutes the speaker's attaching a particular sense to a
word is his using the word--more exactly, using sentences
containing the word--in a particular way. In describing this
use, we can take no advantage of a prior grasp of that sense
by the speaker.... Furthermore, our account of his use of
the word will, by itself, be a sufficient explanation of his
grasp of its sense, that is, in non-Fregean terminology, of
his grasp of the concept ([1981b), p. 52; emphasis in
original).
46 Plato's tribute to Socrates

The first, most primitive, use of general terms is their

predicative use in sentences whose subject is a demonstrative
pronoun: and we may regard that part of their sense which
determines the truth-conditions of such sentences as
constituting their" criterion of application." It is only certain
general terms--those which are the most fundamental in our
language--which have such a use at all. For a general term
to have such a use. it is necessary that we should be able to
understand predication of it which does not relate to a
determinate object, conceived as picked out by the use of a
specific criterion of identity. Both 'smooth' and 'man' have
such a use: 'narrow-minded' does not. We can understand
what it is to predicate narrow-mindedness only when we are
capable of grasping that it is to be predicated only of men,
and therefore of knowing what it is to identify a man
([1981a], pp. 572-573).

Dummett and the analytic philosophers are united against the

proposition that senses are metaphysical objects that determine the
truth-conditions for statements. The analytic philosophers have made
this point repeatedly. Perhaps Quine, however, has given this point its
most well known statement:

One may admit that there are red houses. roses, and sunsets,
but deny except as a popular and misleading manner of
speaking, that they have anything in common. The words
'houses,' 'roses,' and 'sunsets' are true of sundry individual
entities which are red houses, red roses. red sunsets; but
there is not, in addition, any entity whatever, individual or
otherwise, which is named by the word 'redness.' nor, for
that matter, by the word 'househood, , 'rosehood,'
'sunsethood.' ... < One> is no better off, in point of real
explanatory power, for all the occult entities which he posits
under such names as 'redness' ([ 1948], p. 10).

I feel no reluctance toward refusing to admit meanings, for

I do not thereby deny that words and statements are
meaningful. ... I remain free to maintain that the fact that a
given linguistic utterance is meaningful (or significant, as I
prefer to say so as not to invite hypostasis of meanings as
 2.vi The Forms exist in a third realm 47

entities) is an ultimate and irreducible matter of fact; or, I

may undertake to analyze it in terms directly of what people
do in the presence of the linguistic utterance in question and
other utterances similar to it (p. 11; emphasis in original).

< T > he explanatory value of special and irreducible

intermediary entities called meanings is surely illusory (p.
12).

Other analytic philosophers, of course, had made the same point:

If the meaning of an expression is not an entity denoted by

it, but a style of operation performed with it, not a nominee
but a role, then it is not only repellent but positively
misleading to speak as if there existed a Third Realm whose
denizens are Meanings. We can distinguish this knight. as
a piece of ivory, from the part it or any proxy for it may
play in a game of chess; but the part it may play is not an
extra entity, made of some mysterious non-ivory. There is
not one box housing the ivory chessmen and another queerer
box housing their functions in chess games. Similarly we
can distinguish an expression as a set of syllables from its
employment. But its use or sense is not an additional
substance or subject of predication. It is not a non-physical,
non-mental object--but not because it is either a physical or
mental object but because it is not an object. As it is not an
object, it is not a denizen of a Platonic realm of objects
(Ryle [1957], p. 263).

The questions, 'What is length?,' 'What is meaning? " 'What

is the number one?,' etc., produce in us a mental cramp.
We feel that we can't point to anything in reply to them and
yet ought to point to something. We are up against one of
the great sources of philosophical bewilderment: we try to
find a substance for a substantive (Wittgenstein, Blue Book,
p. 1). The mistake we are liable to make could be expressed
thus: We are looking for the use of a sign, but we look for
it as though it were an object co-existing with the sign.
(One of the reasons for this mistake is again that we are
48 Plato's tribute to Socrates

looking for a 'thing corresponding to a substantive. ') (p. 5;

emphasis in original).

These philosophers suppose that making meanings be metaphysical

objects and insisting that these objects do explanatory work is a mistake
that leads to unnatural theories, and they assert that it is the mistake
that underlies Plato's Theory of Forms. For these analytic
philosophers, talk about "Platonic Forms," "Fregean senses,"
"Russellian propositions," etc., is acceptable only as long as one keeps
firmly in mind that it is a way of talking about what actually does
explain the truth-conditions for statements: the reality of sensibles and
the linguistic practices of the community of language usersY If these
philosophers are correct, Plato went wrong when he did not accept this
fact about Forms and came to believe, in Quine's words, that the
Forms are "special and irreducible" objects that are the essences of
justice, beauty, and the other properties that Socrates uses his "What
is F?" question to ask about. He went wrong when he concluded that
Forms exist as substances in a third realm.
Although he probably did not intend to make a serious claim about
the history of philosophy, Wittgenstein's remarks suggest an
explanation for why Plato came to believe that the Forms are
substances. The suggestion is that although Socrates was not wrong to
focus attention on issues concerning definitional knowledge, Plato made
a simple but disastrous mistlke when he accepted the metaphysical
beliefthat the Forms are substances. Because he failed to see that (the
Greek counterparts) of 'beauty,' 'justice,' etc., are really just
nominalized predicates, and not names of special objects, Plato fell to
the desire "to find a substance for a substantive. "24
To show that this interpretation is mistaken, one needs to find a
passage in which Plato argues that the Forms are substances by arguing
against an opponent who accepts a conception of reality in which
essences are not starting-points. If he does not just note that his
opponent fails see that (the Greek counterpart of) 'beauty,' 'justice,'
etc., are names, this should dampen the enthusiasm for any
interpretation that makes him come to believe that the Forms are
substances because he was on the misguided quest "to find a substance
for a substantive." Plato does not disappoint his supporters, as should
already be clear from his discussion of Socrates's turn away from the
 2.vi The Forms exist in a third realm 49

natural scientists. 25 Indeed, he argues for his belief in just about the
way one should expect. He tries to eliminate competing conceptions
of reality. Wittgenstein and other such commentators are unfair to
Plato, and they distort the history of philosophy. On essences and how
they tit into reality, these commentators are wrong to make Plato's
belief that the Forms are substances stem from a simpleminded
linguistic confusion. Plato is a better philosopher.
As I have emphasized, and now shall attempt to bring into sharper
focus, the autobiography is a record of a clash between two very
different conceptions of reality. The dispute, as Plato understands it,
is over how reality must be so that there can be knowledge. Socrates
convinces his interlocutors that the Platonic conception is the winner.
On this conception of reality, the Forms exist as substances.
Although they have not made the connection to Plato's attempt to
eliminate competing ontologies in which essences are not ontological
starting-points, scholars do commonly cite the use of the phrase aim)
KaO' aU7() and its cognates with respect to the Forms as evidence that
Plato believes that the Forms are substances. In his commentary on the
Phaedo, e.g., David Gallop puts the point as follows:

Plato writes of Forms, such as Equality, Largeness and

Health, as if they existed "alone by themselves," in splendid
isolation from the familiar world of equal sticks, large men
and healthy children. It is this "separation" of Fonus from
the sensible objects that distinguishes the fully-fledged
version of the Theory found in the Phaedo. Nowhere is the
contrast drawn more sharply than here. The soul-body
dualism already postulated (64e-65a) is one aspect of the
same dichotomy ([1975J, p. 94).

For stylistic reasons, perhaps, Gallop does not translate the phrase aim)
mO' auro at lOOb6 as 'alone by itself.' Instead, he uses 'itself by
itself.' Gallop does, however, use 'alone by itself to translate it in
connection with the Forms at 78d5-6. He also uses this translation at
64c6 in connection with the soul, and perhaps he helieves, reasonably
enough, that Plato uses the phrase with roughly a single meaning.
Gallop does not explicitly justify his translation of this phrase, so
I assume that he relies on the precedent set by John Burnet in his 1911
50 Plato's tribute to Socrates

edition of the Phaedo. This masterpiece set the stage for subsequent
scholarship.26 Burnet renders the phrase aim) mO' aim> at 64c6 as
'alone by itself,' and he asserts that "< t > he emphatic aVToc; often
acquires a shade of meaning which we can only render by 'alone'" (p.
30). As evidence for this assertion, Burnet maintains that Plato uses
the phrases P,(WT/P mO' aVT~p and aim) mO' aim> with the same
meaning at 67dl and 64c6. Burnet further maintains that 'alone' is the
correct translation of the word p.opoC; as it occurs at 67 d 1.
Burnet may be correct that " < t > he emphatic aVToc; often acquires
a shade of meaning which we can only render by 'alone,'" but 'alone'
is not the only acceptable translation of the word as it occurs in
technical contexts in the Platonic corpus. The word p.opoC; also occurs
in contexts in which 'only' is an acceptable translation, and this
rendering fits the context in the Phaedo and perhaps is even a better
translation than 'alone.' Plato uses the phrases p.fwT/p mO' aVT~p and
ailT~ mO' aVT~p in "Socrates's Defense" (63e8-6ge5). Socrates
explains to his interlocutors why the philosopher does not fear death.
He says that in death the philosopher is easily able to gain wisdom
because the soul reasons best when it is not bothered by the body:

< The soul> reasons best, presumably, whenever none of

these things bothers it, neither hearing nor sight nor pain,
nor any pleasure either, but whenever as much as possible
aim! mO' aV7~v it becomes, disregarding the body ...
(65c5-8).

Socrates does not intend for his interlocutors to conclude that the those
who wish to be true philosophers should try to exist without their
bodies. This would be to urge suicide, and he stands back from this
suggestion (61 b7-62c8). Socrates's point is that the soul of the true
philosopher tries to reason "only according to itself," not according to
information that comes from the senses of the body.27
This translation fits Plato's use of the phrase aV7() mO' ailTo with
respect to the Forms, and his use shows that he believes that the Forms
exist as substances. In connection with Forms, this phrase does not
occur in the early dialogues. It first appears in this context in the
middle dialogues. It occurs once in the Symposium at 211 b 1, but, as
Sir David Ross has noted, the words are not Socrates's, but "are put
 2.vi The Forms exist in a third realm 51

into the mouth of Diotima the wise woman of Mantinea, and their tone
is that of a prophet rather than of a philosopher" ([1953], p. 21). The
most important uses are in the Phaedo at 66a2, 78d5, and lOOb6.
The last use is most telling. It occurs in a passage that occurs just
before Socrates rejects the explanations according to the method of the
natural scientists. He prefaces his rejection with an affirmation of the
Platonic conception of reality. On this conception, the Forms exist
"themselves according to themselves." Socrates remarks:

I shall return to those much harped-on entities, and start

from them, hypothesizing that a beautiful aimj me' auro is
something, and so are a good, a large, and all the rest
(l00b4-7).

Socrates next asks Cebes to "grant him this < hypothesis> and agree
that those things exist" (lOOb7). Cebes, of course, satisfies this request
(lOOcl-2), and Socrates draws out the consequences. He first says that
the ordinary objects of experience are somehow posterior to the Forms,
and he then asks Cebes whether as an explanation of why F-things are
F, he accepts that the (A 1) answer is the correct response to the
preliminary (Q 1) question:

Then look what comes next to those things, and see if you
think as I do. I believe that if anything else is beautiful
besides the beautiful itself, it is beautiful because (oux) of
nothing other than that it participates in that beautiful, and
the same goes for all of them. Do you assent to an
explanation (ai,ria) of this kind (lOOc3-7)?

Cebes accepts this kind of explanation (lOOc8), but in so doing, he does

not just agree to accept explanations in terms of Forms that exist almx
Kete' aim!. He also agrees that no other kind of explanation is
acceptable. He agrees to reject every method of inquiry that is part of
a conception of reality that is inconsistent with the proposition that the
Forms exist as substances and ontological starting-points.
This is what one should expect. Cebes just accepted that "a
beautiful aU7() Kete' aina is something," that the beautiful itself only II

according to itself" contains what it is to be beautiful. He just agreed,

52 Plato's tribute to Socrates

to put the same point in a slightly different way, that a given thing is
beautiful "because of nothing other than it participates in that
beautiful." In case, however, Cebes fails to grasp the significance of
the proposition that he accepts, Socrates quickly reveals its power. In
lOOc9-d8, as I have already shown, he outright rejects explanations
according to the scientists's method of inquiry. Socrates announces that
he "no longer understand < s > nor. .. recognize < s > those other wise
explanations" (lOOd9-1O) according to this method.
The early definitional dialogues and the autobiography in the
Phaedo thus suggest that Socrates helped pave the road to Platonism
and the Theory of Forms. He turned his back on the method of his
predecessors, asked his "What is F?" question, made its answers
necessary for knowledge, and located the answers in Forms. In the
definitional dialogues, this stance manifests itself in the character's
insistence on the priority of definition and his insistence on the priority
of Form. In the autobiography, this character argues against his
predecessors. He argues that if knowledge about matters involving
certain properties is possible, then explanations in terms of the
definitional content of Forms cannot be eliminated in favor of
explanations according to the natural scientists's method of inquiry. He
insists that neither this attempt nor any other attempt to eliminate the
(A 1) answer to the (Q 1) question can provide for possibility of such
knowledge. Plato is firmly convinced that only a conception of reality
in which Forms exist as substances is consistent with the possibility of
knowledge. He believes that only if the Form the F itself exists "only
according to itself" do certain properties constitute the essence of
F-ness and thereby insure that there are the definitional truths that are
necessary for knowledge about matters involving F-ness.
Socrates and Plato were not the only intellectuals to react against
tht: lonians. Almost from the beginning of the naturalist tradition,28
the scientists themselves recognized that the speculations of their
Milesian forefathers must remain speCUlative unless certain
epistemological issues could be resolved. Much of the evidence has not
survived, or survived in only fragmentary form, but statements from
Xenophanes, Heraclitus, etc., show that the thinkers of the day took a
keen interest in the "methods" (/-Lf.(lOOOL = /-LETa + '000L) for acquiring
knowledge. 29 The Milesian cosmological speculations had disturbed
the unreflective confidence in the orthodox conception of reality and the
 2. vi The Forms exist in a third realm 53

place of humans in it. This helped to push questions about method to

the forefront of discussion. The race was on to identify the method
that would produce knowledge and to fit this method into a conception
of reality, for only in this way could confidence in a given claim about
the world and man's place in it be fully justified.
Parmenides's poem is an important milestone in this race. His
thought is opaque, to say the least, but his work remains .extremely
important to scholars because it constitutes a dramatic and seminal
turning-point in the early history of ancient philosophy. He attacked
the very foundations of the naturalist movement, and his attack helped
to open the way for Plato and his Theory of Forms. Plato does not
acknowledge his debt to Parmenides, at least not in Socrates's
autobiography, but there is an interesting parallel between the poem and
this passage from the Phaedo. Parmenides's poem begins with a
prologue and then opens into two parts, the way of "truth" (DK 28 B
1.29) and the way of "opinion" (30). In the prologue, Parmenides sets
off on the "road of the god, which bears the man who knows over all
cities" (2-3) to a place far "from the steps of men" (27).30 Once
there, the goddess distinguishes the two ways that are the subject of the
remaining two parts of the poem. She distinguishes the way of "truth"
from the way of "the opinions of mortals, in which there is no true
reliance" (30). She insists that the "way of truth" is the only way to
knowledge and that the "way of opinion" is somehow wrongheaded and
inconsistent with the possibility of knowledge. Furthermore, and most
importantly, she seems to identify the "way of opinion" with the way
of the Ionian natural scientists. This seems to be the import of her
remarks in the following passage:

They made up their minds to name two forms. of which they

need must not name so much as one--that is where they have
gone astray--and distinguished them as opposite in
appearance and assigned to them signs different one from the
other--to one the aitherial flame of fire. gentle and very
light, in every direction identical with itself. but not with the
other; and the other too is in itself just the opposite. dark
night, dense in appearance and heavy (B 8.53-59).
54 Plato's tribute to Socrates

This passage is obscure and difficult, but the goddess seems to suggest
that the Ionians have gone wrong in their attempt to make sensible
objects, such as portions fire, air, etc., be basic parts of reality.
Parmenides apparently believed that the scientists have fallen into a
fantasy and out of touch with objective reality. 31
In Socrates's autobiography, although he does not argue in the way
that the goddess argues, Plato does proceed in a strikingly similarly
fashion. He rejects the method of inquiry of the natural scientists. He
suggests that this method is part of a conception of reality that is
inconsistent with the possibility of knowledge. Like Parmenides's
mortals, the natural scientists make assumptions that put them out of
touch with reality and put knowledge beyond their reach. Plato insists
on a conception of reality in which essences exist, and he locates them
outside the human sphere and in the third realm of Forms. In no
uncertain terms, he maintains that the (A l) answer is the only answer
that is consistent with the possibility of knowledge. In parallel with
Parmenides's journey, the (Al) answer is the "road of the god, which
bears the man who knows over all cities."
In view of this parallel, no one should be surprised that Plato treats
Parmenides with more respect than he treats anyone but Socrates.
Plato continues the war against the natural scientists that Parmenides
had begun. Parmenides thought that there are two "ways" (000L) to go
"in the pursuit" (Il-ETa) of knowledge, his way and the way of the
natural scientists, but that only his way "attends upon truth" (B 2.2,
2.4), and that the way of the natural scientist is a dead end.
Par men ides thought that his way carried a certain conception of
objective reality, that this reality is somehow a "perfect one." In the
autobiography in the Phaedo, like Parmenides before him, Plato rejects
explanations according to the method of inquiry of the natural
scientists. Plato only accepts explanations in terms of the Forms
according to the method of inquiry that Socrates follows in the early
definitional dialogues. Plato thought that Socrates's method of inquiry
carried a certain conception of reality. Socrates's method presupposes
that essences exist, and other than making them exist as ontological
starting-points, he saw no other way to fit them into a conception of
reality that he believed could account for the possibility of knowledge.
Plato came to believe that Socrates's method of inquiry demanded an
 2.vi The Forms exist in a third realm 55

ontology in which essences exist "themselves according to themselves."

He concluded that the Forms must exist as substances.

vii. Forms and Recollection

Some scholars seem to suggest that Plato concluded that the Forms are
substances, not for the reasons that I have put forward, but because he
accepted the Theory of Recollection (AvaJlVTwt<;). Whether they
actually intend to make this suggestion is not completely clear, but
there is no doubt that their words can leave this impression. Their
words can leave the impression that the Theory of Recollection is a
premise in Plato's argument for the conclusion that the Forms are
substances. Myles Burnyeat, e.g., in criticizing Irwin's account of the
origin of the Theory of Forms, makes the following remarks:

So Plato's great metaphysical vision rests on nothing more

than an unclarity or confusion? I would find that the hardest
paradox of all. The truth is, there is an important dimension
of Plato's thinking about these issues which Irwin has left
out. ... I refer to Plato's provocative doctrine that, in morals
and mathematics and perhaps in other branches of knowledge
as weJl, learning is recoJlection, a recovery from within
oneself of knowledge enjoyed by the soul before it became
incarnate in a human body. ... My contention is that we
shall not understand the Theory of Forms unless we see it as
correlative to a cosmic vision of the soul's existence in
which our present embodiment is but one passing phase
([ 1979], pp. 59-60),'l

Burnyeat clearly rejects Irwin's interpretation, but his positive view

about the development of Plato's thought is less clear. A reader could
take him to assert that Plato concluded that the Forms are substances
because he accepted the Theory of Recollection. This sort of
suggestion is even stronger in some of F. M. Cornford' s writing on the
development of Plato's thought. Witness, e.g., his remarks in the
following passage:

When < Socrates> set out to define what Aristotle calls a

'universal,' such as the Beautiful, he must often have had
56 Plato's tribute to Socrates

occasion to draw the distinction, frequently pointed out in

the early dialogues, between the single character to be
defined and the many things which have that character. ..
([1939], p. 74). Socrates could draw that distinction, and
perhaps must have drawn it, without out going on to assert
that the Beautiful itself has a separate existence, independent
of the many things in which the character appears (p. 74).
Aristotle states quite definitely that the further step was taken
by Plato, who gave these characters an independent existence
and called them Forms (p. 74). The separation (XWpt0"l.l.Oc;)
of the Forms is explicitly effected in the Phaedo (p. 74). It
is entailed by the belief in Anamnesis. This is shown to
involve the separate existence of a conscious and knowing
soul, apart from the body and its senses, before birth--a
conclusion which all parties to the discussion take as
satisfactorily demonstrated, provided that the Forms exist.
If a disembodied soul can know all reality and truth, the
objects of its knowledge must exist apart from sensible
things, such knowledge cannot come to it through the senses
at all. Thus Anamnesis, the separate existence of the soul
before birth, and the separation of Forms from sensible
things, all stand or fall together. The whole of the first part
of the Phaedo is designed to lead the reader to this
conclusion (pp. 74-75).

Cornford first repeats what many accept, that Plato took a step beyond
the Socratic position on the Forms. He then makes a slightly more
controversial claim. He asserts that the Theory of Recollection
commits Plato to this further step. 33
Cornford is correct that Plato took a step beyond the Socratic
position on how the Forms exist, but he makes this point in an
unnecessarily misleading way. He employs the Aristotelian language
of "universal" (wOOf-OlJ) and "separation" (XWPWIl0C;) to state Plato's
conclusion. 34 Cornford says that "< t > he separation (XWPWIl0C;) of
the Forms is explicitly effected in the Phaedo," and although there is
a reading of the dialogue according to which Cornford is correct, his
choice of words is somewhat unfortunate. By the time of the Phaedo,
Plato has concluded that the Forms exist in a way that, later in the
Parmenides, the young Socrates describes as being "apart (xwpiC;)
 2. vii Forms and Recollection 57

themselves according to themselves (aimx KetO' aiJT(~)" (l29d7-8).35

In the Phaedo, however, Plato never uses the word xwpLuj.£fx; with
respect to the Forms. 36 Aristotle describes the development of Plato's
thought in terms of the "separation" of "universals," but this
description is not one that Plato claims for himself.
A much more serious problem in Cornford's remarks, and possibly
also in Burnyeat's remarks, is the strong suggestion that Plato reified
the Forms into metaphysical objects because he accepted the Theory of
Recollection. This interpretation of the development of Plato's thought
has no support. There is no support for it in the Platonic corpus. Nor
is there any support for this interpretation in the Aristotelian corpus.
In his discussion of the origin of the Theory of Forms, Aristotle never
mentions the Theory of Recollection.
This interpretation, furthermore, would reverse the natural order
of explanation. The Theory of Recollection is the theory that the
Socratic question-and-answer dialectic produces knowledge because the
soul is remembering what it knew before it entered the body and when
it was in "contact" (Phaedo 79d6) with the Forms. Plato accepts this
analysis because, among other things, he believes that the Forms are
substances. His belief in the Theory of Recollection is evidence that
he believed that Forms are substances, but it is not a premise in his
argument for the conclusion that the Forms are substances. Plato
concluded that the Forms are substances, not because he accepted the
Theory of Recollection, but because he believed that a conception of
reality in which the Forms exist "themselves according to themselves"
is the only way to provide for definitional truths to be possible objects
of knowledge. The Theory of Recollection is logically posterior to
Plato's conclusion that the Forms exist as substances. It explains why
the Socratic question-and-answer dialectic can result in knowledge of
the definitional content of Forms that exist in a third realm.

viii. Against Penner

In his book The Ascent from Nominalism, Terry Penner argues that
Plato uses the "argument from incorrigible conceptual states" to argue
that the Forms are substances (pp. 13-14,20). He claims to tind this
argument in a passage that occurs well before Socrates's intellectual
58 Plato's tribute to Socrates

autobiography. He maintains that this argument occurs in the Phaedo

at 74a9-c5. As he translates it, this passage reads as follows:

"We say, I suppose, that equal is something--speaking here

not of stick [equal] to stick or of stone to a stone or of any
other such things, but besides all these some other thing, the
equal itself. Do we say that it is something or nothing?"
"We do indeed say so, by Zeus," said Simmias, "assuredly
so. "
"And do we know what it itself is?"
"Yes," he said.
"Where then did we get knowledge of it from? Surely
from the things of which were just now speaking? Having
seen sticks, stones or other such equals, we came to have
this in mind from them, this being other than them.
--Or does it not appear other to you? Look at it in this
way too. Don't equal stones and sticks, the same ones,
sometimes appear equal to one [man) and not to another?"
"Yes, indeed."
"What then? Have the equals themselves ever appeared
unequal to you, or equality [ever appeared to be]
inequality? "
"Oh, never, Socrates."
"Therefore they are not the same [thing]," Socrates said,
"the equals and the equal itself. "

Penner admits that this passage appears to contain an argument that

presupposes the existence of the Form the equal itself, since it appears
to contain an argument for the conclusion that this Form is not identical
with any of the ordinary objects of experience that are equal,37 but he
insists that in this case appearances are deceiving. He maintains that
Plato is actually arguing for the Forms by arguing against "the position
we nowadays call 'nominalism' ... that the only things that exist are
concrete spatio-temporal objects such as people, sticks, stone~ and the
like" (pp. 21-22), and he maintains that Plato uses an identity statement
to formulate the nominalist position. Penner writes:

It seems to me as plain as anything can be that sometimes,

establishing a non-identity statement just is establishing an
 2. viii Against Penner 59

existence statement. To see why establishing the

non-identity < that> the equal itself 4= equal sticks, stones,
etc., just is establishing the existence of the Form the equal
itself, what you have to ask is: who is the enemy Socrates is
shooting down (p. 60; emphasis in original).

< I > t would not have occurred to Plato that it would be

necessary to argue against any species of anti-metaphysician
other than a nominalist. Constructivism and conceptualism
are, from a genetic point of view, fall-back anti-metaphysical
positions, once nominalism has been defeated by the
appearance of Platonism or realism (p. xi).

In fact in identities like these, what we're actually doing is

reducing talk of equality... to talk of equal sticks and
stones ... (p. 61; emphasis in original).

To say that the equal besides the sticks equal to sticks and
stones to stones is nothing would be to assert the thesis of
nominalism: to reduce any equal there may be to stick equal
to stick, stone to stone and any other such things. To say
it's something is say that nominalism is false (p. 62;
emphasis in original).

On Penner's interpretation, Socrates takes his assertions in 74b7-c3

about the ways that equal stones and the Form the equal itself can and
cannot appear to be premises in an argument from the incorrigibility of
certain conceptual states for the conclusion "that equal is something,
speaking not stick to stick or of stone to stone .... "
Although perhaps someone could argue as Penner claims, there is
not much evidence that Plato actually does argue this way in 74a9-c5.
To show that he is making an assertion about "incorrigible conceptual
states" is itself a non-trivial task, but an even more formidable task is
to show that Plato argues that the Forms are substances by arguing
against an identity statement. Penner must show that, when Socrates
asks whether the equal itself appears to him to be "other" (ETEPO/l,
74b7) than the many equals, he is really asking whether talk about
equality can fit into a nominalist ontology. Penner must show that,
when Socrates concludes that the equal itself is "not the same" (ou
60 Plato's tribute to Socrates

nXV70V, 74c4) as the many equals, he is really concluding that talk

about equality cannot fit into a nominalist ontology.
As evidence for his interpretation, Penner cites 476c2-7 in book V
of the Republic. This passage occurs in the context of Socrates's
discussion of the philosopher and his competitor, the lover of
spectacles. To determine who is fit to rule, Socrates makes these two
face a challenge. The challenge, according to Socrates, shows that
only the philosopher can know of a given "city" (7rOALC;) that it is just,
and the reason is that only the philosopher accepts an ontology that is
consistent with the possibility of knowledge about matters involving
justice. In the passage that Penner cites, which occurs before the
Socratic challenge, Plato is kicking a little mud on his opponent. He
stigmatizes him as a dreamer:

As for the man who believes in beautiful things but not in

the existence of beauty itself, nor is able to follow one who
leads him to the knowledge of it, do you not think that his
life is a dream rather than a reality? Consider: is this not
dreaming, namely, whether asleep or awake, to think that a
likeness is not a likeness but is that which it resembles?
I certainly think, Socrates, that the man who does this is
dreaming (476c2-8).

Penner maintains that Plato believes that the lover of spectacles is

"dreaming" because he accepts a certain identity statement:

< W > hat the lovers of sights and sounds say to themselves
in their so-called 'dream state' is obviously just this: the
beautiful itself is identical with the many beautiful sights and
sounds. But since... the lovers of sights and sounds don't
believe in the beautiful itself [476c2-31, ... what possible
interpretation can we give to what the lovers of sights and
sounds say to themselves in their dream--the beautiful itself
is not identical with the many beautiful sights and sounds--if
not precisely the nominalistic claims: that all there is to this
so-called 'beauty itself is just the many beautiful sights and
sounds (pp. 63-64; emphasis in original).
 2. viii Against Penner 61

Penner maintains that the passage from the Republic confirms his
interpretation that, in the Phaedo, Plato is arguing that the Forms are
substances by arguing against an identity statement (p. 64), but it
actually provides no such contirmation.
If Plato used an identity statement to express the position that he
associates with the lover of spectacles, then perhaps the passage from
the Republic would constitute evidence for Penner's interpretation.
Plato, however, does not use an identity statement in this passage. To
characterize the positions of the philosopher or the lover of spectacles,
he never uses (Greek counterparts ot) 'other' or ' same' in identity
statements in any passage in the discussion. Not only does he not use
an identity statement, when Plato says that the lover of spectacles
"think < s > that a likeness is not a likeness but is that which it
resembles," he need not be understood as attributing a belief in some
nominalistic identity. He may mean, and probably does mean, nothing
more than that the lover of spectacles denies that the ordinary sensible
objects of experience are "likenesses" of substances. To Plato's
dismay, the lover of spectacles supposes that these ordinary objects are
what is most real, that they are the substances. 38
Another problem for Penner is that to extract the "argument from
incorrigible conceptual states," the reader has to overlook a more
natural reading of the passage. On this reading, Socrates begins with
an existence claim, that the Form the equal itself exists. After his
interlocutor agrees to this existence claim, Socrates asserts that they
"know what it itself is." His interlocutor again agrees, and Socrates
turns the conversation to the events that help cause the incarnate soul
to have this knowledge. He suggests that it comes from seeing equal
things. He eventually explains that this knowledge comes from seeing
equal things because this makes the soul recollect the Form, but before
he gives this explanation, he pauses to ask whether the equal itself "is
other than" the ordinary objects of experience that are equal. Before
his interlocutor can respond, Socrates argues that the equal itself,
something that they agree exists, has a property that distinguishes it
from all the ordinary objects that are equal. In this way, to forestall
confusion on an issue that was new to the ancient world, Socrates
emphasizes that although the Form the equal itself and the many
ordinary objects that are equal have in common the (Greek counterpart
62 Plato's tribute to Socrates

ot) the word 'equal,' the many equals and the one Form the equal itself
are nevertheless objects that are very different in kind.
This interpretation is consistent with other parts of the Phaedo. At
74dff., e.g., Socrates again helps his interlocutors see that although the
Form and the equal stones and equal sticks, etc., have a name in
common, they are objects different in kind. He helps them see that the
many are not "equal in just the way that itself which is < is equal> "
(d6), but rather "fall somewhat short of that" (6-7), and are not quite
"such as to be as the equal < is equal> " (7).39 Furthermore, 74a9-c5
occurs in a passage in which Socrates is trying to prove that the soul
exists before it enters the body, and in the summary of this argument
at 76d7ff., neither he nor his interlocutors show any indication that
earlier they concluded that the Forms are substances. Socrates notes
that the soul exists pre-natally if, and only if, the Forms are substances,
and instead of reminding Socrates that they have already established
that the Forms are substances, as one might expect if Penner's
interpretation were correct, he says only that he recognizes the truth of
the Platonic conception of reality.
Although I am steadfastly in agreement with Penner's attempt to
give Plato an argument "that he could have had in mind" (p. 44),40
and I believe that his interpretation is intriguing, I do not believe that
there is sufficient evidence to conclude that Plato argues for his belief
in either the way or the place that Penner claims. He does not argue
that the Forms are substances and that nominalism is false by arguing
against an identity statement on the basis of the premise that some of
his conceptual states are incorrigible. He does not argue this way in
the Phaedo at 75a9-c5. Nor does he argue this way anywhere else in
the corpus. On these points, Penner is mistaken. Plato's argument
occurs in the place one might expect. It occurs in Socrates's
intellectual autobiography. Plato's argument is consistent both with his
place in history and with Socrates's insistence on the priority of
definition and the priority of Form in his search for knowledge. Plato
accepted explanations in terms of the definitional content of Forms
according to the leading assumptions in Socrates's method of
intellectual inquiry. He came to believe that these assumptions are part
of a conception of reality in which essences are starting-points, and he
came to believe that explanations according to a method that does not
incorporate these assumptions is part of a conception of reality that is
 2. viii Against Penner 63

inconsistent with the possibility of knowledge. The first explanations

to go down under this burden are the explanations according to the
method of inquiry of the natural scientists.
 3

PHILOSOPHERS AGAINST LOVERS OF SPECTACLES

i. Introductory remarks

In book V of the Republic, in his discussion of the philosopher and the

lover of spectacles, Plato again argues that the Forms are substances.
His argument runs parallel to his earlier argument against the natural
scientists. The philosopher accepts the ontology that Socrates defends
in the Phaedo. The lover of spectacles! opposes the philosopher. The
lover of spectacles accepts an ontology in which essences are not
starting-points. Instead of Forms that exist "themselves according to
themselves" «(nira mO' aura), he accepts "the many conventions
(vop-'p-a) of the many." Socrates argues against this ontology. He
argues that it is part of a conception of reality that is inconsistent with
the possibility of knowledge.

ii. The Socratic challenge

The argument against the lover of spectacles occurs as part of

Socrates's defense of one of the most stunning assertions in the corpus:
that "cities will have no respite from eviL .. unless philosophers rule as
kings" (473cll-d6). Glaucon voices disbelief, and to show that only
the philosopher is fit to rule, Socrates makes the philosopher and his
competitor, the lover of spectacles, face a challenge. Socrates throws
down this challenge to establish that only the philosopher accepts a
conception of reality according to which an inquirer can know of a
given 7ro'l\a; whether it is just. He intends to use this proposition as a
premise in an argument for the conclusion that only the philosopher is
fit to rule. Only the philosopher, Socrates argues, accepts a conception
of reality such that one "can keep looking back and forth, to justice,
beauty, temperance, and all such things as by nature exist, and can
compose human life with reference to thest ... " (501bl-4).
 3. ii The Socratic challenge 65

The challenge to the lover of spectacles is to show that his ontology

is part of a conception of reality that is consistent with the possibility
of knowledge about matters involving justice and the other properties
that the philosopher uses the "What is F?" question to ask about. The
lover of spectacles denies that the Forms are substances. This is part
of what it is to be a lover of spectacles (476a9-d4). Socrates again
emphasizes this difference between the philosopher and the lover of
spectacles, and then he presses his challenge. He presses the lover of
spectacles to answer certain questions:

I want a word with, and an answer from, the good fellow

who does not believe in a beautiful itself or any Form of
beauty itself, which remains eternally the same as itself, but
who believes in the many beautiful things, this lover of
spectacles who will not suffer anyone to say that the
beautiful is one, and so with the just and the rest
(47Se7-479a5).

The questions concern the many things that are beautiful, just, and
pious. In the case of beautiful things, Socrates asks whether the lover
of spectacles can avoid concluding that these things are also ugly. He
asks a similar question about just actions and pious actions:

My dear sir, we shall say to him, of all those many beautiful

things, is there one which will not also appear ugly? And is
there one of those just actions whiclt will not also appear
unjust? And of pious actions, is there one that will not
appear impious (479a5-S)?

Glaucon gives Socrates the answers that he expects. Glaucon responds,

without explanation, that the lover of spectacles cannot avoid these
conclusions:

No, for inevitably those will in a way appear both beautiful

and ugly, and so with the other things included in your
question (479b1-2).

This answer, of course, settles the challenge against the lover of

spectacles. If, with respect to ordinary objects that are F, the lover of
66 Philosophers against lovers of spectacles

spectacles must conclude they are not-F, then clearly he does not know
whether any such object is F. To emphasize the point, however,
Socrates continues his questioning:

What about the many things that are double? Do they

appear any less half than double?
Not one of them.
And big, small, light, and heavy, would we say that no
more will one of these be said than its opposite?
No, but always each will have both (479b3-8).

Socrates is now content to draw his conclusion:

And so is each of the many things more than it is not that

which one says it to be? This is like those double meanings
one is entertained with at banquets, such as the children's
riddle about the eunuch and his throw at the bat, the riddle
whose answer tells what he hit it with and what it was sitting
on. For these things too play both sides, and one cannot
deem clearly that they are or are not or that they are both or
neither (479b9-c5).

Socrates concludes that, in not recognizing that the truth of the Platonic
ontology, the lover of spectacles commits himself to a conception of
reality according to which no one can know how a given ordinary
object of experience stands with respect to F-ness. The philosopher can
look to the Form for the properties that determine F-ness. The lover
of spectacles must look elsewhere, and Socrates believes that there is
no place for him to look. He believes that, in not accepting a third
realm of Forms, the lover of spectacles is forced the absurd result that
every given ordinary object of experience is both F and not-F?
In the place of a third realm of Forms, the lover of spectacles
accepts "the many conventions (JloILLlLa) of the many." Plato rejects
this ontology as a way to ground essences in reality. This is evident
from Socrates's subsequent comments. After pressing his questions,
and receiving the incriminating answers, Socrates concludes that the
lover of spectacles is not a knower, but is an "opiner" or a "fancier. "3
These epistemological titles fit the lover of spectacles because his
 3.ii The Socratic challenge 67

conception of reality eliminates the ground for knowledge, and leaves

room for nothing more than "opinions" or "fancies" (OO~CXL):

We have found that the many conventions (VO/Lt/Lex) of the

many concerning the beautiful and the others are rolled
around between what is not and what really is.
We have.
But we previously agreed that if anything of that kind
appears, it would have to be called opinable, not knowable,
because it is that which is captured wandering in between
being and not being by the intermediate faculty.
We did.
Therefore, those who gaze upon the many beautifuls and
do not see the beautiful itself, and are not able to follow
another who would lead them to it, who see many just
actions but not justice itself, and so with everything, these
people, we shall say, opine everything and know none of the
things they opine.
Of necessity.
What about those who in each case contemplate the things
themselves which are always in every way the same? Do
these have knowledge, not opinion?
That too necessarily follows (479d3-e9).

The lover of spectacles chooses vop,Lp,a over essences that exist as

substances, and Socrates concludes that such an ontology is part of a
conception of reality that is inconsistent with the possibility of
knowledge. The challenge reveals, at least to Plato's satisfaction, that
the lover of spectacles is an opiner, not a knower.
The clash between the philosopher and the lover of spectacles is a
clash between two conceptions of reality. The philosopher accepts a
conception of reality according to which Forms exist as substances. He
insists that the Form the F itself exists "itself according to itself" (aim)
wf)' airro) and contains that which makes F-objects be F. In the
search for knowledge, the philosopher gives the (Al) answer to the
preliminary (Ql) question. The lover of spectacles accepts a very
different conception of reality. Unlike the philosopher, the lover of
spectacles gives
68 Philosophers against lovers of spectacles

(S) The many "conventions" (VOp,Lp,CX) about F-ness

as the answer to the (Ql) question. The lover of spectacles accepts a

conception of reality in which essences do not exist as substances, but
must find their place in nothing more than conventions. 4
Socrates argues that these different conceptions of reality have
different consequences for the possibility of knowledge about matters
involving justice. The challenge for the philosopher and the lover of
spectacles is to mark certain properties as the essences of justice so as
to provide for the possibility of knowledge about matters involving
justice. Because the philosopher accepts the Platonic view that essences
are starting-points, Socrates allows that he meets this challenge. The
lover of spectacles, however, is not so fortunate. Because he explains
essences in terms of conventions, Socrates insists that he fails to meet
this challenge. Socrates argues that this way of incorporating essences
into reality is inconsistent with political knowledge.
Socrates states his argument very quickly, but he makes the central
premises clear enough. In the early definitional dialogues, on matters
involving justice, beauty, etc., Socrates insists on the priority of
definition. He insists that knowledge is possible only if definitional
knowledge is possible. Plato accepted this principle. He further
believed that the (S) answer to the (Ql) question eliminates the
possibility of definitional knowledge. He believed that if the "many
conventions of the many" about F-ness were to replace the one Form
the F itself, then instead of the one objective definitional truth that had
its source in this Form, there would be many "opinions, or "fancies,"
that masquerade as definitional truths. If, instead of the one
definitional truth, there were these many ersatz definitional truths, each
specifying different properties as the essence of F-ness, then the
ordinary objects would be both F and not-F. Plato believed, and tries
to demonstrate, that with respect to a conception of reality in which
essences are not starting-points, knowledge is impossible.

iii. An argument from knowledge

Socrates's argument against the lover of spectacles is part of an

argument for the conclusion that the Forms are substances. 5 The
Socratic method of inquiry presupposes that political knowledge, e.g.,
 3. iii An argument from knowledge 69

is not possible unless justice has an essence. Plato thought that the
Form the just itself is this essence and that it exists as an ontological
starting-point. He reasoned that if there were only the "many
conventions" about justice, then in reality justice would have no
essence, and although language users could fancy themselves as
speaking truths about matters involving justice, in reality no such
statement would have a truth value, and political knowledge and
expertise would therefore be impossible. 6
This sort of argument informs many passages, including Socrates's
autobiography in the Phaedo, but perhaps no such passage is more
famous than the one at the end of the first part of the Parmenides. The
young Socrates has made a poor showing against Parmenides's
arguments against the Theory of Forms, but Parmenides concedes the
need for Forms:

If, Parmenides continued, in view of all these difficulties and

others like them, a man refuses to admit that Forms of
things exist or to distinguish a definite Form in every case,
he will have nothing on which to fix his thought, so long as
he will not allow that each thing has a character which is
always the same, and in doing so he will completely destroy
the power of discourse. But of that consequence, Socrates,
I think you are only too well aware (135b5-c3).7

Parmenides admits, and is "only too well aware," that anyone who
denies that the Forms are substances will thereby "destroy the power
of discourse." Neither character elaborates, but readers of the middle
dialogues should know what "power" they have in mind. Plato
believed that anyone who denies that the Forms exist as substances
must admit that about certain matters no one can use language as a
vehicle to assert truths.8 Plato's readers should know that he believed
that if essences must find their place among "the many conventions of
the many," then, in words from Aristophanes's Clouds, as Pheidippides
learns from the sophists and urges in argument against his father, what
passes for truth about matters on justice is nothing more than the
fancies of those who manage to "talk over the crowd" (1422).
Although his opponents are the natural "cientists, not the sophists,
Plato presses essentially the same argument in a well known passage in
70 Philosophers against lovers of spectacles

the Timaeus. 9 His character, Timaeus, asks whether Forms exist

"themselves according to themselves" or all the talk about Forms is
somehow nothing more than words:

Is there such a thing as Fire just in itself, or any of the other

things which we are always describing in such terms, as
things that exist themselves according to themselves (Olimi
KOIB' OIlm'x)? Or are the things we see, or otherwise perceive
by the bodily senses, the only things that have such reality,
and has nothing else apart from these any sort of being at
all? Are we talking idly whenever we say that there is such
a thing as an intelligible Form of anything? Is this nothing
more than a word (51b7-c5)?

Given the difficulty of the issues involved, and the importance of the
Theory of Forms, one might expect Plato to make Timaeus proceed
very patiently. Instead, Plato is content to have him settle the question
quickly in remarks that seem to presuppose familiarity with prior
dialogues:

If intelligence (vove;) and true opinion (06~0I a'A:qBf]e;) are

two different kinds, then these things--Forms that we cannot
perceive but only think of--certainly exist according to
themselves, but, if, as appears to some, true opinion in no
way differs from intelligence, then all things we perceive
through the bodily senses must be placed as the most
established (51d3-7).

Now we must affirm that they are two different things, for
they are distinct in origin and unlike in nature. The one is
produced in us by teaching, the other by persuasion; the one
is always by a true account, the other is without an account;
the one cannot be shaken by persuasion, the other can be
won over; and true belief, we must allow, is shared by all
mankind, intelligence only by the gods and a small number
of men (51el-6).

The central issue is whether "intelligence" and "true opinion" mark a

genuine distinction and whether it entails that the Forms exist as
substances. Plato is confident that this distinction is genuine and that
 3.iii An argument from knowledge 71

it does entail that the Forms exist as substances, but he does little to
help his readers to share his confidence. In earlier dialogues, however,
especially in the Gorgias, Plato supplies the information necessary to
appreciate the force of his reasoning.
The Gorgias is one of the dialogues in which Plato distinguishes
Socrates's philosophical practice from the practices of the sophists. As
in book I of the Republic, so also in the Gorgias, Socrates faces a
series of opponents and bests them all. His opponents are three
sophists: Gorgias, Polus, and Callicles. Each represents a position
within the sophistical movement. Gorgias is the oldest sophist. He
claims to be an "expert" in the "art" of "oratory" (449a3-5), and
Socrates presses him for an explanation of the nature of oratory.
Gorgias responds that "oratory is a producer of < the kind of>
persuasion that is a disposition for trusting, not a fitness for teaching"
(455a 1). He responds that "the orator is not fit for teaching ... , but is
fit only for persuading" (455a2-5). To illustrate the orator's ability,
Gorgias gives the following illuminating example:

Many a time I have gone with my brother or with other

doctors to call on some sick person who refuses to take his
medicine or allow the doctor to perform surgery or
cauterization on him, and when the doctor failed to persuade
him, I succeeded by means of no other art than oratory
(456bl-5).

In this example, Gorgias portrays the orator as "persuader" who is not

a "teacher." The orator can persuade the layman to accept a given
proposition, but does not use the force of reason. He persuades him
in some way other than by teaching him the justification that the expert
has for believing that the proposition is true.
Socrates agrees with Gorgias' s characterization of oratory, but he
does not conclude that it is necessarily an evil practice. He explains
that oratory, as Gorgias describes it, is a set of devices that aim to
provide the confidence that can make belief appear as knowledge, but
it does not provide the justification that makes true belief be knowledge
(458e6-459c2). For disciplines, such as medicine, with respect to
which recognized experts exist, Socrates allows Gorgias's suggestion
that the orator provides a community service by convincing lay persons
72 Philosophers against lovers of spectacles

to accept certain propositions. For other disciplines, however, Socrates

suggests that the effect of the orator is less salutary. In the case of
politics, e.g., a discipline that he believes has few real experts, the
orator actually appears to the many to be the expert and the teacher.
The orator may even appear to himself to be the expert and the teacher
about matters involving justice. Perhaps because Socrates is there to
refute him, Gorgias does not show himself to suppose that he is such
an expert. He does not go so far as to claim "to know what is just"
(461b6), but Plato firmly plants the suspicion that Gorgias is not always
so reticent and that he and the sophists sometimes mistake the power
of the spoken word for the force of reason.
The Meno reinforces this suspicion. In this dialogue, which in the
traditional ordering is later than the Gorgias, one of Gorgias's former
students assumes that he does have such knowledge. The student is
Meno. He assumes that Gorgias 'knows what virtue is, and he assumes
that when Gorgias was in Athens, he told Socrates what virtue is
(71 b9-d2, 73c6-8, 76a9-b 1). Socrates recalls meeting with Gorgias, but
he jokes that he cannot remember whether Gorgias told him what virtue
is,iO and he asks Meno to "remind" (cXJleXILJl1/U'C;) him what Gorgias
said (71cIO). Meno cannot remember, and after much discussion, and
in one of the most celebrated passages in the corpus, Socrates notes that
"true opinions ... run away ... until someone ties them down by
recollection." The passage reads:

True opinions are a very fine thing as long as they stay in

their place ... , but they are not willing to stay in their place
for a long time, but run away out of a man's soul, so they
are not worth very much, until someone ties them down by
working out the explanation (airia). This, my friend,
Meno, is recollection, as we agreed in what we said before.
When they are tied down, they first of all become pieces of
knowledge, and then permanent; and it is for this reason that
knowledge is more valuable than right opinion, and it is by
being tied down that knowledge differs from right opinion
(97e6-98a8).

Socrates leaves Meno to conclude that Gorgias hoodwinked him. He

leaves him to conclude that Gorgias provided him with a belief and the
 3.iii An argument from knowledge 73

confidence that makes this belief, if true, appear as knowledge, but he

did not actually give him knowledge.
Where Socrates sees trickery and ersatz justification, Polus and the
younger generation of sophists see nothing of the sort. They insist that
justification is nothing but the power of the spoken word. This is clear
in the second part of the Gorgias. After Polus takes the place of his
defeated master, Socrates again focuses the discussion on the nature of
oratory. Polus turns the question back to Socrates, who is only too
happy to answer it. He replies that oratory is "flattery" (464e2), that
it is an "idol" (f.'iOWAOII) of the part of politics that is concerned with
justice, as opposed to legislation (463dl-2, 465c1-3). The orator, he
says, "lacks intelligence (IIOVC;)" (466elO), and he challenges Polus to
"refute" him and "prove that orators do have intelligence, and that
oratory is ... not flattery" (466e13-467al).
Plato believes that oratory appears like the part of politics
concerned with justice, but it actually falls short of this discipline. He
allows the orator a place in society, but insists that it is strictly
subservient to those who have knowledge, such as doctors and
politicians. Gorgias stands on the border of his place, but Polus
transgresses his station, and later when Callicles urges Socrates to give
up philosophy (484c4-485e2) and to engage in practices that "give a
reputation for wisdom" (486c5-6), he is way out of bounds.
Plato thought that Polus and Callicles constitute a dangerous threat
because they do not recognize their rightful place. They pass
themselves off as having political knowledge, but they actually possess
nothing more than the power of the spoken word. Unlike those who
really have political knowledge, the sophists do not possess the force
of reason. 11 Polus tries to claim that orators know "what is just, what
is beautiful, and what is good" (461b6), but Plato leaves his readers to
conclude, and no doubt intends them to conclude, that Polus is badly
mistaken. The orator might acquire a true belief about justice, and he
might instill both this belief and the confidence that makes it appear as
knowledge, but in reality this belief is not knowledge. Plato describes
it is an "idol" of knowledge. It has no "account" (AO"y'OC;, 465a3) and
is held "irrationally" (aAO"y'wc;, 501 a6). 12 In Socrates's words from
the Meno, it is not "tied down" to the Form.
74 Philosophers against lovers of spectacles

The Gorgias thus suggests that the distinction in the Timaeus

between "true opinion" and "intelligence" depends on Forms that exist
"themselves according to themselves" because it is a distinction
between true belief and a certain conception of knowledge. With
respect to justice, beauty, etc., Plato believed that knowledge is
possible only if in a third realm there are Forms for inquirers to
recollect. Recollection retrieves the definitional truths that he believes
are a necessary part of the justification that is necessary for such
knowledge, and recollection is possible only for the philosopher. Only
the philosopher, according to Plato, can have pove;.
When Polus insists that Gorgias is wrong, and that orators do know
"what is just, what is beautiful, and what is good," he not only denies
the (AI) answer to the (QI) question, but he denies that the Forms are
substances, and thereby collapses the distinction between true belief and
intelligence. Plato understood Polus and the younger sophists to
advocate a conception of reality according to which there is no ground
for a distinction between true belief and knowledge. On this
conception of reality, justice is a matter of taste. It is subjective. It is
something that is not subject to error. What passes for truth is merely
the subjective fancy of those who "talk over the crowd."
Plato insisted that in the hands of Polus and the younger sophists,
oratory becomes "flattery" and an "idol" of the part of politics that is
concerned with justice. Plato thought that this flattery is the tool that
the lover of spectacles would use to rule the 7rOALe;. His ontology
constrains him to hold that political expertise is nothing more than such
flattery. He must rule according to "fancies," which, according to
Plato, are all that is possible once one rejects Forms that exist
"themselves according to themselves" and thereby collapses the
distinction between knowledge and true belief. Without their
metaphysical supports, all that remains is fancy.
Plato believed that only those who suppose that the Forms are
substances, the philosophers, not the lovers of spectacles, can
distinguish knowledge and true belief. Those who deny that the Forms
are substances must believe, as Plato says in the Timaeus, that "true
opinion in no way differs from intelligence." They have no choice,
according to Plato, because they eliminate the only way to ground this
distinction in reality. They are not philosophers. They are not lovers
of wisdom. They are nothing more than opiners and flatters. l3
 3. iv The lover of spectacles 75

iv. The lover of spectacles

In book V of the Republic, Plato matches his position on the

truth-conditions for statements that predicate beauty, justice, etc.,
against the position of the lover of spectacles. He believes that the
lovers of spectacles somehow make these truth-conditions relative to
"conventions" and either intentionally or unknowingly commit
themselves to a conception of reality that eliminates the possibility of
knowledge and expertise. Plato thus characterizes the lovers of
spectacles, not as philosophers, but as "lovers of opinions"
(<I)LAOOO~aL):

Shall we then say that these love and welcome that with
respect to which knowledge is possible, as the others do that
with respect to which only opinion (06~o:) is possible? Do
we not remember that we said that these latter loved and
contemplated beautiful sounds and colors and such things,
but they would not allow that beauty itself was something?
We remember.
We shall then not be out of line if we call them lovers of
opinion rather than lovers of wisdom. Will they be very
angry with us for calling them that?
Not if they take my advice, for it is not right to be angry
with those who speak the truth (47gelO-480alO).

Later in the Republic, just as the reader should expect, the lovers of
opinions appear as sophists. They are the real target. The many
thought that the sophists were better suited to ruling than the
philosophers, and many of the sophists agreed.
Plato, of course, had a very different idea. He denied that the
sophists were fit to rule because he believed that their philosophical
commitments eliminated the possibility of political knowledge. He
writes:

Not one of those private paid teachers, whom the people call
sophists and consider to be their rivals, teaches anything
different from the confident opinions ({){rY/J.o:m) which the
many opine (oo~6rw) whenever they are gathered together,
and this he calls wisdom. It is as if its keeper studied the
76 Philosophers against lovers of spectacles

moods and desires of a mighty and powerful beast, how to

approach it, how to handle it, how it becomes most difficult
and most gentle and what makes it so, what sounds it utters
on each occasion, and what sounds addressed to it soothe or
anger it. Having learned all this by tending the brute over
a period of time he calls this wisdom, gathers this
information together as if it were an art (T€XII1/), and turns
to teaching (ot/jcxcfK(:xAicx). The truth is that he knows
nothing about these beliefs and desires, as to which of them
is beautiful or ugly, good or bad, just or unjust, but he gives
them these names in accordance with the brute's fancies
(06~CXL). What it enjoys he calls good, what angers it he
calls bad, and he can give no other account (M'Yo~) of them.
What he must do, he calls just and beautiful, but he has
never observed, nor can he show to another, the nature of
the necessary or the beautiful or the difference between them
(493a6-c6).

Plato associated the sophistical movement with a metaphysical and

epistemological position that he thought was false and dangerous. The
sophists, as he understood them, committed themselves to the
philosophical position that the truth-conditions for statements that
predicate justice, beauty, etc., rest on Ofryp-Cl.7C1..
This position is some kind of anti-realism about the truth-conditions
for statements that predicate the properties that Socrates uses his "What
is F?" question to ask about. The evidence does not determine the
exact kind of anti-realism, perhaps because the sophists were not very
self-conscious about their philosophical commitments, or perhaps
because Plato was not too interested in setting out their position in
detail, but certainly the main contrast between the philosopher and the
lover of spectacles is clear enough. The lover of spectacles maintains
that the essence of justice, beauty, etc., are not ontological
starting-points. He maintains that these essences do not have their
source in a third realm of Forms, but rather must find their place
among nothing more than "the many conventions of the many. "
Given his assertion that Plato uses the "argument from incorrigible
conceptual states" to ascend from nominalism, Penner believes that the
lover of spectacles is some kind of anti-realist, but he also believes that
the text and Plato's position in history determine the exact kind of
 3. iv The lover of spectacles 77

anti-realism. Penner, recall, insists that Plato would not have thought
to argue against any opponent other than the nominalist. If Penner is
correct, then Plato's use of the word VOjlLIJ.CX at 479d4 should be
understood in a way that is consistent with nominalism. As Penner
notes, unless Plato uses this word to refer to "particulars" the lover of
spectacles cannot be a nominalist:

For the many conventions (VOJ1.LJ1.CX) referred to here as what

the many identify beauty with, have seemed to some scholars
to refer not to particulars but rather to types or kinds. On
such a reading, the many (whom we see from the context to
be the lovers of sights and sounds) will not be nominalists.
The many will be identifying beauty with the many different
characteristics conventionally thought to be identifiable with
beauty, for example, the color turquoise, certain
characteristics of shape and so forth (pp. 235-236).

Penner supplies the nominalist reading as follows:

< The "conventions" (VOJ1.LJ1.CX) at 479d4 are> the many

things (plural) thought to be what beauty is (singular). And
the identity between < a> single entity and many entities is
of course just that which the nominalist requires" (p. 236;
emphasis in original).

On his interpretation, the lover of spectacles is the sort of anti-realist

who explains away all apparent references to conventions as a disguised
way of talking about the mental processes of individuals. This is
consistent with Penner's conjecture about the opponents Plato would
have had in mind, and it is consistent with his claim that Plato uses an
"argument from incorrigible conceptual states" to argue for the Forms
by arguing against "the position we nowadays call 'nominalism' ... that
the only things that exist are concrete spatio-temporal objects such as
people, sticks, stones and the like. "
I agree that depending on how the lover of spectacles construes talk
about "conventions," as well as talk about "fancies" and "opinions," he
could be construed as either a nominalist or a conceptualist,14 but that
is where my agreement ends. I deny that Plato ever uses an argument
from incorrigible conceptual states to show that the Forms are
78 Philosophers against lovers of spectacles

substances, and I reject Penner's basic premise that Plato is trying to

ascend from nominalism. I do not, therefore, have to press the
Republic for the lover of spectacles's views on ontological commitment.
Even if it contained an answer, 15 to focus on this question would turn
one's attention away from a more important question.
In Socrates's autobiography in the Phaedo, Plato rejects
explanations according to the method of the natural scientists, not
because they are nominalist explanations, but because they are part of
a conception of reality that is inconsistent with the possibility of
knowledge about the matters in question. Although his reasoning is not
very explicit, Plato seems maintain that any ontology that takes sensible
objects as its starting-points can provide no foothold for the existence
of essences. The reader can ask exactly what ontological status the
natural scientists claimed for essences, but to understand the genesis of
the Theory of Forms, the more relevant question is why does Plato
believe that the explanations that fall out of the natural scientists's
method of inquiry do not provide knowledge.
The same is true with respect to explanations according to the
method of the lovers of spectacles. In book V of the Republic, Plato
rejects explanations according to this method of intellectual inquiry, not
because they are nominalist explanations, but because they do not
provide knowledge. Whether the lover of spectacles is a nominalist or
a conceptualist about 1I0J,LLlla is an interesting question, and Penner may
have the correct answer, but to understand why Plato came to believe
that the Forms are substances, the more relevant question is why he
believed that these 1I0J,LLJ,LCX are insufficient to provide for anything more
than subjective fancies and opinions. The answer to this question
cannot simply be that the lover of spectacles is a nominalist and that
Plato rejected nominalism any more than it can be that he is a
conceptualist and that Plato rejected conceptualism.
The evidence in the Republic is not easy to evaluate, but it does
suggests a plausible answer. It suggests that Plato thought that 1I0J,LLJ,LCX
can provide for nothing more than fancies because no one of the many
1I0J,LLJ,LCX about F-ness can rise above the fray to be the one "by which"
F-things are F. Plato thought that, since "the many conventions of the
many" are really nothing more than the many claims about justice,
beauty, etc., that the many shout when they are together in a mob, and
 3.iv The lover of spectacles 79

since the many are always shouting contradictory things, there are
always equally qualified but incompatible conventions.
If Plato had this belief about "the many conventions of the many, "
it would explain why he believed that the lover of spectacles cannot
know whether a given 7rOALC; is just. The lover of spectacles denies
that essences exist as substances and starting-points. He tries to
understand claims about justice, beauty, etc., in terms of the many
conventions of the many. He supposes that these conventions
determine the truth-conditions for statements that predicate justice.
Since, however, equally qualified and incompatible conventions are in
force, there is equal justification for contradictory statements purporting
to give the essence of justice. To the lover of spectacles, in virtue of
his ontology, the 7rOALC; must appear both just and unjust. This rules
out the possibility of political knowledge and expertise.
Not only does this conjecture fit the textual evidence, it makes
good sense of Plato's reaction to Socrates's complaint against the
sophists. The older generation of sophists tended to confuse the power
of the spoken word with the force of reason. Their successors went
further. They tried to obliterate the distinction. Socrates took steps to
set the situation right. He introduced his question-and-answer method
of discussion to check the power of rhetoric and make persuasion come
through the force of reason. He insisted that knowledge has a
structure, that he and his interlocutors must first answer the "What is
F?" question before they can answer the other questions that interest
them. Plato understood this clash between Socrates and the sophists as
a clash between competing conceptions of reality. He understood the
sophists, especially the sophists in the younger generation, to advocate
a conception of reality according to which claims about justice, beauty,
etc., depend on little more than the fancies of the many. Plato believed
that this conception of reality undermines the claim of politics to be a
"science" (€7rLC1T~P.T/). Without the realm of Forms, Plato thought there
could be nothing more than fancies and opinions.
This conjecture also has the virtue of making the Theory of Forms
be a response to one of the great unsolved problems of Plato's day.
Two groups of intellectuals made this problem salient: the Ionian
natural scientists and the sophists. The early Ionian cosmological
speculations, with their novel conceptions of objective reality,
undermined the unreflective confidence in the traditional methods for
80 Philosophers against lovers of spectacles

acquiring knowledge. This, in turn, provoked interest in these methods

and into how they fit in a general conception of reality. The sophists
added fuel to this fire. The older sophists exhibited some confusion
over the nature of the justification necessary for knowledge. They
tended to confuse the power of the spoken word with the force of
reason. The younger sophists changed this confusion into vehement
ignorance. They rejected the distinction altogether.
In this historical context, Socrates searched for definitions in his
pursuit for knowledge. He thought that having the justification
necessary for knowledge is not the same as having the rhetorical tricks
and devices that the sophists use to persuade. Socrates thought that
knowledge about matters involving justice, beauty, etc., presupposed
knowledge of the essences of these properties. He believed that, on
such matters, knowledge has a structure, e.g., that if there is empirical
knowledge about how a given 7rOA£(; holds with respect to justice, there
also must be formal knowledge of the essence of justice.
Plato accepted these assumptions in Socrates's method and sought
to validate them by fitting them into a conception of reality. He
thought that the Forms must exist as substances because otherwise there
would be no ground in reality for the definitional truths that are
necessary for the possibility of knowledge. He rejected any suggestion
that these definitional truths are no more objective than "the many
conventions of the many." Plato was convinced that if these truths
were subjective in this way, then the younger sophists wrongly turn out
to be correct. On this conception of reality, as Plato understood it,
political knowledge and expertise are impossible. Plato thought, as he
says in the Timaeus, that if the Forms are not substances, the
distinction between true belief and knowledge would collapse and there
would be nothing more than the fancies of individuals.

v. Against Owen

In his seminal and now classic 1957 paper" A Proof in the IIEpL IOEw/I, "
G. E. L. Owen maintains that Plato used an "argument from
compresence" to show that for certain predicates there are Forms that
are substances. His interpretation has stood prominently in the field of
Platonic studies,16 although even some of its supporters have worried
that it uncharitably gives Plato beliefs that seem implausible and
 3. v Against Owen 81

eccentric. l7 If Owen is correct, Plato came to believe that the Form

the F itself is a sample of F-ness and a standard for F-ness.
Owen maintains, first of all, that Plato believed that the predicates
that occur in Socrates's "What is F?" question are "incomplete" and
that only for these predicates do Forms exist. Owen writes:

In Republic VII (523a-525a) numbers are classed with such

characteristics as light and heavy, large and small, on the
score that our senses can never discover any of them KCXO'
cxvT6, in isolation (525dlO): in perceptible things they are
inseparable from their opposites (p. 108; emphasis in
original).

We may say, for convenience, that 'one' as we ordinarily

apply it to things is an incomplete predicate and that,
accordingly as we complete it in this way or that, it will be
true or false of the thing to which it is applied. Now the
same is true, or Plato talks as if it is true, of all these
predicates which in the Republic and earlier works supply
him with his stock examples of Ideas; and conspicuously so
of the logical-mathematical and moral-aesthetic predicates for
which the young Socrates unhesitatingly postulates Forms in
Parmenides l30b-d. In this world what is large or equal,
beautiful or good, right or pious, is so in some respect or
relation and will always show a contradictory face in some
other respect (Rep. 479a-b). As large is mixed with small
(Rep., 524c), so just and unjust, good and bad ... (p. 108;
emphasis in original).

Plato's treatment of these incomplete predicates makes no

essential use of the idea of physical mutability, often though
that idea recurs in characterizing the Forms. Here, it is with
the compresence and not the succession of opposites that he
is expressly concerned (p. 108).18

With these predicates Plato contrasts others which 'finger'

is an example. A finger can be seen KCXO' cxim): sight never
reports it to be at the same time not a finger (Rep., 523d).
This predicate::, then, breeds no contradictions that have to be
resolved by specifying 7rpor:; Ti. And the same is evidently
82 Philosophers against lovers of spectacles

true of 'man,' and of 'fire' and 'mud': all those predicates

for which the young Socrates is unready to admit Forms.
That something is a finger is a matter on which sight is
competent to pronounce (523b, 524d), and it is characteristic
of the sorts of thing to which Socrates refuses Ideas that are
just what we see them to be (Paml., l30d) (pp. 108-109).

As examples of the characteristics expressed by "incomplete"

predicates, Owen cites lightness and heaviness. He apparently believes
that they are "inseparable," or "compressed," in the objects of
experience because each is both light and heavy, relative to different
standards. Owen maintains that Plato "talks as if" only incomplete
predicates occur in Socrates's "What is F?" question and "talks as if"
only for these predicates are there Forms that are substances.
Owen is less clear about why Plato concluded that for incomplete
predicates there are Forms that are "unambiguous Paradigms," but he
seems to claim that Plato came to this conclusion on the basis of the
following beliefs: (i) that the truth-conditions for complete predicates
such as 'is fire' and 'is a finger' are in terms of standards given by
"unambiguous samples" of fire and fingers; (ii) that the truth-conditions
for incomplete predicates have the same form as the truth-conditions for
complete predicates; and (iii) that the "world which contains
unambiguous samples of fire and fingers contains no comparable cases
of goodness or similarity." Witness Owen's remarks:

Zeno's logical puzzles, like the moral antinomies of his

successors, were built on such incomplete predicates, and the
Parmenides of itself would suffice to show that these two
classes of problems lie at the root of Plato's earlier
theorizing. If we hope to resolve such disagreements by
reference to some unexceptional standard, we shall find that
the world which contains unambiguous samples of fire and
fingers contains no comparable cases of goodness or
similarity or equality KCXO' cxim). If we persist, our
unambiguous Paradigms must be relocated elsewhere, in a
lIorrroc; 7IJ1rOC; (p. 109).

Plato concludes, out of dogged persistence, according to Owen, that the

Forms goodness, largeness, etc., exist as both samples of F-ness and
 3. v Against Owen. 83

standards for F-ness. If Owen's interpretation is correct, Plato comes

to believe that the Form the F itself exists as a sample of F-ness that is
F "in itself" (mO' aim), i.e., that can be known to be F without
reference to a standard and is the standard for statements that predicate
F-ness of the ordinary sensible objects.
Although he suggests otherwise, if Owen has no evidence from
Plato's great middle dialogues, the Phaedo and the Republic, he is
without sufficient support. He maintains that his interpretation could
rest on the first part of the Parmenides, an assertion which I believe is
unwise, since these passages provide no solid evidence for his
interpretation, but even if I am wrong, and the young Socrates in the
Parmenides states and argues for the Theory of Forms in the way that
Owen maintains, unless Plato states and argues for the theory in the
same way in the middle dialogues, the reader is free to reject Owen's
interpretation. Plato could, and perhaps does, make Parmenides force
Socrates to fall into cnropLa (= "perplexity" or "lack of EinropLa," i.e.,
free passage) to illustrate the sort of problems that plague those who,
unlike himself in the middle dialogues, state the theory inaccurately.19
Since, on the traditional ordering of the dialogues, Plato clearly puts
the Forms to philosophical work in dialogues that were written after the
Parmenides, this interpretation seems especially likely.
Owen cites two passages from the Republic. He cites Socrates's
discussion of the philosopher in book V and his discussion of education
in book VII, and the first is the most crucial. For this passage to
support Owen's interpretation, the philosopher and the lover of
spectacles must play on a field that is leveled by three main
assumptions. They must assume that the characteristics expressed by
incomplete predicates are compressed with their opposites in the
ordinary objects of experience relative to sensible objects that are made
to function as standards. Where x is a given object of experience, and
F is a given characteristic expressed by a complete predicate, both must
assume that knowledge that x has F is possible only if objects that are
"unambiguous" standards exist with respect to F-ness. They must also
assume that (the Greek counterpart ot) 'is just' is an incomplete
predicate. The lover of spectacles, recall, cannot know that a given
7fOAL~ is just because to him it must appear both just and unjust, and
the reason, on Owen's interpretation, is that he must choose sensible
84 Philosophers against lovers of spectacles

objects as standards and that such objects make for "ambiguous"

standards of justice. The philosopher can know that a given 7f'OI\LC; is
just, and the reason, on Owen's interpretation, is that his more ample
ontology contains some object that is both a non-sensible sample of
justice and an "unambiguous" standard for justice.
As I understand Socrates's discussion of the philosopher and the
lover of spectacles, it provides no direct support for Owen's
interpretation. Socrates does not contrast incomplete predicates with
complete predicates. He does not even mention (Greek counterparts of)
'is a finger' and other such predicates that Owen supposes that Plato
takes to be complete predicates. Socrates does not maintain that, unlike
in the case of complete predicates, the characteristics expressed by
(Greek counterparts of) 'is beautiful,' 'is ugly,' etc., are compressed
with their opposites in the ordinary objects of experience relative to
sensible objects that are made to function as standards. Socrates does
not maintain the philosopher is victorious over the lover of spectacles
because the philosopher has a non-sensible sample of justice that
functions as an "unambiguous Paradigm," whereas the lover of
spectacles has only sensible objects and that such objects make for
"ambiguous" standards. What he says is that the lover of spectacles
cannot know of a given 7f'OI\U; that it is just, and the reason he gives is
that the lover of spectacles refuses to suppose that the Forms are
substances and tries to account for the truth-conditions for statements
that predicate justice in terms of vop.Lp.a. The resulting so-called
"ambiguity" mayor may not have its source in the lovers of spectacles
and the various sensible objects that they choose to function as
standards, but even if it does, this would not constitute a sufficient
reason to accept Owen's interpretation. It would not show that Plato
concluded that the Forms are "unambiguous Paradigms."
The passage that Owen cites from book VII of the Republic is also
without any direct support for his interpretation. In Plato's celebrated
parable of the prisoners in the cave, Socrates describes the situation of
the soul at birth, a time before it has become a philosopher, so that he
can discuss "the effect of education and the lack of it upon our human
nature" (514al-2). The newborn soul, like the lover of spectacles,
believes that there are many F things, but not that the Form the F itself
is a substance. Because ot the "trauma ot birth" (Timaeus 43a4-44b 1),
i.e., because the "circuits in the head," as Plato says, "are deranged at
 3. v Against Owen. 85

birth" (90dl-3), it no longer remembers what it knew while not in the

body, but in "contact" (Phaedo 79d6) with the Forms.
This loss of knowledge need only be temporary. Education helps
the soul accept and remember what it once knew. It is the "leading
around" (Republic 518d3-4) of such a soul away from the lusts and
desires of body and to the contemplation of the Forms. It is the
process by which the incarnated soul becomes a philosopher.. If it does
not succumb to the body's lusts and desires, but engages in the
"upward path" that culminates in having its parts each doing its own
job, i.e., in being just and in living the good life, it can change into a
philosopher. It then can regain what it lost at birth. It can recollect
the Forms that it knew before its incarnation (Phaedo 63e8-6ge5,
Republic 508b9-cl, 519a8-b5, 519clO-dl, Timaeus 90a2-d8).20
To change the orientation of the newly incarnated soul, Socrates
recommends a program of physical and intellectual education. To
"distinguish" the studies which "lead" in the right direction from those
that do not (523a6-7), Socrates distinguishes between those studies in
which "the reports of our senses ... provoke the intellect" (523alO-b3)
from those that do not. To help his interlocutors understand what he
means in saying that some studies are likely to make the senses provoke
the intellect, Socrates provides examples. He remarks:

You will understand my meaning better if I put it this way:

here, we say, are three fingers, the smallest, the second, and
the middle finger (523c4-6). Each of them equally appears
to be a finger, and in this respect it makes no difference
whether it is seen to be at the end or in the middle, whether
it is white or black, thick or thin, and all that sort of thing.
In all this the soul of the many is not compelled to ask the
intelligence what a finger is, for the sense of sight does not
indicate to it that the finger is the opposite of a finger
(523cll-d6). < In the case of largeness and smallness,
hardness and softness, etc., the situation is different.
Consider, first, softness and hardness. > The sense
concerned with the hard is of necessity also concerned with
the soft, and it declares to the soul that it perceives the same
object to be both hard and soft (S24al-4). The soul in tum
is puzzled (S24a6-7). <The same is true of largeness and
smallness. > The sense of sight saw large and small not as
86 Philosophers against lovers of spectacles

separate but as commingled (524c3-4). In order to clarify

this, intelligence is compelled to see large and small not as
commingled but as separate, the opposite way from sight
(524c6-8). In some such circumstance as this, we first think
to ask what is the large and what is the small (524clO-ll).

Socrates says that when the senses report that a given object of
experience is both large and small, etc., the soul is "led" to the Forms
because it is provoked to ask the "What is F?" question. To transform
the soul from a lover of spectacles to a philosopher, he recommends
studies in which such sense reports are likely because he believes that
these studies are likely to make the soul engage in dialectic.
This is as close as Socrates comes to providing direct support for
Owen's interpretation, and as should be evident, his remarks are not
very promlsmg. Perhaps they do suggest that Plato is somehow
implicitly aware of the distinction between complete and incomplete
predicates, but this is too weak to support Owen's interpretation. If
Owen is correct, Plato has remarkable beliefs. He believes that Forms
are the "unambiguous Paradigms" necessary for the truth-conditions for
the statements that predicate the properties expressed by incomplete
predicates, and he believes that the sensibles suffice to provide such
paradigms for the complete predicates. Owen has not provided nearly
enough evidence to attribute these beliefs to Plato.
Not only is Owen's interpretation without textual support, it
depends on a dubious assumption. Plato was working with answers to
difficult questions about truth and knowledge. In an effort to provide
for the possibility of knowledge of propositions about matters involving
justice, beauty, and other such properties, he was trying to explain how
reality must be so that there are truth-conditions for statements that
predicate these properties. If Owen is correct, Plato did not come to
believe that only a third realm of Forms can provide these
truth-conditions. He came to believe that these Forms exist as star
instances of the properties justice, beauty, etc., and that the statements
that these objects have these properties are themselves exempt from
explanation. This is would be very peculiar. Every explanation must
begin with some unexplained propositions, but Owen's claim can seem
plausible only against the background of the false assumption that Plato
believed that substances are particulars.
 3. v Against Owen 87

Owen smuggles this assumption into the background by putting into

the foreground the suggestion that Plato believed that the
truth-conditions for statements that predicate the properties expressed
by complete predicates, such as 'is a finger, 'is fire,' etc., are given in
terms of particular sensible objects that function as unambiguous
standards. This suggestion makes Plato appear to begin with the
famous Aristotelian assumptions that substances are particulars and that
the essences of these objects are somehow given in perceptions. To
accommodate the truth-conditions for incomplete predicates, Plato
advances in a simpleminded way to the belief that Forms exist as both
samples and unambiguous standards. Once, however, that one realizes
that Plato did not have the beliefs about complete predicates that Owen
claims for him, the background assumption that Plato believed that
substances are particulars should fall away.
Without this background assumption, Owen's interpretation is
extremely weak. One can no longer quickly and straightforwardly
conclude that Plato's belief that the Forms are substances is the bizarre
belief that the Forms are special particulars that exist in an "intelligible
place" and that are somehow easy for the soul to know when it is not
in a body. Only very clear and compelling evidence could warrant
attributing this conclusion to Plato, and in book V of the Republic,
Socrates does nothing to show, or even to hint, that he believes that the
Forms exist as "unambiguous Paradigms."
Owen is perhaps correct to suggest, if this is his intention, that
Plato believed that the lIOIlLIlCi put forward by the lover of spectacles
are "ambiguous" in way that the Forms are not. Plato may well have
believed that no convention was uniquely qualified to be the one that
makes F-things be F. Owen is correct, furthermore, if this is his
intention, to suggest that Plato argued for his belief that the Forms exist
in an lI01/TOC; T07rOC; by arguing that the lovers of spectacles accept a
conception of reality that is inconsistent with the possibility of
knowledge. Plato's belief that the Forms are substances, however, is
not the simpleminded mistake that Owen maintains. Plato believed that
on matters involving justice, beauty, etc., if knowledge is possible, then
Forms that are substances somehow provide the truth-conditions for
statements that predicate these properties, but there is not enough
evidence to conclude that he believed that the Forms provide these
truth-conditions by being paradigmatic examples. The text does not
88 Philosophers against lovers of spectacles

show that Plato had the beliefs about complete and incomplete
predicates that Owen claims for him.

vi. Against Vlastos

In his important and intluential 1965 and 1966 papers "Degrees of

Reality in Plato" and "Metaphysical Paradox, " Vlastos offers a different
argument for roughly the same conclusion that Owen tries to
establish. 21 Vlastos notes that Plato sometimes describes the Forms
as somehow more real than the ordinary objects of experience that have
these Forms, and he suggests that Plato intends to draw a certain
contrast between the Forms and the objects of experience with respect
to Socrates's "What is F?" question. Plato intends his readers to
conclude, according to Vlastos, that correct answers to Socrates's
question have their source in the Forms, not in the objects of
experience. The Forms, to use Vlastos's words, are more "cognitively
reliable" than the ordinary objects that they characterize.
This might be reasonable, given a reasonable account of
"cognitively reliable," but Vlastos provides an account that robs his
interpretation of all textual support. Like Owen, he claims that Plato
believed that the Forms exist as samples and standards.
Vlastos tries to extract his interpretation from Socrates's discussion
of the philosopher and the lover of spectacles. As he reads this
passage, Socrates teaches his interlocutors that the attempt to answer
the "What is F?" question by concentrating on objects of experience
that are F at best yields uncertainty and is likely to yield confusion:

As is clear in this last passage < from book V of the

Republic> --the first and also the fullest exposition of
degrees of reality--the intended contrast is between the
Form, F, and instances of it which are reckoned less "pure"
F's than it, because they are not exclusively F, but are F and
not-F: their F nature is adulterated by contrary characters,
so that we could only get a confused and uncertain idea of
what it is to be F, one that would be subject to constant
fluctuations as we encountered instances of F that turned out
to be different in one or more respects from those on which
we based our previous conception of it ([1965], p. 63).
 3. vi Against Vlastos 89

To avoid these problems, Vlastos maintains on Plato's behalf that one

should look to the Forms. These objects somehow wear their essences
on their sleeves:

< T > he real F would be the cognitively reliable F. Thus,

if you want to investigate the nature of gold, coffee,
courage, beauty, you must look to the genuine article; the
other kind will trick you sooner or later, for along with
some F-properties, it has also, perhaps cunningly concealed,
some not-F properties, and if you were to take the latter for
the former your mistake would be disastrous ([19661, p.49;
emphasis in original).

The key to it is Plato's conviction that the Forms are the

objects of knowledge par excellence. They are
incomparably the most rewarding to the mind of all the
things to which it can tum in its search for truth, for their
natures are logically perspicuous, or can be made so with
adequate training in dialectic, and all their properties follow
from their natures in conjunction with the natures of other
similarly luminous and stable objects. Their physical
instances are, by contrast, intellectually opaque and shifty.
They do not display their intelligible structure on their
sensible surface. And when we try to dig it out of them by
inference and extrapolation we cannot be sure that the cluster
of properties anyone of them happens to have here and now
it will still have later on or that other things, to all
appearance similar, will have this same set of properties now
or later. A thing which is F at one time, or in one way, or
in one relation, or from one point of view, will be all too
often not-F at another time, in another way, et cetera. So,
generalizing with that reckless audacity characteristic of
Greek philosophy at its best as at its worst, Plato infers that
this is true of all sensible F's. All of them will be always so
infected, hence none of them could ever be a real F, and if
we take any of them to be such, we will be sure to be
deceived. The Form, conversely, will never deceive, for it
is by hypothesis invariantly and wholly the special F it is its
nature to be. It is, therefore, the real P, the genuine one,
90 Philosophers against lovers of spectacles

which can be trusted absolutely in our pursuit of knowledge

(p. 50: emphasis in original)Y

Vlastos maintains that Plato believes that the Form the F itself is such
that what makes it be F is somehow open for the pre-natal soul to
know. On this interpretation, the pre-natal soul somehow knows that
this Form is F, somehow knows that this Form has other properties,
and is not "deceived" when it concludes that these properties are what
make F-things be F. Once it is in a body, and after it has forgotten
this pre-natal knowledge, the soul with "adequate training in dialectic"
can recollect what it previously knew.
Vlastos realizes that, on his interpretation, Plato has an unusual
belief about justice, beauty, etc., but he insists that the problem is with
Plato, not with his interpretation. He maintains that Plato's
epistemology forces him to this belief about the Forms. In Vlastos's
terminology, Plato holds a so-called "degrees-of-reality doctrine."
Vlastos uses this description to bring into focus what he takes to be the
distance between Plato's reach and his grasp. Vlastos maintains that a
"kinds-of-reality" doctrine "would have served < Plato> much better
as an instrument of categorial inquiry" ([1965], p. 75). He would have
then had "a better chance to see and state correctly the differences
between particulars and universals" (p. 75), but the unfortunate fact of
the matter, according to Vlastos, is that Plato's epistemology causes his
grasp to fall short of his attempt to reach the kinds-of-reality theory:

In the interpretation I have offered of Plato's theory, the

grounds on which sensible particulars are judged to be less
real than their respective Fonus coincide very largely ... with
those categorial features which disqualify them for serving
as objects of a certain kind of knowledge: knowledge which,
Plato says, has "infallibility," or, in less inflated, more
exact, tenus for what he means, logical certainty .... In
recognizing only one kind of knowledge--knowledge of
objects having these, and only these features--Plato had no
choice but to say that only the Fonus were completely, or
purely, or perfectly "real" in the sense I have been
investigating in this paper: cognitively reliable. Thus the
degrees-of-reality doctrine is, in this respect, a lucid
consequence of Plato's epistemology (p. 73).
 3. vi Against Vlastos 91

Vlastos maintains that the only kind of knowledge that Plato recognizes
is the kind that Socrates requests when he asks his "What is F?"
question. He maintains that this shortsightedness forces Plato to come
to believe that the Forms, unlike the ordinary sensible objects of
experience, are samples that the pre-natal soul can use to come to know
the properties that are essences. As Vlastos later says, Plato mistakenly
"misconstrue < s > universals as a higher grade of particulars" (p. 75).
His epistemological beliefs force him to a degrees-of-reality theory,
when he clearly wants a kinds-of-reality theory.
Vlastos has not made a very strong case for his interpretation. He
slips, first of all, from claiming that Plato believed that definitional
knowledge does not have its source in the sensibles to claiming that
Plato denied that empirical knowledge is possible. I do not believe that
Plato ever denied that empirical knowledge is possible, as this would
undermine the right of the philosopher to rule, but even if I am wrong
on this point, Vlastos is certainly wrong to suggest that this
epistemological proposition alone forced Plato to believe that the Forms
exist as samples with special properties. Plato need not reify Forms to
provide for the possibility of definitional knowledge, and he is certainly
not forced to accept the degrees-of-reality theory. Indeed, even if Plato
began with the assumptions that the Forms are metaphysical objects and
that definitional knowledge is possible, he certainly still could reject the
degrees-of-reality theory. These two assumptions would force him only
as far as the kinds-of-reality theory.
Vlastos could close this gap if he could establish that Plato believed
that the pre-natal soul acquired definitional knowledge by knowing that
a given Form is F, knowing that this Form has certain other properties,
and then concluding that these properties are the essence of F-ness. If
Plato had these beliefs, one would certainly expect him to show them
in the Phaedo in his discussion of the Theory of Recollection, but this
expectation is disappointed. Plato is clear in the Phaedo that the soul
first gets knowledge of the Forms before it is born into the body. He
is also clear that when the incarnate soul gains insight into its concepts
it is actually remembering, or recollecting, knowledge that it first
acquired when it existed outside the body and in the realm of the
Forms, but Plato is notoriously taciturn about how the pre-natal soul
first acquires knowledge of the Forms. He is convinced that it
92 Philosophers against lovers of spectacles

happens, but he does not explain how it happens. This fact casts
serious doubt on Vlastos' s interpretation.
Vlastos himself maintains that book V of the Republic contains
Plato's "fullest exposition of degrees of reality," but if he is correct,
then he has no hope of closing the gap between the text and his
interpretation. Vlastos does not attribute to Plato a proposition that any
reasonable person could be expected to believe. He gives him a belief
that few, if any, have ever had. To show that Plato had this belief,
Vlastos requires solid evidence. No such evidence is in the discussion
of the philosopher and the lover of spectacles. Although he is wrong
to assert that Plato only recognized one kind of knowledge, he is
correct that the passage shows that Plato believes that if a given person
limits himself to an ontology in which essences are not starting-points,
he cannot know correct answers to Socrates's "What is F?" question.
This, however, obviously falls far short of showing that Plato believed
that the Forms exist "as a higher grade of particulars." Indeed, it falls
short even if one grants that Plato believed that correct answers to the
"What is F?" question do not have their source in the ordinary objects
of experience. Vlastos is correct to emphasize, if he intends to make
this point, that Plato believed that the Forms are somehow "cognitively
reliable" in a way that the conventions of the lovers of spectacles are
not. He is also correct, if this is his intention, to maintain that Plato
argues that the Forms are substances by arguing against alternative
conceptions of reality in which essences are not starting-points, but
Vlastos is wrong to assert that Plato believed that the Forms exist as
samples and standards. This can seem plausible only if a false
assumption stands in the background. Vlastos seems to assume that
Plato supposed that whatever exists as a substance exists as a particular,
and he seems to use this assumption to bridge the gap between the
evidence in the text and his conclusion that Plato makes the Forms exist
"as a higher grade of particulars." Aristotle believed that substances
are particulars, but one cannot simply read this belief back into Plato.
Vlastos needs evidence, and none is forthcoming from Socrates's
discussion of the philosopher and the lover of spectacles in book V of
the Republic.
 3. vii Against Irwin 93

vii. Against Irwin

Terry Irwin argues for an interpretation that does not charge Plato with
the belief that the Forms exist "as a higher grade of particulars," but
his interpretation is hardly generous. Witness his remarks:

In the middle dialogues Plato does not mention "separation. "

He insists on the non-identity of the Form of F and the many
Fs (Phd. 74a9-12, b4-7; R. 476c9-d3; Symp. 211al-b5).
This might imply nothing more than non-reducibility ....
< This > is a legitimate conclusion from Plato's arguments.
However, "separation" for Aristotle implies a stronger claim
about independent existence... Plato does not clearly
formulate < this claim> in the middle dialogues; but he
suggests it, and Aristotle's view that he holds it is normally,
and rightly, accepted ([1977b), p. 154).

The confusion might be partly explained by indiscriminate

talk of "sensibles" or "what the senses reveal," meaning
sometimes "sensible properties" and sometimes "sensible
objects".... His attitude to separation confirms our claim
that the arguments for < non-reducibility> are his real
arguments for separation; he offers no further argument for
< independent existence> ... (p. 155).

Forms < cannot> be sensible properties; justice cannot, for

instance, be the same property as the property of returning
what has been borrowed, since the latter property makes
actions both just and unjust (by making some just and others
unjust). But some further arguments are needed to suppon
the conclusion (attributed to Plato by Aristotle) that Forms
are separated, and therefore would exist even if no sensible
objects existed. Though Plato may indeed believe in this
independent existence of Forms, he does not argue for it,
and he does not appeal to it in his contrasts between Forms
and sensible things ([ 1989], p. 93).

According to Irwin's interpretation, Plato's conception of reality is born

in the sin of a simpleminded mistake. Irwin maintains that Plato made
this mistake because he did not clearly appreciate the difference
94 Philosophers against lovers of spectacles

between two very different propositions: the ontologically menacing

proposition that justice, beauty, etc., exist separately from the ordinary
sensible objects of experience and the ontologically innocent proposition
these properties cannot be adequately defined in terms of sensible
properties. According to Irwin's interpretation, Plato argued that the
Forms cannot be adequately defined in terms of sensible properties and
concluded that the Forms exist separately and as substances.
To establish his interpretation of the development of Plato's
thought, Irwin tries to show that in texts where Plato seems to be
arguing for the proposition that "the universals < are > separate"
(Metaphysics l078b30-32), to use Aristotle's words, he is actually
arguing for nothing stronger than the ontologically innocent proposition
that justice, beauty, etc., are not definable in terms of sensible
properties. Irwin presses this interpretation for both Socrates's
autobiography in the Phaedo and Socrates's discussion of the
philosopher and the lover of spectacles in book V of the Republic.
Irwin writes:

< In book V of the Republic, Plato argues> that

concentration on sensible properties will yield no knowledge
of justice, beauty and so on. Plato's opponents are "lovers
'If sights and sounds," who deny the existence of a single
form, the beautiful, besides the many sensible properties,
each of which is alleged to be "the beautiful in" various
things. houses, temples, institutions. To find the beautiful
in something, what makes something beautiful, the
sight-lovers must find a property which is always beautiful,
never not-beautiful (cf. Phd. 102d5-8). But any of the
sensible properties alleged to be the "many beautifuls" will
also be not-beautiful (478e6-479alO); the bright color (cf.
Phd. lOOdl) alleged to be the beautiful in temples will be
not-beautiful in some temples; debt-paying, alleged to be
justice in some actions, will also sometimes be unjust (cf.
Rep. 331c-d, 538d6-e3); the pleasures which are alleged to
be the good will sometimes be bad--some will be good and
others bad (505c1 0-11). These properties allow no
knowledge of the beautiful, the just or the good; for
knowledge. we must describe a separated Form irreducible
to sensible properties ([1977a], p. 9).
 3. vii Against Irwin 95

Republic V explicitly insists, as the Phaedo implied, that

reference to separated Forms is necessary for knowledge ... ;
and it shows, as the Phaedo also implied, that the separated
Forms are required because of faults in the senses and
observation. Plato argues that those who love sights and
sounds, and recognize "many beautifuls" but no single
Form, The Beautiful, cannot escape the Phaedo's objections
to defective explanations ([1977b], p. 151). The sight-lover
must recognize a single property ... , apart from observable
properties (p. 152).

According to Irwin, Socrates's complaint against the explanations

according to the lovers of spectacles is that they wrongly believe that
justice, beauty, etc., are definable in terms of sensible properties. 23
Irwin maintains that this same point is earlier "implied" in the
autobiography in the Phaedo. He maintains that this point is the
implied lesson of Socrates's discussion of the "defective explanations"
of the natural scientists. These explanations are defective, on Irwin's
interpretation, because they are attempts to define justice, etc., in terms
of sensible properties. Irwin writes:

The deficiency of sensible properties supports Plato's

rejection of mistaken efforts at explanation. If someone tries
to explain why things are larger by reference to some
observable property (e.g., "by a head," Phd. 96d7), he will
fail; for many things are larger than other things without
being larger by a head, and things may be smaller, as well
as larger, by a head. Observable operations (e.g., "add
two") equally fail to explain what makes one set more
numerous than another (96el). These accounts are meant to
be ridiculous--no one would try to explain largeness and
numerousness that way. But Plato uses them to suggest that
some apparently less ridiculous explanations are no more
defensible. He rejects accounts of the beautiful referring to
bright colors, shapes, "or anything else of this kind" (1 OOc9)
([ 1977a], p. 8).

The rejected accounts mention some observable features

(e.g. "by a head") of what makes x larger than y (Phd.
96d7-el), or some observable operation (e.g. "add two") to
96 Philosophers against lovers of spectacles

make x more numerous than y (96el-3). An observable

feature 0 of x may show us that x is F, but 0 will not
provide the right account of x' s being F, for the correct
account should apply always and only to F's, whereas (a) 0
might be present in non-F's (IOOe8-101b2), and (b) 0 might
be absent in F's (97a5-b3) ([1977b), p. 151).

Irwin maintains that Socrates cites the "intentionally ridiculous"

attempts to appeal to sensible properties to explain what makes large
things large, numerous things numerous, etc., in an effort to make the
general point that his "What is F?" question is about properties that are
not identical with sensible properties. If Irwin's interpretation is
correct, Socrates is not making an ontological claim at all. He is not
arguing that the Forms are substances. Socrates is simply trying to
help his interlocutors understand what his predecessors apparently
failed to understand: namely, that beauty, justice, and the other
properties he uses his "What is F? question to ask about, are not
definable in terms of sensible properties.
As I read the autobiography, Plato is not just making the point that
Irwin claims for him. When Socrates says "farewell" to "having a
blooming color," etc., as aLTim for why beautiful things are beautiful
(IOOc9-d3), he does not describe the rejected answers as attempts to
define beauty in terms of sensible properties. He describes them as
"wise" answers that "trouble" him. No doubt part of what troubled
Plato about at least some of these attempts at definition is that they are
open to counterexample. Perhaps Plato also believed that some of
these counterexamples are possible because these "wise" answers are
attempts to define beauty in terms of sensible properties, when in fact
beauty is identical to no such property, but after dismissing "having a
blooming color," etc., what his character actually goes on to say is that
nothing other than the Form "itself according to itself" (auro mO'
auro) is that "by which" (instrumental dative) beautiful things are
beautiful. This suggests that what disturbed Plato about the "wise"
answers of the natural scientists has nothing to do with whether justice,
beauty, etc., are definable in terms of sensible properties.
With respect to matters involving justice, beauty, etc., Plato
rejected these "wise" answers because he believed that they are born in
a method of inquiry that is part of a conception of reality that is
 3. vii Against Irwin 97

inconsistent with the possibility of such knowledge. Plato rejected the

natural scientists's method of inquiry in favor of the leading
assumptions in the method that Socrates pursues. In the definitional
dialogues, Socrates insists that in the search for knowledge he and his
interlocutors must first answer the "What is F? question before they can
answer the other questions that interest them. Furthermore, in these
same dialogues, Socrates refuses to consider a definitional answer to his
"What is F?" question until he has first secured the preliminary
agreement that the Form the F itself is that "by which" F-things are F.
In his autobiography, Socrates insists on the priority of Forms that exist
"themselves according to themselves," and he rejects explanations
according to the method of the natural scientists as outright
non-starters. Perhaps this does not amount to a completely explicit
argument that the Forms are substances, but neither is it just an
innocent way of asserting that every attempted analysis of justice, etc.,
in terms of sensible properties is open to counterexample.
Book V of the Republic provides more evidence against Irwin's
interpretation. Plato again argues for the Forms by arguing that
alternative accounts suffer from defects that seem to have nothing to do
with sensible properties. The lover of spectacles fails to meet the
Socratic challenge because he insists on explanations of the
truth-conditions for statements that predicate beauty in terms of "the
many conventions (v0Il-Lll-a) of the many concerning the beautiful"
(479d3-4), not because he tries to define beauty in terms of sensible
properties. In the passage at 493aff., e. g. , where the lover of
spectacles appears in the guise of the sophist, Plato complains that
"what < the mob> enjoys he calls good, what angers it he calls bad,
and he can give no other account of them" (493c2-4). These accounts
are not definitions in terms of sensible properties.
Contrary to Irwin's interpretation, Plato is not trying to show that
justice is not definable in terms of sensible properties. His lesson is
much more striking and philosophically interesting. At issue is the
proper explanation for why a given 7rOALC; with certain properties
counts as just. The definition he puts forward earlier in book IV shows
that Plato believed that justice is a non-sensible property, but this fact
is not relevant to his discussion of the philosopher and the lover of
spectacles in book V. In book V, Socrates teaches a lesson about the
source of the truth-conditions for statements that predicate justice.
98 Philosophers against lovers of spectacles

Plato believed that explanations in terms of "many conventions of the

many," unlike explanations in terms of the definitional content of the
one Form that exists "itself according to itself" (aim) mO' aUTo), are
part of a conception of reality that has the very unpalatable skeptical
result that political knowledge and expertise is impossible. This is why
Plato concluded that the Forms are substances.
Another problem for Irwin's interpretation is his suggestion that
Plato conflates sensible objects and sensible properties. This suggestion
is implausible. Plato might not have appreciated the difference, but this
is not what one would expect from someone who never tired of making
his character explain that the "What is F?" question is not a question
about ordinary particulars that are F. Irwin must provide evidence for
his suggestion, and the only evidence that he musters presupposes that
his interpretation is true. Irwin is apparently convinced that Plato
argues that justice, etc., are not definable in terms of sensible
properties and invalidly concludes that the Forms are substances. To
explain this development, Irwin hypothesizes that Plato conflates
sensible objects and sensible properties. This is a move of last resort,
and I do not believe that it is warranted.
The philosophical import of the Phaedo and Republic is certainly
not straightforward, but I see no good reason to conclude that Plato is
confused in the way that Irwin maintains. Plato believed that the
Forms are substances, not because he fell into a confusion over sensible
objects and sensible properties, but because he believed that no other
conception of reality validates the leading assumptions in the Socratic
method for acquiring knowledge about matters involving beauty,
justice, and other such properties. Plato believed that conceptions of
reality in which essences are not starting-points are inconsistent with
the possibility of such knowledge. Plato indicates this belief in the
two great dialogues from the middle period. He indicates it in the
autobiography in the Phaedo in his discussion of Socrates's encounter
with the natural scientists's method of inquiry, and he indicates it in
book V of the Republic in his discussion of the philosopher and the
lover of spectacles. 24
 4

HERACLITEAN FLUX

i. Introductory remarks

Aristotle tries to explain the genesis of the Theory of Forms in terms

of certain of Plato's beliefs about definition, knowledge, and the
Heraclitean thesis that the sensibles "flow" (Metaphysics 987a32-b7,
1078b9-1079a4, and 1086a31-bll). His explanation is incomplete, in
that it make no mention of the sophists, but it does serve to more
sharply focus Plato's argument against the natural scientists. His
remarks suggest that Plato believed that, since the sensibles are
flowing, no ontology that takes them as starting-points can provide for
the existence of essences. According to Aristotle, this is why Plato
concluded that the Forms exist separately and as substances.

ii. The Aristotelian explanation

The relevant question to answer about Plato and his belief in Forms,
when it is expressed in Aristotle's terminology,l is why he made the
universals exist separately and thereby made them exist as substances.
Aristotle states part of his answer in the following passages:

Socrates was busying himselfwith ethical matters, neglecting

the world of nature as a whole. He was seeking the
universal in these ethical matters, and he fixed thought for
the first time on definitions. Plato accepted the importance
of definition, but because he thought that the common
definition could not be about any sensibles, as they were
always changing, he thought this happens on account of
other things, and not the sensibles. Things of this other sort
he called ideas (987bl-8).
tOO Heraclitean flux

< Plato and the Platonists > did not make the substances the
same as the sensibles. They thought that in the sensibles the
individuals were flowing and that none of them remained,
but that the universal exists beside (7rcxpa) these things and
is something different from them. This, as has already been
said, was set going by Socrates through his definitions, but
he did not separate them from the particulars, and he was
right not to make a separation. This is clear from the
results. Without universals, to take knowledge is
impossible. Separation is responsible for the difficulties
arising about the Ideas (1086a36-b8).

Aristotle reports that Socrates "was seeking the universal" in his search
for knowledge. He reports that Plato accepted this practice, but
because he further supposed that the sensibles are somehow "flowing,"
he concluded that the universals are separate.
Aristotle does not say why Socrates and Plato believed that
definition is important in the search for knowledge, but Plato's readers
should have no trouble supplying the answer. In the early definitional
dialogues, Socrates proceeds in a way that suggests that, with respect
to beauty, justice, etc., he believes that the inquirer must know the
essence of F-ness before he can know the answers to the other
questions involving F-ness that interest him. Plato may not accept a
strong form of the priority of definition, but he certainly believes that
if no one can answer Socrates's "What is F?" question, and knowledge
of the essence of F-ness is impossible, then knowledge about matters
involving F-ness is equally impossible.
Although his remarks are obscure in many ways, Aristotle is
clearly correct that the road to Platonism begins with "Socrates through
his < search for> definitions. " Socrates turned away from
explanations according to the method of the natural scientists, and he
resisted the sophists's tendency to confuse the power of the spoken
word with the force of reason. Socrates suggested that definitional
knowledge is part of the justification that is necessary for knowledge.
He also suggested that the Forms insure that there are definitional
truths to be possible objects of knowledge. In the early dialogues, in
his search for knowledge, the character insists on the (A 1) response to
the preliminary (Ql) question. In the Phaedo, and then again in the
 4.ii The Aristotelian explanation 101

Republic, Plato defends this insistence on the priority of Form. He

decides against the ontologies that he associates with the natural
scientists and the lover of spectacles, and he concludes that only if the
essence of F-ness exists as a substance and a starting-point can there be
knowledge with respect to matters involving F-ness.
Many Platonic scholars suppose that Aristotle is correct to maintain
that some sort of flow doctrine plays a central and essential role in the
development of Plato's thought about Forms. Norman Gulley offers
an especially strong statement of this position:

In the Metaphysics Aristotle states that an important factor

in the development of Plato's theory of Forms was an
acquaintance, through Cratylus, with the Heraclitean doctrine
of universal 'flux,' which led Plato to the conviction that,
since all sensible things are always in flux, they cannot be
the objects of knowledge, and, consequently, that there must
be other things to serve as objects of knowledge which are
not sensible and not subject to changes (987a32-4;
1078bI2-16). There are no adequate grounds for
questioning the general truth of Aristotle's statement ([1962],
p.26).

Gulley confidently proclaims that on the Theory of Forms, there are

"no adequate grounds for questioning the general truth of" Aristotle's
account. 2 Whether his confidence is warranted depends on what
propositions Aristotle's remarks express.
Although perhaps they are not equally likely, Aristotle's remarks
admit of more than one interpretation. On one interpretation, Aristotle
claims that Plato concluded that the universals are separate because he
believed that the sensibles are flowing, believed that there are referents
of the subject terms in the propositions that constitute_ knowledge, but
believed that flowing things cannot be among them. Aristotle's
remarks can, I think, seem to contain these suggestions. Witness, e.g.,
his thoughts in the following passages from the Metaphysics:

In his youth Plato became familiar with Cratylus and the

Heraclitean doctrines, that all sensibles are always flowing
and that knowledge about them is not possible, and these
views even in his later years he accepted (987a32-bl).
102 Heraclitean flux

The belief about Forms came about to those who spoke

about them because they were persuaded by the Heraclitean
arguments that all sensibles are always flowing. They held
that if knowledge and thought are to be about anything, there
must be remaining some different natures, besides (1rapa)
the sensibles, for there is no knowledge about flowing things
(1078b12-17).

In these passages, Aristotle may seem to make Plato argue from

(1) knowledge has objects

(2) the sensibles are flowing
and (3) if (2) is true, the sensibles are not objects
of knowledge

to the conclusion that

(4) the universals exist separately.

On this interpretation, Aristotle maintains that Plato was a skeptic

concerning knowledge about the ordinary objects of experience. He
concluded the universals exist separately and as substances because
otherwise there would be nothing to know. 3
Depending on the analysis of "flowing," this interpretation divides
into two more specific interpretations. Premise (2) could express the
weak claim that the ordinary objects are changeable, or it could express
the very strong and paradoxical claim that they are always changing in
every way. These readings shift the burden of proof between premises
(2) and (3). The weak reading puts the burden on premise (3). The
strong reading puts it on premise (2).
In either case, there is a good reason to suppose that Aristotle is
dead wrong about Plato. The reason is in the central books of the
Republic. These books show that Plato did not turn his back on the
world of the ordinary objects of experience. On the contrary, he
argued that only the philosopher, not his competitor, the lover of
spectacles, is a candidate to rule because only the philosopher has an
ontology that is part of a conception of reality according to which there
can be political knowledge and expertise. 4 No premise about change
 4.ii The Aristotelian explanation 103

figures in this argument. Plato's complaint is that, instead of accepting

a conception of reality according to which essences exist as substances,
"themselves according to themselves," the lover of spectacles
recognizes only the "many conventions of the many."
Indeed, despite the passages from the Metaphysics, the context in
which Aristotle places the Theory of Forms suggests that he did not
believe that Plato was a skeptic. He characterizes the theory as one
among many attempts to put forward objects as "causes," the
knowledge of which constitutes the "wisdom" (aocPLO:) that the
philosopher seeks (981b28). Aristotle says that Plato supposed that
"the Forms are causes of the other things" (987b18-1O) and that they
"provide the essences of each of the other things" (988b4-5). In the
light of these remarks, Aristotle must at least admit that Plato should
believe that the Forms provide for the possibility of knowledge about
the ordinary sensible objects. Aristotle himself claims that "causes are
spoken of in four ways, in one of these ways of speaking we say that
the cause is the substance and the essence" (983a26-27) and that "it is
knowledge of each thing when we know the essence of that thing"
(1031b6-7). Aristotle, moreover, argues that the Forms can not be
substances by arguing that such objects would contribute nothing to the
"knowledge" of the sensibles "because they are not the substances of
these things because they are not in them" (991a12-13, l079b15-17).
These remarks strongly suggests that Aristotle admits that Plato not
only should, but actually did believe that the Forms provide for the
possibility of knowledge with respect to the sensibles.
If this is correct, and it seems to be, Aristotle is not uncharitable.
He allows that Plato intended for Forms to provide for the possibility
of such knowledge. He does not complain that the Theory of Forms
does not solve some problem that Plato does not intend it to solve. He
does not complain that Plato thought that he could make the Forms be
causes and essences but that he need not enlist their help to account for
the possibility of knowledge with respect to the sensibles.
Aristotle's complaint is not an uncharitable quibble. It is that Plato
failed to understand that if the Forms are substances, and exist
separately from the sensibles, then they cannot help provide for the
possibility of knowledge about these objects. Aristotle explains that
Plato failed to understand that nothing in another realm can help
provide for the possibility of knowledge about objects that are in the
104 Heraclitean flux

realm of ordinary experience. "Separation," he says, "is responsible

for the difficulties arising about the Ideas" (1086b8). 5
Another problem with these interpretations is that, in each case,
they force Plato to accept a very implausible premise. On the weak
understanding of "flowing," premise (3) can only seem to result from
a very simpleminded error. Hardy puts the point as follows:

It looks at the outset as if liability to change could not of

itself suffice to preclude knowability, and as if a very simple
fallacy might give rise to the suggestion that it could. It
might be said that, if x is something changeable, the
proposition 'x is characterized by y' cannot express
knowledge since it may not be true to-morrow that 'x is
characterized by y.' But knowledge can never become false.
To this it seems sufficient to answer that, in the case
supposed, nothing has 'become false'; since it will not
anyhow be false to-morrow that x is characterized by y
to-day ([ 1936], p. 12).

Plato is not immune to philosophical error, but given his brilliance,

very clear evidence is necessary before one can conclude that he is
guilty of such a simple mistake.
A similar point holds for the strong reading of "flowing."
Although Plato could in principle believe that the ordinary objects are
in radical flux, this proposition has immediate consequences that are
obviously absurd, and these consequences are in clear conflict with his
most distinctive philosophical claims. If the ordinary objects of
experience are in radical flux, whether or not the Form the just itself
exists, and whether or not the philosophers or the lovers of spectacles
are the politicians in power, the 7ro);.,~ is nevertheless always both just
and not just. There thus seems to be no point to arguing, as Plato does
at length in the Republic, that only the philosopher is fit to rule because
only the philosopher agrees that the Form the just itself exists as a
substance. Indeed, argument itself would be impossible. Whether or
not the Forms exist, if the ordinary objects are in radical flux, every
token sentence both expresses a given proposition and does not express
that proposition. Plato could have missed these consequences, but only
clear and compelling evidence could show that he was so shortsighted. 6
 4.iii Careless assimilation 105

iii. Careless assimilation

If Plato does not believe that the sensibles are in radical flux, and
Aristotle claims otherwise, then Ari~totle is mistaken. One way to try
to explain his mistake is to show that, in his explanation of the genesis
of the Theory of Forms, Aristotle "carelessly assimilated" various
thoughts from different parts of the Platonic corpus. In her
commentary on books M and N of the Metaphysics, Annas suggests
that Aristotle is guilty of such an assimilation:

The nearest Plato comes to accepting something like the flux

argument is at Timaeus 51-2, where he does accept that the
physical world is in flux and that therefore objects of
knowledge cannot be found in it ([1976], p. 153).

Aristotle is then at best giving what he takes to be the

leading idea behind Plato's theory, rather than pointing to
what Plato himself argues. It is possible that he is thinking
mainly of the Timaeus argument (the Timaeus is the dialogue
he most frequently cites) and that this leads him to ignore
the fact that the argument most prominent in the middle
dialogues is in fact quite different from the flux argument,
though it could be quite easy to confuse them. The line of
thought stressed at Phaedo 74-5 and Republic 479-480, and
referred to at Parmenides 128-9, depends on features of
predicates of which 'equal' and 'just' are examples, and
whose striking aspect is that one can, in the case of any
particular to which it is applied, also apply the contrary with
equal justification (what is just from one point of view is
unjust from another, and so on). By contrast, the Form of
Justice is that to which only 'just,' and never 'unjust,' can
be applied. Forms must be distinct from particulars because
only to the latter, never to the former, can one of these
predicates be applied together with its opposites. Aristotle
may have carelessly assimilated this 'argument from
opposites' to the flux argument, which involves the
succession of opposites and thus provides a support for the
other argument, but is clearly different from it and not
implied by it. Unless Aristotle has made this assimilation it
is hard to see why he ignores a line of thought which is
106 Heraclitean flux

much more prominent as serious backing for Forms in the

middle dialogues than considerations of flux and change (pp.
153-154).

Annas follows Owen in the analysis of the argument in the Republic

against the lover of spectacles. 7 She suggests that because Aristotle
was very familiar with the Timaeus, and because in this dialogue Plato
comes closest to giving the "argument from flux," Aristotle "carelessly
assimilated" the two arguments.
Annas is wrong to follow Owen, and she is wrong to suggest that
Plato anywhere accepts the extreme Heraclitean doctrine that the
sensibles are in radical flux, but perhaps "assimilation" is at least a
possible explanation of how someone familiar with the Platonic
COrpUS,8 could arrive at the Aristotelian explanation ofthe development
of Plato's thought. Perhaps the explanation results from assimilating
the argument against the lovers of spectacles and the argument in the
Cratylus and the Theaetetus against the Heracliteans. I do not believe
that Aristotle is guilty of this assimilation, but as I shall show, it is an
easy mistake to make.

iv. Heraclitus and Protagoras

Although he may allude to it elsewhere, Plato's only explicit

discussions of the flow doctrine occur in the Cratylus and the
Theaetetus. In these dialogues, Plato seems to put the flow doctrine
together with the Protagorean measure doctrine to constitute a
conception of reality that he takes to be an unsuccessful competitor to
the Theory of Forms. In this context, the flow doctrine appears to be
the proposition that everything is subject to change, including the
essences of justice, beauty, and the other properties that Socrates uses
his "What is F?" question to ask about. Plato clearly rejects this
proposition about the essences of justice, beauty, etc., but his
arguments seem directed against a different and much more paradoxical
proposition. His arguments seem directed against the proposition that
everything is always changing in a every way.
Plato rightly argues that this proposition is absurd, and he never
says, or even hints, that the absurdity would be eliminated if the
universal quantifier in this proposition were restricted to range over the
 4.iv Heraclitus and Protagoras 107

ordinary objects of experience. In these two dialogues, nor, for that

matter, in any other place in the corpus, does Plato show himself to
agree with Cratylus and the extreme Heracliteans that the ordinary
objects of experience are always changing in every way.
The Cratylus begins with Socrates interrupting a dispute over what
is very vaguely described as the "correctness of words" (383a4-5).
Hermogenes asks Socrates to moderate a dispute over what determines
whether a given sequence of sounds or letters constitutes a word
(383a4-b2). Hermogenes believes that words are simply a "portion of
the human voice that men agree to use" (a6-7), but Cratylus believes
that whether a given sequence of sounds or letters is a word is
somehow "determined by nature" (a7 -b I ).
After he enters the discussion, Socrates quietly changes the topic
of conversation to a more interesting subject. Instead of what
determines whether a given sequence of sounds or letters constitutes a
word, he asks Hermogenes what determines whether a given utterance
or sentence expresses a truth (385a6-1O). Hermogenes does not notice
the change in topic, and instead of helping him directly, Socrates
decides to let him fumble through a question-and-answer session. He
gets him to dismiss the Protagorean doctrine that "man is the measure
of all things" as an answer to this question (385e4-386d2), and he then
again turns the conversation to a new question.
Socrates now wants to know who are the experts in the uses of
words, and the answer he puts in Hermogenes's mouth is reminiscent
of Socrates's own claim in the Republic that only the philosopher is fit
to rule over the 1fOALC;. Hermogenes agrees that words are
"instrument < s > for teaching and distinguishing beings (ouaLm)"
(388b13-cl) and that the "legislator" (JlOIlOOhoC;, 388el) is the artisan
who introduces the acceptable uses of words to the ordinary language
users in the general public. He agrees that the legislator is no common
craftsman, but is very highly skilled, and that the "dialectician"
(cnaAEKToc;, 390c11), the person who knows how to ask and to answer
questions, is the only person with the requisite knowledge and expertise
to direct the legislator in his work (388b7-390d6).
In his discussion of the Protagorean measure doctrine, although he
does not explicitly mention the Heraclitean flow doctrine, Socrates
seems to oppose these two doctrines to two claims that, in other
108 Heraclitean flux

dialogues, he makes about the Forms. In the Phaedo and the Republic,
Socrates accepts a realist theory of the essences of beauty, justice, and
the other properties that he uses his "What is F?" question to ask about.
In the Crary[us, Socrates appears to have this theory in mind. He twice
stands behind the remark that the "being (ovCTLa) of things" is both
"firm" and "by nature." The passages read:

Does it appear to you that things hold this way, that the
being of them is private to each, as Protagoras said, saying
that man is the measure of all things--that as things appear
to me, such they are for me, and as they appear to you, such
they are for you--or does it seem to you that things have a
certain firmness of being of their own?
There have been times, Socrates, when I have been driven
in my perplexity to take refuge in what Protagoras says;
however, I do not believe that things hold this way
(385e4-386a7).

< If, Hermogenes, you reject what Protagoras says, then>

it is clear, indeed, that the things themselves have a firm
being of their own, not relative to us, nor dragged up and
down according to our appearance, but according to
themselves relative to their own being as they are by nature.
That is how it seems to me, Socrates (386d9-e5).

In these passages, Socrates presents Hermogenes with a disjunction.

On one side, which he clearly believes is the correct side, there is

(i) "being by nature" and "firm being."

These are features that characterize the Forms. On the other side,
which Socrates clearly believes is the incorrect side, there is

(ii) "being relative to _" (or: "conventional being")

and "fluctuating being. "

Socrates does not explicitly talk in terms of "fluctuating being, " but he
no doubt takes this to correspond to "firm being."
 4. iv Heraclitus and Protagoras 109

This contrast appears in other places in the corpus. In the

Republic, e.g., in his explanation of why only philosophers are fit to
rule, Socrates maintains that the just itself and the other Forms exist by
nature:

< As the philosopher-rulers> work. they would keep

looking back and forth to justice, beauty, moderation. and all
such things as by nature exist. and they would compose
human life with reference to these ... (501bl-4).

In the first part of the Theaetetus, where Socrates considers whether

"knowledge is perception" (151e2-3), the measure doctrine and the flow
doctrine appear as a pair and presumably in opposition to the Theory
of Forms. They appear together, e.g., in the speech that Socrates
makes in Protagoras's defense:

Whatever in any city is regarded just and fine is just and

fine. in that city and for so long as that convention maintains
itself, but the wise man replaces each pernicious convention
by a wholesome one. making this both be and seem just
(l67c4-7).

If you take my advice. you will sit down with us without ill
will or hostility, in a kindly spirit. You will genuinely try
to find out what our meaning is when we maintain that all
things are in motion and that for each person and each city.
things are what they seem to them to be (168b2-6).9

They also appear together in Socrates's description of certain of the

beliefs that were common in his day:

< Some> men are ready to insist that no one of these

things, just and unjust, religious and irreligious, etc., has by
nature any being of its own. In respect to these, they say.
the opinion in common becomes true at the moment when it
is made and so long as the opinion stands, and even those
who are not prepared to go all the way with Protagoras take
some such view of wisdom on these matters (172b2-7).
110 Heraclitean flux

We were speaking of the people who assert a being that is

in motion, and who hold that for every individual things are
always whatever they seem to him to be; and we said that
they were prepared to stand upon their principle in almost
every case--not least in questions of what is just and right.
Here they are perfectly ready to maintain that whatever any
community decides to be just and right, and establishes as
such, actually is what is just and right for that community
and for as long as it remains so established (177c6-d2).10

Socrates makes Heraclitus and Protagoras stand together, and he makes

them fly banners that oppose the two most distinctive features that Plato
gives to the Forms. Heraclitus declares that "being is in motion," and
Plato argues that "being is firm." Protagoras declares that "being is
relative to _," and Plato argues that "being is by nature. "
Plato does not explicitly introduce the flow doctrine until the latter
half of the first section of the dialogue. This section begins with
Socrates suggesting to Hermogenes that because he rejects the measure
doctrine he must admit that Cratylus is the winner in their dispute over
the "correctness of words" (390d7-e4). Hermogenes is rightly
nonplused, but does not see how to make his case, so he asks Socrates
to provide an example (390e5-391a3). Socrates should give him a
lesson in dialectic, but he apparently judges that Hermogenes is not
ready for philosophy. He suggests instead that Hermogenes, like his
brother before him, ought to get his education from the sophists. ll
Hermogenes balks at this suggestion, since he has already rejected
Protagoras's measure doctrine, and Socrates agrees to help him learn
from Homer and the poets according to the method of "etymology"
(iTVjLOAo,,(La = ETVjLO~ ("true") + AO"(O~ ("account"».
In pursuing this method of inquiry, which seems to amount to
extracting meaning from the letters that language users use to spell their
words, Socrates first mentions Heraclitus and his doctrine in connection
with the Doric word for "being":

Those who said wC1ia [= the Doric form of the Attic word
OVC1icx ("being"») must have inclined to the opinion of
Heraclitus, that all things are made to flow and none to
remain, for to them the cause and ruler of all things is the
 4. iv Heraclitus and Protagoras 111

hurrier (TO'WOOVII), and was therefore rightly called waio:

(401d4-7).'2

After several more such accounts, Socrates suggests that he now sees
that the flow doctrine is the assumption that guided these ancients and
also guides the contemporary wise men:

By the dog, I have not a bad notion which came into my

head only this moment. I believe that the ancient givers of
words are altogether just as the many contemporary men of
wisdom, who, in their search for how the things hold, get
dizzy by being turned around, and then to them it appears
that the things are whirling around and always changing.
They do not cite their internal condition as the cause of these
beliefs, but claim that the things themselves are thusly by
nature, that none of them is steadfast nor firm, that they
flow ... (411b3-c4).

Socrates says that the ancient givers of words and contemporary men
of wisdom in their search for "how the things hold," got turned around
and became dizzy, and because to them the world seemed to be
whirling around, they declared that nothing has a "firmness of being,"
but that everything "flows." Socrates does not say how these inquirers
are "turned around," but the answer is clear from the Phaedo and the
Republic. They are not looking to the Forms.
In the second section of the dialogue, as one might expect, Socrates
marks Cratylus as one of these contemporary so-called "men of
wisdom. " Cratylus is not, at least not initially, as agreeable as
Hermogenes. He rejects Socrates's claim that properly using words is
an "art" in which only a few are skilled, and he insists that all uses are
equally acceptable (428e6-429bll). He does not offer the Heraclitean
flow doctrine as justification for his views on the correctness of words,
but Plato seems to intend the reader to draw this conclusion.
This intent seems to show itself at the end of the dialogue. In an
apparent effort to refute the flow doctrine to preserve his thesis about
the uses of words, Socrates repeats his claim about the ancient givers
of words and introduces the Forms:
112 Heraclitean flux

If they gave words having in mind the thought that the

things were all always changing and flowing--at least I think
this is what they thought--and, if as it happens, this is not
how things are, they themselves are both such as to have
fallen into a kind of whirlpool and to drag us down after
them. Consider, Cratylus, that which I often dream. Shall
we say that there is a beautiful itself, a good itself, and one
each of the beings thusly?
It seems so to me, Socrates.
Let us consider this itself, not if some face is beautiful, or
something of this sort, and it seems that they all flow, but let
us ask whether the beautiful itself is always such as it is?
Necessarily (439cl-d7).

Socrates gets Cratylus to agree that the beautiful itself is not flowing,
but "is always such as it is." He does this, presumably, as a prelude
to explaining why Cratylus and others who accept that everything is
always flowing have "fallen into a kind of whirlpool. "
The reader would expect Socrates to explain that because the
Heracliteans reject the Platonic conception of reality and assert that
everything is subject to change, including the "beings" of justice,
beauty, etc., they accept a conception of reality that is inconsistent with
the possibility of knowledge. This, however, is not quite what
happens. Socrates does explain that the Heracliteans accept a
conception of reality that is inconsistent with the possibility of
knowledge, but in his argument, he makes them accept the paradoxical
proposition that everything is always changing in every way.
This strategy informs Socrates's subsequent remarks. He first
argues that if a given thing is in such flux, then, among other things,
no one can know how it holds. 439d7 continues as follows:

Can one rightly call it, if it is always slipping away, first

that this is, then that such as, or must not at the same time
that we are speaking it straightaway becomes other and slips
away and no longer thusly holds?
Necessarily.
How could that be something which never holds in the
same way? If it ever holds in the same way, then in that
time it is clear that nothing changes; if it always holds in the
 4.iv Heraclitus and Protagoras 113

same way and is the same, how could this change and move,
not departing from the form of itself?
In no way.
Nor could it be known by anyone. At the moment when
he who seeks to know is upon it, it becomes other and
different, so that which it is and how it holds can no longer
be known, and clearly there is no knowledge that knows
what holds in no way.
It is as you say (439d8-440a5).

Socrates next seems to argue that if everything is always changing in

every way, then there are no conditions that would count as
instantiating knowledge:

But we cannot even say that there is any knowledge, if all

things are changing and nothing remains. If this itself,
knowledge, does not change and cease to be knowledge, then
always knowledge would remain and be knowledge. If the
Form itself of knowledge changes, then at the same time as
it changes into a kind other than knowledge it would not be
knowledge, and if it always is changing, then there always
would not be knowledge, and by this reasoning there will be
neither nothing to know nor anything to be known. But if
there is always that which knows and that which is known,
if there is the beautiful, the good, and one each of the other
beings, then I do not believe there is any likeness between
these of which I am now speaking and flux and motion
(440a6-c 1).

These passages are obscure, to say the least, but they suggest that
Socrates argues against a straw man. To show that the Heracliteans
have fallen into a "whirlpool," Socrates seems to make them understand
their flow doctrine in a way that is unnecessarily weak and that is not
the most likely competitor to the Theory of Forms.
Although he proceeds more cautiously in the Theaetetus, the
outcome is the same. The main question is "What is knowledge?," and
in the first section of the dialogue Socrates plays midwife to
Theaetetus's answer that "knowledge is perception" (151e2-3). He
notes that Theaetetus has come out with a proposition that one might
114 Heraclitean flux

expect of an older and more "sophisticated" person. He says that

Theaetetus has answered as Protagoras would (l51eB-152a3), and the
conversation turns to whether the measure doctrine is true.
Socrates does not quickly dismiss the doctrine, as he does in the
Crarylus, but he rejects it for the same reason. He argues that it
wrongly entails that no man can be wiser than another (Crarylus
3B6c2-d2; Theaetetus 179alO-b5). After he rejects the measure
doctrine, Socrates turns his attention to the tlow doctrine:

We shall have to consider and test this moving being, and

find whether it rings true or sounds as if it has some flaw
(179d2-4).

Since, according to Theodorus, the Heracliteans are not clear about

what they mean when they declare that everything is always flowing
(17ge2-1BOc6), Socrates agrees to distinguish the kinds of flowing that
they could have in mind. He distinguishes spatial movement and
alteration (lBl bB-d7), and he and Theodorus agree that the Heracliteans
mean that everything is always tlowing in both ways (lB1dB-e4).
Socrates argues that, in this case, the flow doctrine has an absurd
consequence. The reasoning is not very clear, but Socrates seems to
argue that if at a given time a given thing is changing in every way,
then it is both F and not-F. His argument seems to depend on the
following three premises:

(a) if x is changing in every way, x is becoming not-F

and becoming F;
(b) if x is becoming not-F and becoming F, x is not yet
not-F and not yet F;
(c) if x is not yet not-F and not yet F, x is F and
is not-F.

Socrates and Theodorus first agree that if a given thing were just
moving through space, and not changing in every way, then throughout
at least part of its journey it would be a certain way, say, white. They
also agree that if it were also changing in every way, then it would not
be white:
 4. iv Heraclitus and Protagoras 115

If they were only moving through space and not altering,

we should be able to say of what sort the moving things
flow?
That is right.
Since, then, not even this abides, that what flows flows
white, but it changes, so that there is a stream of this very
thing, the whiteness, and it is passing over into another
color, lest it be convicted of standing still in this respect, can
one ever call it any color, so that he calls it correctly?
How can that be done, Socrates? Nor yet anything of that
sort, if, inasmuch as it is a stream, it is always slipping
away as you speak.
And what are we to say about a perception of any sort;
such as, seeing or hearing? Are we to say that as seeing or
hearing it remains in itselt?
Necessarily not if all things are in motion (l82c9-e2).

Socrates next appears to argue that the point holds for predicates and
their negations, that if everything is always flowing, then, in virtue of
(a) - (c), both not-F and F are true of the same subject:

Then it has no right to be called seeing any more than not

seeing, nor any other perception more than not, if it be
admitted that all things are in motion in every way?
No, we may not.
And so our answer to the question, What is knowledge?,
gave something which is no more knowledge than not.
It seems as if it did.
A fine way this turns out to be of making our answers
right. We were most anxious to prove that all things are in
motion, in order to make that answer come out correct, but
what has really emerged is that, if all things are in motion,
every answer, on whatever subject, is equally correct, to say
both that it is thus and that it is not thus--or if you like
becomes, as we don't want to use any expressions which
bring our friends to a stand-still.
You are quite right (l82e3-183a8).
116 Heraclitean flux

Theodorus agrees with Socrates, and one would expect the conversation
to be over. Socrates has shown that the flow doctrine is absurd, that
it entails that everything is both F and not-F.
Socrates, however, is not yet completely satisfied and decides to
press his case further. 183a8 continues as follows:

Except, Theodorus, that I said thus and not thus. One

must not say this thus, for this thus would no longer be in
motion. Nor again not thus, for this is not a motion. The
exponents of this theory need to establish some other
language, since, as it is, they have no words that are
consistent with their hypothesis--unless it would perhaps suit
them best to use not thusly mostly, being said in an
indefinite way (183a9-b5).

His point, apparently, is that the flow doctrine is so extremely radical

that it cannot be stated in the language of the day. This language
supports inferences such that any attempted statement of the idea that
stands behind the flow doctrine would entail that everything is F and
not-F, and the extreme Heracliteans cannot accept this consequence,
since in this case there would be a way in which things hold. In this
case, contrary to the Heraclitean doctrine that everything is always
flowing in every way, things would always have the property of being
F and not-F.

v. Two mistakes

Given Plato's argument against the Heracliteans, and its similarity to

his argument against the lover of spectacles, perhaps Aristotle failed to
distinguish the two arguments. Socrates, recall, forces the lover of
spectacles to admit that "each of the many things < is no> more than
it is not that which one says it to be" (479b9-1O). This is the very
same conclusion that, in the Cratylus and the Theaetetus, Socrates
forces the Heracliteans to accept. Aristotle, perhaps, could have
"carelessly assimilated" these two arguments.
The arguments, of course, are distinct. Socrates forces his
respective opponents to the same absurd conclusion on the basis of two
different assumptions. The lover of spectacles does not begin with the
 4.v Two mistakes 117

radical flow doctrine. He may accept this doctrine, but in the context
of the Socratic challenge, he denies that the Forms exist in reality as
substances. Instead of accepting that the essences of justice, beauty,
etc., are starting-points in reality, the lover of spectacles only
recognizes "the many conventions (POJLLJLCX) of the many" (Republic
479d3-4). The Heracliteans may be lovers of spectacles, but Socrates
does not force them into absurdity on this basis. He appeals to the
radical flow doctrine. He argues that "if all things are mOtion, every
answer, on whatever subject, is equally correct, to say both that it is
thus and that it is not thus" (Theaetetus 183a4-6).
If Aristotle failed to distinguish these two arguments, this alone
would not fully explain how he came to the mistaken views about the
development of Plato's thought. It would not explain why he believed
that Plato supposed that the sensibles are in radical flux. Aristotle must
be guilty of some further confusion, and once again, Annas makes
remarks that could in principle point in the right direction:

Aristotle's account of the theory of Forms can be

questioned. It is true that in the dialogues Plato does often
talk about Forms in a way which emphasizes their stability
and fixed nature in contrast with the changeable and
therefore unsatisfactory nature of physical objects (for
example, Symposium 207c-208b, Phaedo 78e-79d, Republic
485b, 508d, 534a). But the flux argument as Aristotle
presents it is nowhere to be found explicitly in the dialogues.
Indeed, some parts of the dialogues suggest that Plato could
not have accepted it. In the Theaetetus, which Aristotle
knew (cf. 10 lOb 11 ff.), Plato refutes the Heracleitean theory
that physical things are always in flux in every way ... (p.
153).

Annas correctly notes that Plato often contrasts the eternal and
unchangeable Forms with the transitory and changeable ordinary
objects of experience, and she quietly hints that Aristotle may have
misunderstood the intended contrast.
Plato does contrast Forms with the sensibles, and, indeed, a reader
could misunderstand this contrast. Plato does not make his intent as
clear as one might hope. Witness, e.g., the way he puts the contrast
in following passages from the Phaedo:
118 Heraclitean flux

Then let us go back to those entities to which we turned in

our earlier argument. Does the being itself, whose being we
give an account of in asking and answering questions,
always hold similarly and the same, or is it other at other
times? Does the equal itself, the beautiful itself, what each
thing is itself, that which is, ever admit of any change
whatever? Or does what each of them is, being uniform
alone by itself, hold similarly in the same way, and never
admit of any kind of alteration in any way or respect
whatever?
It must be unvarying and constant.
But what about the many beautiful things, such as men or
horses or cloaks or anything else at all of that kind, or
equals, or all things that bear the same name as those
objects? Do they hold in the same way, or are they
altogether opposite (7rap TOVpaPT£oP) of those others, and in
respect to themselves and to others are they almost (we; €'7rOe;
EL7r€LP) in no wise in the same way?
That is their condition. They never hold similarly
(7SclO-e5).

Now, Cebes, were we not saying a moment ago that

whenever the soul uses the body as a means to study
anything, either by seeing or hearing or any other
sense--because to use the body as a means is to study a thing
through sense perception--then it is dragged by the body
into the things that never hold in the same way and it
wanders about itself, and is troubled and dizzy, as if
stupefied, because of contact with such things?
Certainl y, Socrates (79c2-S).

Plato does not say that the sensibles are in radical flux, but his words
are not especially clear. A reader could misunderstand the intended
contrast. He could conclude that the Forms are never changing in any
way and that the sensibles are always changing in every way.
If Aristotle failed to distinguish the argument against the lover of
spectacles and the argument against Cratylus and the extreme
Heracliteans, this further confusion would explain how he could be
incorrect about the OrIgin of the Theory of Forms. On this
interpretation, Aristotle believed that Plato supposed that the sensibles
 4. v Two mistakes 119

are in radical flux. Aristotle also conflated the argument against the
lover of spectacles with the argument against the Heracliteans. He
believed that, in each case, Plato argued that if everything is in radical
flux, then knowledge is impossible. On the basis of these two false
beliefs, and on the basis of the true belief that Plato agreed with
Socrates that knowledge is possible, the argument from flux could have
been born. Aristotle could have concluded that Cratylus and Socrates
come together in Plato to issue in the Theory of Forms.
If Aristotle did make these two mistakes, and I do not claim that he
did, then, to borrow Plato's metaphor, the Metaphysics passages are
"whirlpools" that threaten to "drag down" those who seek to understand
the genesis of the Theory of Forms. These passages would incorporate
a misreading of the development of Plato's thought, and this misreading
would have stood as one of the longest standing misunderstandings in
the history of Platonic scholarship. The argument from flux, on this
interpretation of Aristotle's remarks, is not an argument that Plato
claims for himself. It is an Aristotelian assimilation of various thoughts
that Plato expresses in various parts of the corpus.

vi. Forms are not "in" sensibles

Although Aristotle could in principle have made these mistakes, an

alternative interpretation of his remarks is possible. Aristotle is much
more philosopher than historian. His explanation of the development
of Plato's thought is not likely to contain the sort of detail that would
please an historian. Aristotle is not especially interested in historical
detail. His concern is with the philosophical issue. He intends his
remarks to focus the fundamental difference between their two
metaphysics. Aristotle passes over in silence the points of agreement.
In the context of his remarks, he recognizes no necessity to voice his
agreement with Plato against the lover of spectacles. Like Plato,
however, Aristotle does not allow claims about justice, beauty, etc., to
depend on "opinions" and "fancies." In the context of his remarks,
Aristotle recognizes no necessity to agree that detinitional truths are
necessary for the possibility of knowledge. Plato and Aristotle both
work within the Socratic tradition. In his explanation of the origin of
the Theory of Forms, Aristotle's primary concern is show where Plato
went wrong. He insists that Plato went wrong in his conclusion that
120 Heraclitean flux

essences fit into reality by existing separately and as substances.

Aristotle explains that" < s > eparation is responsible for the difficulties
arising about the <Theory of> Ideas" (l086b8).
Thus, in his presentation of the "argument from flux," Aristotle's
motivation is philosophical. He is trying to highlight what he takes to
be the underlying mistake in Plato's attempt to fit essences into reality.
He seems to argue that the mistake is to suppose that, because the
sensibles are "flowing," they cannot provide a stable foothold for the
existence of essences, and that therefore essences must fit into reality
as substances and ontological starting-points. Aristotle makes Plato
argue from

(5) knowledge has starting-points

(2) the sensibles are flowing
and (6) If (2) is true, the sensibles are not starting-points
of knowledge

to the conclusion that

(4) the universals exist separately.

On the basis of the conditions for knowledge and the instability of

ordinary objects of experience, Aristotle makes Plato draw a conclusion
about how essences exist. He makes him conclude that they are not
"in" sensibles, but exist separately and as substances.
This interpretation has many important virtues. It is consistent with
the obvious fact that both Plato and Aristotle are philosophers of rare
depth and insight. Aristotle does not unfairly saddle Plato with more
problems than he deserves. He does not make Plato be a skeptic
concerning knowledge about the sensibles. Nor does he make Plato
accept the extremely implausible and paradoxical doctrine that these
objects are always changing in every way. Furthermore, given his
close proximity to Plato, his evident ability as a critic, and his deep
interest in the problems that motivate the Theory of Forms, Aristotle
is not likely to be wrong about Plato. On this interpretation, Aristotle
does not misread Plato. His remarks in the Metaphysics do not stem
from a "careless assimilation" of unrelated arguments that Plato
expresses in different parts of the corpus.
 4.vi Forms are not "in" sensibles 121

Another important virtue of this interpretation is that it helps to

shed light on Plato's reasons for concluding that the Forms are
substances. In the autobiography in the Phaedo, in his own attempt to
chart the genesis of the Theory of Forms, Plato argues for Socrates's
method of inquiry and argues against the method of the Ionian natural
scientists. His remarks in this passage are compressed, but his point
seems, at the very least, to be that the natural scientists recognize no
need to find a place in reality for essences because in their search for
knowledge they do not recognize the priority of definition. Aristotle's
explanation helps to amplify this line of thought. It suggests that Plato
believes that even if the natural scientists wanted to recognize the
priority of definition, they could not satisfy this desire because there is
no place for "intelligible" (VOTJTO~) objects in their conception of
reality. As Aristotle reports in book I of the Metaphysics, the natural
scientists in the Ionian tradition take as their starting-points certain
"sensible" (cxlaf)TJTo~) objects, such as portions of water, air, etc., and
they try to explain everything else that exists as modifications of these
basic objects. Plato rejects this approach as inadequate. He believes
that this conception of reality makes knowledge impossible because it
cannot account for the existence of the definitional truths that are
necessary for knowledge. The argument from flux suggests that he
believes that because the sensibles are "flowing," they cannot provide
a stable foothold for the existence of essences.
Yet another important virtue of this interpretation is that it makes
Aristotle's explanation be consistent with Plato's thought in the
Timaeus. The Timaeus revisits some of thoughts first aired in the
Phaedo, and as I shall show in the next chapter, Plato has his
character, Timaeus, argue that about matters involving fire, water, and
the other stuffs that are traditional in Greek cosmology, the Ionian
cosmologists accept a conception of reality that makes knowledge
impossible because it makes definitional knowledge impossible. He
argues that the Ionian scientists cannot answer the "What is F?"
question by pointing to nature and attempting to give an ostensive
definition. He maintains that the sensible objects of such demonstrative
pointings do not have enough "firmness" (49d7), and are not
"steadfast" (e3) enough, to provide "accounts" (M,),ot) that are
122 Heraclitean flux

"trustworthy and firm" (49b5) enough to provide for the possibility of

knowledge about matters involving fire, water, and the other stuffs.

vii. Against traditional interpretations

Many important scholars hold a different view. They maintain that

Plato was a skeptic concerning knowledge about the ordinary objects
of experience because, under the influence of Cratylus and the extreme
Heracliteans, he came to believe that the ordinary objects of experience
are always changing in every way. Sir David Ross, e.g., puts forward
this interpretation. He maintains that the evidence for it is nearly
omnipresent in the dialogues:

According to < Aristotle at Metaphysics 987a30> , Plato's

philosophy 'in most respects followed the Pythagoreans,' but
was modified by two other influences:--(1) an early
acquaintance with Heraclitean views, as represented by
Cratylus, and a consequent conviction that as sensible things
are always in flux, they cannot be the objects of
knowledge; ... ([1924), p. xlv).

The recognition of the flux of all sensible things and the

consequent impossibility of knowledge of them is present
through-out the dialogues as the underlying assumption
which does not need to be often emphasized because it is so
unquestioningly taken for granted (p. xlvii).

Plato was, then, as Aristotle says, convinced of the truth of

Heraclitus' doctrine that the world of sense is in constant
flux, and that what is in constant flux cannot be known. But
he was equally convinced that knowledge exists, and
therefore that there must be non-sensible entities which are
the objects of knowledge. Thus the theory of Ideas was built
on the foundation of Heracliteanism ([1953], p. 157).

Ross, indeed, is so confident that he is repeating a commonplace that

he does not bother much with evidence from the Platonic corpus. He
only cites three passages from the middle dialogues:
 4. vii Against traditional interpretations 123

According to Aristotle < in the Metaphysics at 987a32-b7 > ,

Plato's earliest philosophical association was with Cratylus
the Heraclitean, and he retained the belief that all sensible
things are in constant flux: but when he had come under
Socrates's influence he held that because of their
changeability not they but something else must be the object
of knowledge. This is just what we find in the Cratylus < at
439b4-440c1 > (p. 20).

Even without Aristotle's testimony we should know that

Plato had been influenced by Heracliteanism. There is a
striking passage in the Symposium < at 207d2-208b6> in
which he speaks of the transitoriness of all things
human--not only of hair, flesh, bones, blood, and the whole
body, but of habits, traits of character, opinions, desires,
pleasures, pains, fears, knowledge. In the Phaedo < at
78c10-79all> the same thought is repeated, and there the
transitoriness of such things is contrasted with the
eternalness of Fonos, equality itself, beauty itself (p. 156).

Ross cites the passages from the Symposium and the Phaedo to show
that the int1uence of Cratylus and the Heracliteans is not based solely
on Aristotle's testimony. Plato, however, never even mentions the
Heracliteans in these two passages, and given that they hardly demand
the interpretation that he puts on them, Ross must do more than merely
cite these passages if he is to support his position.
Although he is the most famous, Ross is not the only scholar to
endorse this interpretation of Plato. John Brentlinger, e.g., writes:

That < Plato> did accept < the flux doctrine> is certain
([1972aJ, p. 62). For the texts in which this is unqualifiedly
asserted, see Cratylus 439b, and Phaedo 78e and 79c. A
later, more detailed exposition, but one which leaves open
whether it is Plato's position, is to be found at Theaetetus
181cff (n. 5 on p. 62).

The flux position is that all particulars change in all respects

at all times. This is the position Plato sets forth in the
Theaetetus as the position of the Heracleiteans, and I take it
to be identical to the position held in the Phaedo, in which
124 Heraclitean flux

he describes sensibles as "never the same" (ouocxp.wr; KcxTa

Taho Phaedo 78e, 79c) (n. 3 on p. 61).

A short passage in the Cratylus (439b-440e) is the only place

prior to the Theaetetus where we find a moderately detailed
statement of Plato's position on change. The passage
contains a number of different arguments, and is rather
disjointed and obscure, yet is clear on three main points: (a)
Plato wants to assert that particulars are in flux; (b) he
believes that whatever is in flux is unknowable and even
incapable of being the object of ostensive reference; and (c)
he believes that whatever is in flux is intrinsically
indefinite--i.e., has no properties whatsoever ([1972b), pp.
133-134).13

Robert Bolton makes similar assertions, and he adds book V of the

Republic to the list of alleged support for this interpretation:

At the end of Book V < of the Republic> Plato draws a

distinction between two types of entities--what "purely is"
and what "at the same time both is and is not" (478d5-7).
The entities of the latter type make up the realm of
becoming. And Plato says of them that "they no more are
whatever anyone may affirm them to be than not"
(479b9-1O). Plato does not mean by this that these objects
have contradictory predicates true of them. The claim is
weaker than that. They may be beautiful in a way (7fWC;,
479bl) but if so they will also be ugly in a way. But,
strictly, they cannot even be both beautiful in a way and
ugly in a way. Whatever they may be said to be they are
only in a way, and in a way not (c3-5). Nothing one can
say of them is ever sufficiently qualified to be strictly true.
In this respect the doctrine of the Republic coincides with the
doctrine of the extreme Heracliteans ([1975), pp. 77-78;
emphasis in original).

< I > t is Plato's explicit aim in this passage that one cannot
have any knowledge of sensible objects. The only reason
which he gives for this contention, moreover, is that these
objects no more are than are not whatever one may say they
 4. vii Against traditional interpretations 125

are (479b). And this reason suffices only if we take it

literally (p. 78; emphasis in original).

< In the Phaedo, > the being-becoming distinction is

introduced by means of the contrast between those things
which are 'always constant and unchanging' and those which
are 'different at different times and never constant' (78c).
To put the contrast in this way does not by itself commit
Plato to the extreme flux doctrine ... , < b > ut subsequently
Plato goes much further than this and says that these entities
are 'exactly opposite' (-Trav TOVVCXV7iov, e2-e3) to those
things which never change in any way (d6-7). They are 'so
to speak, never constant with respect to those other things or
to themselves in any way' (e3-4). They are not only never
unchanging (/ly/OE7rOTE KaTCt mimx) they are never in any
way unchanging (OVO€7rOTE ovocxp.w<; KaTCx mim:x). And this
is the extreme thesis. Plato does qualify the claim that these
objects are never unchanging in any way by the addition of
the phrase 'so to speak' (we; ;;'7r0<; E't7rELV). And this phrase,
like its companion we; EL7rELV, often has the restrictive force
of 'generally speaking' in Plato. But these phrases, like 'so
to speak,' may also be used in ways that do not put such a
qualification on what is said. The phrase may be used as a
stylistic filler included for emphasis and best translated by a
phrase such as 'in fact.' That Plato is using the phrase in
this way is required by the context. He has just said of the
objects in question that they are 'exactly opposite' (7raV
TOVVCXV7iov) to those things which 'never in any way admit
of any change' (p. 82).

If Bolton's interpretation is correct, Plato accepted some sort of

"extreme" form of the flow doctrine and concluded that knowledge
about the ordinary objects is impossible.
Bolton rests his interpretation in part on the discussion of the
philosopher and the lover of spectacles in the Republic, but serious
obstacles stand in the way of his reading. In this passage, as he reads
it, "Plato's explicit aim ... < is > that one cannot have any knowledge
of sensible objects." This aim is exceptionally bold, and to accomplish
it, Bolton makes Plato appeal to an equally strong premise. He claims
that Plato comes to this skeptical conclusion on the basis of the
126 Heraclitean flux

paradoxical proposition that the ordinary sensible objects of experience

"no more are than are not whatever one may say they are."
Plato could in principle accept this paradoxical proposition, but it
would undermine one of the most important points in the Republic. It
would undermine his claim against the lover of spectacles. This claim
is based on the fact that the lover of spectacles accepts a conception of
reality according to which there can be no knowledge with respect to
whether a given 7rOALC; is just. This is why the lover of spectacles is
not fit to rule. He is "not very different from the blind" (484d4).
Only the philosopher "can keep looking back and forth, to justice,
beauty, temperance, and all such things as by nature exist, and can
compose human life with reference to these ... " (501b1-4).
The problem for Bolton's interpretation is not that it is contrary to
a principle of charity because it reveals that Plato is inconsistent,14 but
rather that Plato just does not say what Bolton claims for him. Plato
does not argue anywhere in the corpus that "one cannot have any
knowledge of sensible objects." In book V of the Republic, contrary
to Bolton's interpretation, Socrates argues that the lover of spectacles
cannot have knowledge about the sensible world. Because the lover of
spectacles denies that the Forms are substances, and thereby identifies
the one Form with the many conventions, he cannot know of any given
7rOALe; that it is just. For the properties that he uses his "What is F?"
question to ask about, Socrates insists that whoever denies that the
Form the F itself exists "itself according to itself" (aim) mO' aUTO)
cannot know of a given object of experience that it is F. Without this
Form, according to Plato, there is nothing to know because there is no
fact about what features something must have to be F.
Ross, Brentlinger, and Bolton cite passages in the Phaedo to
support their interpretations, but no where in this dialogue does Plato
give any clear indication that he believes that "all particulars change in
all respects at all times." As Bolton notes, the phrase we; E7rOe; E~7rE1p
has different uses. It can be used with "restrictive force." This
phrase is "often used to limit too strict an application of a general
statement, especially 7rae; ('all') and ovoEie; ('none')" (Smyth [1956],
section 2012.b; translations added in parenthesis). The phrase also can
be used in a way that it does not have this restrictive force, and Bolton
maintains that Plato uses it this way at 78e4. This use is "required by
 4. vii Against traditional interpretations 127

the context," according to Bolton, because Plato just said that the
Forms and the sensibles are 1faJl TOVJlaJlTioJl (e2-3).
If the "context" is fixed as Bolton maintains, it surely does nothing
to settle the question. Plato could use the phrase w~ E1fO~ Et1fEiJl in
either way, depending on how he intended his earlier use of 1faJl
TOUJlaJlTioJl, which need only be translated as 'altogether opposite,' not
as 'exactly opposite,' as Bolton translates it. If Plato intended to
restrict the quantifier, as I think, then 'almost' is an appropriate
translation. 15 If he intended leave the quantifier wide open, as Bolton
thinks, then 'in fact' is an appropriate translation.
To fix the context in a way that really settles the matter, it must
contain Socrates's subsequent remarks on the longevity of the body.
In these remarks, Socrates tries to convince Cebes that although the
human body is not eternal and unchangeable, it can remain intact for
a very long time:

Now you are aware that when a man has died, the part of
him that is seen, his body, which is situated in the seen
world, the corpse as we call it, although liable to be
dissolved and fall apart and to disintegrate, undergoes none
of these things at once, but remains as it is for a fairly long
time--in fact for a considerably long time, for if someone
dies with his body in beautiful condition, and in the flower
of youth, why the body that is shrunken and embalmed, like
those who have been embalmed in Egypt, remains almost for
an immensely long time, and even should the body decay,
some parts of it, bones and sinews and all such things, are
still almost immortal.. .(80c2-d2).

This passage should remove absolutely all doubt about how Plato
intends to use the phrase w~ I:'1fO~ E'L1fELJI at 78e4. Socrates and Cebes
agree that "the body remains as it is for a fairly long time" (c5-6).
Plato did not intend to use the phrase to mean "in fact. "16 He did not
believe that the sensibles are always changing in every way.
Ross and Brentlinger cite 439b-440e in the Cratylus to support their
interpretations, and although they merely state their position, and do
not argue that this passage shows that Plato believes that sensibles
"change in all respects at all times," other scholars have argued for this
128 Heraclitean flux

reading. Gulley cites this passage, and he argues that it shows that
Plato believes that, because they are in radical flux, the sensibles have
"no determinate characteristics and can have no significant description
applied to" them. He writes:

< T > he argument begins and ends with an assertion of the

thesis that non-sensible realities exist, and it is clear that
Plato considered the assumption of their existence essential
in opposing the flux doctrine. Wby is this assumption
considered essential? It is obvious that Plato does not think
it enough, in opposing the flux doctrine, to put forward the
logical argument that the doctrine has the apparently absurd
consequence that nothing can be known, but thinks it
necessary to select one special class of objects, the
non-sensible 'realities,' and to emphasize that these cannot
known if they are in flux. And his reason for thinking this
necessary must be that he considered that if these
non-sensible 'realities' were in flux then it would in fact be
the case that nothing can be known. His argument thus
implies his acceptance of the flux doctrine as true for
sensibles (pp. 71-72).

The position is, then, that Plato, at least implicitly and

almost certainly explicitly also, is accepting the flux doctrine
as true for sensibles. He had, of course, earlier in the
Phaedo made clear his view that sensibles are constantly
changing and that his is reason for condemning perception as
incapable of yielding truth (78c-80b, 83b). And that he
accepted the flux doctrine as true for sensibles is stated by
Aristotle in two passages in the Metaphysics (987a32-b7,
1078b12-17) where he is explaining that Plato's acceptance
of the flux doctrine as true for sensibles led him to postulate
non-sensible Forms in order to save the possibility of
knowledge (p. 73).

Thus the evidence of the Cratylus is by no means the only

evidence for Plato's acceptance of the flux doctrine. Yet it
does deserve special emphasis and consideration. For it is
the only place in tbe dialogues where Plato clearly shows
 4. vii Against traditional interpretations 129

that he considered the specifically Heraclitean doctrine of

flux to be true for sensibles (p. 73).

< I> n the Cratylus Plato affirms that no signitkant

description can be applied to what is in flux; one cannot
describe it as 'that' or as 'of such a kind,' because it is
never in any determinate condition (pp. 73-74). <T>he
Cratylus clearly shows that Plato accepted the Heraclitean
doctrine of flux as true for sensibles, and hence accepted as
true for sensibles the consequences that what is in flux has
no determinate characteristics and can have no significant
description applied to it (p. 75).

Gulley maintains that the fact that Plato emphasizes the Forms in his
argument against Cratylus at 439c l-440c 1 would be otherwise
inexplicable if Plato did not believe that the only point in dispute is
whether these objects are in radical flux. Gulley maintains that Plato
is concerned to "save the possibility of knowledge" and agrees with
Cratylus that there is no knowledge with respect to the sensibles, since
they are caught up in radical flux.
Gulley's argument is very weak. First of all, the passages from the
CratyLus are much too unclear to provide justification for a such
substantive claim. If, as Gulley claims, the end of the CratyLus "is the
only place in the dialogues where Plato clearly shows that he
considered the specifically Heraclitean doctrine of flux to be true for
sensibles," then the evidence is inconclusive. Secondly, unless Plato
later changed his mind, the CratyLus does not show that he believes that
an explicit introduction of a third realm of Forms is necessary to refute
the flux doctrine. In the Theaetetus, e.g., Socrates convinces his
interlocutors to reject the flux doctrine, but he does not introduce the
Forms. There is no reason, furthermore, to suppose that Plato
emphasizes the Forms because he believes that he and the Heracliteans
are otherwise in agreement. He could easily believe that the most
important disagreement is over these objects.
This interpretation is intrinsically likely, since if he were to accept
the paradoxical view that the sensibles were in radical flux and that
nothing is true about them, he would undermine the main question in
the Cratylus. The topic of conversation is about "words," and although
130 Heraclitean flux

Socrates never asks his interlocutors to agree that language users

typically know the type of a given word-token, surely this is an
assumption that all parties take to be evident. If Plato really believed
that all sensibles were in so much flux that no one could know anything
about them, I doubt that he would have so much enthusiasm to write
at such length about the "correctness of words."
Gulley is mistaken. The Cratylus does not "clearly" show that
Plato believed that "Heraclitean flux doctrine is true for sensibles" and
that they consequently have "no determinate characteristics."
Irwin accepts the broad outline of Aristotle's explanation of the
Theory of Forms, but unlike Ross, Gully, Brentlinger, Bolton, and
other like-minded scholars, he tries to supply the missing detail so that
Plato is neither a skeptic nor a radical Heraclitean about the ordinary
objects of experience. Irwin writes:

Some have thought there is ample evidence that Plato

believes the sensible world is constantly changing and
therefore unknowable, just as Aristotle appears to say. But
this alleged evidence is faulty ([1977a], p. 2).

According to Aristotle, Plato's Heracleitean views, applied

to Socrates' search for definitions stimulated him to separate
the Forms; since sensible things are always in flux, there can
be no knowledge of them ... ([1977b), p. 148). The
separation of the Forms relies, as Aristotle says, on a
doctrine of flux (p. 153).

Aristotle comments that Plato separated the Forms as objects

of knowledge because sensible things are always in constant
change. Heracleitean flux, and so unknowable .... I want to
show that Aristotle is right, but only if we reject the most
common interpretation of his remarks on flux ([1977a), p.
1).

The key to Irwin's more charitable proposal is novel interpretation of

the flow doctrine that Aristotle attributes to Plato. Irwin rejects what
he takes to be "the most common interpretation" of this doctrine. He
suggests, instead, that it should be understood as a proposition about
the compresence of opposite properties:
 4.vii Against traditional interpretations 131

< Heraclitus's> "unity of opposites" includes ... things with

compresent opposite properties--the road up and down, the
straight and crooked writing, the food which is good (for
some people) and bad (for others) (Heracleitus, B59-61).
Let us call this aspect-change (a-change): x a-changes iff x
is F in one aspect, not-F in another, and x is in the same
condition when it is F and when it is not-F (e.g., x is big in
comparison with y, small in comparison with z) (p. 4;
emphasis in original).

Irwin suggests, furthermore, that Plato would not go wrong if he were

to accept the flow doctrine so construed:

Plato can fairly claim that sensible properties are in flux,

because opposite moral properties are compresent in them,
and they a-change from being beautiful (e.g.) to being ugly
in different situations--bright color changes from being
beautiful in this temple to being ugly in that one (p. 10).17

Irwin thus squares Aristotle's explanation of the origin of the Theory

ofthe Forms with his own explanation of Plato's thought. Irwin argues
that Plato has fallen prey to a simple confusion, that he argues for the
ontologically innocent proposition that justice, beauty, etc., are not
definable in terms of sensible properties, but invalidly jumps to the
metaphysical conclusion that the Forms are substances.
I fully agree with Irwin's attempt to understand Aristotle's
explanation in a way that makes Plato be neither a skeptic nor a radical
Heraclitean about the ordinary objects of experience, but I do not
believe that his interpretation is very plausible. To save Plato from the
twin evils of skepticism and radical flux. Irwin subjects him to a
straightforwardly bad argument for the central proposition in the
Theory of Forms. Plato fell into this bad argument, according to
Irwin, because he did not fully appreciate that "sensibles" include both
sensible properties and sensible particulars ([1977b], p. 155). Plato
could in principle have been confused in this way, but the dialogues
provide very little evidence for this interpretation.
Indeed, as I have already shown, the Phaedo and the Republic
suggest a very different interpretation. They suggest that Plato came
to believe that the Forms are substances, not because he fell into a
132 Heraclitean flux

confusion over sensibles, but because he came to believe that no other

way of fitting essences into reality would validate the leading
assumptions in Socrates's method of intellectual inquiry. Since
Aristotle's remarks in the Metaphysics passages can easily support a
reading that confirms this interpretation, this is clearly how they should
be understood.
 5

FORMS AND THE FOUR KINDS OF STUFF

i. Introductory remarks

Near the beginning of the second main part of the Timaeus (47e-69a),
Plato once again argues against an opponent who accepts an ontology
in which essences are not starting-points. l His opponents are the
Ionian cosmologists. As starting-points in their ontology, they accept
neither the Receptacle nor Forms. The starting-points in their ontology
are sensible objects, such as the portions of fire, water, etc., that are
traditional in Greek cosmology. Plato has his character, Timaeus,
argue against this ontology. 2 He argues that the Ionian ontology is
part of a conception of reality that is inconsistent with the possibility of
knowledge about matters involving stuffs of the four kinds.

ii. Eternal Forms versus transitory sensibles

After the first remarks, and in the prelude to the first main part of the
dialogue, the topic of conversation turns to a question about "the whole
firmament" (28b2) or "cosmos" (3).3 The question is whether this
"thing" has come into being or exists eternally (27c5). Timaeus tries
to determine the answer by deduction from a more general framework.
This framework consists of two propositions about objects and how
they exist. Timaeus maintains that "sensible" (ala(}1JT{x;) objects come
into being and that "intelligible" (V01JTOr;) objects exist eternally.
Timaeus introduces this metaphysical framework in one of the most
important passages in the corpus. It reads:

Now, in my opinion, we first of all have to make this

distinction: what is that which is always, and has no birth
('YEV€O"t<;"), and what is that which comes into being
[always),4 and never is? The one is grasped in thought with
the help of reason, i!, always in the same way; the other
opined in belief with the help of non-rational perception,
134 Forms and the four kinds of stuff

comes into being and passes out of being, but really never
is (27d5-28a4}.5

Armed with "this distinction," Timaeus is ready to decide whether the

cosmos has come into being or exists eternally. He makes this decision
in the following passage:

The universe--or cosmos or whatever name it should receive,

this let us name it--we must first inquire about it, which we
laid down that it is necessary to inquire about all things,
whether it was always, not having a source of birth, or came
into being, beginning from some source. Came into being;
for it can be seen and touched and it has body, and all such
things are sensible, and we said that the sensible, grasped in
belief with the help of perception, comes into being and is
brought forth by birth (28b2-c2).

Timaeus decides that the cosmos has not always existed. He argues
that because it can be sensed. i.e., seen, touched, etc., it is a sensible
object. He argues that because all sensible objects have come into
being, the cosmos has come into being.
These two passages are important for Platonic scholarship, not so
much because they contain this argument that the cosmos is sensible
and has come into being, but because they show Plato once again
dividing objective reality into an intelligible realm and a sensible realm.
This division should be very familiar from the Phaedo. Socrates gets
Cebes to agree that there are intelligible and unchangeable Forms, and
he gets him to agree that there are sensible and changeable things "that
bear the same name as" the Forms (78e2). In these two passages from
the Timaeus, Plato does not explicitly mention either the Forms or the
ordinary sensible objects of experience. Later in the dialogue,
however, in passages in which he again discusses objects that come into
being and those that exist eternally, he is more explicit. Witness his
remarks in 48e2-49al and 52al-7:

Our new starting-point in describing the universe must be a

fuller classification than we made before. We then
distinguished two kinds; but now a third must be pointed
out. For our earlier discourse the two were sufficient: one
 5.ii Eternal Forms versus transitory sensibles 135

postulated as a paradigmatic Fonn, intelligible and always in

the same way; second, an imitation (1lI.p:T/lla) of the
paradigm (-mpaoH'Ylla), having birth and being visible.

We must agree that one kind is the Form that holds in the
same way, not brought forth by birth and indestructible,
neither receiving anything else into itself from elsewhere nor
itself entering into anything else anywhere, invisible and
otherwise imperceptible, this is that which intelligence
considers. The second is that which bears the same name
and is like (OIlOLOr:;) it, sensible, brought forth by birth,
carried about always, comes into being in a certain place and
again from thence passes out of being, grasped in belief with
the help of perception.

In these passages, as earlier in the Phaedo, Plato has his character

contrast the Forms with the ordinary objects of experience. Timaeus
says that the Forms are eternal and intelligible. He says that the
ordinary objects are transitory and sensible, and he says that these
sensible objects take their names from the Forms.
If this were all Timaeus said about the Forms and the ordinary
objects of experience, his conception of reality would not be
significantly different from the one that Socrates put forward earlier in
the Phaedo, but, as perhaps is already evident from 48e2-49al,
Timaeus has much more to say about the most basic categories of
existence. He is about to place an object on philosophical center stage
that Socrates did not mention, or even suggest, in the Phaedo. In order
to try to give a more accurate description of the cosmos, Timaeus is
about to introduce a new "starting-point" and "third kind" into his
ontology. This new starting-point is the Receptacle.

iii. Starting-points in ontology

Plato introduces the Receptacle in an effort to solve a certain problem

concerning the stuffs ofthe four kinds that are "starting-points" (apxo:i)
in traditional Greek cosmologies. To solve this problem, Plato
modifies his pre-Timaeus conception of objective reality. 48e2-49al
continues as follows:
136 Forms and the four kinds of stuff

A third we did not then distinguish. thinking that the two

would suffice. but now. it seems, the argument compels us
to attempt to bring to light and to describe a form difficult
and obscure. What nature must we conceive it to possess
and what part does it play? This more than anything else:
that the Receptacle is. as it were, the nurse of all birth
(49al-6).

The third kind is the thing of space, being always, not

receiving destruction. providing a seat for all that has birth
(52a8-bI).

Plato believes that he needs some special object in addition to the

objects that Socrates accepted earlier in the corpus and that Timaeus
accepted earlier in the dialogue. ,
The problem whose solution requires the Receptacle is about the
"birth" of the stuffs of the four kinds. He says that "men" have false
beliefs with respect to these stuffs, and he intends to set them straight.
He remarks:

To this day no one has explained their birth. but as if men

knew what fire and each of the others is. we speak of them
as starting-points (apxai). positing them as letters of the
universe. when in fact anyone who has ever so little
intelligence should not rank them even so low as syllables
(48b5-c2).

Timaeus complains that although their "birth" is in need of explanation,

the many commonly and wrongly suppose that these stuffs are among
the starting-points and "letters" of the universe.
Timaeus does not explicitly blame natural scientists in the Ionian
tradition, but the prominence of their theories partly explains why the
many have fallen into their false beliefs. The Ionian (j>vuLOM,,(OL started
with the four kinds of stuff and used them to explain the essences of all
other material kinds. They maintained that, in various combinations,
portions of the stuffs constitute all other kinds of material objects. 6
Timaeus intends to argue against this ancient orthodoxy.
To demote the four kinds from the rank that the Ionians accord
them, Timaeus provides a formal explanation of their existence. He
 5. iii Starting-points in ontology 137

shows that sensibles of these kinds are not "letters" in the universe, but
are explained in terms of more basic objects. One way he could
proceed is to explain, in the style of the Ionian scientists, that portions
of stuff of the four kinds are constituted from portions of even more
basic kinds of stuff, but there are compelling reasons for supposing that
this is not how he actually proceeds. He never mentions any more
basic kinds of stuff, and he says that the Demiurge makes the cosmos
out of the four kinds (31b6-33b1, 52d2-53b7, 69a6-c2).7
These reasons are perhaps proof enough, but there is another, even
more telling reason for supposing that Timaeus and the Ionian scientists
are not simply engaged in a family dispute over what stuffs are basic.
Timaeus realizes that if he were to posit more basic kinds of stuff to
constitute the traditional four kinds, he would not properly put to rest
the request for explanation that he presses in 48b5-c2. He does not
depart from the Ionians on the basis of an empirical discovery. He
does not deny that the four stuffs are starting-points because he and his
friends have discovered some kind of stuff that the Ionian researchers
have overlooked. Timaeus departs from them on the basis of a belief
about the possibility of knowledge and the nature of explanation. He
believes that only intelligible objects, not sensible objects, are fit to be
starting-points in a conception of reality that is consistent with the
possibility of knowledge. Timaeus believes that intelligible objects are
the starting-points for explanations of sensible objects.
Timaeus maintains that the Receptacle "shares in the intelligible"
(51a7-b1, 52bl-2), and together with the relevant Forms, which are
also "intelligible" (51 c5, 52a4), it figures in his formal exp lanation of
the existence of the four kinds of stuff. He does not explain its role
very clearly.8 He suggests that it is somehow like the role that the
female generator plays in human sexual generation:

We must conceive of three things: that which comes into

being, that in which it becomes, and that from which in
whose likeness that which comes into being is born. Indeed,
we may fittingly compare the Receptacle to a mother, the
from which to a father, and the nature between them to their
offspring (50c? -d4).
138 Forms and the four kinds of stuff

In the ancient world, according to a standard theory of sexual

generation, the male provides the form and the female provides stuff
that somehow receives this form.9 Plato makes Timaeus claim that the
metaphysical situation is analogous.
The Receptacle, of course, is not analogous to the female generator
because both provide stuff. The four kinds are the basic kinds of stuff,
and the Receptacle is part of Timaeus's formal explanation of these
kinds. His suggestion seems to be that the Receptacle provides a
region to have sensible properties and that the Forms provide the
logical claims that the presence of certain properties is equivalent to the
presence of certain stuffs.1O Witness his remarks:

< T > he part of the Receptacle that is igneous appears as

fire, and the part that is aqueous as water, and as earth and
air as such parts receive imitations of these (51b4-6).

By providing such formal explanations, Timaeus demonstrates that

portions of the four kinds of stuff are not starting-points. He explains
these sensible objects in terms of intelligible objects. He explains, e.g.,
that portions of fire are regions of the Receptacle that possess certain
properties and that the Form the fire itself marks the presence of these
properties as equivalent to the presence of some fire. 11
The text is compressed,12 but Timaeus seems to argue for his
explanations in a way that should be familiar to Plato's readers. In his
autobiography in the Phaedo, and in book V of the Republic, Socrates
argues that his opponents mistakenly embrace an ontology that is
inconsistent with the possibility of knowledge about certain matters.
Timaeus seems to believe that the Ionians make the same mistake. He
argues that if, as is true in their ontology, the starting-points for reality
are sensible objects, such as the portions of fire, air, etc., then to
"know what fire and each of the others is" (48b6-7) is impossible, and
if such knowledge is impossible, then equally impossible is knowledge
that a given portion of stuff is, say, fire.
Indeed, given the Ionian ontology, the introductory remarks in the
Timaeus encourage the reader to expect such an argument. Critias says
that Timaeus "is better at astronomy than the rest of us and has made
knowledge of the nature of the universe his chief study (27a3-5), but
he does not mean that he is a natural scientist with no philosophical
 5. iii Starting-points in ontology 139

trammg. Earlier in this same conversation, Socrates says that Timaeus

has "returned from the heights of philosophy (20a4-5). Plato's readers
should be surprised, therefore, if Timaeus is not concerned with the
ontology necessary to answer the "What is F?" question. Earlier in the
Phaedo, Plato pays tribute to Socrates for helping to canonize this
approach to intellectual inquiry.
Although they have not made the connection to Socrates's
arguments in the Phaedo and the Republic, other scholars have
suggested that Timaeus is concerned with the ontology that is necessary
for there to be answers to the "What is F?" question. Edward Lee
made this point years ago in his 1967 paper "On Plato's Timaeus,
49D4-E7." He writes:

With this analysis, we can now also trace some relations

between < 49d4-e7 > and its wider context in the Timaeus.
Plato had earlier complained that people think they know
well enough "what the elements are" (48B5-7 1'01' ... w~
€iOOaLI' 7rUP OTL 7rOTE € aTL v, KTA.). The view he means, no
doubt, was that these elements are just the material stuffs
that we perceive.... On that view, anyone who seeks a
M'Yo~ of the elements--by asking, for instance, Ti 7rOT' ;;'CTTL,
TO 7rVp:--might (in principle, at least) be answered by
presenting him with an actual sample of fire, pointing to it,
and saying, TOVTO 7rUP EaTL (pp. 24-25). In short, they hold
that there can be "ostensive definition" of the elements'
names (n. 61 on p. 25). In 49B-E, Plato repudiates this
view on the semantic ground that no proper M'Yo~ can be
had in reference to phenomenal substances (p. 25).

On Lee's interpretation, Timaeus denies that adequate "accounts"

(AO)'OL) for the terms 1£'Vp ('fire'), V-owp (,water'), can be given by
"ostensive definition." Lee says that Timaeus denies that a "proper
AO)'OC; can be had in reference to phenomenal substances." I reject the
rest of his interpretation, but I fully agree on this point. 13
The passages necessary to establish my interpretation are elliptical
and difficult, but against the background of Plato's previous arguments
in the Phaedo and in the RepUblic, I believe that the evidence is clear
enough. In these previous dialogues, Plato has his character argue
against those who try to provide for the possibility of knowledge with
140 Forms and the four kinds of stuff

an ontology in which essences are not starting-points. I believe that a

similar argument occurs in the Timaeus. Timaeus argues that the
Ionian ontology is inconsistent with the possibility of knowledge about
matters involving fire, water, etc., because it cannot provide for
adequate accounts for the terms that identify these kinds.
Timaeus begins with the assertion that "the argument compels"
(49a3) him to introduce the Receptacle as somehow "the nurse of all
birth" (5-6), and to help his interlocutors "more clearly" (7) understand
this conclusion and his argument, he works through a problem
concerning "fire and its consorts" (bl-2). He presents this problem in
the following passage:

Of them, to say which is the sort of thing that one must

really call water rather than fire, and which anyone, rather
than some other or every other, in such a way as to employ
a trustworthy and firm account (M'Yo~), is hard (2-5).

Although Timaeus is concerned with nouns, not adjectives such as

OI.KO:LO~('just'), the problem that he describes appears essentially the
same as the one that Socrates earlier put to the lover of spectacles.
Timaeus does not work against a named opponent, and he does not
work with a lot of fanfare about knowers versus opiners, but the
problem that he describes appears to concern knowledge, and he seems
to argue that those who reject the Receptacle and the Forms cannot
solve this problem. He argues that those who accept such an ontology
cannot explain the accounts for the terms 'vowp (,water'), 1fVp ('fire'),
etc., in a way that makes them "trustworthy and firm" enough to
provide for the possibility of knowledge.
Before Timaeus shows how he meets this challenge, he seems to
intend his interlocutors to see that the Ionian cosmologists cannot meet
it. This seems to be his intention in two examples, the cycle of
becoming example and the gold example. These example situations
involve portions of the four kinds of stuff and objects of various
geometrical shapes:

What we now call water we see, so we think, condensing

and becoming stones and earth, dissolving and dispersing
again this same thing < becoming> wind and air... from
 5.iii Starting-points in ontology 141

water < comes> earth and stone once more, thus in a cycle
passing on their birth, so it seems (49b7-c7).

< Suppose> someone were molding all shapes in gold, and

never stopped remolding each into every other (50a5-7).14

With respect to each of these examples, Timaeus makes an assertion

about a situation in which change occurs. His intent does not appear
to be to argue that because objects change over time, no one can pick
out some object as an object of a given kind. How this is possible is
the problem that he has just mentioned at 49b2-5 and that he has just
said is "hard" (49b5), but not impossible to solveY Timaeus's intent,
rather, appears to be that such demonstrations are impossible if the
accounts are not "trustworthy and firm."
Timaeus seems to conclude that the Ionian cosmologists accept a
conception of reality according to which one cannot know of a given
object what it should be called:

Since none of them ever appears the same, of which sort of

them can one confidently assert to be this (rouro), whatever
it may be, and not another, without being ashamed of
himself. It is not possible (49c7-d3).

If someone were to point at some one of them, and ask what

it is, by far the safest answer with regard to truth is to say
that it is gold, but the triangle and whatever other shapes
came to be in the gold, never to say these are, for they
change during placing ... (50a7-b4).

Since the Ionians do not accept the Forms, they cannot put forward
such objects to determine that regions of the Receptacle with a certain
property are portions of a certain kind of stuff. They must provide
some other way of investing the terms 'fJowp (,water'), 7rVP (,fire), etc.,
with accounts that are consistent with the possibility of knowledge, and
Timaeus seems to believe that they cannot solve this problem.
Timaeus seems to believe that the Ionians cannot solve this problem
about "fire and its consorts" (49b 1-2) because they cannot provide
accounts that specify a "such-and-such" (TOLOVTOV)16 that recurs in all
142 Forms and the four kinds of stuff

and only the places and at the times that portions of a given stuff come
into existence. 49b7-d3 and 50a7-4 continue respectively as follows:

By far the safest course is for us to propose to speak

concerning these in the following way. Always that which
we see coming into being at different times and in different
places, for example. fire, not this, but what is such-and-such
on each occasion to call fire, nor water this, but what is
such-and-such always < to call water>, nor another
< this >, as if it had some firmness, the things at which we
point using the expressions this and that, we think we make
something clear. For it runs off, not staying behind for of
this or this or to this, or for any utterance that marks them
out as beings that are steadfast. These each not to say. but
rather what is such-and-such on each occasion always
recurring like (0110£0<;) to say, for indeed fire is what is
throughout such-and-such, and thusly the whole of what has
birth (49d3-e7).17

but if what is such-and-such he should be willing to accept,

with some degree of safety, be well content (50b4-5).

The point in these passages is not at all straightforward, but in the light
of Socrates's previous arguments against the natural scientists and the
lover of spectacles, a reasonable conjecture is that Timaeus intends to
suggest that to provide "trustworthy and firm" accounts for the terms
vowp ('water'), 7rUP ('fire'), etc., the Ionian cosmologists cannot fall
back on ostensive definition. They cannot claim that language users
create such accounts by pointing to nature and uttering sentences such
as TovTo vowp fun ('This is water') and TovTo 7rVP fun ('This is
fire').18 Timaeus suggests that this practice is not sufficient to provide
"trustworthy and firm" accounts because it is not sufficient to invest a
given term with an account that specifies the TOLOVTOV (such-and-such)
that recurs in all and only the places and times that portions of the
relevant kind of stuff comes into existence.
Timaeus is correct, if this is his point, but he is hardly charitable.
Terms for kinds of stuff cannot be- meaningfully introduced into
language by simply holding up a token word and pointing to the world.
If the Ionians thought otherwise, they were mistaken. The procedure
 5.iii Starting-points in ontology 143

is much more complicated, but Plato makes Timaeus show absolutely

no interest in trying to uncover it.
One reason, perhaps, for this lack of interest is that Plato just did
not think that any kind of observable linguistic practice could be
sufficient to establish the objectivity, the ground for the existence of a
distinction between correct and incorrect sincere application of a
concept. that is necessary for the possibility of knowledge. In the
Republic, certainly. with very little argument. he has Socrates assert
that if essences must tind their place among the "conventions" (I!Op,~p,o:)
of the lover of spectacles, then objectivity and knowledge disappear and
all that remains are the fancies of individuals and groups. Plato may
not have not seen any promise in the attempt to ground objectivity in
the linguistic practices of language users. 19
Another possible interpretation of the passages, which attributes to
Plato a slightly cruder way of making essentially the same point, is that
Timaeus rejects the Ionian ontology because he believes that if one
were to accept it, then he would have to make objects that are subject
to change playa role that can only be adequately played by objects that
are not subject to change. On this interpretation, because of their
ontology, the Ionian scientists make the same mistake that Socrates's
interlocutors make in the early definitional dialogues. They put
forward an example in the place of a definitional answer to the "What
is F?" question. Timaeus demands an answer of the form TOWVTOI!
uDwp fan (,Such-and-such is water'), but the Ionian scientists try to
satisfy him with an ostensive definition. They point to nature and utter
sentences such as TOVTO ~Dwp fan (,This is water'). To invest the
terms uDwp ('water'), 1fiJp ('fire'), etc., with accounts that are
"trustworthy and firm" enough to provide for the possibility of
knowledge, the Ionians must rely on demonstrative pointings. Timaeus
insists that these pointings cannot result in adequate accounts. The
problem, according to Timaeus, is that the Ionians point to objects that
lack "firmness" (49d7) and are not "steadfast" (e3).
Timaeus maintains that only in terms of a third realm of Forms can
there be accounts for the terms vDwp ('water'), 1fiJp ('fire'), etc., that
are "trustworthy and firm" enough to provide for the possibility of
knowledge. He does not mention the Forms in these two passages, but
he tips his hand in 49d3-e7 with his use of Op,OWI!. He intends for the
relevant Forms to provide the information necessary to fill the TOWUTOI!
144 Forms and the four kinds of stuff

schema in sentences of the form TOLOVTOP uowp €un ('Such-and-such is

water') and TOLOVTOP 7rVP fun (,Such-and-such is fire'). The Ionian
ontology, Timaeus argues, is part of a conception of reality that is
inconsistent with the possibility of knowledge:

Is there such a thing as Fire just in itself, or any of the other

things which we are always describing in such tenns, as
things that exist themselves according to themselves (au7C"X
Ka()' aimfi)? Or are the things we see, or otherwise perceive
by the bodily senses, the only things that have such reality,
and has nothing else apart from these any son of being at
all? Are we talking idly whenever we say that there is such
a thing as an intelligible Fonn of anything? Is this nothing
more than a word (Slb7-cS)?

If intelligence (voue;) and true opinion (06~a O:A1/()~e;) are

two different kinds, then these things--Fonns that we cannot
perceive but only think ·of--cenainly exist according to
themselves, but, if, as appears to some, true opinion in no
way differs from intelligence, then all things we perceive
through the bodily senses must be placed as the most
established (5Id3-7).

Now we must affinn that they are two different things, for
they are distinct in origin and unlike in nature. The one is
produced in us by teach.mg, the other by persuasion; the one
is always by a true account, the other is without an account;
the one cannot be shaken by persuasion, the other can be
won over; and true belief, we must allow, is shared by all
mankind, intelligence only by the gods and a small number
of men (Slel-6).

Timaeus argues that the Forms the fire itself, the water itself, etc.,
must exist "themselves according to themselves" (aim:i Kaf)' aUT&')
because with respect to issues involving the four kinds, "intelligence"
is distinct from "true belief." Thus, in the Timaeus, just as previously
in the Phaedo and in the RepUblic, Plato argues that the Forms are
substances by arguing that only ontologies in which essences are
starting-points are part of a conception of reality that is consistent with
the possibility of knowledge.
 5. iv Against Cherniss 145

iv. Against Cherniss

Any interpretation of 47e-53c must be judged against alternative

possibilities, and there is no shortage of such interpretations. These
passages have stubbornly resisted analysis, and a wide diversity of
interpretations has an uneasy co-existence in the literature.
Harold Cherniss' s influential 1954 paper" A Much Misread Passage
of the Timaeus (49c7-50b5)" suggests a natural way to divide this
literature. In his very thorough critical review of previous translations
and interpretations, Cherniss vigorously dissented from the traditional
resolution of the ambiguity in 49d-e. Furthermore, he supposed that
the passage should be interpreted on the assumption that Plato accepts
the Heraclitean doctrine that the sensibles are always flowing in every
way and are thus not possible objects of knowledge.
These two aspects of Cherniss's interpretation provide the lines of
division. The most general division is with respect to an ambiguity in
49d-e. On this point, scholars are either for the traditional resolution
or are for some version of Cherniss' s proposal. Within in this division,
the alternative interpretations again divide with respect to the
Heraclitean flow doctrine. On this point, the alternatives do not divide
so neatly. Some scholars reject Cherniss's assumption, but otherwise
they do not present a unified front.
Cherniss argues, as I mentioned, that Platonic scholars have
typically mistranslated and misunderstood Timaeus 49c7-50b5. The
crucial phrase in the passage is d5-7. The following is an austere
translation of the words in the order in which they occur in the Greek
phrase, and without the quotation marks that translators typically add
to the text:

... not this, but what is such-and-such on each occasion to

call fire, nor water this, but what is such-and-such always ...

The Greek behind this phrase is ambiguous. The words TOI)70 ('this')
and TOLOVTOP (,such-and-such') could either be primary or secondary
objects of the verb 7rpo(JCX:YOPVEtP ('to call').
According to the traditional resolution, which Cornford, e.g., reads
into his translation in his 1937 book Plato's Cosmology, the words
146 Forms and the four kinds of stuff

70U7O and 70WV70/l are secondary objects of the verb 7rPO(]cxyopUEL/I.

Cornford thus translates d5-7 as follows:

... we should speak of fire, not as "this," but as "what is of

such and such of quality." nor of water as "this," but always
as "what is such and such a quality" ...

Those who accept this resolution suppose that Plato's point, very
roughly, is that the untutored have made a category mistake. They
have failed to realize that fire, water, etc., are qualities, not stuffs that
constitute particular objects.
The other possibility is to take the words 70070 and 70WV70/l as
primary objects of the verb. Cherniss adopts this reading. He
translates d5-7 as follows:

... not to say "this is fire" but "what on any occasion is such
and such is fire" nor "this is water" but "what is always
such and such is water" ...

Cherniss cites many reasons for his reading, but his philosophical
reason is that Plato accepted the extreme Heraclitean position about
sensibles and the impossibility of knowledge.
Although I did not translate the passage this way, but let the
ambiguity come through into the English, I agree with Cherniss that
this is the way to resolve the ambiguity. Both readings are possible,
but the Phaedo and the Republic provide the context necessary to
indicate the intended assertion. In the light of Plato's previous
arguments against the natural scientists and the lovers of spectacles,
Timaeus appears to be trying to show his interlocutors that "firm and
trustworthy" accounts for the terms uowp (,water'), 7rVP ('fire'), etc.,
do not come into existence simply by pointing to nature and uttering
sentences such as Tov70 vowp fun ('This is water').
I depart from Cherniss not on his resolution of the ambiguity, but
on the radical interpretation he uses to support this resolution. He
maintains that Plato intended to convince his readers that portions of
the four kinds of stuff "cannot be called anything distinct from anything
else. " Not appreciating this point, maintains Cherniss, is the
 5. iv Against Cherniss 147

"fundamental mistake" that undermines the interpretation of most

previous scholars. He remarks:

The fundamental mistake made by these and most

interpreters, however, is their assumption that Plato must
here be saying what name or kind of name the phenomenal
"phases," "moments," or "occurrences" should be called,
whereas he has already said that these transient moments of
flux cannot be called anything distinct from anything else (p.
122).

Cherniss admits that Plato did allow for reference to something sensible
in the extension of the nouns 7rVP ('fire'), Gowp (,water'), etc., but he
maintains that Plato denied that these references are to portions of stuff:

The distinctive names, naively and improperly applied to

phenomena, denominate in each case "the such and such,
whatever the correct formula may be, that is always identical
in each of all of its occurrences" (p. 128).

What are identified by the formula TO' TOtOVTOJl... are

manifested by coming to be in the receptacle from which
again they disappear. Later these distinct and self-identical
characteristics that enter and leave the receptacle are called
p,'p,r,p,CYTOt... and their nature is said to be that of an
image.... They are consequently not ideas (p. 128).

Cherniss supposes that Plato believed that if the nouns to refer, they
refer to "self-identical characteristics." Language users can point and
utter Tou7o vowp Eun ('This is water'), but if they express a truth, they
are not talking about some "phenomenal" portion of water. They are
talking about some self-identical characteristic.
On Cherniss's interpretation, not only did Plato believe that if such
statements are true, they are about self-identical characteristics, he
believed that no one can say anything true about a given portion of
"phenomenal" stuff. The reason that there can be no true statements
about these stuffs is not that they do not exist. Cherniss maintains on
Plato's behalf that phenomenal portions of stuff are regions of the
Receptacle that are "affected" by the self-identical characteristics:
148 Forms and the four kinds of stuff

On the other hand, < the distinct and self-identical

characteristics> are not the same as the transient phenomena
either, for phenomena are the apparent alterations of the
receptacle as a result of their continual entrance into it and
exit from it (50C3-4). Phenomenal fire is the region of the
receptacle that has at any moment been affected by fire,
phenomenal water the region that has been affected by
water, and so according as the JLLJL~JLCXm enter into the
receptacle (51B4-6, cf. 52 D4-El). The intensity and limits
of the apparent affections of the receptacle are continually
changing and so are indeterminable as fire, water, or
anything else (p. 129).

Of the phenomenal flux itself nothing more can be said than

that it is the resultant of these entrances and exits in the
receptacle (p. 130).

One might well expect Plato to agree that upon seeing, say, the
self-identical characteristic water come to be in a given region of the
Receptacle, one could point to phenomenal water and truly utter TaUTO
{jbwp tun ('This is water'), but Cherniss makes him deny that this is
possible. On Cherniss's interpretation, Plato believed that the
phenomenal stuffs are in radical flux and that nothing true can be said
about what is in such flux.20
Several problems stand in the way of Cherniss's interpretation.
One problem, which others have noted, is that Timaeus does not
distinguish sensible phenomenal stuff and sensible self-identical
characteristics in his ontological classifications at 48e2-49a6 and
52al-b 1. In addition to the Forms and the Receptacle, he just mentions
sensibles. This oversight would surprising if Cherniss's interpretation
were correct. If Plato really believed in these two kinds of sensibles,
one would expect Timaeus to list them. 21
Another problem is that Cherniss's interpretation is so very
uncharitable. Plato is content to overlook the tension, if not outright
incoherence, in his assertions that the self-identical characteristics are
recognizable, that the phenomenal stuffs are regions of the Receptacle
that are "affected" by these characteristics, but that language users
cannot truly say that a given such regiun contains a given such stuff.
Plato is not worried about this tension, according to Cherniss, because
 5. iv Against Cherniss 149

he is convinced that portions of the four kinds of stuff are in radical

flux and that nothing true can be said about them.
Timaeus does say that the phenomena "slip away" and "do not
abide" (4ge2-4), but this is slim evidence for attributing such a
paradoxical view to Plato. In passages from the Craty/us and the
Theaetetus in which he uses similar language (439d8-12; 182d 1-7), he
is arguing against the Heraclitean proposition that everything is always
flowing in every way. In these passages, in the course cif arguing
against the Heracliteans, Plato does nothing to show that he accepts a
restricted version of this doctrine. Indeed, he leaves his readers to
conclude, and no doubt intends them to conclude, that no way of
restricting the range of its quantifier can save it from absurdity. If, in
the Timaeus, without any explanation, he were to suppose that the
phenomenal stuffs were always flowing in every way, and introduce a
new kind of sensible so that language users would have something to
talk about, this would come as a complete surprise.
Although Cherniss is correct that Timaeus rejects sentences such as
TOVTO vowp fun ('This is water') in favor of sentences such as TOWVTOV
vowp fun (Such-and-such is water'), his interpretation is not very
plausible because he wrongly accepts a radical interpretation of
Aristotle's assertion that Plato was "persuaded by the Heraclitean
arguments that all sensibles are always flowing" (Metaphysics
1078b 13-15). Timaeus does not maintain that the phenomenal stuffs
are in radical flux and that the terms 'uowp (,water'), etc., have to be
predicated of different objects. His assertion, rather, is that simply
pointing to a given portion of such stuff cannot provide "trustworthy
and firm account < s >" because it cannot provide the appropriate
TOWVTOV (such-and-such). Timaeus's reason is somewhat obscure, but
if it is expressed in 4ge2-4, as Cherniss seems to believe, then it is that
the Ionians wrongly try to point to objects that are subject to change,
that "slip away" and "do not abide," in an effort to accomplish a task
that cannot be accomplished without the help of the Forms, "beings that
are steadfast." It has nothing to do with the radical Heraclitean
proposition that Plato rejects in the Cratylus and the Theaetetus. On
this point, Cherniss's interpretation is seriously in error.
150 Forms and the four kinds of stuff

v. Against Cornford

Cornford defends an important interpretation on the traditional reading

of Timaeus 49d5-7, but unfortunately he presents this interpretation in
a way that can be confusing. He maintains, at least initially, that Plato
believed that before the Demiurge put his hand to the cosmos, language
users could not use the word 7oiiTo ('this') to demonstrate portions of
stuff of the four kinds. At this point in the history of the cosmos,
language users can only talk in a way that does not require the
existence of objects with a so-called "constant identity." The reason,
apparently, is not that their language is impoverished. It contains the
nouns that provide for reference to material objects, but before the
Demiurge goes to work, the cosmos contains no such objects. In his
commentary on 49a-50a, Cornford writes:

Plato is now asserting that "fire" is properly only a name for

a certain combination of qualities or "powers," which appear
and disappear and are always varying. Such groups of
qualities, though perpetually shifting, are sufficiently alike
to be indicated by names; but in referring to fire we ought
not strictly to say "this (thing)," because the phrase suggests
something which preserves a constant identity ([1937]. p.
181).

In contrast with this stream of fluctuating qualities stands

that in which they make their transient appearances. The
Receptacle is the only factor in the bodily that may be called
"this," because it has permanent being and its nature does
not change. What this Receptacle is, we do not yet know.
Later on, when the Demiurge intervenes to introduce an
element of rational order, he will form the primary bodies
by fashioning for them geometrical shapes. But here we are
considering the bodily as it was "before" the Heaven was
made. We are not to imagine the qualities here described as
existing in particles of any shape, regular or otherwise.
There is nothing yet but a flux of shifting qUalities,
appearing and vanishing in a permanent Receptacle (p. 181;
emphasis in original).
 5. v Against Cornford 151

Cornford's claim is not self-evident, but he seems suppose that Plato

believed that Yawp TOLOUTOV iun (Water is a such-and-such), not that
Yawp Toi170 fun (Water is a this). Before the Demiurge acts for the
best, the only object "in the bodily" is the Receptacle. There were no
drops of water, chunks of earth, or other objects with a "constant
identity" for language users to talk about.
Cornford seems to rest his interpretation on Timaeus's subsequent
assertion about the primordial cosmos and the work of the Demiurge.
Cornford says that "< w > e are not to imagine the qualities here
described < in 49a-50a> as existing in particles of any shape." This
appears to be his gloss on Timaeus's remark that the kinds were first
"without reason and measure" (53a8) and "except for possessing some
traces of themselves, were altogether in such a condition as is likely
when deity is absent" (b2-4). Cornford says that" < I> ater on, when
the Demiurge intervenes to introduce an element of rational order, he
will form the primary bodies by fashioning for them geometrical
shapes." This appears to be his gloss on Timaeus's subsequent remark
that the Demiurge intervenes to "dress them in shapes and numbers"
(4-5) so that they have "reason and measure. "
In view of this commentary on 49a-50a, Cornford appears to
maintain that Plato is trying to describe the cosmos before and after the
Demiurge acts for the best. After this event, because the Demiurge has
made the "primary bodies," there no need for the prohibition against
claims such as Yawp TOI170 Eun (Water is a this). Before this event,
the prohibition is necessary because there are no drops of water or
other such objects with a so-called "constant identity."
In fact, however, despite his words, this turns out not to be
Cornford's interpretation at all. Contrary to the impression he may
give in his remarks on 49a-50a, he supposes that Plato does not really
believe in the Demiurge. Cornford supposes that the Demiurge is
somehow a "mythical" character in Plato's cosmology. In this way,
Cornford opts for an interpretation that sets him with Xenocrates and
against Aristotle. 22 He indicates his intention in remarks that occur
early in his commentary:

Though Aristotle chose to criticize Plato's statement in its

apparently literal meaning, his colleague Theophrastus
recorded the Academic interpretation as at least possible.
152 Forms and the four kinds of stuff

This question is. of course. bound up with the question

whether the Demiurge, as such. is mythical. If he was not
really a "maker." then there was no moment of creation.
We shall presently argue in support of this tradition. For the
present we may accept the Academic tradition (p. 26).

These remarks must be keep in mind if one is to understand Cornford' s

commentary on 49a-50a and on other passages in the Timaeus.
In his commentary on 47e-48e, e.g., Cornford appears to assert
that the Oemiurge creates portions of the four kinds of stuff from
certain two-dimensional geometrical shapes:

The unexplained existence of the four elements had been

taken as the starting-point for cosmogony, their properties
and behavior assumed. "as if men knew what fire and each
of the other is." Plato at once denies them the status of
elements. and promises to "explain their generation" from
prior and simpler beginnings. He intends to construct the
geometrical shapes of the four primary bodies from triangles
which he takes as elementary (p. 162; emphasis in original).

Since, however, Cornford believes that the Demiurge is in fact a

"mythical" figure, these remarks cannot be taken at face value. No
merely mythical figure has the power to actually "construct the
geometrical shapes of the four primary bodies. "
The proper interpretation begins to appear in his commentary on
52d-53c. Cornford maintains, in this commentary, that Timaeus did
not intend to assert a literal proposition about some change that the
Oemiurge caused in the past:

Such is the chaos taken over by the designing intelligence.

How is it to be interpreted? It is now generally agreed that
this disorderly condition can never have existed by itself at
a time before order was introduced .... It follows that chaos
is. in some sense, an abstraction--a picture of some part of
the cosmos. as it exists at all times. with the works of
Reason left out. "such a condition as we should expect for
anything when deity is absent from it." Now if you abstract
Reason and its works from the universe what is left will be
 S. v Against Cornford 153

irrational Soul, a cause of wandering motions, and an

unordered element of the bodily, itself moving without plan
or measure. This bodily element is represented as consisting
of qualities with active powers moving within the Recipient
which contains them, not as yet limited by the element of
definite quantity, number, measure, shape (p. 203).

With respect to S7c, Cornford maintains that Timaeus puts forward the
theory in terms of "shapes and numbers" as an instrumental theory:

But what is to be made of the picture of plane surfaces being

broken up and the fragments drifting about till they find
others to combine with? It cannot be taken literally (p.
229).

Plato's elementary factors are pictured as triangular surfaces,

whose regrouping provides for a sudden change from water
or air to fire. Such an account seemed to him "likely," as
covering phenomena which Democritean atomism could not
satisfactorily explain; but he did not mean it to be taken as
a literal statement of what "really" happens (p. 230).

These remarks help bring Cornford's real view into focus. He

supposes that Timaeus intends to express propositions that are "in some
sense, an abstraction--a picture of some part of the cosmos ... with the
works of Reason left out." Contrary to what one might first think, he
does not really maintain that Timaeus denies that chunks, drops, and
other such portions of the four kinds of stuff have always existed. Nor
does he really maintain that in the past, when he "dressed them in
shapes and numbers" (S3b4-5), the Demiurge transformed the four
kinds from "qualities" into "primary bodies." Cornford' s real view is
that 49a-SOa is part of Plato's attempt to explain how the cosmos would
be if "the works of Reason" were "left out."
I am sympathetic to Cornford's suggestion that the theory in terms
of "shapes and numbers" is some sort of instrumental theory, since
surely Plato did not really believe that portions of the four kinds of
stuff are literally constituted by abstract, two-dimensional mathematical
entities,23 but I find no good reason to suppose that Timaeus' s point
in 49a-SOa has anything to do with the theory of changes in terms of
154 Forms and the four kinds of stuff

"shapes and numbers" that he later discusses in 53c-57c. The text does
not provide sufficient evidence to conclude that Timaeus's point is "in
some sense, an abstraction." Timaeus does not explicitly maintain that
portions of the four kinds of stuff can be abstracted into two parts. He
does not explicitly maintain in 49a-50a that, in abstraction, each has
"an unordered element of the bodily, itself without plan or measure."
Nor does he explicitly maintain later in 53c-55c that, in abstraction,
each has an "element of definite quantity, number, measure, shape."
The way that Cornford links these two passages, if this is his intention,
has no very clear basis in the text. Timaeus says nothing to suggest
that he is trying to accomplish the fantastic goal that Cornford sets for
him. He says nothing to suggest that he intends to explain what
portions of stuff of the four kinds would be like if one were to
"abstract Reason and its works from the universe. "24 His concern in
49a-50a seems to more straightforward. His concern seems to be to
argue against the Ionians. He seems to maintain that the stuffs are
regions of the Receptacle with certain properties that the relevant
Forms specify, and he seems to argue that if the four kinds were
starting-points, as the Ionians maintain, then no one could know the
kind of a given portion of one of these stuffs.25

vi. Against other interpretations

Cornford's translation pre-dates Cherniss's attack on the tradition, but

subsequent scholars have come to its defense. Norman Gulley is
perhaps the most int1uential of these scholars. He defends an
interpretation on the traditional translation of 49d-e, and his
interpretation is significantly different from Cornford's. Cornford
maintains that Plato is presenting an "abstraction," but Gulley asserts
that Plato proscribes the use of TOVTO ('this') because he believes that
"perpetual change" entails that the cosmos contains no material objects:

< T > he argument of < 49d-e > is, briefly, that since the
visible world is one of perpetual change, it is necessary to
distinguish between a right and a wrong way of describing
it. "This" or "that" (T601: Ked TOVTO) is always wrong, since
these tenus suggest a reference to something substantial and
pennanent, whereas in fact the sensible world is a world of
 5. vi Against other interpretations 155

transient, yet recurrent, qualities or groups of qualities

(subsequently called "copies" or "likenesses" of the eternal
realities--5OC, 51A), which are properly described as "of
such and such a kind" (TOLOUTOV). Thus the fact that the
visible world is in continual flux does not entail that it is
devoid of detenninate and recognizable characteristics, but
it does entail that there are no substantial and permanent
"things" in it ([ 1960], p. 54).

Gulley sides against Cherniss on the ambiguity in 49d5-7, but he agrees

with him that the Heraclitean flow is relevant to the proper
interpretation of the passage.
Unlike Cherniss, however, Gulley supposes that Plato shows
himself to stand back from earlier assertions in the Cratylus and the
Theaetetus about the implications of the Heraclitean flow doctrine.
Gulley supposes that Plato had argued against this Heraclitean doctrine
by arguing that whatever is always flowing in every way "is devoid of
determinate and recognizable characteristics." By the time of the
Timaeus, however, Plato changed his mind about the implications of
radical flux. Gulley maintains that Plato came to believe that whatever
is always flowing in every way can have recognizable characteristics,
but cannot be a "thing." He writes:

< B > oth the Theaetetus and the Crary/us argue that the fact
that the visible world is in continual flux does entail that it
is devoid of detenninate and recognizable characteristics,
and make it clear that it is as illegitimate to apply the term
"of such and such a kind" to any part of it as it is to apply
the terms "this" or "that" (p. 54; emphasis in original).

< I > n the Timaeus Plato is contradicting the assertions of

the two other dialogues and is no longer willing to accept
what he had earlier propounded and accepted as the
implications of the theory that the sensible world is in flux
(p. 55).

On Gulley's interpretation, Plato supposes that the sensibles are in

radical flux and that this fact requires that many ordinary claims are
false. He supposes, e.g., that it requires that fire, water, etc., do not
156 Forms and the four kinds of stuff

constitute "things." Plato does not, however, suppose that the flux
doctrine entails that the sensible world "is devoid of determinate and
recognizable characteristics." In the place of claims such as Yawp
TOVTO fan (Water is a this), Plato is now willing to allow claims such
as Y[)wp TOLOVTOP fan (Water is a such-and-such).
To show that Plato's readers should expect this change from the
Cratylus and the Theaetetus to the later view in the Timaeus, Gulley
asserts that Plato came to realize that his views in the middle dialogues
are inconsistent. In the Timaeus, according to Gulley's interpretation,
Plato is trying to bring his various views into line. He continues:

This is a reflection of the greater consistency of doctrine

about the status of sensible "images" of Forms which is
found in the later dialogues. A major inconsistency in the
middle dialogues is that side by side with a theory which
gives the sensible image a fundamental part to play in the
recovery of knowledge there is a theory of perception which
condemns the sensible world as an aid to knowledge ... (p.
55).

In Timaeus 49d-e, according to Gulley, Plato publicly corrects an

earlier mistake about the implications of flux. He no longer accepts
that if the sensibles are in radical flux, then they are "devoid of
determinate and recognizable characteristics. "
Although perhaps Plato's thought could develop in the way that
Gulley supposes, there is not sufficient to show that it actually does
develop in this way. Aside from the rather implausible beliefs that he
attributes to Plato, and the lack of evidence in the Timaeus, the greatest
problem for Gulley's interpretation is that there is simply no dialogue
in which Plato shows himself to hold that the sensibles are in radical
Heraclitean flux. Under the influence of what I believe is a mistaken
interpretation of Aristotle, Gulley and many other Platonic scholars
have supposed otherwise. The fact of the matter, however, is that
Plato simply never shows himself to accept the paradoxical proposition
that the ordinary objects are in so much Heraclitean flux that no one
can know anything about them. This radical Aristotelian interpretation
is fundamentally wrong about the origin of Plato's Theory of Forms,
and this fact stands in the way of Gulley's interpretation of the
 5. vi Against other interpretations 157

Timaeus. He has not shown that Plato is recanting past views on "the
implications of the theory that the sensible world is in flux" so that he
can square his epistemology with his metaphysics. 26
In his defense of the traditional translation in his 1975 paper "Plato
and talk of a World in Flux: Timaeus 49a6-50b5," Donald Zeyl wisely
steers clear of the Heraclitean flux doctrine. He proposes that Plato
discusses the use of the words TOVTO ('this') and TOLOOTOV
(,such-and-such') in an effort to inform his readers that what, e.g., may
very well appear as a drop of water, or a chunk of earth, is actually not
an "entity in its own right," but rather is "a recurrent attribute of
something eLse." Zeyl writes:

< W > hen we refer to a given phenomenon by the term

"fire," we must not think that we are referring to what is
rouro, but rather what is ro' rOLOVrOJl . ... Because rouro
describes a thing as being an entity in its own right, a
permanent subject which, while possibly undergoing various
modifications, yet retains its identity. And nothing in the
phenomenal world is entitled to such status. The description
ro'rOLOVrOJl, on the other hand, merely describes its referent
as being an attribute of something else. Thus Plato's
justification for the reference of "fire," etc., to phenomena
is the new logical role that he assigns to those terms: they
are to be viewed as picking out a recurrent attribute of
something else. In other words, these terms are to be
construed as logically (though not grammatically) adjectival
(pp. 146-147; emphasis in original).

On Zeyl's interpretation, Plato urges that there are no drops of water,

or other objects constituted from the four stuffs, but that what may
appear, say, as a drop of water is actually a region of the Receptacle
in which a certain characteristic inheres.
Zeyl maintains that Plato urges this "new logical role" because he
is worried over the possibility of substantial change and believes that
conceptual reforms are necessary so that "we < do not> involve
ourselves in absurdities and contradictions." Zeyl writes:

< 0 > n the old M)'o~, i.e., on the old way of construing
our references to phenomena and to the elements in primis
158 Forms and the four kinds of stuff

we involve ourselves in absurdities and contradictions, since

this M'Yo~ construes the references as identifying references.
On its terms, when I call the stuff in my pond "water" I do
not predicate "water" of the stuff I am referring to (as I
would ifI described it as "cold" or "wet"); rather, I identify
it as water. And, in the cosmologies of Plato's
predecessors, the ultimate bearers of identity references, the
ultimate or basic subjects (the linguistic counterpart of their
ontological role as basic substrata), were elements. But
now, when I, on the terms of this M'Yo~, identify the stuff
in my pond as water, what am I going to say when it has
turned to air? Shall I now identify it as air? But it is the
same stufft If I do, what happens to my former
identification? Am I not forced (at least) to contradict
myself? Perhaps I might give it both (or, given the cycle,
all four) names. But then how do I distinguish the "water"
of my pond from the "air" into which it has evaporated?
Maybe Anaximenes was right: it's all air. But why air?
What sufficient reason is there to pick one element rather
than another? This M'Yo~, then, is neither reliable nor firm
(pp. 128-129; emphasis in original).

Here < in pp. 48 to 52 of the Timaeus> Plato wrestles with

the perennial problem of permanence through change, trying
to identify the substratum which underlies even the most
basic kind of change, inter-elemental change ([1987], p. 125;
emphasis in original).

On Zeyl's interpretation, so that language users will not involve

themselves in absurdities, Plato urges them to realize that if they point
to nature and utter sentences such as Tov7o ~owp MpC!. 'YEP~(J07CI.L ('This
water will become air'), the words vowp (,water') and a~p ('air') have
the logical role of adjectives. He urges them to realize that, contrary
to the surface grammar, they have asserted that a certain watery portion
the Receptacle will subsequently be airy. 27
Zeyl's interpretation is more likely than the interpretations that I
have considered, but it is not without problems. He maintains that
Plato is puzzled about how there can be "inter-elemental change" if
stuffs of four kinds are starting-points, but Plato gives little or no
 5. vi Against other interpretations 159

indication that he is plagued by this problem. Zeyl maintains that Plato

is puzzled because he believed that all change requires that something
persists through the change, but if water passes out of existence and air
comes into existence in its place, then nothing persists through the
change. For this reason, he introduces the Receptacle to be the object
that persists through "inter-elemental change." He decided that such
change is alteration, not generation or corruption.
If Zeyl were correct, one might expect Timaeus to explain why the
use of the language of generation and corruption with respect to the
four kinds involves one "in absurdities and contradictions." One
might, e.g., expect him to explain that it entails generation from
nothing, but Timaeus offers no such explanation. Furthermore, one
might expect him to refrain from using this language in contrasting the
sensible and intelligible objects. One might expect Timaeus to refrain
from explaining to his interlocutors that the sensibles have "birth,"
"come into being," and "pass out of being," but he shows absolutely
no such restraint. He freely uses this language with respect to sensibles
in 27d5-28a4, 28b2-c2, 48e2-49al, and 52al-7. One might expect
Timaeus to refrain from saying that the Receptacle is a necessary
condition for the "birth" of all sensibles. Again, however, he shows
no such restraint. He uses this language in 49a5-6 and 52b I. If Zeyl
were correct, this is not what one should expect. 28
Another reason to doubt Zeyl's interpretation is that it puts Tilr'.aeus
48b-51e in a context that breaks an important link to the past. If Zeyl
is correct, there is no connection between Timaeus's argument in this
passage and Socrates's previous arguments in the autobiography in the
Phaedo and in book V of the Republic. These arguments are against
opponents who accept ontologies in which essences are not
starting-points, and Timaeus seems to offer roughly the same argument
in 48b-5Ie. This passage clearly begins with Timaeus engaging in a
discussion of stuffs of the four kinds and their position in the Ionian
ontology, and this is an ontology in which essences are not
starting-points. Furthermore, this passage clearly ends with Timaeus
and his interlocutors accepting an ontology in which essences are
starting-points. Timaeus and his interlocutors agree that the Forms are
among the most basic of all existents. Zeyl could be correct that
Timaeus rejects the Ionian ontology because he is concerned with
"inter-elemental generation." This reading is not incoherent, but I
160 Forms and the four kinds of stuff

believe that it is much more likely that, like Socrates before him,
Timaeus is primarily concerned with the possibility of knowledge. He
is concerned to show that the Ionian cosmology is inconsistent with the
possibility of knowledge because it is part of a conception of reality in
which essences as not starting-points.

vii. Against Frede

Michael Frede provides a two part explanation for why Plato came to
believe that the Forms are "onto logically ... prior to the ordinary objects
of experience." The first part states that Plato maintains that there are
essences. Frede is not concerned to develop this part of his
exp lanation, although his remarks suggest that he supposes .that Plato's
belief in essences stems from Socrates's search for definitions. Frede's
concern is with the more difficult task of explaining why Plato went
from believing in essences to supposing that essences are "prior" to the
ordinary objects. Frede intends the second part of his explanation to
account for this development in Plato's thought. Witness Frede's
remarks in the following passages:

Plato, from some point onwards, came to take the view, that
in addition to the ordinary objects of experience, there exist
entities of an altogether different kind, ontologically and
epistemologically prior to the ordinary objects of experience.
There is good reason to think that Plato, in the course of
time changed his mind as to the precise nature of these
entities and their relation to the objects of experience. But
what I am concerned with here is not this further evolution
of his view, but rather how Plato came to have it in the first
place ([1988], p. 37).

This is a vexed question, but a first rough answer might

come in two parts: (i) Plato assumes that there is such a
thing as the essence or the nature of an F, specified by the
correct answer to the question 'what is an F?' Obviously
this is a crucial and questionable assumption to make. But
I will not consider here the reasons why Plato makes it.
Equally obviously, as we can see in Aristotle's case, to make
this assumption is not yet to assume that there exists,
 5. vii Against Frede 161

separately from the objects of experience, Platonic ideas.

This requires a further step. And it is this further step
which the second part of the answer is supposed to explain.
(ii) Plato also has a view of the ordinary objects of
experience which makes it impossible for him to assume that
these natures or essences are somehow to be identified with,
or to be found in, the ordinary objects of experience (p. 37).

According to Frede, Plato acquired a "view of the ordinary objects of

experience" that explains why he went from believing in essences to
supposing that they are "prior" to the ordinary objects. This view
made "it impossible for < Plato> to assume" that essences "are
somehow to be identified with, or to be found in, the ordinary objects
of experience." Since Plato was convinced that essences somehow fit
into reality, he concluded that they exist in a way that is
"ontologically ... prior to the ordinary objects of experience."
The view of the ordinary objects that Frede attributes to Plato is
striking. If Frede is correct, Plato believes that the ordinary sensible
objects of experience somehow masquerade as something they are
not. 29 Frede maintains that Plato shows himself to have this
remarkable belief at the outset of the Timaeus in 27d5-28a4. To justify
this interpretation, Frede first presents a "superficial reading" of this
passage that he neither rejects nor wholeheartedly endorses:

A superficial reading of this difficult text suggests the

following view: in characterizing the objects of experience
as constituting a realm of becoming Plato characterizes them
as being in every respect subject to change. Natures or
essences, on the other hand, are not subject to any change;
they are not the kind of thing which can change, come into
being, or pass away. Hence, if there is nothing about the
ordinary objects of experience which is not subject to
change, ordinary objects of experience cannot have natures,
let alone be natures. Hence, the natures or essences which
there are must constitute a separate realm. Now, I don't
think that such a superficial understanding of the text is
wrong. But I do think that a closer reading will also get us
more closely to the heart of the matter (p. 38).
162 Forms and the four kinds of stuff

On this superficial reading, Plato leaves his readers to conclude that an

ontology that includes the sensibles is not necessarily sufficient to
include essences. He leaves them to see that this conclusion follows
from two main premises: (i) that "there is nothing about the ordinary
objects of experience which is not subject to change," and (ii) that
essences "are not subject to any change."
To push "more closely to the heart of the matter" of what he takes
to be Plato's deep point in 27d5-28a4, Frede offers a novel
interpretation of the meaning of the Greek words 011 ('being') and
'YL'YIIOIl€1I01l (,becoming')30 as they occur in this passage in the
following phrase:

... TO' 011 beL, ,,(€IIECHII oe' OUK i!X.01l,

[aeL], 011 Of OUO€7rOTe (27d6-28al) .

... that which always is and does not involving any

becoming, .. , that which always is becoming, but never is
(Frede's translation).

Frede maintains that the superficial reading is superficial because in this

passage the word 'YL'Yllop..elloll cannot mean "coming into being." Frede
maintains that Plato uses 011 and 'YL'YIIOIl€1I01l in this phrase to mark a
distinction between appearing to have the property and having the
property:

I want to suggest that what Plato is saying when he is talking

of a realm of becoming in the Timaeus... < is > that for any
predicate "F" which we attribute to the < ordinary> objects
of experience, these objects only temporarily take on and
display the character of an F without ever being an F. A
real F, by contrast, is one which displays the marks of an F
because of the nature it has ... (p. 48).

The difference that Frede intends to draw between

"being an F"
and "temporarily tak < ing > on and display < ing> the
character of an F"
 5. vii Against Frede 163

is, in his words, the difference between

the nature or character or basic disposition... < and > a

temporary disposition, the way something looks, the
impression it gives. It is the difference between being pretty
or beautiful and looking pretty or beautiful, between being
old and acting or looking old (p. 46).

If Frede is correct, Plato had some very eccentric beliefs. He believed

that the Form the F itself is really F and that the ordinary objects which
are called F are actually only masquerading as F. He believed that the
Form is really F because it contains the essence that determines which
properties make sensible objects be F. Plato believed that the ordinary
objects of experience only masquerade as F. They lack this essence
and somehow only have the look, or appearance, of being F.
Frede provides an example that helps to illuminate this distinction
between "being" and "becoming" that he tries to read into 27d5-28a4.
He says that Plato believed that what one takes to be Socrates, e.g., is
such that on closer inspection it only appears to be a man, that the
features in virtue of which it appears to be a man turn out on closer
inspection to themselves be part of an even more elaborate masquerade,
and that "ultimately" there is only "matter which is nothing in its own
right and which can only be described metaphorically as part of some
randomly shifting chaos." Frede writes:

Why should we not say that Socrates has such a nature as to

have the marks of a human being and hence really is a man?
It is not clear what the answer is supposed to be. There
seems to be some wavering on Plato's part. But if we come
to the Timaeus the answer seems to be the following: if we
look at Socrates to find what it is that is a man and to see
whether it is a man of its own nature, all we find is some
material which only temporarily takes on the form of man,
and ultimately, if we push the analysis further, some matter
which is nothing in its own right and which can only be
described metaphorically as part of some randomly shifting
chaos. So underlying the marks or looks which the world
around us displays there are no natures or essences which
164 Forms and the four kinds of stuff

produce these marks. To put the matter grossly: there is no

reality underlying appearances (p. 50).

Frede maintains, if I understand him, that Plato believed that the one
essence that determines which properties make objects be men is
somehow not "in" the many individual men. This deficiency is
apparently how "there is no reality underlying appearances." One can
look there for the essence, but it is not there to be found.
Frede does not just maintain that Plato believed that essences are
not in the ordinary objects of experience. Frede sets out to explain
why Plato came to have this belief. Frede explains that this belief
about the ordinary objects stems from Plato's more basic belief that
these objects are impostors. Frede maintains that the so-called
"language of becoming" shows that because Plato believed that the
ordinary objects are "deficient," that they just have "looks," he came
to believe that they do not contain essences. This seems to be the
claim in the following passages:

But if this is what Plato means when he assumes that the

objects of experience only constitute a realm of becoming,
then it seems that this assumption is quite problematic. For,
first of all, it involves the assumption that objects of
experience do not have a nature or an essence. But this only
captures part of what is problematic. For we might be quite
ready to grant that ordinary objects do not have natures or
essences, that Socrates, for example, does not have a nature
such as to display the marks of a human being or a nature
such as to display the marks of something just. But we still
would not describe the matter in such a way as to suggest
that Socrates. in not having the nature of a human being or
the nature of a just thing, was missing or lacking something
as a result of which he merely, as it were, had the look of
a human being or the look of a just thing. But exactly this
is suggested, if not implied by the language of becoming, as
now understood (p. 49).

In characterizing the objects of experience as constituting a

realm of becoming because they lack natures or essences
which could account for their appearance, Plato describes
 5. vii Against Frede 165

them as if they have to be seen as something which is

missing something which it should have, which it is
pretending to have, which it misleads us into thinking it has,
which only has the look of the real thing, which is only an
imitation of the real thing, which tries or strives to be like
the real thing (cf. Phaedo 74d9ft). And looking at the
objects of experience in this way we are, of course, pushed
in the direction of looking for the real F, of which particular
F's only have the look, just as we might be pushed to look
for the really good person if we find out that we ourselves
at best only have the look of a good person. We certainly
have to have at least the notion of a really good person in
order to think of ourselves as just at best displaying some of
the marks of a good person (pp. 50-51).

Hence it does seem to me that Plato's assumption of a realm

of paradigmatic ideas is not independent of his view of the
ordinary objects of experience. His assumption ... that the
objects of experience lack a nature so as to be anything,
properly speaking, they just have a look or appearance of
being something and hence have to be understood in terms
of, or rather in contrast with, those things which really are
something because they have a nature so as to be this way,
and not just give the appearance of being this way. Without
an adequate understanding of this view about the objects of
experience one will not understand why Plato ever came to
postulate ideas of the kind he did postulate (p. 52).

On Frede's interpretation, Plato believed that the sensibles masquerade

as F, thought that things masquerade as F only if they do not have in
them the essence that determines the properties that make objects be F,
thought that something masquerades as F only if something really is F,
and concluded that this real F is the Form the F itself. 31
Plato could in principle have this "view of the ordinary objects of
experience," but clear evidence would be necessary before one can
reasonably attribute this view to any great philosopher. I believe that
the Timaeus contains no such evidence. Although Frede does not think
that his superficial reading of 27d5-28a4 is "wrong," it is certainly not
clearly correct. Timaeus never says, at least not explicitly, that the
ordinary objects of experience are subject to change in every way. His
166 Forms and the four kinds of stuff

aim is to show that the cosmos has a beginning, and to achieve this
aim, he distinguishes transitory objects from eternal objects. Frede
uses the superficial reading as a stepping stone to a reading in which
Timaeus draws an exclusive contrast between "being" and "becoming."
On the superficial reading, the ordinary objects of experience are in no
way not subject to change. On the more sophisticated reading, these
ordinary objects are in "no way being." Frede writes:

< W > hat seems to be intended < is > a clear and

straightforward contrast between being and becoming.
Something either is being, in which case it is in no way
becoming, or it is becoming, in which case it is in no way
being. That this is what Plato has in mind is suggested by
later passages in the dialogue (p. 40).

Frede maintains, furthermore, that Plato returns to this contrast

"repeatedly" and warns his readers against believing that objects in the
so-called "realm of becoming" have "any being whatsoever." Frede
writes:

Plato later in the Timaeus repeatedly returns to the contrast

between objects in the realm of being and the objects in the
realm of becoming. And he warns us against talking of the
latter in a way which would suggest that they had any being
whatsoever. Thus we are not supposed to talk of them as
'this' or 'that' (roc5E, roiYro, 4gel and passim). In particular,
we are not supposed to use the expression 'is' of them. This
expression, we are told, is only appropriate for the objects
of the realm of being. Thus we are not even supposed to
say, properly speaking, of what is becoming that it is
becoming (37e-38b, esp. 38bl-2) (p. 40; emphasis in
original).

I agree that 27d5-28a4 contains an important contrast, but it is not the

contrast that Frede claims. The contrast matches eternal and intelligible
objects against sensible and transitory objects. The subsequent parallel
passages show that the Forms are among the intelligible and eternal
objects (48e2-49al, 52al-7). These passages also show that the
ordinary objects of experience are among the sensible and transitory
 5. vii Against Frede 167

objects. The perplexing phrase at 27d6-28al, which might seem to

support Frede's interpretation, does not occur in these subsequent
passages. Nor does it occur elsewhere in the corpus, despite the fact
that in the Phaedo and elsewhere Plato contrasts the Forms with the
sensibles. Frede thus seems twice wrong. 27d5-28a4 does not contain
the "clear and straightforward contrast between being and becoming"
that Frede maintains, and Plato does not "repeatedly" return to this
contrast in the subsequent parallel passages.
The other two passages that Frede explicitly cites are equally
unconvincing. Frede maintains that by proscribing the use of TOUTO
('this') in demonstrating ordinary objects of experience, Timaeus is
denying that these objects have "any being whatsoever." I do not
accept the way that Frede resolves the ambiguity in the Greek, but I
agree, if this is Frede's point, that Timaeus is concerned to deny that
an ontology restricted to the ordinary objects of experience can provide
"trustworthy and firm accounts" for terms that identify the four kinds
of stuff. This, however, falls considerably short of showing that Plato
has the view of the ordinary objects of experience that Frede attributes
to him. It does not show, e.g., that Plato believes that what one takes
to be Socrates only masquerades as a man.
In 37e-38b, the other passage that Frede explicitly cites, Timaeus
appears equally unconcerned with this view of the ordinary objects.
Timaeus's remarks naturally divide into the following two passages:

For there were no days and nights, months and years, before
the firmament came into being, but the demiurge devised
their birth at the same time that he put the firmament
together. All these parts of time, and the was and the will
are forms of time that came into being, which we
unthinkingly and wrongly transfer to eternal being. We say
that is was is and will, but according to the truth is alone
belongs to it, and was and will are only properly used of
becoming that proceeds in time. They are motions, but that
which always holds in the same way immovably can become
neither older nor younger by time, nor can it have come to
be in the past, come to be now, or in the future. Nothing
belongs to it of all that belongs to the moving things of
sense, but these came into being as forms of time, which
168 Forms and the four kinds of stuff

images eternity and revolves according to number

(37el-38a8).

In addition to them still there are these, that which is past is

past, that which is happening is happening, and again that
which will happen is what will happen, and the non-existent
is non-existent, no one of these we say accurately. But this,
perhaps, may not be the right moment for a precise
discussion of these matters (38a8-b5).

In neither of these passages is Timaeus very clearly concerned with the

view that Frede attributes to Plato. Timaeus's concern in the first
seems to be with philosophical puzzles that stem from the use of tensed
language to talk about objects that exist timelessly. His concern in the
second passage seems to be with the puzzles that stem from not
properly distinguishing talk about timeless propositions and talk about
the dated events that render these propositions true or false. His point
in neither passage is straightforward, but Timaeus certainly does not
appear to be trying to help his interlocutors to grasp Frede's "clear and
straightforward contrast between being and becoming."
Frede may also intend to maintain that Phaedo 74d-e supports his
interpretation. In this passage, with respect to a given person who sees
a given ordinary object, Socrates says that this person thinks to himself
that this thing "wishes to be such as some other of the beings"
(74d9-10), but is actually "inferior" (e2) to these beings. In the
Timaeus, Plato has his character make similar remarks. With respect
to the basic categories of existence, Timaeus says that the ordinary
sensible objects of experience are actually "imitation<s> " (48e6),32
"image < s > " (52c1), and "semblance < s > " (c3) of the Forms.
These remarks are perhaps suggestive, but they do not provide
much in the way of support for Frede's interpretation. Frede needs to
show that Plato believed that the ordinary objects of experience are
"imitators, that they only have "looks, and therefore do not contain
II II

essences, but one need not understand the metaphor of imitation in this
way. Frede himself proves this point in his masterful paper on
Aristotle's conception of first philosophy. In Aristotle's metaphysical
system, if Frede is correct, "lower forms of being somehow imitate
higher forms of being, but this form of imitation is not that same as
II
 5. vii Against Frede 169

the masquerading that Frede attributes to Plato. In Aristotle, as Frede

understands him, the imitating is some other relation. Witness Frede's
remarks:

Aristotle seems to asswne not only that ultimately everything

depends on God for its being, but also ultimately nothing is
intelligible unless it is understood in its dependence on God.
And this in various ways. It is not just that everything
depends on God as its first cause. There is also the notion,
reflected in Aristotle in various ways, that lower forms of
being somehow imitate higher forms of being. Animals
procreate; this is their way of sharing in the eternal. The
heaven eternally rotates to imitate, as well as it can, the
unchanging nature of the unmoved mover. This suggests a
scale of perfection in which the less perfect is to be
understood in terms of the more perfect and ultimately the
unmoved mover, as if everything was like him in the limited
way it could be ([1987b], p. 89; emphasis added).

This Aristotelian cosmology is foreign to Plato, but he did believe that

the sensibles have to be understood in terms of their dependence on
intelligibles. A central tenet in Platonism is that intelligible objects, not
sensible objects, are starting-points in reality. Plato believed that the
Forms are "ontologically prior" to the ordinary sensible objects of
experience, and he uses the metaphor of imitation to help illuminate
this dependence. 33 Contrary to Frede's interpretation, however, given
that the metaphor need not be so understood, the corpus does not
provide sufficient reason to conclude that Plato supposed that
masquerading as something they are not is the way that the ordinary
sensible objects "imitate" the Forms.
The text is certainly not always easy to understand, but I do not
believe that Frede is correct about Plato's reasons for believing that the
Forms are "ontologically... prior to the ordinary objects of
experience. " There is not enough evidence in the Timaeus, or
elsewhere in the corpus, to show that Plato believed that the ordinary
objects of experience masquerade as F and believed that to masquerade
as F is not to contain the essence that determines which properties
make objects be F. Frede needs clear and compelling evidence to
attribute such remarkable beliefs to Plato, and the passages that he cites
170 Forms and the four kinds of stuff

fall considerably short of constituting such evidence. They do not

come close to showing that Plato believed that the ordinary objects of
experience masquerade as Forms, that they have "looks," but do not
contain the essences that the Forms contain.
 6

CONCLUSION

i. Introductory remarks

Plato, but not Socrates, concluded that the Forms are substances. On
this point, there is general agreement. The dispute is over the
explanation for this dramatic development in Plato's thought. Although
many scholars have offered explanations of the origin of the Theory of
Forms, no explanation has seemed clearly correct. I have tried to show
that this appearance is not a deception. I have tried to expose the
shortcomings in the orthodox explanations and to establish another
explanation in their stead. If I have succeeded, philosophers inside and
outside the field of Platonic studies will have to abandon much of the
received wisdom about Plato.

ii. The received wisdom

Plato has from style among twentieth century philosophers, especially

among those in the analytic tradition that stems from Wittgenstein. If
these philosophers bother look to Socrates and Plato for inspiration,
they look to Socrates. The message is that Socrates was a
proto-analyst, a counterpuncher who matched counterexample against
false claims to knowledge and who pursued logical analysis and clear
thinking about justice, beauty, and other such properties that figured so
prominently in the human affairs of fifth century Athens.
The word on Plato is less enthusiastic. He is condemned as a
metaphysician, a philosopher who insisted on the truth of propositions
that Socrates would have quickly set aside as beyond "human wisdom"
(Apology 20d8). Plato's contribution to philosophy is summar~ed in
an epitaph. There is no apology. No one explains that he grasped a
problem that Socrates missed or was only dimly aware. Plato is
dismissed as a metaphysician who believed that there are Forms in a
third realm and that the soul first gets knowledge of these objects
172 Conclusion

before it is incarnated. The student of the history of philosophy is left

to conclude either that Plato had no reasons for these beliefs or that his
reasons are so strange that they are not worth stating. In either case,
Plato takes the blame for a crime that he did not commit.
This epitaph should be regarded as nothing more than a bad
caricature, but the lack of credible alternative explanations has turned
it into something of a commonplace. Socrates was no metaphysician.
This much is true, but the implied reason is false. Contrary to the
hopes of some analytic philosophers, Socrates did see that the
enterprise is doomed to failure. He did not stand back from the Theory
of Forms and the Theory of Recollection because he believed that
metaphysical theories are meaningless. Socrates did not consider the
philosophical problems that Plato intended these theories to solve, but
this fact is rarely emphasized in the secondary literature.
In the twentieth century, no explanation has been the more
influential than one that Owen presents in his 1957 paper "A Proof in
the IT€PL Io€wp." He correctly sees that the clash between the
philosopher and the lover of spectacles is a clash between two
conceptions of reality, but he matches this insight with a groundless
claim. He claims that Plato believes that the Forms are "unambiguous
Paradigms" that exist in a P01170~ 767ro~, Le., that in some "intelligible
place" there are instances of justice, beauty, etc., and that these
instances function as standards for determining the truth of statements
that predicate these properties of the ordinary objects of experience.
If Owen is correct, Plato comes to accept this belief about Forms
because he insists on extending a certain, rather implausible, semantics
for so-called "complete predicates," such as 'is a finger' and 'is a
portion of fire,' to cover the sort of predicates that figure in Socrates's
"What is F?" question. He believes that the truth of statements that
predicate complete predicates are judged against ordinary sensible
objects that function as unambiguous standards. For the predicates in
Socrates's "What is F?" question, since he cannot find unambiguous
standards in the familiar world of ordinary objects of experience, he
insists that he has found them in a third realm of Forms.
There have been several attempts to refine Owen's interpretation,
but perhaps Irwin's is the most influential. He tries to preserve Owen's
insight into book V of the Republic, and he tries to make this insight
consistent with the Aristotle's explanation of the origin of the Theory
 6.ii The received wisdom 173

of Forms. Irwin agrees with Owen's claim that the "compression of

opposites," not "the succession of opposites," is central to the correct
interpretation of the clash between the philosopher and the lover of
spectacles, but he sets aside Owen's attempt to make a semantics of
complete predicates be the proximate cause of the Theory of Forms.
On Irwin's interpretation, Plato falls into a confusion over sensibles.
Plato agrees with the Heracliteans that the sensibles are in flux, but
fails to see that he agrees only because he believes that the Forms and
their opposites are compressed in them. This shortsightedness causes
Plato to drastically overstep his premises. It causes him to draw the
metaphysical conclusion that the Forms exist separately from sensible
objects. From the premise that justice, beauty, etc., are not definable
in terms of sensible properties, Plato invalidly concludes that these
Forms exist separately and as substances. On Irwin's interpretation,
the Theory of Forms stems from a straightforward blunder.
In view of these and other such explanations of the development of
his thought, no one should be too surprised that Plato has fallen from
style among the majority of twentieth century philosophers. Plato and
his Theory of Forms appear to have little or nothing going for them.
The theory seems to stem from simpleminded confusions or extremely
implausible beliefs. It seems to be an embarrassment that Aristotle and
subsequent more sober-minded philosophers quickly and correctly
abandoned. This is the import of the leading explanations of the
development of Plato's thought.

iii. The development of Plato's thought

I have tried to say more for Plato and his theory. To establish my,
explanation of the development of his thought, I have tried to clear
away certain orthodox interpretations and to put others in their place.
This first was Vlastos's interpretation of Socrates's autobiography in
the Phaedo. Vlastos wrongly turns this passage into an anticipation of
Aristotle's Doctrine of the Four Causes, and he thereby falsifies Plato's
reasons for concluding that the Forms are substances. The Theory of
Forms is part of a stream of thought that includes the cosmological
speculations of the early ~1ilesian natural scientists, Parmenides's
dramatic attempt to establish an alternative road to knowledge, and
Socrates's insistence on the priority of definition and the priority of
174 Conclusion

Form in his search for knowledge about matters involving justice,

beauty, and other such properties. Plato takes Socrates's practice to
constitute a clean break from the practices of the natural scientists.
Contrary to Vlastos' s interpretation, he does not try to make room for
both methodologies. He sees these different ways of searching for
knowledge as alternatives, and he decides that Socrates's method is the
"philosophical," or scientific, way to search for knowledge.
In addition to deciding in favor of Socrates and against the natural
scientists, Plato shows himself to grasp a point that his mentor failed
to grasp. He realized that to justify Socrates's new method of
intellectual inquiry, he must fit its two leading assumptions into a
general conception of reality. Socrates supposed that the possibility of
knowledge about matters involving justice, etc., presupposes the
possibility of definitional knowledge of these properties. He supposed
that these definitional truths somehow have their source in Forms, but
he did not try to justify his assumptions. He did consider how Forms
fit into reality. Plato did. He struggled with this philosophical
problem, and he concluded that the Forms exist as substances.
Book V of the Republic is the other place in the middle dialogues
that Plato argues that the Forms exist as substances, but to appreciate
the full import of this passage, one must reject Owen's interpretation
and its descendants in the work of Vlastos, Irwin, and other scholars.
The sophists are the other intellectuals in Plato's past, and they are his
enemy in the Republic. Plato agreed with Socrates that if there can be
political knowledge and expertise, justice must have an essence. Plato
also supposed that the sophists, especially the younger generation of
sophists, advocate a conception of reality according to which there is
no ground for this formal truth about justice. He thought that on a
conception of reality according to which essences are not
starting-points, but rather must find their place among of "the many
conventions (vOJLLJLa) of the many," there would be no real distinction
between knowledge and true belief. He thought that on such a
conception of reality, instead of one objective definitional truth that has
its source in a Form that exists "itself according to itself" (aUTO
KaO'min)), there would be many subjective, mutually inconsistent
"fancies," or "opinions," that masquerade as this definitional truth. To
avoid this consequence, and to insure that the 1rOA£(; need not appear
both just and not-just, the philosopher recognizes the truth of the
 6.iii The development of Plato's thought 175

Platonic conception of reality. He insists that the Forms are

substances. Plato thought that only on such a conception of reality
could there really be political knowledge and expertise.

iv. Aristotle on Plato

One should expect Aristotle to confirm my explanation, but many

scholars read Aristotle in a way that is inconsistent with it. They take
him to argue that the Theory of Forms is part of a reactionary backlash
to the Milesian tradition. These scholars suppose that Plato concluded
that the Forms are substances because he accepted paradoxical
Heraclitean claims that few have ever accepted and that were on the
intellectual fringe in the ancient world. In contrast to the Milesian
optimism about the possibility of knowledge about the nature of the
cosmos, Cratylus and the extreme Heracliteans were much more
pessimistic. They maintained that knowledge is impossible because
everything is caught up in radical flux. Cratylus even went further.
He "did not think it right to say anything but only moved his finger"
(Metaphysics 101OaI2-13). Many scholars maintain that the views of
Cratylus and Socrates come together in Plato to issue in the Theory of
Forms. They understand Aristotle to explain that Plato accepted this
extreme Heraclitean doctrine about sensibles, and that because he
accepted from Socrates that knowledge was possible, he concluded that
there are unchangeable Forms in a third realm.
This interpretation should have always seemed impossible. With
respect to the sensibles, Plato is not always as clear as his readers
might hope, but he is clear enough to cast doubt over this explanation
of the origin of the Theory of Forms. It gives Plato a belief in a
clearly false proposition, and the corpus provides no reason to suppose
that he was so shortsighted that he did not recognize it as such. Not
only does Plato never claim this proposition for himself, it is
inconsistent with some of his central and most well known
philosophical positions. If Plato denied that knowledge about the
sensible world is possible, he would have no reason to argue, as he
does at length in the Republic, that only the philosopher is fit to rule
because only he accepts a conception of reality according to which
political knowledge and expertise is possible. Only the philosopher
recognizes that the true politician looks "back and forth, to justice,
176 Conclusion

beauty, temperance, and all such things as by nature exist, and

composes human life with reference to these ... " (501bl-4).
Irwin recognized that an alternative reading of Aristotle's account
is necessary, and his interpretation has taken on the air of orthodoxy.
According to Irwin's interpretation, Plato does not succumb to the twin
evils of radical flux and skepticism about the ordinary objects of
experience. Irwin maintains that, in accepting the Heraclitean claim
that the sensibles are flowing, Plato holds the plausible view that
sensible properties are flowing because opposites moral properties are
compressed in them, not the implausible view that the sensible
particulars are always changing in every way. In this way, with his
novel interpretation of Plato's understanding of the flux doctrine, Irwin
tries to unite the best of Owen's interpretation with Aristotle's
testimony that the influence of Socrates and the Heracliteans came
together in Plato and gave birth to the Theory of Forms.
I have argued that there is a better way to understand Plato and
Aristotle. Irwin is correct to see that Plato did not fall prey to the twin
evils of radical flux and skepticism, but he is wrong to make the
Theory of Forms stem from a simple confusion about sensibles. The
Phaedo and the Republic suggest that Plato concluded that the Forms
are substances, not because he conflated sensible objects and sensible
properties, but because he came to believe no other ontology is part of
a conception of reality that is consistent with the possibility of
knowledge. Aristotle's remarks can be understood in a way that
confirms this reading of the dialogues, and I have argued that this is
how they should be understood. In the autobiography in the Phaedo,
in addition to rejecting the natural scientists's method of inquiry in
favor of the method that Socrates pursues in his search for knowledge,
Plato concludes that the Forms exist as substances and ontological
starting-points. Plato's reasoning is not as clear as one might hope, but
Aristotle's remarks help bring it into clearer focus. The natural
scientists make certain sensible objects, such as portions of water, fire,
etc., be ontological starting-points in their cosmologies. Aristotle's
remarks suggest that Plato believed that Socrates's method of inquiry
could not be part of this conception of reality. Aristotle's remarks
suggests that Plato believed that because sensible objects are "flowing,"
they can provide no stable foothold for the existence of the essences
that are a necessary condition for the possibility of knowledge.
 6.iv Aristotle on Plato 177

According to Aristotle, this is way Plato concluded that the Forms are
separate and exist as substances.

v. Illuminating a perplexing passage

My interpretation can stand on Socrates's autobiography in the Phaedo

and on the discussion of the philosopher in book V of the Republic, but
it has explanatory power that reaches far beyond these important
passages from the middle dialogues. It also helps to illuminate a
difficult and much debated section of the Timaeus in which Plato sets
himself against the Ionian cosmologists.
The background assumptions that many scholars have brought to
bear on this passage have forced their interpretations to go astray.
Cherniss, e.g., assumed that Plato accepts the extreme Heraclitean
doctrine that the sensibles are in radical flux and that there can be no
knowledge with respect to these objects. Others have made similar
assumptions. Gulley, e.g., assumed that although in previous dialogues
Plato accepted this Heraclitean doctrine, he now accepts a qualified
version. Neither scholar, however, is correct. Indeed, this whole
approach is wrong. It is predicated on a mistaken interpretation of
Aristotle's explanation of the origin of the Theory of Forms.
Once this mistaken interpretation is set aside, and the reader recalls
Socrates's previous encounters in the Phaedo and the Republic with the
natural scientists and with the lovers of spectacles, the proper
interpretation of this passage from the Timaeus begins to come into
focus. In these previous dialogues, Socrates's opponents advocate a
conception of reality in which essences do not exist as ontological
starting-points. The lover of spectacles, e.g., tries to put the many
vOILLlLa in the place the one Form that exists mho mO' aUTO. Timaeus
faces the same sort of opponent in the Ionian cosmologists. The Ionian
scientists do not recognize the truth of the Platonic conception of
reality. They fail to see that only "intelligible" (V07/TOr;) objects are fit
to be starting-points. The starting-points in their ontology are
"sensible" (aiU07/TOr;) objects. They are portions of stuff of the four
kinds that are traditional in Greek cosmology.
Timaeus argues against this ancient orthodoxy. Like Socrates in
the Phaedo and the Republic, Timaeus argues that the natural scientists
178 Conclusion

accept a conception of reality that is inconsistent with the possibility of

knowledge. Timaeus maintains that the lonians cannot invest the terms
7rVP ('fire'), vowp (,water'), etc., with MYOL that are "trustworthy and
firm" enough to provide for the possibility of knowledge about matters
involving these stuffs. Timaeus supposes that the lonians are
constrained to give inadequate replies to the "What is F?" question. He
supposes that, in virtue of their ontology, they must point and utter
sentences such as TOVTO 7rVP ~UrL ('This is fire') and TOVTO uowp fUrL
('This is water'), and he rejects such responses as inadequate. Timaeus
insists on responses of the form TOLOVTOJI vowp fUrL ('Such-and-such is
water') and TOLOVTOJI 7rVP f.UrL (,Such-and-such is fire') because only
Forms can provide the information necessary to fill the TOLOVTOJI
schemata in a way that knowledge is really possible. About matters
involving fire, water, etc., Timaeus believes that there are truths to
know. He rejects the Ionian ontology as inconsistent with this fact.
Timaeus maintains that it is part of a conception of reality according to
which there is no ground for a distinction between knowledge and true
belief. He maintains that on the Ionian conception of reality, there can
be nothing more than fancies and opinions. To avoid this unpalatable
consequence, like Socrates before him, Timaeus recognizes the truth of
the Platonic conception of reality. He recognizes that the Forms exist
in reality as substances and ontological starting-points.

vi. Plato's conception of reality

The Platonic conception of reality has little support among twentieth

century philosophers, but I hope that I have shown that the difference
of opinion is over substantive metaphysical and epistemological issues.
The Theory of Forms does not have its origin in a simpleminded
mistake. It is Plato's solution to a real problem. Socrates started a
revolution. He introduced a new "method" (/LE(}OOO~ = /LETa + ooo~)
of intellectual inquiry. Socrates turned his back on the practices of his
predecessors and decided to follow a new "way" (ooo~) of inquiry "in
the pursuit of" (/LETa) knowledge. Plato applauded him for his
decision, but he also realized something that Socrates was perhaps only
dimly aware, and may not have recognized at all. Plato realized that
the "philosopher" (<f)LMuo1>o~) could have complete confidence in this
new method of inquiry only if he could understand how it fit into a
 6. vi Plato's conception of. reality 179

general conception of reality. The Theory of Forms was born in this

realization. It is the direct result of Plato's attempt to validate the
leading assumptions in the method of intellectual inquiry that Socrates
uses in his pursuit of knowledge. Plato agreed with Socrates on the
priority of definitions, and he came to believe that this fact has
significant and far reaching implications for how reality must be. Plato
thought that reality must contain Forms if there are to be definitional
truths, and because he was unsatisfied with the alternative ontologies
that he associated with the natural scientists and the sophists, he
concluded that the Forms must exist in reality as substances and
ontological starting-points. He thought that reality could be no other
way if knowledge and expertise, including political knowledge and
expertise, were to be possible at all. This is why Plato concluded that
the Forms are substances.
 NOTES

Chapter 1

1. 'Form' and 'substance' are technical terms. 'Form' takes its meaning as
a translation of the Greek nouns e~ooc; and iOEa as these words occur both in
the context of the search for definitions and in the context of the more
theoretical discussions in the middle and late dialogues. (E[ooc; and iOEa come
from a root in the verb e·iow. This verb means "to see." In ordinary contexts,
the nouns often mean "shape" or "figure.") 'Substance' takes its meaning as
translation of the noun ovuia as it occurs in philosophical contexts in Plato and
Aristotle. (Ovuia comes from a participle of the verb eilLi, which means "to
be." In ordinary contexts, ouuia often means "that which is one's own,"
"property," or "estate. ") In his classic 1911 edition of the Phaedo, John
Burnet speculates that the Pythagoreans first pressed the word ouuia to serve
as a technical term and that Plato took his use from these philosophers (p. 34).
Burnet's evidence, however, is rather thin. For criticism, see Ross [1924],
pp. xxxiii-xlviii, and Grube [1935], pp. 291-294.

2. Miles Burnyeat notes that explaining why Plato concluded that the Forms
are substances is one of the most difficult and important tasks in Platonic
scholarship. "< Plato> demands an Equal that is not relative, a Beauty that
is not transitory and comparative ... and he locates them in a higher realm of
transcendent Forms. Here is the true beginning of the great metaphysical
vision. But what is its justification? ... To my mind this is the hardest and
most pressing question about Platonism, which surprisingly few scholars try
to answer" ([1979), p. 57).

3. I assume, furthermore, that the Platonic corpus is a relatively

straightforward indication of Plato's thoughts. Although it is not universal,
this assumption is commonplace among contemporary scholars. Witness, e.g.,
Richard Kraut's remarks in his introduction to The Cambridge Companion to
Plato. "Our best chance of understanding Plato is to begin with the
assumption that in each dialogue he uses his principal interlocutor to support
or oppose certain conclusions by means of certain arguments because he,
Plato, supports or opposes those conclusions for those reasons. This
methodological principle is not an a priori assumption about how Plato must
be read, but is rather a successful working hypothesis suggested by an
intelligent reading of the text and confirmed by its fruitfulness. Reading Plato
in this way allows us to make use of whatever material we have in the
 Notes to Chapter 1 181

dialogues to contribute to our understanding of them.... The fundamental idea

is that unless we have good evidence to the contrary, we should take Plato to
be using the content of his interlocutors' speeches, the circumstances of their
meeting, and whatever other material he has at his disposal, to state
conclusions he believes for reasons he accepts" ([1992], pp. 29-30). For
further helpful advice about how to pursue the history of philosophy, see Frede
[1987a).
4. 'Essence' is the usual English translation of the phrase Ti ~v 'e[vCtt as it
occurs in philosophical contexts in Aristotle. He uses the phrase to designate
what is articulated by a definition. Given its prominence in the Topics, this
phrase and its technical use may have its origin in Plato's Academy.

5. I believe that this principle is Socratic in origin, and is not a Platonic

invention, but nothing in my argument turns on this issue. For an extensive
recent discussion of the "priority of definition" in the early dialogues, see
Benson [1990].
6. I believe, as I have stated, that Socrates had no ontological theory about
the Forms, but my argument only requires the weaker claim that Socrates did
not believe that the Forms are substances.

7. On the historical Socrates, I agree with the orthodox view that Kraut sets
out in his introduction to The Cambridge Companion to Plato. "The most
plausible answer, one that is now widely accepted by many scholars, is this:
In the Apology and in several other works that search for ethical definitions but
show no deep interest in mathematics and make no inquiry into metaphysics.
we have a portrait of the historical Socrates; but then. as Plato continued to
develop in his thinking, he retained Socrates as the main interlocutor of his
dialogues, even though the doctrines of this more ambitious 'Socrates' go well
beyond anything dreamt of by that philosopher. And this interpretation of
Plato's development accords with the distinction Aristotle makes between the
real Socrates and the Socrates who is a mouth piece for Plato; the former. he
says, professed ignorance and inquired about ethical matters but not 'the whole
of nature' < Metaphysics 987b2>; to the latter he attributes no such
limitations, but instead regards him as a thinker who speculated about a wide
range of issues and fell into utter confusion when he posited a realm of
separately existing Forms and the Form of the Good central to ethical theory"
([ 1992a], p. 4).
182 Notes to Chapter 1

8. Bertrand Russell uses similar language to describe the realist position.

"Having now seen that there must be such entities as universals, the next point
to be proved is that their being is not merely mental. By this is meant that
whatever being belongs to them is independent of their being thought of or in
any way apprehended by minds" ([1912], p. 97). After giving the proof, he
concludes that "we must admit that ... < universals belong> to the independent
world which thought apprehends but does not create" (p. 98).

9. For other ways of describing this development in Plato's thought, witness

the following remarks. "If we now ask, 'What is a Platonic Form?', the first
thing to say is that the notion of a Form is the notion of that which a definition
defines (cf Phaedo 75cd). Hence 'X itself does often import a reference to
a Form. But the mere identification of something which a definition defines
is not enough to constitute a theory of Forms. It is one thing to set out to
define justice, quite another to say that justice as such exists on its own,
separately from just men and just actions. We cannot properly speak of the
Platonic Theory of Forms until this has been said. In dialogues like the
Phaedo and Republic, which propound the Theory of Forms as a theory about
the epistemological and ontological presuppositions of the Socratic method of
definitional inquiry, the Forms are transcendent natures, existing independently
of the empirical world, unchanging and divine, to be grasped only by
reflective reasoning. In other dialogues, however, especially the early Socratic
dialogues, 'X itself is contrasted with its particular instances and definitions
are sought, without any of this high-powered theoretical background"
(Burnyeat [1990], p. 38; emphasis in original). "In the so-called Socratic
dialogues concerned with definition, and specifically with the Euthyphro and
Meno, Plato introduces the notion of a form (eloor;, ioea) as the objective
correlative of these objects. The theory of the middle dialogues provides the
characteristically Platonic ontology, where the Forms are described as unique,
eternal, immune from change, without bodily or spatial location, and
accessible only to the intellect, not to the senses. The Symposium announces
this ontological theory for the first time, with reference to a single Form of
Beauty. The Phaedo then states the theory generally, for all Forms; and the
Republic applies this doctrine to a whole range of problems in epistemology,
political theory, and education" (Kahn [1973], p. 154). "Socrates said that
you cannot discuss moral questions like how to act justly, or aesthetic
questions like whether a thing is beautiful, unless you have previously decided
what you mean by the concepts 'justice' and 'beauty.' (This we know from
Xenophon and Aristotle no less than Plato.) Until these are fixed, so that we
 Notes to Chapter 1 183

have a standard in our minds to which the individual actions or objects may
be referred, we shall not know what we are talking about, and discussion may
be frustrated because the parties are attaching different meanings to the same
words. If this is right, Plato said, then we must believe that such a thing as
justice or beauty really exists, for otherwise what is the use of trying to define
it? It is no good looking for a universal standard if it is only imaginary. He
taught therefore the existence of a 'Form' or 'Idea' (in Greek €looc:; or l.oea)
of these and other concepts, which was not a mere concept existing in our
minds but had an eternal and unchanging nature independent of what human
beings might think it was. This is the famous XWPLCTP.6c:;, or affirmation of the
separate existence of the Ideas--separate, that is, both from the particular
instances of them in the world and from our thought of them; and belief in
such independently existing Forms constitutes what is known as 'Plato's
Theory of Ideas'" (Guthrie [1969], pp. 352-353; emphasis in original).

10. Martin West makes the same point. "We run < the pre-Socratics>
together as 'philosophers,' but they had no generic name for themselves.
Philosophy is of course a Greek word--it meant originally something like
,devotion to uncommon knowledge' --but it did not acquire a specialized sense
or wide currency until the time of Plato" ([1988), p. 127).

11. Heraclitus seems to use the word in this way. He says that "men who
love wisdom (¢LA0f16¢ovc:; Civopac:;) must be good inquirers ([(TTopac:;) into
many things" (DK 22 B 35). Diogenes Laertius reports that Pythagoras was
the first to call himself a ¢LAOUO¢OC:; (Lives of the Philosophers 1.12).
12. Charles Kahn provides a clear explanation of the etymology of the use of
the noun LUTopia. "As we can see from other early samples, Greek prose was
at first employed primarily for the publication ofIonian LUTOpirr for presenting
the results for systematic 'inquiry' or 'research' on a variety of subjects from
astronomy to biology, including historical research in connection with the
description of lands and people (as in the travel book of Hecataeus, a Milesian
contemporary of Heraclitus). The old Ionic term'LUTOpi1'f soon became fixed
in its narrow application to 'history' in our sense, because it was this type of
investigation that first gave birth to major works of prose literature: the
Histories of Herodotus and Thucydides" ([1979], p. 96). The use of the term
[UTopia for systematic inquiry is preserved in the English term 'natural
history.' This linguistic fact is at least in part due to Aristotle's long dominant
intluence in the practice and philosophy of science. The title of one of his
books on animals is Twv 7r€PL'Tei Na I.UTOPL{i)V (,On the History of Animal.,;,').
184 Notes to Chapter 1

13. The phrase 7r€PL CPU(J€WC; i(J70ptCXV occurs in the Phaedo at 96a8. In his
Clarendon Plato Series edition, David Gallop uses 'natural science' in his
translation ([ 1975]). In his edition of the Phaedo, Burnet says "that it is the
oldest name for what we call 'natural science'" ([1911), p. 99).

14. Unless otherwise noted, all references to the Platonic Corpus are to
Burnet's Platonis Opera.

15. As Vlastos has correctly noted, Socrates's autobiography "is one of the
great turning-points in European natural philosophy ... " ([1969), p. 297).

16. This, in part, explains why Aristotle makes the science of physics be
"second philosophy." The full explanation is quite involved, but Plato's
demotion of the natural scientists from the rank of philosopher is a crucial
historical antecedent for Aristotle's position. For a helpful discussion of
Aristotle's conception of "first philosophy," see Frede [l987b).

17. Guthrie may have a similar point in mind. "< The Theory of Forms>
was seen by later philosophers, from his immediate pupil Aristotle onwards,
to constitute a new conception of reality, and further reflection, and discussion
with his colleagues in the Academy, revealed serious difficulties in it to Plato
himself. Yet in his own mind, I am sure, it would seem to constitute no more
than the obvious and only possible defense of Socrates's teaching, and to do
no more than bring its implications into the open. 'Before you can know that
a thing is beautiful, you must be able to say what beauty is.' It could not have
been long before someone raised the pertinent question: 'Yes, Socrates, but
are you asking us to define what is real or not'? For, if beauty, justice and
such are only creations of our imaginations, surely you are wasting our time. '
By trying to answer this question in a sense favorable to Socrates rather than
to Gorgias, how could Plato have thought he was acting otherwise than from
motives of piety ... " ([ 1969). p. 353). Guthrie describes Socrates's debate with
Gorgias and the sophists in subsequent remarks. "One feature in the thought
and speech of his contemporaries seemed to Socrates particularly harmful.
Whether in conversation, in political speeches or in the oratory of the
law-courts, they made constant use of a great variety of general terms,
especially descriptive ethical ideas--justice, temperance, courage, ap€7~ and
so forth. Yet at the same time it was being asserted by Sophists and others
that such concepts had no basis in reality. They were not god-given virtues,
but only 'by convention,' varying from place to place and age to age .... From
this theoretical soil grew the pride of youthful rhetoric in its ability to sway
men to or from any course of action by mastery of the persuasive use of
 Notes to Chapter 1 185

words. In such an atmosphere it was not surprising that there was much
confusion in the meanings attached to moral terms. Socrates noted this, and
disapproved of it. If these terms correspond to any reality at all, he thought,
then one meaning must be true and the others false. If on the other hand the
Sophists were right, and their content was purely relative and shifting, it must
be wrong to go on using the same words for different things and they ought
to go out of use. He himself was convinced that the first alternative was true,
and that it was illegitimate and unhelpful for an orator to exhort the people to
adopt a certain course of action as being the wisest or most just, or for
advocates and jury to debate whether an individual had acted well or badly,
justly or unjustly, unless those concerned were agreed upon what wisdom,
justice and goodness are. If people are not agreed on that, but though using
the same words mean different things by them, they will be talking at
cross-purposes, and their discussions can make no progress either intellectually
or--when ethical terms are in question--morally. Here Socrates was raising for
the first time a fundamental question of philosophy, the question by what right
we use general terms, including all nouns except proper names, and what is
the factual content of such terms, and Aristotle was right to see that this was
so. At the same time, as Aristotle also recognized, he did not see it as a
logical or ontological question, but simply as an indispensable requirement for
what to him was much more important: the discovery of the right way to live"
(pp. 431-432; emphasis in original).

18. Owen, Vlastos, and other scholars in the Oxford tradition have dominated
Platonic scholarship in the United States in recent years. Nicholas White puts
this point clearly. "The philosophical study of Plato during the last
half-century has been dominated by a conception of it that arose mainly in
Oxford before World War II and then spread to this country. The conception
has been represented by figures like Gilbert Ryle, Richard Robinson, G. E.
L. Owen, and Gregory Vlastos. The set of problems established then
continues to resist most attempts at revision" ([1991], p. 318). These scholars
approach Plato from within the analytical tradition. This can cause
misreadings. Vlastos's interpretation of the Phaedo illustrates this point. He
tries to see Plato as using one of the techniques that is part of the
stock-in-trade of analytical philosophy. Vlastos tries to see Plato as solving
philosophical problems by using the Rylean technique of exposing unapparent
senses of words. This analytic approach also brings an anti-metaphysical basis
that can make dismissive readings too easy to accept. White, again, puts the
point clearly. "< M > ost work on ancient metaphysics in the last few decades
186 Notes to Chapter 1

has been too anti-Platonic to articulate effectively what drives Plato's thinking"
(p. 320).

19. I accept the traditional view that the Timaeus is among Plato's latest
dialogues, that it comes after the Theaetetus, and that Theaetetus comes after
the Parmenides. See, e.g., Ross [1953], p. 10. The traditional view comes
under attack in Owen [1953]. For criticism of Owen's argument, and a
vigorous defense of the traditional dating, see Cherniss [1957]. See also Prior
[1985], pp. 168-193 and Fine [1988), pp. 374-383. For a helpful discussion
of the methods for dating Plato's dialogues, and the various orderings that
result, see Brandwood [1990].
20. Some twentieth century philosophers do accept Plato's most general
claims about knowledge and reality. See, e.g., Nagel [1986], chapter V.
 Notes to Chapter 2 187

Chapter 2

1. Although the position has little or no support among contemporary

scholars, in the past some scholars have argued that the Theory of Forms, as
it is presented in the middle dialogues, was invented by the Pythagoreans, held
by the historical Socrates, and popularized by Plato. See, e.g., Burnet [1911),
pp. xliii-xlviii and [1930), pp. 308-309. For criticism of Burnet's argument,
see Ross (1924), pp. xxxiii-xlviii and Grube [1935), pp. 291-294.

2. Socrates's intellectual autobiography is one of the most controversial and

most discussed passages in the Platonic corpus. For some of the many recent
discussions, see Taylor (1969), Vlastos (1969), Burge [1971), Cresswell
[1971), Stough (1976), and Annas (1982).

3. Another part of Socrates's method is the elenchus. For some discussion,

see part of I of Richard Robinson's Plato's Earlier Dialectic.

4. Some scholars have found Socrates's insistence on the priority of

definitional knowledge to be puzzling, even wrongheaded, and some have
charged him with committing a fallacy. Peter Geach, e.g., takes Socrates to
claim "that if you know you are correctly predicating a given term 'T' you
must 'know what it is to be T,' in the sense of being able to give a general
criterion ofa thing's being T" ([1966), p. 371). He maintains that this claim
is "the Socratic fallacy, for its locus classicus is the Socratic dialogues" (p.
371; emphasis in original). For helpful discussion of the textual evidence in
the various pre-Phaedo dialogues, see Beversluis [1974], [1987], Burnyeat
[1977], Nehamas [19871. and Santas [1972].
5. In translating passages from the Platonic corpus, I stay close to standard
and easily accessible translations. In translating passages from the Republic,
I follow Grube's translation ([1974]). In translating passages from the
Phaedo, I follow Gallop's translation ([1975]). Unless otherwise indicated, I
follow the translations in Hamilton [1961].
6. Although the Hippias Major was accepted in antiquity as a Platonic
dialogue, some scholars now debate its authenticity. I accept the dialogue as
genuine. I agree with John Malcolm and Paul Woodruff that the Hippias
Major is a transitional dialogue which falls between the Euthyphro and the
Phaedo (Malcolm [1968], pp. 189-190; Woodruff [1982], p. 93). For an
argument that the Hippias Major is not a Platonic dialogue, see Kahn [1985].

7. See, e.g., Lysis 223b4-8, Chamlides 176a6-8, Euthyphro 6e2-6, Protagoras

360e7-361a3, and Meno 71a3-8, 80a4-b5.
188 Notes to Chapter 2

8. For the use of the instrumental dative in Greek, see Smyth [1956], sections
1503-1520. Burnet puts the correct sort of philosophical gloss on Socrates's
use of this locution in connection with the search for definitions. "The
instrumental dative is regularly used of the 'form' to express the fact that the
universal makes the particulars what they are" ([1924], pp. 36-37). Russell
uses 'in virtue of which' in the following passage to make the same point. "If
we ask ourselves what justice is, it is natural to proceed by considering this,
that, and the other just act, with a view to discovering what they have in
common. They must all, in some sense, partake of a common nature, which
will be found in what is just and in nothing else. This common nature, in
virtue of which they are all just, will be justice itself, the pure essence the
admixture of which with facts of ordinary life produces the multiplicity of just
acts" ([1912], pp. 91-92).

9. In his discussion of the Theory of Forms, Ross suggests that the Euthyphro
contains Plato's first uses of (Greek counterparts of) the terms 'Form' and
'Idea' for the objects of Socrates's "What is F?" question. "It seems probable
that the Euthyphro is the first dialogue in which either of the words ,oeoand
eloor; appears, in its special Platonic sense; and both appear there. The
passages are <5dl-5 and 6d9-e6> " ([1953), pp. 12-13).

10. Despite Socrates's insistence on the priority of Form, the Hippias Major
does not provide enough evidence to conclude that Plato believes that the
Forms are substances. Woodruff makes a similar claim. "Ontology is
peculiar to philosophers, and to philosophers of a certain stamp. To construe
ontology more broadly would be to trivialize the question of whether Socrates
leaned in that direction. Those who would enlarge ontology to cover every
question about what there is would make ontologists of us all, and thus blur
the history of philosophy. I must therefore dismiss as irrelevant the
considerable evidence at Hippias Major 287cd that commits Socrates to the
existence of things like justice and fine. Such commitments have nothing to
do with ontology and are not controversial. Any discussion presupposes the
existence of the subjects it discusses; but few discussions commit themselves
to an ontological theory about their subjects. Most people who talk of forestry
have not doubt that trees exist, but at the same time have not the inkling of an
ontological theory about them: in talking of trees they do not adopt an
ontology of physical objects continuous in space and time. Commitments to
existence are not commitments to specific ontological theories" ([ 1982], pp.
163-164; emphasis in original). "This I take to be the point of Quine's
doctrine of ontological relativity. Until talk of trees is interpreted against a
 Notes to Chapter 2 189

background of theory, it is not committed to any specific ontology" (n. 7 on

p. 179).
11. R. E. Allen argues that Socrates has an ontological theory, but Allen
seems not properly to distinguish between the relatively innocent proposition
that essences are necessary for knowledge and the more metaphysically
menacing proposition about how essences exist. "Now, what does all this
imply? It seems to imply something which is properly called a theory of
Forms. That theory is, in the first place, a technical theory, a body of rules
governing the practice of a useful art, that of dialectic. Thus, the question,
'What is holiness?' cannot be answered by examples, < but> . .. must be
answered by an analysis of the essence of holiness, because holiness is a
Form. So it is a technical theory--a logical theory, if you will, in some sense
of that much used and abused term. But it is also a metaphysical theory. It
assumes the existence of Forms, as universals, standards, and essences. That
assumption is stated or implied throughout the early dialogues. It is to be
remembered that one mark of being is power. As essences, Forms do not, so
to speak, just sit there. They do honest work. They affect the career of the
world, being that by which things are what they are. There is an argument in
the Hippias Major which makes this point explicitly. Socrates leads Hippias
to agree that justice is something (Elm TL 70V70) and that this is true of wisdom
too, for 'things which are just and wise and so on would not be such by them,
if they were not something.' Because beautiful things are beautiful by beauty,
Hippias is compelled to agree that beauty is something too. Socrates then goes
on to ask, What is it? (287C-D). The argument is an excellent one. If beauty
is that by which beautiful things are beautiful, and beautiful things exist,
beauty exists. Beauty is not a word, a thought, or a concept. It is an existing
thing, for the things it makes beautiful are existing things, and they are not
made beautiful by our words or thoughts or concepts. This conflicts with the
commonly held view that there is no commitment to the existence of Forms
in the early dialogues, and that talk of them there is 'merely a matter of
language '" ([ 1967], pp. 328-329; emphasis in original). For further critical
discussion of Allen's claims, see Woodruff [1978].

12. Platonic scholars have not grasped the importance of this passage. They
claim that Socrates is again concerned to insist that the natural scientists give
answers that are open to counterexample. The following remarks are typical.
"What Socrates intends to explain is what we have learned from Aristotle to
call 'formal' causality, but he has no technical terminology ready to hand and
therefore makes his meaning clear by examples. If we ask why something is
beautiful, we may be told in one case, 'because it has a bright color,' in
190 Notes to Chapter 2

another 'because it has such-and-such a shape.' The point that Socrates wants
to make is that such answers are insufficient. There must be one single reason
why we can predicate one and the same character, beauty, in all these cases.
Having a bright color cannot be the cause of beauty, since the thing we call
beautiful on the strength of its shape may be not colored at alL .. " (Taylor
(1926), p. 202; emphasis in original). "It is obvious why such 'wise reasons'
as color and shape are to be rejected. No given color or shape is either a
necessary or a sufficient condition for a thing's being beautiful. The reason
why the Form answer would be 'safest' (d8, el, cf. 101d2, 105b7) is that any
other answer could be refuted with counter-examples: a certain color or shape
might be present in a thing, and yet that thing might not be beautiful, or might
be ugly; and other things might be beautiful, even when that color or shape
was lacking" (Gallop [1975], p. 183). "In <Plato's> view, ifx is F and not
F, it cannot explain why anything is F; it cannot, in other words, be that in
virtue of which anything is F. Since some sensible properties of F suffer
compresence, reference to them does not explain why anything is F, and so
they cannot be what F-ness is. Since explanation is possible, in these cases
things are F in virtue of a non-sensible property, the form of F. So Plato
concludes that 'if anything else is beautiful besides the beautiful itself, it is so
for no other reason than it participates in the beautiful' (Phd. 100c4-6). Or
again, it is not because of 'bright color or shape or anything else of that sort'
(lOOdl-2) that anything is beautiful; rather 'it is because the beautiful that all
beautiful things are beautiful' (lOOd7-8). For sensible properties suffer
compresence in so far as bright color, for example, is sometimes beautiful,
sometimes ugly" (Fine [1993), p. 58).
13. " < The school of Parmenides and Melissus> saw, what no one
previously had seen, that there could be no knowledge or wisdom without
some unchanging entities ... " (Aristotle, On the Heavens 298b22-24). "It had
been asserted that everything in the world was subject to generation and
nothing was ungenerated, but that after being generated some things remained
indestructible while the rest were again destroyed. This had been asserted in
the first instance by Hesiod and his followers, but afterwards outside his circle
by the earliest natural philosophers. But what these thinkers maintained was
that all else is being generated and is flowing, nothing having any stability,
except one single thing which persists as the basis of all these transformations.
So we may interpret statements of Heraclitus of Ephesus and many others"
(26-34).
 Notes to Chapter 2 191

14. Socrates digresses into his intellectual history to explain and establish
premises in his proof that the soul is immortal and that the philosopher should
not fear death.

15. The Phaedo is a purported record of Socrates's last conversation on the

day of his execution, and it is the fourth in a tetralogy of dialogues that
purportedly describe Socrates's final days.

16. Other scholars have also made the comparison to Descartes's works.
Witness, e.g., the following remarks. "The autobiography pages of < the
Phaedo> are thus the ancient counterpart of Descartes' Discours de La
methode pour bein conduire sa rasion with the interesting difference, (1) that
though both philosophers are concerned to simplify philosophy by getting rid
of a false and artificial method, Descartes' object is revive the very
'mechanical' interpretation of nature which Socrates rejected, and (2) that
Socrates left it to the piety of another to do for his mental history what
Descartes did for himself" (Taylor [1926], n. 1 on p. 200). "Socrates'
account of his intellectual history is... a striking counterpart in ancient
philosophy of Descartes's Discourse on Method, despite its rejection of
'mechanistic' explanations that Descartes was to revive. Cf. A. E. Taylor, P.
M. W. 200, n. 1. Like Descartes, Socrates professes to be confused by the
senses and to abandon their use. Both are pioneers of a new philosophical
method. Both seek metaphysical foundations for mathematics and natural
science. And both formulate basic certainties that fortify their religious
convictions. Moreover, the autobiographical form of Socrates' story, as of
Descartes's, disguises the true rigor of its author's thought" (Gallop [1975],
p. 169).
17. In stating his interpretation, Vlastos slides from talk about the senses of
the word ai7ia to talk about kinds of at7iaL. I do the same in restating his
interpretation, but I also know that talk about senses of the word ai,7ia and
talk about kinds of ai7iaL are not two ways to talk about the same thing. For
a helpful discussion of this point, see Matthews (1972).

18. Other commentators have interpreted the Phaedo similarly. The

following two claims are typical. "< Plato> is concerned to show that there
are important forms of explanation that have nothing to do with causal agency"
(Burge [1971], p. 4). "Consequently it is clear that in our Phaedo passage
Socrates is not advancing a new theory of causation but in effect is considering
the kinds of answers that may appropriately given to various Why-questions"
(Mates [1973], p. 139).
192 Notes to Chapter 2

19. For an index of the occurrences of the word atria and its cognates in the
Platonic corpus, see Brandwood [1976], pp. 23-24.

20. Like Vlastos and other scholars, Annas slides back and forth between talk
about senses and talk about kinds.

21. This is one of the great unsolved problems in pre-Socratic scholarship.

For some discussion, see Irwin [1989], pp. 29-32.

22. For a discussion of Anaxagoras's influence on Plato's thought in the

Phaedo, see Brentlinger [1972a].
23. There is, perhaps, less agreement on this point than I suggest. Quine and
Ryle seem to push for nominalism. For argument that this push for
nominalism is misguided, see Searle [1969], pp. 103-106.

24. Many apparently believe that Plato concluded that the Forms are
substances because he was subject to this misguided desire. The following
statements are typical. "The Platonic mistake about the nature of forms goes
with a liberal use of what we may class together as abstract-singular
expressions like 'X-ness' or 'the attribute of being X'; these expressions are
not just grammatically but also logically, argumentatively, handled as though
they were proper names. I do not say that such abstract expressions looking
like proper names should be totally banned; it would make things very difficult
for philosophers. (I myself used 'human nature' in this way in the last
paragraph.) But I do say that anyone who uses them ought to be ready to
replace them on demand by use of the concrete predicates from which they are
derived. (Thus: for 'neither oneness nor manyness is a mark of human nature
itself read 'whether there is one man or many men is irrelevant to what X
must be if X is a man,' or something like that.) Sometimes this replacement
is stylistically better, sometimes not. But it must be possible; a sentence with
an irreducible abstract 'proper name' in it (say: 'Redness is an eternal object')
is nonsense" (Geach [1969], pp. 46-47; emphasis in original). "We may
ourselves prefer to resist the temptation to posit real entities like forms, or
universals, to correspond to the structures of language. Plato himself,
however, is implicitly committed to a broad theory of language as naming"
(Rowe [1984], p. 59).
25. Some scholars maintain that in the middle dialogues Plato never argues
for the Forms. Witness the following remarks. "In the early and middle
dialogues, Plato does not argue for the claim that the Forms exist" (Prior
[1985], p. 10). "<The Theory of Forms> is nowhere defended <in the
Phaedo>, but is simply accepted without argument by all parties" (Gallop
 Notes to Chapter 2 193

[1975], p. 97). "The existence and immutability of the Forms is not

established in the final section of the Cratylus: it is taken for granted at the
outset (439D). As in the Phaedo and the Republic, the existence of fixed and
eternal Forms is here postulated as an indispensable premise or Hypothesis, not
defended as the conclusion of an argument. (Only in the Timaeus 51B-D does
Plato present, or rather sketch, an explicit argument for the existence of
Forms.) Nevertheless, the last section of the Cratylus comes perhaps as close
as Plato ever does to showing just why immutable Forms are required for
rational discourse and knowledge, as the Parmenides assumes (135B-C). The
argument suggested in Cratylus 439D ff. is somewhat confused, and this
confusion perhaps reflects Plato's deep reluctance to present the doctrine of
Forms as a conclusion rather than as the starting-point (apx~) for philosophic
reasoning" (Kahn [1973), p. 169; emphasis in original).

26. As R. Hackforth correctly notes in the introduction to his 1955 translation

and commentary on the Phaedo, the "<s>tudy of the Phaedo has, for
English readers at least, been naturally associated for almost half a century
with the name of John Burnet..." (p. ix).

27. Gail Fine makes a similar point. "Plato speaks of the philosopher as
trying to be auroe; wO' auro" so far as possible (65b-d); I do not think he
means that the philosopher is trying to exist without his body (even though, to
be sure, he regards philosophy as a preparation for death). Rather, he means
that the philosopher tries to be uninfluenced by the body so far as possible; he
tries to reason without being influenced by sense perception, which is
misleading in the search for truth. To be a&ro KaB' aIm), that is, is to be
uninfluenced by, unmixed with, anything alien" ([1984], p. 60).
28. In his sketch of the early history of intellectual inquiry, Aristotle says that
the naturalist tradition begins with Thales and the Milesians. Aristotle
distinguishes this school of thought from the "school of Hesiod and the
mythologists" (Metaphysics 983b20-21, l000a29-20; On the Heavens
298b25-29). Irwin puts the right gloss on Aristotle's distinction. "Between
the age of Homer (mid-eighth century) and the age of Socrates (late fifth
century), the Greeks began systematic rational study of the natural order and
the moral order. Aristotle distinguishes those who talk about gods and offer
poetic or mythological accounts from those who offer rational accounts and
that can be studied seriously .... He calls the second group' students of nature'
or 'naturalists' (If>U(JWAO),OL), as opposed to Hesiod and his followers, because
they abandon mythology to ask a new question, about the nature (If>V(JLC;) of
things" ([1989), p. 20).
194 Notes to Chapter 2

29. For helpful discussion, see Hussey [1990]. See also Hussey [1972].

30. I follow the text and translation in Kirk [1983].

31. "It is emphasized < in DK 28 B 8.53-9 that to allow two subjects of

discourse instead of one> is not a matter of truth, but of a mistaken decision
on the part of the 'mortal men.' Because it is a mistake to allow nvo forms
instead of one, the giving of names to both of them is merely a matter of
empty convention. Each of them must have not merely a name, but a set of
characteristic predicates by which it is to be distinguished from the other; these
too are supplied by arbitrary convention ... " (Hussey [1972], p. 128; emphasis
in original). "Mortals have placed their trust in a thoroughly deceptive order
created by their own acts of naming giving; having done the naming, they
believe that their names are names of real distinctions in things" (Nussbaum
[1979], p. 74). "Parmenides' metaphysics and epistemology leave no room
for cosmologies such as his Ionian predecessors had constructed nor indeed for
any belief at all in the world our senses disclose to us" (Kirk [1983], p. 241).
"Mortal opinions do not reflect the discovery of objective truth: the only
alternative is to interpret them as products of conventions elaborated by the
human mind. Now it follows that nothing about the world can explain why
mortals should have such conventions or why they should invest them with the
specific content they give them. Hence the currency of these conventions can
only be represented as due to arbitrary fiat" (p. 256). "In deference to the
Eleatie arguments the world of our experience is thought of as mere
appearance, and theories become, for the first time, reductive: they tell us
what is really there (atoms and void, for example) and the world of experience
is consigned, mysteriously, to mere convention" (Annas [1988], p. 279).

32. I believe that Burnyeat is right to reject Irwin's explanation. The Theory
of Forms does not stem from a simpleminded mistake. For my discussion of
Irwin's interpretation, see Chapter 3.

33. Cornford is not alone. "< T > he doctrine of anamnesis clearly implies
the separate existence of Ideas ... " (Ross [1953], p. 25). "<W>hat is the
precise nature of the ideal standards or Forms? The Phaedo theory of
recollection clearly falls down unless they have separate existence, in a
non-sensible form" (Gulley [1962], p. 29). Not everyone, however, agrees.
Fine argues "that although Anamnesis and Affinity are compatible with
separation, neither requires it" ([1984], p. 74).
34. The proper analysis of "separation" in this context is disputed. I. M.
Crombie seems to me to be on the right track. "Aristotle says that whereas
Socrates was interested in defining universals, he left it to Plato to treat them
 Notes to Chapter 2 195

as 'separable' things. What exactly this means is not clear. On the whole in
Aristotle's language a 'separable' thing seems to be a 'substance' --something
like the Cheshire Cat (as opposed to its simile) which can exist on its own"
([1963], p. 254). For some recent discussion, see Fine (1984), [1985],
Morrison [1985a), [1985b], and Vlastos [1987].
35. Fine has argued for a different interpretation. "Aristotle is probably
correct to say that at least some Forms, in some dialogues, are separate. But
he and others are incorrect to suggest that Plato, beginning with the Phaedo,
heralds separation as a new feature of Forms" ([1984], p. 33). "Aristotle is
misleading to suggest that separation is a key feature of Forms, one Plato
argues for, beginning with the Phaedo. Though Forms may well be separate,
even in the Phaedo, neither Aristotle's account of separation, nor his belief
that it is a matter of explicit concern to Plato, should be accepted" (p. 81).
"For < Aristotle> writes as though separation is the big differentiator between
Plato and Socrates, and this is not true; commitment to separation is as muted
in the middle dialogues as lack of commitment to it is in the Socratic
dialogues. Though separation may indeed divide Plato from Socrates, it does
not play the central role in their contrasting stories that Aristotle suggests" (p.
85).

36. For an index of the occurrences of these words in the Platonic corpus, see
Brandwood [1976], 965-966.

37. Gallop interprets the passage this way. "This passage is of utmost
importance for understanding the Theory of Forms, since it argues that the
Form Equal and by implication many other Forms (cf. 75clO-d5), must be
non-identical with their sensible instances .... The non-identity of the Form
Equal with its sensible instances is inferred from the fact that the latter do, but
the former does not, possess a certain property. Thus, the argument is of the
following form: sensible equals have the property P; but the Form Equal does
not have that property; therefore sensible equals and the Form Equal are not
the same thing" ([ 1975], p. 121).

38. Hardy offers a similar interpretation of Plato's remarks. "In book V

Plato compares opinion to a state of dreaming or of being dreamy
(OJlf.LpWTTELJI), and says that to be dreamy is 'to think that something which is
like something else is not like but the same' (476c). His meaning is clear
thought it is incorrectly expressed. The opinionative man cannot have the
'occurrent' thought that particulars are like forms; for he has not yet been
'reminded' of forms. He is a man conversant with copies and unaware that
there is in the universe anything more real than they" ([\936], pp. 58-59.
196 Notes to Chapter 2

39. Gallop offers a similar interpretation. "Does the present passage, in fact,
improperly predicate 'equal' of the Form? Undeniably, it implies that the
word 'equal' is grammatically predicated of it. For it must be taken to mean
that sensible equals seem to us to fall short of being equal in the same way as
the Form Equal is equal. The italicized words, though not in Burnet's text at
74d6, clearly have to be understood .... But in what way 'is' the Form the
Equal equal? ... A possible solution is to understand it not as attributing
equality to the Form, but simply as an identity statement: the Form Equal is
(identical with) Equal. Sensible equals are therefore not equal in the way that
It IS. For they are not identical with the Form, but only (to use the
terminology introduced later) 'participate' in it. They 'fall short' of it. not in
failing to be exactly equal, a claim for which the present passage provided no
argument whatever, but rather in that they are not identical with it, as argued
at 74b7-c6. They 'strive to be like it,' but fall short .... On this view, the
judgement that sensible equals 'fall short' of the Form Equal will not be that
of a plain main confronted with logs that he regards as not quite equal. It will
be the judgement of a philosopher, who has recognized that the Form is
distinct from its sensible instances, and who, on sensing the latter, reflects
upon how different they are from the former" ([1975], pp. 128-129; emphasis
in original). Allen made a similar point years earlier. "<S>urely the force
of the metaphor of imitation, and of the xwpLup.6C;, is to indicate that the
deficiency in question is that of one type of thing with respect to something of
another type: 'deficiency' is here a categorial distinction ... " ([ 1960], p. 52;
emphasis in original).

40. Penner is concerned to combat so-called "diagnostic" interpretations that

have dominated the scholarly literature on Plato. "Now in attributing to Plato
a view that he could have had in mind, the present interpretation contrasts with
other well-known interpretations of this and other similar passages. On these
other interpretations, Plato is committed to positions which he could not have
consciously taken up, since their immediate implications are so crazy he could
only have arrived at those positions as a result of some grammatical error or
some seriously misleading analogy suggested by our language, the operation
of which was sufficiently unconscious that Plato did not notice it. Such are the
'diagnostic' interpretations offered by Gregory Vlastos and G. E. L. Owen,
two of the most influential Plato scholars of recent years" (p. 44).
 Notes to Chapter 3 197

Chapter 3

1. Some scholars conjecture that although he does not name him, Plato has
Antisthenes in mind. For some discussion, see Grote [1885], pp. 163-165.
See also Adam [1905], pp. 337-338, 341.

2. Socrates slides from "appear F and not-F" in 479b3-5 to "is no more said
to be F than not -F" in b6-9-c5, but I do not think this marks any canfusion on
his part. For a contrary suggestion, see White [1979], p. 161. .

3. Heraclitus seems to use the verb OOK€W to make a similar point. Kahn
brings this out clearly in his translations ([1979]), which I follow. "Although
the M'Yo<; is shared, the many live as though their thinking were a private
possession" (DK 22 B 2). "Most men do not think things in the way they
encounter them, nor do they recognize what they experience, but they believe
their own opinions" (DK 22 B 17). The Greek for "believe their own
opinions" is €wVToiO'L OOK€OVO'L, and this phrase, as Kahn notes, can equally
well mean "imagine for themselves" or "seem to themselves." Compare
Xenophanes's remarks in the following fragment: "As for certain truth, no
man has known, nor will any know it, concerning the gods and about all the
other things that I am saying. For however much he might chance to say what
has actually been brought to pass, still he himself does not know; it is opinion
that is constructed in all cases" (DK 21 B 34; Hussey (1990), p. 18; emphasis
in Hussey's translation).

4. In his edition of the Republic, Adam offers a similar interpretation. "The

words < 7(i TWI' 7rOAAWI' 7rOAAa I'OP.LP.OI at 479d3-4> refer to general rules,
standards, canons, believed in by the multitude (cf. 7(i TWI' 7rOAAWI' OiY'YP.OITO/
VI 493 A), who have on every single subject many such standards (7rOAAa
I'OP.LP.OI), mutually inconsistent and uncoordinated, because they do not know
that TO' KOIMI', TO' 01/,01001' etc., are each of them 'EI''' ([ 1905], p. 343). Hardy
offers a similar interpretation. "An opinion or formula (I'OP.LP.OI'), such as that
of Polemarchus about justice, has some basis in facts but breaks down when
confronted by further facts. The statement that it 'tumbles about between what
is and what is not' (Republic, 479d) might be interpreted as simply a highly
rhetorical way of saying this, viz., that it is false but has some foundation in
facts. If this is so, instead of speaking of the objects of opinion as between
being and not being, Plato ought to have spoken of opinions as more or less
false, or more or less removed from the truth" ([1936], pp. 31-32). Gosling
and Crombie seem to have something similar in mind. Gosling suggests that
198 Notes to Chapter 3

the phrase means "something like 'customary rules about the beautiful and the
just''' ([1960], p. 120). Crombie says that his interpretation "owes much to
discussions with Mr. Gosling" ([ 1963), n. 1 on p. 53), and, indeed, he makes
roughly Gosling's point in much more detail. "It is the multifarious
conventional opinions of ordinary men which he argues must roll about
between 0" and l1:ri ()'" and these are what we should call beliefs. How it can
be that beliefs can perform the extraordinary feat of 'rolling around between'
(this perhaps means 'occupying some point anywhere on the scale between')
'the existent and the non-existent' has I hope been made clear above. I
certainly cannot imagine how the subject-matter of beliefs (the things that we
have beliefs about) could do anything of the kind. That beauty is a matter of
bright coloring is perhaps one of the multifarious conventional opinions of
ordinary men about beauty, and the 'quasi-fact' which this clause stands for
(namely beauty's being a matter of bright coloring) is not a fact or a reality,
but on the other hand it is not a complete figment either. It is between being
a fact or reality and being a figment, just as the corresponding state of 06~a
is between grasping perfectly and being completely out of touch" (p. 58).
5. In his commentary on the Republic, White claims that Plato only intends
to show that Forms and sensibles are distinct. "Plato's main concern < in
476d-480a> is not to argue for the existence of Forms, and he does so only
very indirectly and briefly. In the passage in which he might appear to be
giving an argument for their existence, 479a-b, his aim is really to show
instead that sensible objects--i.e., what the lovers of sight and sounds are
concerned with--are 'between what is and what is not,' and not to prove that
there is such a thing as 'what is' (Le., Forms). In effect, therefore, the
passage is supposed to show only the distincmess of sensibles from 'what is,'
whereas it assumes that there are such things as 'what is'" ([1979], pp.
159-160; emphasis in original).

6. Gosling and Cherniss make similar points. "To say that there is a Form
of F is more than to say that 'F' has meaning and can be defined; it is to claim
that statements of the form 'X is F' are either true or false, and their
acceptability is to be judged objectively and not by reference to the tastes of
individuals or groups. In short, the claim that there is a Form of F is
tantamount to the claim that E1rLUrTfp'7I is possible in a certain area" (Gosling
[1973], p. 174). "The phenomena for which Plato had to account were of
three kinds, ethical, epistemological, and ontological" (Cherniss [1936), p.
16). "The necessity for an absolute standard of ethics which would not depend
upon the contradictory phenomena of conventional conduct but would be a
measure of human activities instead of being measured by them was forcibly
 Notes to Chapter 3 199

demonstrated by the phght mto wluch Democntus had fallen" (p 17) Plato
came to beheve that "the defimuons reqUisite to normative etlucs are possible
only on the assumptIOn that there eXist, apart from the phenomena, substanuve
objects of these defimuons wluch alone are the source of the values attaclung
to phenomenal eXistence The posslblhty of etlucal dlstmctlOns, then, Imphes
obJecuve differences wluch can be accounted for only by the hypotheSIs of
substantive Ideas" (p 18)

7 Frege makes a slIDllar pomt m The Foundatlons of Anthmetlc "If

everythmg were m contmual flux, and notlung mamtamed Itself fixed for all
tIme, there would no longer be any posslblhty of gettmg to know anytlung
about the world and everytlung would be plunged m confuSIOn" (p Vll)

8 Goshng, agam, makes a slIDllar pomt "Plato IS not mterested m what

makes language possible, and so not m problems of meamng or of umversals
He IS mterested m dIscovenng whether €7rLCTrr,/-I.7I IS possible, and that m Ius
view mvolves showmg that the umverse IS a system of a certam sort
AssertiOns that the Forms are necessdry for MyoC; (cf Parmentdes 135b-c) see
them as necessary for argument not speech" ([ 1973], P 197)

9 In translatmg passages from the Ttmaeus, I follow Cornford [1937]

10 "Socrates admIts that long speeches are beyond him and concedes to
Protagoras on that score But when It comes to dIalectical dISCUSSIOn and
understandmg the gIVe and take of argument, I would be surpnsed If he Yields
to anyone Now, If Protagoras admIts that he IS Socrates's mfenor m
dIalectic, that should be enough for Socrates But If he contests the pomt, let
hIm engage m a questIon-and-answer dIalogue and not spm out a long speech
every tIme he answers, fendmg off the Issues because he does not want to be
accountable, and gomg on and on until most of the lIsteners have forgotten
what the question was about, although I guarantee you Socrates Will not forget,
no matter how he Jokes about hiS memory So I t1unk that Socrate~ has the
stronger case" (Protagoras 336b8-d5)

11 Edward Hussey puts the pomt well m Ius excellent commentary on the
pre-Socratics "The assumptIOns and methods of the sophists were attacked
m the next century by Plato, whose cntICIsm by ItS bnlhance and power has
mfluenced all subsequent estlIllates But even m their own time the sophists
wIll have had artIculate opponents, most notably Socrates, whose method of
dISCUSSIOn by questIon and answer IS a WdY ot diSSOCiatIng two thmgs that the
~OphiStS were prone to confuse the force of rea~on and the power ot the
spoken word So tar as the sophists can be diVided mto an earlier and a later
generatIon, the change between the two groups IS mtellIglble as a response to
200 Notes to Chapter 3

criticism from Socrates and others; also, possibly, to the harsher realities of
the Peloponnesian War. The later sophists have shed the comfortably tolerant
and conservative assumptions of Protagoras and his contemporaries, and are
driven to propound radical theories of morals and politics (about which a little
will be said later). The difference is marked between Plato's treatment of the
older sophists, who are handled with some affection and respect, even while
their philosophical shallowness and their minor faults of character are being
revealed, and his portrayal of members of the younger generation--such as
Polus in the Gorgias and Thrasymachus in the Republic--who are shown as
thoroughly unsympathetic" ([1972], pp. 117-118).

12. In the Republic, Socrates makes similar points. "And have you noticed,"
he asks Adeimantus, "that opinions without knowledge are ugly things? The
best of them are blind. Does it seem to you that those who opine some truth
without intelligence (pour;) are no different from blind people who yet follow
the right road?" (506c6-9). Adeimantus agrees with Socrates. He says, in
reply, that such people "are no different" (10) from such blind people.

13. A. E. Taylor may have in mind a similar interpretation. "The founding

of the Academy is the turning-point in Plato's life, and in some ways the most
memorable event in the history of Western European science. For Plato it
meant that, after long waiting, he had found his true work in life. He was
henceforth to be the first president of a permanent institution for the
prosecution of science by original research. In one way the career was not a
wholly unprecedented one. Plato's rather older contemporary Isocrates
presided in the same way over an establishment for higher education, and it
is likely that his school was rather the older of the two. The novel thing about
the Platonic Academy was that it was an institution for the prosecution of
scientific study. Isocrates, like Plato, believed in training young men for
public life. But unlike Plato he held the opinion of the 'man in the street'
about the uselessness of science. It was his boast that the education he had to
offer was not founded on hard and abstract science with no visible humanistic
interest about it; he professed to teach 'opinions,' as we should say, to provide
the ambitious aspirant to public life with 'points of view,' and to train him to
express his 'point of view' with the maximum of polish and persuasiveness.
This is just the aim of 'journalism' in its best forms, and Isocrates is the
spiritual father of all the 'essayists,' from his own day to ours, who practice
the agreeable and sometimes beneficial art of saying nothing, or saying the
commonplace, in a perfect style .... Plato's rival scheme meant the practical
application to education of the conviction which had become permanent with
 Notes to Chapter 3 201

him that the hope of the world depends on the union of political power and
genuine science" ([1926), p. 5; emphasis in original).

14. According to Quine in "On what there is," the conceptualist "holds that
there are universals but they are man-made" (p. 14) and the nominalist
"object < s > to admitting abstract entities at all, even in the restrained sense
of man-made entities" (p. 15). Not everyone, however, understands the
positions in quite this way. "Traditionally there have been two versions of a
non-realist theory of universals. Universals have been seen as essentially
linguistic entities, the 'general names' of grammatical classification.
Alternatively they have been seen as consisting not so much in naming what
we call common features of the external world as in thinking of such features
as common. 'Nominalism' is an ambiguous label. Sometimes it has been
used to cover any non-realist theory and is therefore equivalent to 'post rem.'
Sometimes it has been used to identify universals with the general words--say
adjectives, common nouns and verbs--of language. In this narrow sense
Aristotle is no more a nominalist than Aquinas or indeed Ockham. The
traditional label which should be attached to him is that of conceptualism. The
common features that we attribute to the external world are not names, in so
far as names do not belong to language but to this or that language. They are
to be found in what is for Aristotle logically prior to language, namely
thoughts, of which 'concepts' is a synonym" (Lloyd (1981), p. 2).

15. Penner tries to find textual support for his claim in the way that Socrates
characterizes the lover of spectacles. Socrates does not explicitly allow the
lover of spectacles to believe in anything other than particulars. He makes the
lover of spectacles accept the many things that are F, but reject the Form the
F itself. "The many VOiJ.tiJ.CX at 03-4 must surely be the many beautifuls at EI
which in tum can surely not be other than the many beautifuls [A3, 5) that the
'good fellow' [AI) who is the lover of sights believes in. What are the many
beautifuls the lover of sights believes in? Surely the many beautiful sounds,
colors, shapes and so forth" (p. 236). Although Penner claims otherwise,
Socrates's characterization is too weak to show anything but that the lover of
spectacles is not a realist about the truth-conditions for statements that
predicate beauty and other such properties. Socrates neither explicitly insists
that the lover of spectacles is a nominalist nor explicitly denies that he is a
conventionalist. Plato takes no stand on this issue.

16. Witness, e.g., the following claims. "He thought that understanding what
it is to be f is acquaintance with a sample, specimen or paradigm
thing-that-is-f. He noticed that there is a difference between, for example,
understanding what it is to be like Socrates and understanding, simply and
202 Notes to Chapter 3

without qualification, what it is to be like. But if the latter is acquaintance

with a sample, then it can differ from the former only by a difference in the
sample, which must thus be a thing that is like but not like Socrates. By
generalization from this case he concluded that it is necessary to construe
understanding, simply and without qualification, what it is to be like as
acquaintance with some paradigm that is simply and without qualification, like;
and such a paradigm, since it is not to be found in the physical world, must
be a form" (Kirwan [1974], pp. 112-113; emphasis in original). "My view is
that, at least in the Phaedo and the middle books of the Republic, Plato
envisaged Forms for these incomplete predicates only. In the Phaedo
(100B-I05C) Socrates distinguishes between sensible particulars, their
properties or characters, and the Forms. These Forms, and all the Forms
mentioned elsewhere in the Phaedo and in the middle books of the Republic,
correspond to incomplete predicates. Plato is introducing a new class of
objects which the soul, confused by the contradictory reports of the senses (cf.
Rep 524 BD), has to contemplate if it is to grasp what beauty, justice, or
tallness are in themselves. Unlike sensible particulars, the Form and the
characters of beauty are completely, essentially, beautiful. If a particular is
beautiful, it will also be ugly in another context; that very same particular will
be ugly, without undergoing any change. But the Form and the characters of
beauty would cease to be what they are if they were ever qualified by
ugliness" (Nehamas [1975], pp. 108-109; emphasis in original).
17. "It has been repeatedly pointed out that Plato's line of thought here goes
astray, because of a failure on his part to take account of the workings of, in
particular, relational predicates. Wanting an unqualified sample of 'large,'
tout court, but finding the sensible world full only of samples of 'large
compared to,' or like predicates, he looks elsewhere for the unqualified
instance which he wants" (White (1976), p. 68). "Some will be inclined to
think that the view just now described is too clearly mistaken, and even
bizarre, to be rightly ascribed to Plato. Such an impression, I think, is the
result of an overexposure to contemporary philosophical discussions ... " (p.
70).
18. With these remarks, although he does not mention Aristotle, Owen
politely sets aside Aristotle's account of the genesis of Plato's Theory of
Forms.
19. For excellent commentary on the Parmenides, see Allen [1983].
 Notes to Chapter 3 203

20. R. M. Hare diagnoses Plato's mistake as follows. "It is an example of

the perceptive genius of that great logician, that in spite of being altogether at
sea concerning the source of our philosophical knowledge ... he discovered the
very close logical analogies between philosophical discoveries and
remembering. He was wrong in supposing that we are remembering
something that we learned in a former life--just as more recent mythologists
have been wrong in thinking that we are discerning the structure of some
entities called 'facts.' What we are actually remembering is what we learnt
on our mothers' knees, and cannot remember learning" ([1960], p. 161;
emphasis in original).
21. Vlastos acknowledges that he was influenced by Owen's views on Plato.
" < M > y greatest debt is to Professor G. E. L. Owen, whose views on this
topic I came to know from one of his unpublished MSS, and from discussion
with him in Oxford in 1960" ([1965], n. 6 on p. 59).
22. Archer-Hind may have this same sort of interpretation in mind. "Plato
was a true scholar of Herakleitos: he saw that in things which abide not, but
ever fluctuate and fleet away, there can be no stable truth nor basis of
knowledge. Knowledge is of that which abides firm and changes not, if there
exists such in the universe. And now Plato despairs no longer of finding this
existence, he sees it in the principle of universals. But not in the universals
as he received from Sokrates; a change must pass upon them before they will
serve his end" ([1894), p. xxxviii). "<The Sokratic universal is a concept
that> is simply a thought in our mind, it has no existence of its own: it is, as
Protagoras might tell us, doubly unsubstantial; for it is formed from the
impressions produced by an ever-changing object upon a subject that is never
constant: the image of a flitting insect in running water is not more shadowy
than the perceptions from which our definition is formed. Knowledge
demands for its object a constant self-existent verity. This led Plato to the
hypostatisation of the universal. In place of the mental concept derived from
particulars he gives us an essential idea prior to particulars, whereof it is the
cause" (p. xxxix).
23. Fine follows Irwin. "The sightlovers do not acknowledge Forms; all their
accounts or explanations of beauty, justice, and the like, will be phrased in
terms of sensibles. They will define beauty, for example, as the brightly
colored; their accounts will refer to and be based on such observable
properties. But we know from earlier steps, and from elsewhere, that such
properties are F and not F; some cases of bright coloring are beautiful, others
not" ([1978), p. 136).
204 Notes to Chapter 3

24. His book came to my attention too late for me to give it full
consideration, but Moravcsik seems to offer a similar interpretation. "There
is a third argument in the larger section of Book 5 of the Republic that deals
with the 'great contrast' between understanding and opinion. In this argument
(475d-476b) the objects of the opinions and beliefs of the lovers of sights and
sounds are contrasted with the objects of understanding. Further, the latter are
linked to the Forms by the remark that the Fonns have the unity appropriate
to objects of understanding (476a2-6). The argument for the existence of the
Forms can thus be reconstructed as (1) < t > here is understanding and it
requires objects with a certain kind of unity; (2) < t > he Forms are entities
with the required unity; < therefore> < t> he Forms exist as objects of
understanding .... We should note here that this argument can cover both the
range posited by the first and that corresponding to the second of the
arguments surveyed above" ([1992], p. 68).
 Notes to Chapter 4 205

Chapter 4

1. In the context of his attempts to explain the genesis of the Theory of

Forms, Aristotle often puts the words Ka(JOAOV (,universal ') and ioea ('idea')
to different uses. He uses the first in his description of Socrates's search for
definitions, and he uses the second for the metaphysical objects that, by the
time of the middle dialogues, Plato supposed Socrates was asking about.
Aristotle maintains that "Socrates did not make the universals be separate, nor
the definitions, but they did, and called such entities Ideas" (Metaphysics
1078b30-32). Plato typically uses a form of frOOe; ('form') or tOea ('idea') for
what Socrates uses his "What is F?" question to ask about. He never uses
Ka(JOAOV in the corpus (Brandwood [1976)), and he may not use the word at
all. He does, however, once use the phrase KaTa OAOV in the Meno at 77a6,
and this use is an important antecedent to Aristotle's frequent use of Ka(JOAOV.

2. Platonic scholars, however, are not all agreed on this point. Indeed, some
make just the opposite claim. Annas, e.g., insists that "Aristotle's account of
the theory of Forms can be questioned" ([1976], p. 153).

3. Grice, Pears, and Strawson seem to offer this interpretation. "< I > t may
be that Plato believed in the supreme reality of those eternal changeless
entities, the Forms, partly because he wished to preserve the possibility of
scientific knowledge and yet was threatened with having to reject it; for
philosophical reflection made it appear that knowledge of perceptible things
was impossible, since perceptible things were constantly in process of change.
In order, then, to resolve this quandary, to find a subject-matter for scientific
knowledge, Plato was led to introduce, and to exalt, the Forms. But the idea
that knowledge of changeable things is impossible, which gives rise to the
quandary, may itself arise from a preference: from the feeling that to be
unchanging (and so permanent and stable) is so much better than to be
changing (and so fleeting and insecure) that only knowledge of unchanging
things is worthy of the name of knowledge" ([ 1957], pp. 20-21; emphasis in
original).

4. Other scholars make a similar point. Witness the following statements.

"It is a mistake to draw a sharp distinction, as is sometimes done, between a
06~a that apprehends 'YL'YI'OJ.l.fl'a and a 06~a that vaguely and imperfectly
apprehends ollm" (Bluck [1961], p. 36). "<T>o take the distinction in
Republic V between TO' oo~aC1ToII and TO' 'Y"WC1TO" as meaning that the
€7rLC1rr,J.l.WII cannot make valid judgements about questions of practical morality
206 Notes to Chapter 4

is take the distinction out of its context and to use it in a way that Plato can
never have intended. The general purpose of the Republic V passage is to
reject mere 'culture' in favor of hrLCTn,P.T/ as a qualification for ruling the
state, so that Plato must have thought of €7rLCTn,P.OPEC; as being particularly well
equipped to deal with administrative problems in the sensible world" (pp.
37-38; emphasis in original). "< In book V of the Republic, > although Plato
in some way correlates knowledge with Forms, and belief with sensibles, he
does not say that there is knowledge only of Forms or belief only about
sensibles. All he argues is the weaker claim that to know, one must, first of
all, know Forms; restricted to sensibles, one cannot achieve knowledge. This
makes Forms the primary objects of knowledge, but not necessarily the only
ones; knowledge begins, but need not end, with knowledge of Forms" (Fine
[1978], p. 122). "< Socrates> does argue, at the close of the passage < in
book V of the Republic>, that whoever knows will know Forms, since it is
only by reference to them that correct accounts are forthcoming; if one is
restricted to sensibles, like the sightlovers, the most one can attain is belief.
But although knowledge begins with Forms, it need not end with them ... " (p.
139). "< T > his skeptical result would be quite surprising in the context of
the Republic, which aims to persuade us that philosophers should rule, since
only they have knowledge, and knowledge is necessary for good ruling. If
their knowledge is only of Forms--if, like the rest of us, they only have belief
about the sensible world--it is unclear why they are specially fitted to rule in
this world. They don't know, any more than the rest of us do, which laws to
enact" (Fine [1990], p. 86).
5. Annas understands Aristotle's objection similarly. "Forms cannot
contribute to knowledge of things, because they are not their substance or
reality. For Aristotle definitions of what a thing is, which define its reality,
are the first principles of scientific knowledge. The only reason why Forms
cannot be the reality of things is that they are not 'in' them: Aristotle's point
seems to be that Platonic Forms cannot fulfill the useful role of Aristotelian
formal causes, because they are 'separate' and not, like Aristotelian forms,
inherent in certain kinds of matter. This is more a confrontation of two
philosophical positions than an argument. Aristotle adds that if Forms were
literally in things they would make a real contribution to the natures of things,
but that this is not a real option, for the theory of Forms presupposes that it
is impossible" ([1976], p. 160).
6. Another reason to doubt that this explanation is correct is that otherwise
Aristotle attacks the Tbeory of Forms in an uncharacteristically obtuse way.
One direct and easy way for Aristotle to disarm Plato's argument for the
 Notes to Chapter 4 207

conclusion that the universals exist separately would be simply to reject as

obviously false the premise that the ordinary objects are always changing in
every way, but this is something Aristotle never does. Given that he is in
general a very efficient critic, and given the enthusiasm with which he scotches
the Theory of Forms, such an oversight would be extremely surprising. Irwin
makes a similar claim. "< Aristotle> argues against Cratylus ... , < so > if
Aristotle ascribes to Plato this extreme Cratylean view ... , why does he not
attack Plato too? He never suggests that Plato abandoned the doctrine of
sensibles in flux and separated Forms; and he would hardly agree that the
incoherence of extreme Heracleiteanism about the observable world is avoided
by the postulation of unchanging Forms separated from the particulars ...
< H > is failure to criticize Plato... is uncharacteristically and mistakenly
charitable" ([1977a], p. 12).

7. "A central thought is that the Form F is what has the quality F essentially;
this is the heart of the most extended argument for Forms, which recurs in
different guises (Phaedo 74-6, Republic 475-80, 523-5). When we say of
things in our experience that they are just, or equal, we can equally well
ascribe to them the opposite of that quality, for a variety of reasons, such as
applying a different standard. This possibility, it is claimed, shows that in our
experience there are no beautiful things that are not also ugly, no just actions
that are not also unjust. So (unless we are to infer, which Plato never does,
that the use of these terms is always relative to some standard) they cannot be
applied without the possibility of the opposite also applying, within our
experience, but only to the Form of F which is essentially F and never not-F,
and which the F things and actions in our experience 'partake' (in so far as we
can correctly call them F) but also 'fall short of (in so far as we can also say
of them that they are not F)" ([ 1988], pp. 289-290).
8. The Aristotelian corpus stems from an edition that Andronicus created
many generations after Aristotle's death, and some of the material he
incorporated may have come from senior members of the Lyceum. This is far
from certain, but given the evidence concerning the history of Aristotle's
library, it is certainly a real possibility. For discussion of Aristotle's library
and his school, see Lynch [1972) and Grayeff [1974].

9. My translation of passages from the Theaetetus follows Levett's translation

in Burnyeat [1990].

10. Plato has his character make a similar point in the Laws. The Athenian
reports that" < they assert that> right docs not at all exist by nature, but men
are perpetually disputing about it and altering it, and whatever alteration they
208 Notes to Chapter 4

make at any time is at that time authoritative, by design and convention, not
in any way by nature" (88ge6-890a2).

11. "At times, Theaetetus, I come across people who do not seem to me
somehow to be pregnant. Then I realize that they have no need of me, and
with the best will in the world I undertake the business of match-making; and
I think I am good enough at guessing with whom they might profitably keep
company. Many of them I have given away to Prodicus, and a great number
also to other wise and inspired persons" (Theaetetus 151bl-6).

12. Plato also claims that "Heraclitus somewhere says that all things give way
and nothing stays, and likening existing things to the stream of a river, he says
that you would not step twice into the same river" (402a8-1O), but this
attribution is probably not accurate. In an uncontested fragment from the
beginning of his book on nature (Aristotle, Rhetoric 1407bll), Heraclitus
makes the following remarks. "Although this account (AO-Y0C;) holds forever,
men ever fail to comprehend, both before hearing it and once they have heard.
Although all things come to pass in accordance with this account, men are like
the untried when they try such words and works as I set forth, distinguishing
each according to its nature (¢vcnc;) and telling how it holds (exw). But other
men are oblivious of what they do awake, just as they are forgetful of what
they do asleep" (DK 22 B 1, after Kahn [1979], but replacing 'holds' for 'is'
as the last word of the second to last sentence). Heraclitus says nothing in this
fragment to suggest that he believes that rivers, or any other such objects,
cannot survive the passage of time. His stated goal is to show how each thing
"holds" with respect to its "nature." This nature is determined by the AO')'OC;,
and although he could have believed that the AO')'OC; entails that rivers and all
other ordinary objects of experience are such that they cannot survive the
passage of time, this seems very unlikely.

13. Brentlinger rejects Aristotle's explanation of the origin of the Theory of

Forms. A "weakness in the Aristotelian interpretation of separation is that
there is no hint in it of a philosophical reason why Plato should have accepted
the flux doctrine" ([1972a], p. 62; emphasis in original). "<I>t is very
peculiar indeed if Plato, the exponent of rationality, accepted this extreme,
highly paradoxical view, so basic to his philosophy as a whole, as an
unexamined assumption. However, Aristotle's interpretation gives no reason
why he accepted it. We are told only that he was familiar with the doctrine
from his youth--hardly a reason why he should accept it as a mature
philosopher" (pp. 62-63). Brentlinger offers "a different explanation,
according to which the flux theory is not a motive for the separation doctrine,
but a consequence of it" (p. 63; emphasis in original).
 Notes to Chapter 4 209
14. Bolton is happy enough to accept that Plato is inconsistent. "Could Plato
have believed that sensible objects are in no way detenninate or stable and,
hence, in no way accurately describable? Doesn't he claim in the Phaedo that
equal objects are at least deficiently equal (74d-e)? Doesn't he also claim that
such objects are reminiscent of the fonn of equality (75a-b)? And does this
not require a certain stability? If so how could Plato have failed to realize it?
The answer to these questions is that, given what he clearly says, we must
conclude that Plato's position was incoherent.. .. It is perhaps a curious fact
that Plato did not perceive from the beginning the incoherence of his position.
But he is in good company. From Parmenides to Kant to Wittgenstein
philosophers have been unable to maintain silence about that which they took
t') be literally ineffable" (pp. 83-84).

15. In his translation of the Phaedo, Gallop uses 'practically.'

16. Prior presses for a similar interpretation. "Fonns are also constant and
invariant, never changing in any respect (Phaedo 78d), whereas phenomena
are 'entirely opposite to those and so to speak never in the same state either
relative to themselves or to one another' (78e2-4). Again, it is not clear
precisely what is meant here. The constancy of the Fonns is clear enough,
but when Plato asserts that phenomena are never in the same state, does he
mean that every property of every phenomenal object is in change at each
moment, or merely that some property of every object can be found which is
changing at a given moment? The fonner view we may call 'radical
Heracliteanism'; it is the position Plato attributes to Heraclitus in the
Theaetetus and which he shows here to be incoherent (182a-183c). Some
scholars have attributed this view to Plato himself, at least in the middle
dialogues; but, in so far as Plato believes in some similarity between
phenomena and the Fonns, it would perhaps be unwise to saddle him with a
view of phenomena which makes any similarity impossible, given that a less
extreme interpretation of his claim in this passage is possible" ([ 1985], pp.
30-31; emphasis in original).
17. Fine follows Irwin. "It is relational, not real, change that chiefly
concerns Plato in arguing for Forms--that what is large, for example, is also
small, or that what is heavy is also light. Such change, moreover, attaches not
only to sensible particulars--Theaetetus is tall (in relation to Socrates) and short
(in relation to Milo )--but also, and primarily, to sensible properties--the action
type of standing finn in battle is both brave and cowardly. The flux of
sensibles on which Plato focuses attention in arguing for Fonns, then, is
compresence of opposites as it attaches to sensible particulars and, more
especially, to sensible properties" ([1984], p. 66).
210 Chapter 5

1. Taylor argues that Plato uses his character, Timaeus, to express a

fifth-century Pythagorean doctrine, not to express his own views ([1928], p.
11), but his argument is not widely accepted. For convincing criticism of
Taylor's view, see Cornford [1937], pp. v-ix.

2. "With the exception of Taylor, most ancient and modem interpreters have
agreed that the character Timaeus represents Plato's views on cosmology, just
as Socrates in the middle dialogues speaks for Plato" (Prior [1985], p. 87).
3. 'Cosmos' derives from the Greek noun KOCTp.Or;. This noun is formed from
the verb KOCTP.€W, which means "to set in order," "to marshal," or "to
arrange." For helpful discussion of the etymology, see Taylor [1928], pp. 65-
66.
4. This occurrence of the word aEt ('always') is poorly attested in the
Platonic manuscripts. For discussion, see Hackforth [19591, Whittaker [19691,
[19731, Robinson [1979], and Dillon [19891.
5. Unless otherwise indicated, except for a few passages that are based on
Cornford [19371, the translations of passages from the Timaeus are mine.
6. "These so-called 'elements,' or some one or more of them, had been
regarded by Ionian science and by popular thought as the original principles
(apxai) of all things. The earliest Ionians had chosen water or air as the one
original condition (apx~) from which a manifold world had emerged, and also
as the fundamental form (q,VCTLr;) of which all things at all times ultimately
consist. Empedocles had taken all four and clearly endowed them with the
status of elements, irreducible and immutable factors which are merely mixed
and rearranged in space to yield all the variety of compounds. The
unexplained existence of the four elements had been taken as the starting-point
for cosmogony ... " (Cornford [1937], pp. 161-162; emphasis in original).
7. "Plato agreed with Empedocles to the extent of believing that fire, air,
water, and earth are the only distinct corporeal substances which are not
compounds of other distinct corporeal substances. He also accepted, with
Empedocles, that they are irreducibly four in number, no two of them being
in reality identical" (Mills [19681, p. 156).

8. For a helpful discussion of the Receptacle, see Kung [19881.

9. For helpful discussion, see Furth [19881.
 Notes to Chapter 5 211

10. In 11.3 of Generation and Corruption, Aristotle maintains that fire is stuff
that is hot and dry, air is stuff that is wet and hot, water is stuff that is cold
and dry, and earth is stuff that is cold and dry. For an interesting conjecture
about this sort of scheme in Heraclitus's thought, see Hussey [1972], pp.
51-53.

11. In this way, aside from its metaphysical commitments, Plato's view is
reminiscent of modern views. Dummett gives such a modern view. "We
cannot give a correct representation of our language if we do not recognize,
as underlying that level at which we may be said to refer to objects,and to
predicate things of them, a more fundamental level at which there is as yet no
reference to or discernment of objects" ([1981b], p. 216). "<T>he picture
of reality as an amorphous lump, not at the outset articulated into objects, is
a correct one ... " (p. 218). "Such an articulation may be accomplished in any
one of many different ways: we slice up reality into distinct individual objects
by selecting a particular criterion of identity" ([1981a], p. 563). "Any
reference to objects, properly so called, and any predication of objects,
involves the tacit or explicit invocation of a criterion of identity ... " ([1981b],
p. 218). "Countable general terms, in order to be able to be used to form
definite descriptions, have first to be assigned a further use, namely in
statements of identification (where this is understood to mean sentences of the
form 'This is the same X as that: and the like). This involves learning when
it is right to point in two different directions on the same occasion, and say,
'This is the same X as that'; and when it is right to say things like, 'This is
the same X as the one which we saw on such-and-such a previous occasion. '
That feature of the sense of the countable general term 'X' which determines
the truth-conditions for such sentences may be called the criterion of identity
associated with 'X' ... " ([1981a), p. 573).

12. "Plato's Greek <in Timaeus 49c-50b> is in places notoriously hard to

translate, and as a result there has been much divergence between the
interpretations that have been offered of the passages's philosophical
significance" (Mills [1968], p. 152).

13. Lee attempts to defend "the main line" of Cherniss' s translation and
interpretation (p. 1), and for reasons that I shall make clear, I do not accept
Cherniss's interpretation.
14. For detailed and helpful discussions of the gold example, see Lee [1971)
and Mohr [1978].
212 Notes to Chapter 5

15. Zeyl correctly emphasizes this point. He insists that "Plato is drawing
our attention to something that is difficult, not something that is impossible or
illegitimate" ([1975], p. 128; emphasis in the original).
16. The Greek word TOWUTOC; ('such-and-such') is a demonstrative pronoun
that is used in reply to the interrogative 1roioC; ('of what sort'). For discussion
see Smyth [1956], sections 333-340.

17. This is one of the most discussed passages in the corpus. For some of
the many discussions, see Cherniss [1954), Gulley [1960], Lee [1967], Mills
[1968], Zeyl [1975], Mohr [1980), and Gill [1987].

18. The Greek is ambiguous, but for reasons that I shall make clear, I agree
with Cherniss and others that the nouns 1rVP ('fire') and tiowp (,water') are
secondary objects of the verb 1rpo(1cx,,(opeuEtP ('to call'). For argument for this
resolution of the ambiguity, see Cherniss [1954), n. 3 on pp. 115-116; Lee
[1967], n. 9 on p. 4; Mills [1968], p. 154; Reed [1972), pp. 65-66; and Mohr
[1980], n. 5 on p. 87. For argument for taking the nouns to be primary
objects, see Zeyl [1975], pp. 131-136.
19. Dummett makes a similar point about Frege. "His recognition that senses
are not inner mental contents was one of his principal motives for locating
them outside the mind in a special compartment of reality independent of our
apprehension of it.... Only so did he think he could safeguard the objectivity
of sense: and that was precisely because he did not have the idea of identifying
the sense of a word with its observable use" ([1981b], p. 52). "We have seen
that the objectivity of sense is sufficiently guaranteed by its being expressed
within the common language: it was not necessary for Frege, in order to
safeguard that objectivity, to view it as having an existence independent even
of the means of expressing it" ([1981a], p. 680).

20. In his interpretation of the Timaeus, Lee follows Cherniss on this point.
See [1966], pp. 359-360 and [1967], pp. 17-18.

21. "Cherniss's view appears to entail that Plato's theory recognizes four
types of levels--the Forms, the Receptacle, the copies of Forms and
phenomenal appearances. Yet Plato himself is emphatic on the point that in
this paragraph his concern is to make out a threefold distinction--to introduce
just one extra item (viz., the Receptacle) to be added to the twofold distinction
already drawn at 27d-28a)" (Mills [1968), p. 154; emphasis in original). "The
introduction of a fourth ontological stratum, however, runs up against Plato's
emphatic and repeated assertion that he is dealing with only three strata (cf.
Mills 154, 170)" (Mohr [1978), p. 106). "It is well known that Cherniss'
 Notes to Chapter 5 213

reading commits Plato to a fourfold division of entities (Form, Receptacle,

immanent character or property and instance of property), whereas Plato
insists (51e-52b) on a threefold division (cf. Mills, 1968, pp. 153-154, 170)"
(Prior [1985], n. 15 on p. 124). "It is puzzling that items which are meant to
do so much work simply escape the net of Plato's classification" (Gill [1987],
p.42).
22. The controversy is very familiar to Platonic scholars. Hackforth
describes it as follows. "As is well known, there have been, from the very
first age of Platonic study, two directly opposed opinions on the main point.
Aristotle seems to regard it as plain, incontrovertible fact that Plato described
the universe as having come to be; whereas his contemporary Xenocrates, the
third Head of the Academy, held that the account put into the mouth of
Timaeus ought to be interpreted as analysis of the world's existing structure
in the guise of a story of its construction in the past, Plato's object being to
facilitate exposition ... " ([1959), p. 9).
23. There are some scholars who do not reject the Demiurge as a mythical
figure, but nevertheless favor a more literal interpretation. Witness, e. g.,
Vlastos's remarks. "On one occasion he complains that those who had posited
earth, water, air, and fire as the CTTOtXEia of the universe had not penetrated
very deeply: what they had taken as the letters of the alphabet of nature should
be ranked, he says, not even as low as syllables for this purpose. Here he
must have had Empedocles in view, for it was he who had made earth, water,
air, and fire the elements of the natural universe, endowing them with
absolute, Parmenidean, unalterability to make them qualify for this purpose.
So in offering us his own CTTOtxeLa Plato shows that he has enrolled in the
same program of physical elementarism. Like Empedocles, Anaxagoras, and
the atomists, he is undertaking to identify those recurrent elements whose
shufflings and reshufflings would explain the lawlike constancies of the flux"
([1975], pp. 67-68). "<An important difference, however, is that the>
indivisibles of his physics were still more remote from the bodies of
sense-experience: they were not even bodies, but only bounding surfaces of
bodies. His atoms are two-dimensional" (p. 69).

24. For further criticism of this "Academic" interpretation, see Vlastos [1939)
and [19641.

25. Although Mohr follows Cherniss against Cornford and the traditional
translation, his interpretation is very much like Cornford's. He offers a
"double aspect" interpretation. "I take the main point of the much debated
passage 49b-e to be the establishment of a contrast between phenomena taken,
214 Notes to Chapter 5

on the one hand, as being in flux, about which, as the result of their being in
flux, nothing whatsoever may be said, and the phenomena taken, on the other
hand, as images. As images of Ideas, the phenomena may be said to be
phenomena of a certain sort (TOtOVTOV). The development of this double aspect
of the phenomena (taken as flux and taken as images of Ideas) is the main
point, as I see it, of 47e-52c" ([19781, pp. 99-100).

26. For another attempt to see Plato's remarks in Timaeus 49a-50a as a

response to his earlier remarks in Theaetetus 182c-183b, see Gill [1987].

27. Although Zeyl accepts the traditional translation, he puts forward

essentially the same interpretation as the one that Mills put forward in 1968.
"What, then, are we to say of those experiences, every day events in the lives
of all men, in which we see something becoming now one, now another, of
the four? For example, whenever we boil water, we see something which,
though it begins as water, becomes by the end of the process air, or at any
rate a substance of which air is a major component; and again, when coal or
wood is ignited, something which begins as earth, or as predominately earthy,
finishes as a flame composed of fire. Confronted with such facts as these we
speak of water's becoming air and of earth's becoming fire. But if that which
comes to be air were water at the time of its so becoming, it would then be
both water and air at once, which has been admitted to be quite impossible ....
It thus becomes clear that in any change in which one of the Empedoclean four
is perceived as becoming another, the persisting substrate of the change--that
which retains its identity throughout, though assuming now this character, now
that--cannot itself be any of the Empedoclean four. What, then, is it? ... It is
in fact space--not a body or stuff, but the room which bodies are said to
occupy" (pp. 156-157). Prior accepts the traditional translation ([1985]. pp.
109-100, n. 15 on p. 142), and he puts forward an interpretation that is
essentially the same as Mills's and Zeyl's. "The apparent transmutation of the
elements--like Plato's claim that they are not 'letters but 'syllables' or worse
in the grammar of creation--again reveals the need to find an unchanging,
elementary underpinning to serve as the substantial base from which these
complex, constantly changing 'elements' can be constructed. This substantial
base is the Receptacle" (p. 109). "The Receptacle plays the role of
substratum, a subject, a substance, of which entities in the category of
Becoming are predicated or to which they are attributed" (p. 113). "<I>t
seems fair to characterize the phenomenal images which make up the category
of Becoming as accidental properties of the Receptacle" (p. 115). This sort
of interpretation seems to go back to Taylor's 1928 commentary on the
Timaeus. "The 'cyclical' transition of the 'four' familiar bodies into one
 Notes to Chapter 5 215

another was the standing doctrine of the earlier science, before the criticism
of Parmenides had driven home the conviction that what really is cannot
merely 'arise from' and 'vanish into' something permanent.. .. His pluralistic
successors have the task laid on them of saying what the 'permanent' which
persists through change is. Empedocles had found it in the 'roots' as opposed
to their perishable compounds .... Timaeus also is concerned to find a solution
of the problem, but his solution goes deeper than that of Empedocles" (p.
314). "< H > e pronounces against Empedocles' way of facing the problem
by simply asserting that the four 'roots' are nature's invariant; they are only
passing phases in the life of nature (49dl-3). His way of expressing this is to
say that nothing which we see changing its character (aAA07E CxAATI
,,(L,,(VOfJ.EVOV, d5) ought to be called 'this' (70V70); it should only be called
70WU70V, 'this-like.' We must never use the words 'fire' and the like which
would suggest that they are permanent 'things,' i.e., we ought not to use
substantives, or pronouns which are equivalent to them, in speaking of
"(L"(VOfJ.Eva. We should use only adjectives and adjectival pronouns, as that
suggests rightly that what the words stand for is only a 'phase' of something,
merely adjectival qualification of the really permanent" (pp. 315-316).
28. Empedocles denies that "mortal things" have "birth" and "death" (DK 31
B 8), and seems to say that although he is content to use the terms, so as to
comply with "custom," he does not take himself to be committed to their
customary implications (B 9). For some discussion, see Kirk [1983], p. 292.

29. Rowe makes a similar claim. "It is only in this sense that Plato impugns
the 'existence' of the world, that is, in the sense that its parts are not securely,
fully, genuinely what they claim to be" ([1984], p. 58).
30. The word OV is a participle of the verb E'LfJ.L. which means "to be." The
word ,,(L,,(VOfJ.EVOV is a participle of the verb "(i"(vofJ.aL, which means "to
become."
31. "I am basically in agreement with Professor Frede's conception of the
metaphysical picture involved in the Platonic distinction between a realm of
being and a realm of becoming, and am already on record as endorsing the
view that (according to this contrast) the inhabitants of the realm of
becoming are not endowed with essences, or essential natures, whereas the
inhabitants of the realm of being are so endowed. The essence of an F is
specified by a definition (that is to say, a correct answer to a 'What is X?'
question), and as such a definition applies (strictly speaking) only to the
form of F-ness, and hence the Fornl of F-ness is the real F. The many
sensible things that we call F are not really F, and are called F not because
216 Notes to Chapter 5

that is their nature, but rather because they stand in some appropriate
relation (participation, or whatever) to the real F. Frede is quite right in
pointing out that the many sensible F's merely display the character of an
F, without being F essentially, without being F of their own nature" (Code
[1988], p. 53; emphasis in original). "Whereas Socrates assumed that the
definition of the F applies to all of the F particulars, and that each of those
particulars is 'such as' the essence (overia) of the F itself, Plato believed
that the many F particulars are all deficient in being 'such as' the F itself,
and that striving to be such, they none the less fall sort (see Phaedo 74).
This is the famous Platonic separation of the universal" ([1985], pp.
425-426; emphasis in original). "Plato accounts for and explains the
application of language to items in the sensible realm in terms of what those
items Have (at least in some key cases, in terms of the separable Forms in
which they participate), not in terms of what they Are--for there is nothing
definable or scientifically knowable that they Are. The Forms, but not the
sensible particulars, are endowed with essential natures--the very natures
that the Forms are" (p. 427; capitals in 'Are' and 'Have' in original).

32. Cf. Timaeus 50c5 and 51b6.

33. Allen makes a similar claim. "Plato's metaphor of imitation brilliantly

expresses a community between different orders of objects. different levels
of reality.... < It> expresses a fundamental thesis of his ontology. that
particulars differ from Forms, as resemblances differ from originals, in
degree of reality. For particulars 'fall short' of their Forms, and are
'deficient' with respect to them" ([1960], p. 51). "Particulars and Forms
are not merely different types of things; they are types of things which
differ in degree of reality, for the one is wholly dependent on the other.
Particulars have no independent ontological status; they are purely relational
entities, entities which derive their whole character and existence from
Forms. Because their being is relational. adjectival. dependent..." (p. 57;
emphasis in original). Lee also makes a similar claim. "The 'insubstantial
image' is not an entity related to a Form; it is a product of a
relation--perhaps just the holding of a relation--between something else and
the Form" ([1966], p. 365; emphasis in original).
 BIBLIOGRAPHY

Adam, 1. The Republic of Plato. Cambridge University Press, 1905.

Allen R. E. "Participation and Predication in Plato's Middle Dialogues."

Philosophical Review, 69 (1960), 147-164. Reprinted in Allen [1965], pp.
43-60.

--Studies in Plato's Metaphysics. Routledge & Kegan Paul, 1965.

--"Plato's Earlier Theory of Fonus." Paper read at Princeton University,

March 14, 1967. Reprinted in Vlastos [1971], pp. 319-334.

--Plato's Parmenides. University of Minnesota Press, 1983.

Annas,l. Aristotle's Metaphysics, Books M and N. Oxford University Press,

1976.

--" Aristotle on Inefficient Causes." Philosophical Quarterly, 32 (1982),

311-326.

--"Classical Greek Philosophy." In Boardman [1988], pp. 277-305.

Anton, 1. Essays in Ancient Philosophy. State University of New York Press,

1971.

Archer-Hind, R. D. The Timaeus of Plato. Macmillan and Co., 1888.

--The Phaedo of Plato, 2nd ed. Macmillan and Co., 1894.

Bambrougb, R. New Essays in Plato and Aristotle. Humanities Press, 1965.

Benson, H. "The Priority of Definition and the Socratic Elenchus." Oxford

Studies in Ancient Philosophy, 8 (1990), 19-65.

Beversluis, 1. "Socratic Definition." American Philosophical Quarterly, 11

(1974), 331-336.
218 Bibliography

--"Does Socrates Commit the Socratic Fallacy." American Philosophical

Quarterly, 24 (1987), 211-223.

Bluck, R. S. Plato's Meno. Cambridge University Press, 1961.

Boardman, J., et al. The Oxford History of Greece and the Hellenistic World.
Oxford University Press, 1988.

Bolton, R. "Plato's Distinction between Being and Becoming." Review of

Metaphysics, 29 (1975), 65-95.

Brandwood, L. A Word Index to Plato. W.S. Maney, 1976.

--The Chronology of Plato's Dialogues. Cambridge University Press, 1990.

Brentlinger, J.
"Incomplete Predicates and the Two-World Theory of the
Phaedo." Phronesis, 17 (1972), 61-79.

--"Particulars in Plato's Middle Dialogues." Archiv fur Geschichte der

Philosophie, 54 (1972), 116-152.

Burge, E. "The Ideas as Ai,rLa in the Phaedo." Phronesis, 16 (1971), 1-13.

Burnet, J. Platonis Opera, 5 vols. Oxford University Press, 1900-1907.

--Plato's Phaedo. Oxford University Press, 1911.

--Plato's Euthyphro, Apology ofSocrates, and Crito. Oxford University Press,

1924.

--Early Greek Philosophy, 4th. ed. Adam & Charles Black, 1930.

Burnyeat, M. "Examples in Epistemology: Socrates, Theaetetus, and G. E.

Moore." Philosophy, 52 (1977), 381-398.

--Review of Terry Irwin's Plato's Moral Theory. New York Review of Books ,
27 September 1979, 56-60.

--The Theaetetus of Plato. Hackett Publishing Company, 1990.

 Bibliography 219

Camap, R. "Empiricism, semantics, and ontology." Revue Internationale de

Philosophie, 4 (1950), 20-40.

Code, A. "Aristotle: Essence and Accident." In Grandy [1985], pp. 411-439.

--"Reply to Michael Frede's 'Being and Becoming in Plato. '" Oxford Studies
in Ancient Philosophy. Supplementary Volume, (1988), 53-60.

Cherniss, H. "The Philosophical Economy of the Theory of Ideas. " American

Journal of Philology, 57 (1936),445-456. Reprinted in Vlastos [1971bl, pp.
16-27.

--"A Much Misread Passage of the Timlleus (49c7-50b5)." American Journal

of Philology, 75 (1954), 113-130.

--"The Relation of the Timlleus to Plato's Later Dialogues." The American

Journal of Philology, 311 (1957), 225-266.

Cornford, F. Plato's Cosnwlogy. Routledge & Kegan Paul, 1937.

--Plato and Parmenides. Routledge & Kegan Paul, 1939.

Cresswell, M. "Plato's Theory of Causality: Phaedo 95-106." Australasian

Journal of Philosophy, 49 (1971), 248-249.

Crombie. I. M. An Examination of Plato's Doctrines, vol. II: Plato on

Knowledge and Reality. Routledge & Kegan Paul, 1963.

Dillon, J. "Tampering with the Timlleus: Ideological Emendations in Plato,_

with special reference to the Timlleus." American Journal of Philology, 110
(1989), 50-72.

Dummett, M. Frege: Philosophy of Language, 2nd. ed. Harvard University

Press, 1981.

--The Interpretation of Frege's Philosophy. Harvard University Press, 1981.

Everson, S. Companions to Ancient Thought, vol. I: Epistenwlogy.

Cambridge University Press, 1990.
220 Bibliography

Fine, G. "Knowledge and Belief in Republic V." Archiv fur Geschichte der
Philosophie, 60 (1978), 121-139.

--"Separation." Oxford Studies in Ancient Philosophy, 2 (1984), 31-88.

--"Separation: a Reply to Morrison." Oxford Studies in Ancient Philosophy,

3 (1985), 159-166.

--"Owen's Progress." Philosophical Review, 97 (1988), 373-399.

--"Knowledge and Belief in Republic V-VII." In Everson [1990], pp. 85-115.

--On Ideas. Oxford University Press, 1993.

Frede, M. Essays in Ancient Philosophy. University of Minnesota Press,

1987.

-- "Introduction: the Study of Ancient Philosophy." In Frede [1987], pp.

ix-xxvii.

-- "The Unity of General and Special Metaphysics: Aristotle's Conception of

Metaphysics." In Frede [1987], pp. 81-95.

-- "Being and Becoming in Plato." Oxford Studies in Ancient Philosophy.

Supplementary Volume, (1988), 37-52.

Furth, M. Substance, Form and Psyche: an Aristotelian Metaphysics.

Cambridge University Press, 1988.

Gallop, D. Plato, Phaedo. Oxford University Press, 1975.

Geach, P. "Plato's Euthyphro: an Analysis and Commentary." Monist, 50

(1966), 369-382.

--God and the Soul. Routledge & Kegan Paul, 1969.

Gill, M. "Matter and Flux in Plato's Timaeus." Phronesis, 32 (1987), 34-53.

Gosling, I. "Republic V: Tei 7ro"A"Aa mAa." Phronesis, 5 (1960), 116-128.

 Bibliography 221

--Plato. Routledge & Kegan Paul, 1973.

Grandy, R., et al. Philosophical Grounds of Rationality. Oxford University

Press, 1985.

Grayeff, F. Aristotle and his School. Duckworth, 1974.

Grice, H., et al. "Metaphysics." In Pears [1957], pp. 1-22.

Grote, G. Plato and the Other Companions of Socrates, vol. I. John Murray,
1885.

Grube, G. M. A. Plato's Thought. Methuen & Co., 1935.

--Plato's Republic. Hackett Publishing Company, 1974.

Gulley, N. "The Interpretation of Plato, Timaeus 49D-E." American Journal

of Philology, 81 (1960), 53-6i.

--Plato's Theory of Knowledge. Methuen & Co., 1962.

Guthrie, W. K. History of Greek Philosophy, vol. III. Cambridge University

Press, 1969.

Hackforth, R. Plato's Phaedo. Cambridge University Press, 1955.

--"Plato's Cosmogony (Timaeus 27dff.)." Classical Quarterly, 9 (1959),

17-22.

Hamilton, E., et al. Plato. Princeton University Press, 1961.

Hardy, W. A Study in Plato. Oxford University Press, 1936.

Hare, R. "Philosophical Discoveries." Mind, 69, (1960), 145-162.

Hussey, E. The Presocratics. Duckworth, 1972.

--"The beginnings of epistemology: from Homer to Philolaus." In Everson

[1990], pp. 11-38.
222 Bibliography

Irwin, T. "Plato's Heracleiteanism." Philosophical Quarterly, 27 (1977),

1-13.

--Plato's Moral Theory. Oxford University Press, 1977.

--Classical Thought. Oxford University Press, 1989.

Kahn, C. "Language and Ontology in the Cratylus." In Lee [1973], pp.

152-176.

--The Art and Thought of Heraclitus. Cambridge University Press, 1979.

--"The Beautiful and the Genuine: a discussion of Paul Woodruff, Plato,

Hippias Major." Oxford Studies in Ancient Philosophy, 3 (1985), 261-287.

Kirk, G. S., et al. The Presocratic Philosophers, 2nd ed. Cambridge

University Press, 1983.
•
Kirwan, C. "Plato and relativity." Phronesis, 19 (1974), 112-129.

Kraut, R. The Cambridge Companion to Plato. Cambridge University Press,

1992.

--"Introduction to the study of Plato." In Kraut (1992), pp. 1-50.

Kung, J. "Why the Receptacle is Not a Mirror." Archiv fur Geschichte der
Philosophie, 70 (1988), 167-178.

Lee, E. "On the Metaphysics of Image in Plato's Timneus." Monist, 50

(1966), 341-368.

--"On Plato's Timneus 49d4-e7." American Journal of Philology, 88 (1967),

1-28.

--"On the Gold Example in Plato's Timneus (50a4-b5). In Anton (1971), pp.
219-235.

--Exegesis and Argument. Van Gorcum & Company, 1973.

 Bibliography 223

Lloyd, A.C. Form and Universal in Aristotle. Francis Cairns, 1981.

Lynch, P. Aristotle's School. University of California, 1972.

Mace, C. A. British Philosophy in Mid-Century. George Allen & Unwin,

1957.

Malcolm, 1. "On the Place of the Hippias Major in the Development of

Plato's Thought." Archiv fur Geschichte der Philosophie, 50 (1968), 189-195.

Mates, B. "Leibniz and the Phaedo." Studia Leibnitiana Supplementa, 12

(1973), 135-148.

Matthews, G. "Senses and Kinds." Journal of Philosophy, 69 (1972),

149-157.

Mills, K. W. "Some Aspects of Plato's Theory of Forms, Timaeus 49cff."

Phronesis, 13 (1968), 145-170.

Mohr, R. "The Gold Analogy in Plato's Timaeus (50a4-b5)." Phronesis, 23

(1978), 243-252. Reprinted in Mohr [1985], pp. 99-107.

--"Image, Flux, and Space in Plato's Timaeus." Phoenix 34 (1980), 138-152.

Reprinted in Mohr [1985], pp. 85-98.

--The Platonic Cosmology. E. 1. Brill, 1985.

Moravcsik, 1. Plato and Platonism. Blackwell, 1992.

Morrison, D. "Separation in Aristotle's Metaphysics." Oxford Studies in

Philosophy, 3 (1985), 125-158.

--"Separation: a Reply to Fine." Oxford Studies in Philosophy, 3 (1985),

167-174.

Negal, T. The View from Nowhere. Oxford University Press, 1986.

Nehamas, A. "Plato on the Imperfection of the Sensible World." American

Philosophical Quarterly, 12 (1975), 105-117.
224 Bibliography

--" Socratic Intellectualism." Proceedings of the Boston Area Colloquium in

Ancient Philosophy. 2 (1987). 275-316.

Nussbaum. M. "Eleatic conventionalism and Philolaus on the conditions of

thought." Harvard Studies in Classical Philology. 83 (1979). 63-108.

Owen. G. E. L. "The Place of the Timaeus in Plato's Dialogues." Classical

Quarterly. 3 (1953). 79-98.

--"A Proof in the I1epL IoewJI." Journal of Hellenistic Studies. 77 (1957).

103-111.

Pears. D. The Nature of Metaphysics. Macmillan & Co .• 1957.

Penner. T. The Ascent from Nominalism. D. Reidel, 1987.

Prior, W. Unity and Development in Plato's Metaphysics. OpenCourt, 1985.

Quine. W. V.O. "On What There Is." Review of Metaphysics. 2 (1948),

21-38. Reprinted in Quine [1980], pp. 1-19.

--From a Logical Point of View. 2nd. ed., rev. Harvard University Press,
1980.

Reed. N. H. "Plato on Flux. Perception. and Language." Proceedings of the

Cambridge Philological Society, 18 (1972), 65-77.

Robinson, T. "The Argument of Timaeus 27dff." Phronesis, 24 (1979).

105-109.

Robinson, R. Plato's Earlier Dialectic. Cornell University Press, 1941.

Ross. D. Aristotle's Metaphysics. 2 vols. Oxford University Press, 1924.

--Plato's Theory of Ideas. Oxford University Press, 1953.

Rowe. C. 1. Plato. St. Martin's Press. 1984.

Russell, B. The Problems of Philosophy. Home University Library, 1912.

 Bibliography 225

Ryle, G. "The Theory of Meaning." In Mace [1957], pp. 239-264.

Santas, G. "The Socratic Fallacy." Journal of the History of Philosophy, 10

(1972), 127-141.

Searle, J. Speech Acts. Cambridge University Press, 1969.

Stough, C. "Forms and Explanation in the Phaedo." Phronesis, 21 (1976),

1-30.

Symth, H. Greek Grammar. Harvard University Press, 1956.

Taylor, A. E. A Commentary on Plato's Timaeus. Oxford University Press,

1928.

--Plato: the Man and His Work. 6th ed. Meuthen, 1949.

Taylor, C. "Forms as Causes in the Phaedo." Mind, 78 (1969), 45-59.

Vlastos, G. "The Disorderly Motion in the Timaeus." Classical Quarterly, 33

(1939),71-83. Reprinted with minor changes in Allen [1965), pp. 379-399.

--"Creation in the Timaeus: is it a Fiction?" In Allen [1965], pp. 401-419.

--"Degrees of Reality in Plato." In Bambrough (1965), pp. 1-19. Reprinted

in Vlastos [1981). pp. 43-57.

--" A Metaphysical Paradox." Proceedings of the American Philosophical

Association, 39 (1966), 5-19. Reprinted in Vlastos [1981], pp. 58-75.

--"Reasons and Causes in the Phaedo." Philosophical Review, 78 (1969),

291-325.

--The Philosophy of Socrates. Anchor Books, 1971.

--Plato, vol. I. Anchor Books, 1971.

--Plato's Universe. University of Washington Press. 1975.

226 Bibliography

--Platonic Studies, 2nd. ed. Princeton University Press, 1981.

--"'Separation' in Plato." Oxford Studies in Philosophy, 5 (1987), 187-196.

West, M. "Early Greek Philosophy." In Boardman [1988], pp. 126-141.

White, N. Plato on Knowledge and Reality. Hackett Publishing Co., 1976.

--A Companion to Plato's Republic. Hackett Publishing Co., 1979.

--"Review of Terry Penner's The Ascent from Nominalism." Philosophical

Review, 100 (1991), 318-321.

Whittaker, I. "Timaeus 27d5ff." Phoenix, 23 (1969), 181-182.

--"Textual Comments on Timaeus 27c-d." Phoenix, 27 (1973), 387-388.

Woodruff, P. "Socrates and ontology: the evidence of the Hippias Major."

Phronesis, 23 (1978), 101-117.

--Plato, Hippias Major. Hackett Publishing Co., 1982.

Zeyl, D. "Plato and Talk of a World in Flux: Timaeus 49a6-50b5." Harvard

Studies in Classical Philology, 79 (1975), 125-148.

--"Comments on Robinson." Proceedings of the Boston are Colloquium in

Ancient Philosophy, 2 (1987), 120-125.
 PHILOSOPHICAL STUDIES SERIES
Founded by Wilfrid S. Sellars and Keith Lehrer

Editor:

KEITH LEHRER, University of Arizona

Board of Consulting Editors:

Jonathan Bennett, Allan Gibbard, Robert Stalnaker, and Robert G. Turnbull

I. JAY F. ROSENBERG, Linguistic Representation, 1974.

2. WILFRID SELLARS, Essays in Philosophy and Its History, 1974.
3. DICKINSON S. MILLER, Philosophical Analysis and Human Welfare. Selected
Essays and Chapters from Six Decades. Edited with an Introduction by Lloyd D.
Easton, 1975.
4. KEITH LEHRER (ed.), Analysis and Metaphysics. Essays in Honor of R. M.
Chisholm. 1975.
5. CARL GINET, Knowledge, Perception, and Memory, 1975.
6. PETER H. HARE and EDWARD H. MADDEN, Causing, Perceiving and
Believing. An Examination of the Philosophy of C. 1. Ducasse, 1975.
7. HECTOR-NERI CASTANEDA, Thinking and Doing. The Philosophical
Foundations of Institutions, 1975.
8. JOHN L. POLLOCK, Subjunctive Reasoning, 1976.
9. BRUCE AUNE, Reason and Action, 1977.
10. GEORGE SCHLESINGER, Religion and Scientific Method, 1977.
11. YIRMIAHU YOVEL (ed.), Philosophy of History and Action. Papers presented
at the first Jerusalem Philosophical Encounter, December 1974, 1978.
12. JOSEPH C. PITT, The Philosophy of Wilfrid Sellars: Queries and Extensions,
1978.
13. ALVIN I. GOLDMAN and JAEGWON KIM, Values and Morals. Essays in
Honor of William Frankena, Charles Stevenson, and Richard Brandt, 1978.
14. MICHAEL 1. LOUX, Substance and Attribute. A Study in Ontology, 1978.
15. ERNEST SOSA (ed.), The Philosophy of Nicholas Rescher: Discussion and
Replies, 1979.
16. JEFFRIE G. MURPHY, Retribution, Justice, and Therapy. Essays in the
Philosophy of Law, 1979.
17. GEORGE S. PAPPAS, Justification and Knowledge: New Studies in Epistemol-
ogy, 1979.
18. JAMES W. CORNMAN, Skepticism, Justification, and Explanation, 1980.
19. PETER VAN INWAGEN, Time and Cause. Essays presented to Richard Taylor,
1980.
20. DONALD NUTE, Topics in Conditional Logic, 1980.
21. RISTO HILPINEN (ed.), Rationality in Science, 1980.
22. GEORGES DICKER, Perceptual Knowledge, 1980.
23. JAY F. ROSENBERG, One World and Our Knowledge of It, 1980.
24. KEITH LEHRER and CARL WAGNER, Rational Consensus in Science and
Society, 1981.
25. DAVID O'CONNOR, The Metaphysics of G. E. Moore, 1982.
26. JOHN D. HODSON, The Ethics of Legal Coercion, 1983.
27. ROBERT 1. RICHMAN, God, Free Will, and Morality, 1983.
28. TERENCE PENELHUM, God and Skepticism, 1983.
29. JAMES BOGEN and JAMES E. McGUIRE (eds.), How Things Are, Studies in
Predication and the History of Philosophy of Science, 1985.
30. CLEMENT DORE, Theism, 1984.
31. THOMAS L. CARSON, The Status of Morality, 1984.
32. MICHAEL J. WHITE, Agency and Integrality, 1985.
33. DONALD F. GUSTAFSON, Intention and Agency, 1986.
34. PAUL K. MOSER, Empirical Justification, 1985.
35. FRED FELDMAN, Doing the Best We Can, 1986.
36. G. W. FITCH, Naming and Believing, 1987.
37. TERRY PENNER, The Ascentfrom Nominalism. Some Existence Arguments in
Plato's Middle Dialogues, 1987.
38. ROBERT G. MEYERS, The Likelihood of Knowledge, 1988.
39. DAVID F. AUSTIN, Philosophical Analysis. A Defense by Example, 1988.
40. STUART SILVERS, Rerepresentation. Essays in the Philosophy of Mental
Rerepresentation, 1988.
41. MICHAEL P. LEVINE, Hume and the Problem of Miracles. A Solution, 1979.
42. MELVIN DALGARNO and ERIC MATTHEWS, The Philosophy of Thomas
Reid,1989.
43. KENNETH R. WESTPHAL, Hegel's Epistemological Realism. A Study of the
Aim and Method of Hegel's Phenomenology of Spirit, 1989.
44. JOHN W. BENDER, The Current State of the Coherence Theory. Critical Essays
on the Epistemic Theories of Keith Lehrer and Laurence Bonjour, with Replies,
1989.
45. ROGER D. GALLIE, Thomas Reid and 'The Way ofldeas', 1989.
46. J-c. SMITH (ed.), Historical Foundations of Cognitive Science, 1990.
47. JOHN HElL (ed.), Cause, Mind, and Reality. Essays Honoring C. B. Martin,
1990.
48. MICHAEL D. ROTH and GLENN ROSS (eds.), Doubting. Contemporary
Perspectives on Skepticism, 1990.
49. ROD BERTOLET, What is Said. A Theory of Indirect Speech Reports, 1990
50. BRUCE RUSSELL (ed.), Freedom, Rights and Pornography. A Collection of
Papers by Fred R. Berger, 1990
51. KEVIN MULLIGAN (ed.), Language, Truth and Ontology, 1992
52. JESUS EZQUERRO and JESUS M. LARRAZABAL (eds.), Cognition,
Semantics and Philosophy. Proceedings of the First International Colloquium on
Cognitive Science, 1992
53. O.H. GREEN, The Emotions. A Philosophical Theory, 1992
54. JEFFRIE G. MURPHY, Retribution Reconsidered. More Essays in the
Philosophy of Law, 1992
55. PHILLIP MONTAGUE, In the Interests of Others. An Essay in Moral
Philosophy, 1992
56. J.- P. DUBUCS (ed.), Philosophy of Probability. 1993
57. G.S. ROSENKRANTZ, Haecceity: An Ontological Essay. 1993
58. C. LANDESMAN, The Eye and the Mind. Reflections on Perception and the
Problem of Knowledge. 1993
59. P. WEINGARTNER (ed.), Scientific and Religious Belief 1994
60. MICHAELIS MICHAEL and JOHN O'LEARY-HAWTHORNE (eds.),
Philosophy in Mind. The Place of Philosophy in the Study of Mind. 1994
61. W.H. SHAW, Moore on Right and Wrong. The Normative Ethics of G.E. Moore.
1995
62. T.A. BLACKSON, Inquiry, Forms, and Substances. A Study in Plato's
Metaphysics and Epistemology. 1995