The Idea of the Good in Platonic-Aristotelian Philosophy

THE
IDEA OF THE G O O D
IN
PLATONIC-ARISTOTELIAN
PHILOSOPHY

HANS-GEORG

GADAMER

T R A N S L A T E D A N D WITH AN
INTRODUCTION AND ANNOTATION
BY P. C H R I S T O P H E R S M I T H

YALE UNIVERSITY P R E S S
NEW HAVEN A N D L O N D O N

The preparation of this volume was made possible

in part by a grant from the Program for
Translations of the National
Endowment for the Humanities,
an independent federal agenqr.
Copyright 1986 by Yale University.
All rights reserved.
This book may not be reproduced, in whole
or in part, in any form (beyond that
copying permitted by Sections 107 and 108
of the U.S. Copyright Law and except by
reviewers for the public press), without
written permission from the publishers.
Designed by Sally Harris
and set in Sabon type by Eastern Graphics
Printed in the United States of America by
Vail-Ballou Press, Binghamton, N.Y.
Library of Congress Cataloging-in-Publication Data
Gadamer, Hans Georg, 1900The idea of the good in
Platonic-Aristotelian philosophy.
Translation of: Die Idee des Guten
zwischen Plato und Aristoteles.
Includes index.
1. PlatoContributions in ethicsAddresses, essays,
lectures. 2. AristotleContributions in ethics
Addresses, essays, lectures. 3. Ethics, Ancient
Addresses, essays, lectures. I. Title.
B398.E8G2913 1986
170
85-22710
ISBN 0 - 3 0 0 - 0 3 4 6 3 - 6 (alk. paper)
The paper in this book meets the guidelines for
permanence and durability of the Committee on
Production Guidelines for Book Longevity
of the Council on Library Resources.
10

9 8 7 6 5 4 3

CONTENTS

Translator's Introduction vii

Preface 1
I The Question at Issue 7
II Socratic Knowing and Not-Knowing 33
III The Polis and Knowledge of the Good 63
IV The Dialectic of the Good in the Philebus 104
V Aristotle's Critique of the Idea of the Good 126
VI The Idea of Practical Philosophy 159
Index 179

TRANSLATOR'S
INTRODUCTION

Though shorter than Truth and Method} and, as Gadamer readily acknowledges, not a completely unified study. The Idea of the
Good in Platonic-Aristotelian Philosophy^ must be counted
nonetheless among Gadamer's most important books. For one
thing, it gives us an extended example of the hermeneutical or interpretive techniques for which Gadamer has become so well
known; for another, it provides us with remarkable new insights
1. Hans Georg Gadamer, Wahrheit und Methode (Tbingen, 1965), henceforth WM. This work has appeared in translation as Truth and Method (New
York, 1975). Translations of passages cited here are my own.
2. Gadamer, Die Idee des Guten zwischen Plato und Aristoteles (Heidelberg,
1978). The zwischen is difficult to render in ordinary English. "Between," naturally, will not do, and Gadamer intentionally avoids the more common "von . . .
zu" (from . . . to) and "in." He found the circumlocution I have used here appropriate for two reasons, which should become clearer in the course of this introduction. Briefly, he gives priority not to the individual authors as such, but to the
thought we find expressed in their works. Hence he tends to say "the Socratic
question" and "Platonic philosophy" rather than Socrates' question and Plato's
philosophy. Furthermore, he includes expositions of texts whose authorship may
be contested, for example, the Magna Moralia, but whose content is clearly Platonic or Aristotelian "heritage." He also wants to stress that one should not approach these thinkers "developmentally," which is to say, in a way that makes it
appear that Aristode broke with Plato. Instead, he contends, it is best to think of
a shared, continuous tradition to which both belong. In short, the choice of
zwischen is meant, above all, to set off Gadamer's approach from that of Werner
Jaeger. See Jaeger, Aristoteles, Grundlegung einer Geschichte seiner Entwicklung
(Berlin, 1923), translated as Aristotle: Fundamentals of the History of his Development (London, 1948).

vu

xxviii

TRANSLATOR'S

INTRODUCTION

into what Platonic-Aristotelian philosophy was about. But, most

significantly, since it is concerned in large part with the relationship between theory and practice and with "the good," it introduces us to the ethical dimension in Gadamer's thinking.
Truth and Method^ of course, had a great deal to say about
moral philosophy, and particular attention is given there to
Aristotle's idea of phronesis^ or moral reasonableness. But Gadamer's intent in that book was to point out that interpretation
of works of art and historical texts could not proceed according
to the method of the natural sciences, and that the truths contained in these works and texts were of a different sort from
those discovered in science. In Truth and Method he turns to
phronesis primarily because Aristotle shows that it, too, must be
distinguished from scientific reasoning insofar as it has its own
distinct way of relating to the truth {aletheuein). Thus Aristotle's
moral understanding serves in Truth and Method as an example
of the general theory of nonscientific understanding that Gadamer wishes to elaborate. Here, in contrast, phronesis is investigated for its own sake, and Gadamer turns expressly to the
good in human life, the theme of his first book, Platos dialektische Ethik:^ Hence we find in The Idea of the Good in PlatonicAristotelian Philosophy not only a continuation of the things we
3. Platos dialektische Ethik, Phnomenologische Interpretationen zum
'Philebus', (Hamburg, 1931) and republished in Platos dialektische Ethik (Hamburg, 1968), henceforth PDE, pp. xiii-178.
Other works by Gadamer bearing on the subject matter of this book are:
"Zur Vorgeschichte der Metaphysik," in Um die Bergriffswelt der Vorsokratiker (Darmstadt, 1968), pp. 364-90.
"Antike Atomtheorie," in ibid., pp. 512-33.
"Plato und die Dichter," in PDE, pp. 181-204, translated as "Plato and the
Poets," in Dialogue and Dialectic (New Haven, 1980), henceforth DD, pp.
39-72.

xxviii T R A N S L A T O R ' S

INTRODUCTION

have come to expect in Gadamer's work, his interpretive artistry

and his reliance upon, and extraordinary grasp of, Greek
thought, but also an investigation of moral philosophy its
premises, its subject matter, and its applications.
With regard to the principles of Gadamer's hermeneutics, we
find two especially well displayed here. First, particular attention
"Platos Staat der Erziehung," in PDE, pp. 205-20, translated as "Plato's Educational State," in DD, pp. 7 3 - 9 2 .
"Dialektik und Sophistik im VII. platonischen Brief," in PDE, pp. 221-48,
translated as "Dialectic and Sophism in Plato's Seventh Letter,^ in DD, pp.
93-123.
"Amicus Plato Magis Arnica Veritas," in ?D, pp. 249-69, translated with
same title in DD, pp. 194-218.
"Vorgestalten der Reflexion," in Kleine Schriften, 4 vols. (Tbingen, 196777) henceforth KS, vol. 3, pp. 1 - 1 3 .
"Platon und die Vorsokratiker," in KS, vol. 3, pp. 14-26.
''Logos und Ergon im platonisichen 'Lysis,'" in KS, vol. 3, pp. 5 0 - 6 3 , translated as "Logos and Ergon in Plato's Lysis," in DD, pp. 1 - 2 0 .
"Platons ungeschriebene Dialektik," in KS, vol. 3, pp. 2 7 - 4 9 , translated as
"Plato's Unwritten Dialectic," in DD, pp. 124-55.
"Uber das Gttliche im frhen Denken der Griechen," in KS, vol. 3, pp. 64
79.
"Gibt es die Materie? Eine Studie zur Begriffsbildung in Philosophie und Wissenschaft," in Convivium Cosmologicum (Basel, 1973), pp. 93109.
"Die Unsterblichkeitsbeweise in Platons Thaidon'," in Wirklichkeit und Reflexion (Pfullingen, 1973), pp. 145-61, translated as "The Proofs of Immortality
in Plato's Phaedo,"" in DD, pp. 21-38.
"Idee und Wirklichkeit in Piatos Timaios'" in Sitzungsberichte der Heidelberger Akademie der Wissenschaften (Heidelberg, 1974), translated as "Idea and
Reality in Plato's Timaeus,*' in DD, pp. 156-93.
"Vom Anfang bei Heraklit," in Sein und Geschichtlichkeit (Frankfurt, 1974),
pp. 166-75.
"Philosophie und Religion im griechischen Altertum," in Festschrift paper for
Wilhelm Anz, 1975, unpublished.
See also numerous references to Plato and Aristotle in Warheit und Methode
and in Hegels Dialektik (Tbingen, 1971), translated as Hegel's Dialectic (New
Haven, 1976), henceforth HD.

xxviii

TRANSLATOR'S

INTRODUCTION

is paid in each case to the kind of text we are interpreting. That

attention is especially called for here, since any attempt to understand what Plato and Aristorie are saying must confront the
problem of the disparate nature of the extant texts, the fact that
we have fictional dialogues from Plato and a mixture of treatises
and lecture notes from Aristotle. Even the differences among the
texts from Aristotle must be carefully taken into account. Gadamer points out, for instance, that the Magna Moralia^ since it is
in the form of lecture notes, or perhaps even class notes by someone else, cannot be read with the logical and compositional expectations that one brings to the other ethical treatises, for, as
notes, they depend on what Aristode would have said in his lectures to fill them out. Most important of all textual differences,
however, is the distinction between Plato's mythical, metaphorical way of putting things and Aristode's "cautious" conceptualizationsof natural phenomena, on the one hand, and the phenomena of our practical life, on the other. It is precisely the
failure to observe this distinction that has misled us into taking
what Plato says in the dialogues as statements and then comparing these to the supposedly equivalent statements made by Aristotle. As Gadamer points out (WM 444), anyone who has seen
what comes back in recorded "statements''in a court proceeding, saywill see at once that taking what someone says as a
statement is more likely to obscure what he or she was getting at
than to display it. M/sinterpretation will be inevitable.
The most unfortunate consequence of comparing Plato and
Aristotle in this way has been that the two thinkers have been
put at odds not only with each other, but also with themselves.
Both are said to have "developed" beyond, and subsequently
taken back, their earher positions, and this thesis has been supported by contrasting statements from later works with statements from earlier works. As Gadamer sees it, there must be

xxviii T R A N S L A T O R ' S

INTRODUCTION

something wrong here. For one thing, before the modern period
(Galileo) no one thought of putting the two thinkers in opposition to each other/ For another, the ancient tradition reports no
such modification of the views of either one of them. Hence it is
far more promising, Gadamer argues, to assume that they were
consistent, with themselves and with each other, in "getting at"
the same thing, albeit in quite different modes of discourse. The
task then becomes to find out what that same thing is and to display it as it shows up in Plato's and Aristotle's diverse presentations. Here two influences on Gadamer's thought become evident at once: above all Husserl, of course, but also, I suggest,
Rudolf Bultmann. Gadamer will use the ^'workman-like" techniques of Husserl's phenomenological description to bring to
light the common subject matter of Platonic-Aristotelian
thought. And in Plato's case that means that he will have to be
"demythologized," to use Bultmann's termthat is, what he
says metaphorically will have to be translated into conceptualizations of the phenomena of bur existence, the kind of conceptualizations that Aristotle was attempting.
But it is not only inattentiveness to textual differences that is
responsible for the misunderstanding that pits Plato and Aristotle against each other. The misunderstanding is also the result
of failure to establish the occasion for what is said in each case.
With that we come to the second hermeneutical principle exemplified in this book. The principal problem with reducing any
form of discourse to a series of statements is that the context or
setting, which alone gives what is said its meaning, is thereby
omitted. Naturally, we face this problem with any written text:
in order to know what was said, we need to know what it was
4. The modern period is faced with a new problem: how to overcome teleoiogical science. Galileo is thus led to play off the Pythagorean-mathematical in
Plato against Aristotelian physics. See Gadamer, Dialogue and dialectic, p. 195.

xxviii

TRANSLATOR'S

INTRODUCTION

said in response to, and, more often than not, written texts do
not tell us this expressly. They have come down to us as excerpts, as it were, even if, prima facie, they are complete in themselves. Hence, for someone who would understand them, the
task is to put them back in their setting. Said another way, one
must give priority to the question being addressed over the answer to it contained in the text; precisely what is unsaid and tacitly presupposed must be brought to the fore. Thus Gadamer replaces propositional logic's testing of statements with "the logic
of question and answer" (cf. WM 351 ff.).
Gadamer applies this principle of interpretation in several
ways. On .the other hand, he often takes into account those
things which are meant precisely by not being said, that is, "indirect" or "negative" meanings, as he calls them. The paradigm
here is satire, of which he finds a good deal in Plato's Republic;
the absurdly impractical proposals that Plato makes there can
only be understood as satirical inversions of the perversions of
political life in the Athens of that time. That is to say, they can
only be understood "e contrario."^ But, more generally, he finds
it imperative to treat anything that Plato or Aristotle says as an
answer to a particular question posed. If we are to understand
what they are saying, this question must be uncovered and
specified.
It turns out that Plato and Aristotle are answering questions
posed somewhat differently, and that is the primary reason for
5. The phrase "e contrario" epitomizes one important sense of the word "dialectical" in Gadamer: one must read satire "dialectically," which is to say, as
the contrary of what is meant. But not only satire. Gadamer's analysis of Hegel's
"Verkehrte Welt" (inverted-perverted world) provides a masterful exposition of
this principle of dialectical reading (see Gadamer, HegeVs Dialectic^ pp. 35-53).
And Socratic irony, of course, is another prime example of something which
must be understood e contrario.

xxviii T R A N S L A T O R ' S

INTRODUCTION

the apparent differences in their thinking. Though Gadamer

does not cite the passage expHcitly, Aristotle gives us a decisive
clue in Metaphysics, Alpha 6, where he says that Plato extended
Socrates' concern with invariant moral truths to the whole of nature. Gadamer's thesis is that Plato is responding in the whole of
what he says to a near-total moral dissolution in his society and
to the concurrent sophistic rhetorical techniques that confound
any real moral reasoning and substitute for it more or less disguised rationalizations of cupidity and the unbridled will to
power. This advent of sophism is the occasion for his thought.
Consequently, Plato turned to mathematics, for he saw there a
kind of reasoning that was self-evidently invulnerable to sophistic "tricks," and which in its structure must be similar to the
unshakable reasoning by means of which Socartes held to what
he knew to be right. Aristotle, on the other hand, is responding
to theoretical questions, for his thinking is prompted by wonderment {thaumazein) not political alienation. Hence he wishes to
display the analogical structures of things in a variety of realms,
among them the world of nature and, in distinction to that
world, the human practical world. The "descriptive caution" appropriate to his task leads him to divorce these realms, to distinguish sharply between the methods and objects of inquiry appropriate to each, and to fault Plato precisely for not making these
distinctions, for example, for importing the ontological idea
of the good into practical considerations where it has no
application.
Thus, if Plato argues theoretically against the Pythagoreans'
failure to distinguish between the perceived thing and the noetic
"number"their failure, that is, to separate the pure arithmetical relationship from the imperfect example of ithe does so, in
the final analysis, in order to head off confusing sophistic talk
and to be able, like Socrates, to hold to what is morally true de-

xxviii

TRANSLATOR'S

INTRODUCTION

spite appearances to the contrary that some clever speaker may

conjure up. But there is no occasion for Aristotle to make such a
separation {chorismos), since he is addressing questions raised
by the particular phenomena he is investigating, and here, too,
he is consequently critical of Plato. His starting point, unlike
Plato's, is the specific "this-whatever" {tode ti) and the fact "that
..
{bote . . .). Thus Aristotle's critique of Plato is not so much
indicative of serious philosophical differences as of the fact that,
given the questions he was raising, he could not find any application of Plato's conclusions, especially those concerning his "separated" idea of the good.
Now if one sorts these things out properly, the result is surprising: in the end there is a "same thing," Gadamer finds, which
both Plato and Aristotle are elaborating, despite the different occasions for their doing so and their different ways of doing it
namely, the world as it becomes clear to us in our speaking of it.
Plato and Aristotle both belong to the unified tradition of logos
philosophy. In the Phaedo Plato's Socrates turns away from naturalistic accounts of things to the logoi, our ways of speaking;
similarly, Aristotle founds his investigations on ''ps legetai/'
how something is spoken of. Furthermore, both Plato and
Aristotle concern themselves with the invariant eidos^ the form
which answers the central question that arises in our speaking of
anything, namely, 'Hi estin?" ("what is it?"). To be sure, Plato
turns to the mathematical in his initial response to the moral
threat of sophism, but it is clear as early as the Phaedo^ in which
Plato's Socrates engages two mathematically oriented Pythagoreans in discussion of the eide (forms), that the cosmos too is to
be understood in terms of our moral experience. When Socrates,
in leading up to the hypothesis of the eidos (form), describes his
turn to the logoi (96a ff.), he says that he was pursuing an account of nature {peri physes historia)^ and that he was attracted

xxviii T R A N S L A T O R ' S

INTRODUCTION

to Anaxagoras because the latter, so it seemed, wanted to account for natural things in terms of what is best (beltiston) for
them (97c). Hence Anaxagoras's nous (mind) promised, albeit
deceptively, to overcome the deficiencies of merely physiological
causal accounts. Aristotle, on the other hand, begins with "what
is by nature," the physei onta, in responding primarily to theoretical questions. Yet ultimately Aristotle too may be said to
have extended Socrates' and Plato's question concerning arete
(virtue) and the good or best (ariston) in his investigations of the
universe: the "physicists" {physikoi), he says, with only two
kinds of explanation at their disposal from what, and by the
agency of whatcould not properly grasp the cosmos, in which
things occur for the sake of (heneka) what is good.
It remains now to specify briefly how Gadamer arrives at these
conclusions. I offer here an overview of Gadamer's argument
since it often seems highly compressed and, in any event, is
sometimes difficult to follow once one is immersed in the details
of it.
In regard to Plato, what is emphasized above all is the continuity of his thought. Gadamer adheres to the traditional chronology assigned to Plato's dialogues but not to the prevailing twentieth-century theory that this chronology can be arrived at by
establishing turning points at which Plato is supposed to have
distanced himself from positions he had taken previously. On
the contrary, he maintains, there is continuity in the line of
thought unfolded in Plato's work. What varies is only the way he
chooses to get at his subject matter.
To begin with there are the refutational, or elenchtic, dialogues, in which Socrates confronts the great sophists of his
timeProtagoras, Gorgias, Callicles, Thrasymachusand displays the emptiness of their claims to be able to teach virtue. In
truth theirs is a technical mentality, and what they teach is only a

xxviii

TRANSLATOR'S

INTRODUCTION

techne (art) of succeeding. The conclusion to be drawn from

these dialoguesat times explicit, but more often implied negatively or indirectlyis that knowledge in virtue is somehow different from knowledge in techne. This shows up above all in the
fact that virtue cannot be taught. And there is a further truth that
emerges here, again largely unsaid, that the traditional ideas of
arete upon which the sophists rely and which might be learned
by imitating a paragon have become groundless and susceptible
of sophistic dissimulation. Therefore, to withstand seduction by
self-interest and by otherwise insatiable desires for sensuous
gratification and power, to which sophism panders, one must
now be able to give justification for what is good, that is, justification for one's choices of what is right as opposed to what is
wrong. But nobody seems to know how to do that, least of all
those sophists who claim that they do.
The primary concern in these elenchtic dialogues, then, is a
practical one, and that makes the Phaedo and the Republic^ in
which the concern seems to shift to epistemological and ontological matters, appear to mark a definite transition in Plato's
thought. For in these next works the "ideas" are introduced, and
in the Republic even the "idea of ideas," the idea of the good,
which would seem to serve as a first principle of both true
knowledge and true reality. In the twentieth-century traditions
of Plato interpretationparticularly the neo-Kantian, which
found support for its theory of science in this "stage" of Plato
would have it that the question Plato is addressing here is a new,
theoretical one no longer related to the Socratic question concerning arete. But Gadamer sees a serious oversight here. A careful reading, he argues, shows that Plato is still dealing with the
same issue. Gadamer maintains that both the hypthesis of the
eidos in the Phaedo and the Republic's discussion of a training in
dialectic leading up to the idea of the good do in fact follow from

xxviii T R A N S L A T O R ' S

INTRODUCTION

the concern with arete. (The problem here is one of long standing, for Aristotle himself argued that Plato's introduction of the
question of arete in his lecture "On the Good" led to an incoherence, which would make it appear that the idea of the good is a
purely ontological doctrine that has no connection with the
practical question of arete.)
In regard to the Republic^ Gadamer points out that the extension of the discussion beyond book 4 does seem to have something arbitrary about it, in that book 4's discussion of the unity of
the virtues in knowledge is not continued, nor is what follows
tied in to, or derived from, this discussion. The concern is now
the theoretical education of the guardians, an education which in
book 6 will lead them out of the impure world of practical matters, the shadowy "cave," into the pure world of the forms.
Whether intended or not, the discussion here would seem to result in an absolute antithesis of theory and practice. Moreover,
there seems no doubt that, as a way of life, the former is far superior to the latter in Plato's eyes.
Gadamer maintains, however, that if one asks what question
is actually being addressed in the allegory of the cave, one sees
that the concern is not just, or even primarily, a theoretical, scientific one, but in fact an existential-practical question of holding steadfastly to the truth in the face of tests or refutations
(elenchoi)^ much as a soldier holds his ground in battle or a wrestler stands firm against attempts to throw himthis is the language Plato uses here. And what are these tests of mettle? The
guardians will find themselves in a conflict between honest execution of the duties of their office for the public weal and the
"pursuit of power after power," as we in the English-speaking
world might put it, using the words of Thomas Hobbes. The
guardians, in other words, will be tested by the flattery of both
their own desires for gratification and sophistic sycophants, flat-

xxviii

TRANSLATOR'S

INTRODUCTION

tery that might seduce them into sacrificing their integrity. It is

precisely this flattery that they must be educated to withstand.
In short, it turns out that the theory of dialectic in which they
are to be trained is a way of distinguishing and, above all, of distinguishing practically between right and wrong, good and bad.
To do that, the guardians must know the idea of the good, for
that idea is exactly what those in the cave do not know. The
cave-dwellers' technai (arts) give them the knowledge "how
to'' do something, knowledge of the means to an end, but not
knowledge of the end itself, the hou heneka^ the "what for." And
without this knowledge, knowledge of the good, the guardians,
like the cave-dwellers, will succumb to the particular urgings of
their self-interest. Thus, knowledge of the good turns out to be
not just a theoretical insight, not just a matter of logos (reasoning). It is a principle in ergon (deed) and makes possible constancy in the choice of the life that one leads (Aristotle: prohairesis tou biou).
And now the real meaning of the hypothesis of the eidos in the
Phaedo becomes evident. As opposed to what the neo-Kantian
school might have seen in it, the hypothesis of the eidos is not at
all the "scientific" postulation of a universal idea that is to be
verified by the facts. As in the Republic^ the issue in the Phaedo
is how to head off whatever might mislead us, how to head off
sophistic talk, and this can be done only if we know what the
eidos of the thing we are speaking of includes and what it excludes, what pertains to it and what does not. Hypothesizing the
eidos thus has nothing to do with empirical verification but, instead, with making precisely these eidetic distinctions. The
method of hypothesis is thus dialectic, and it allows us to hold
undisconcertedly to what is meant, just as dialectic in the Republic enabled the guardians to hold undisconcertedly to what is
good. The paradigm for steadfastness in both cases remains

xxviii T R A N S L A T O R ' S

INTRODUCTION

Socrates, who was not to be dissuaded from what he saw to be

right, no matter what arguments concerning his own ''advantage" might be adanced. His arete was such that he would not
even accept the escape from death offered to him by his friends.
In this way it is possible to establish a continuous line leading
to the 'later' Philebus. It will be noted that this dialogue too begins with an ethical issuewhether the good life is one of pleasure or of intellect or of some third thingand it also involves
its interlocutors in the task of giving justification for their contentions concerning the good life, justification that will make
their arguments invulnerable to the sophistic arts of confusing
and confounding people with sleights of hand that interchange
the one and the many/ The way for us to head off such eristic
"tricks"the "Promethean fire," as Socrates calls itis to
know just what number (arithmos) of things is included in the
"one" thing we are speaking of, and thus what is to be excluded
as extraneous. This knowledge correlates precisely with what
was said about the hypothesis of the eidos in the Phaedo,
Still, the introduction of the arithmos clearly leads to ontological considerations that transcend the practical questions that initially occasioned this dialogue. The question about human existence now becomes a question about the structure of the whole
cosmos, a structure of which human existence is just one example. Hence the Philebus brings into sharper focus the problem
that had already surfaced in the Republic concerning the relationship between the theoretical-ontological and the practical.
The guardians, we learn there, are to be trained in dialectic,
which is to say, in the knowledge that comes after they have
6. One should not overlook the fact that the Sophist, a later dialogue whose
principal concern seemed to some (for example, F. Comford) to be with ontological questions, in fact focuses on sophistic deception, and that the heady ontological issues it raises continue to be occasioned by the threat of sophism.

xxviii

TRANSLATOR'S

INTRODUCTION

passed through training in the theoretical dianoetic sciences and

that would appear to be the culmination of the latter. Put another way, in the Republic dialectic and knowledge of the good
are theoretical knowledge about reality, about being, and seem
at first glance far removed from any practical concerns. And in
the Philebus too the question again becomes the same one raised
by Aristotle in regard to Plato's "On the Good," namely, what
such purely theoretical matters as the doctrine of number, the
"one" and the "indeterminate two," the peras (limit) and the
apeiron (unlimited, indefinite)all of which appeared in "On
the Good" as well as in the Philebus could possibly have to do
with the Socratic question about arete.
An answer will be forthcoming, Gadamer suggests, if one first
makes clear just what these theoretical doctrines mean, but clarity can be obtained only if one translates Plato's myths and metaphors into conceptual language. Here the techniques of Husserl's
descriptive phenomenology can be put to good use, for the task
is to display the phenomenon that Plato is getting at. In the
Philebus Plato speaks of the good as the structure of a "mixture"
someone might brew, the "potion" as it were, which is our human life. The various "ingredients," the kinds of pleasure and intellect, are not to be mixed indiscriminately, but well. That they
are not means that a limit must be set: not an indeterminate
number, but only "so and so many," a specific arithmos, will be
admitted. A "good" mixture is thus one that has limits set to it,
and the good itself is limitedness (measuredness, or metriotes) in
the midst of constantly threatening indeterminacy and limitlessness. Hence not only an ontological principle of the cosmos is
applied when nous sets a limits {peras) to the indefinite-unlimited {to apeiron) but a principle of the right life as well. As
measuredness, the good in the Philebus^ we now see, is precisely

xxviii T R A N S L A T O R ' S

INTRODUCTION

Aristotle's mean between the extremes, which is to say, an ontological principle with very clear practical relevance.
In anticipating Aristotle's criticisms, it should be noted that, as
measuredness, the good in Plato must be distinguished, or "separated" intellectually, from the mixture itself. But if one abstracts
from this metaphor, one finds that, as measuredness, the good is
what Hegel calls ein Moment^ namely, an aspect of something
which does not exist separately from it. In short, it is in the thing
of which it is the structure. Thus, when we say that it is christon
(separate), we are not denying that it is in the thing. We are saying only that it must be distinguished from the thing in our
thinkingsomething that Aristotle would be the last to deny.
For he too raises the question ti estin ("what is it?"), a question
aimed at distinguishing the invariable Moment^ the "what" {ti)
in any "this-whatever" (tode ti).
These striking convergences of Plato's thought with Aristotle's
leave us puzzled about Aristotle's critique of Plato. Why would
he criticize Plato if in fact he and Plato are saying the same thing?
The question raised here is twofold: on the one hand, we must
ask what the ontological intent of his critique of Plato's idea of
the good is; on the other, we need to know the moral-theoretical
intent of that critique.
In resolving this twofold question Gadamer finds it best to
hold to the hermeneutical technique of displaying the phenomena with which Aristotle's three ethical treatises are concerned the ''mens auctoris" is of so Httle importance that the
issue of whether all three are by Aristotle himself need not even
be raised. The task is to repeat in ourselves {nachvollziehen) the
steps in the arguments, and thereby to bring to light the thing
they are about, what they are aiming at, or, as Husserl and Gadamer following Husserl put it, their ''Intention/' And it can be

xxviii

TRANSLATOR'S

INTRODUCTION

established right away that whoever their author might be, they
are indeed "about" the same thing.
At the core of the argument in all three is the contention that
Plato improperly fuses the ontological and the practical in his
applications of the idea of the good. Consequently, it is argued
that from the start a clear separation must be made between the
practical good and the idea of the good in whatever explanatory
function it might have in the universe, and that ethics must be restricted to a consideration of the former. Aristotle justifies this
restriction by pointing out that the idea of the good is of no use
in the various technai, but this argument, far from weighing
against Plato, only affirms Plato's contention that the idea of the
good is not accessible to technical knowing, and that knowing
the good and knowledge in the technai are radically distinct, a
conclusion which, to be sure, was not at all the one "intended"
by the Aristotelian line of thought here.
Furthermore, despite his aversion to introducing physics and
metaphysics in the ethical treatises, Aristotle constantly finds his
arguments drawing him in the direction of ontological considerations which by rights he ought to pursue, but he is prevented
from pursuing them by the way in which he was put his question. In the first place, he resorts to the category argument, with
the idea of showing that, just as there is no one "being" in itself
but only a number of ways in which "is" can be said of something, so too there is no idea of the good, separate {christon)
and for itself, but only a number of ways in which the same word
"good" is used. As Gadamer points out, putting things this way
raises "uncomfortably more" issues than it should. For when he
is expressly pursuing questions of ontology, as he is in the CategorieSy Aristotle maintains that the relationship of the other categories, or ways of using "is," to the central category of "sub-

TRANSLATOR'S INTRODUCTION

xxiii

Stance" must be clarified, and it would follow that, similarly, the

senses of "good" would need to be related to a central sense of
the word. Like "is," "is good," which is to say, one and the same
expression, is not used in many different ways coincidentally. By
rights one would have to ask just what it is that all these uses
have in common, but, given the "occasion" and "intention" of
Aristotle's argument here, such a question cannot arise.
In the second place, in the ethical treatises Aristotle plainly
equates the idea of the good in Plato with the one in Plato's doctrine of the ideal numbers, the one and the indeterminate two, or
dyad. He then proceeds to argue that just as it is mistaken to turn
the principle of number into a number itself, so Plato is mistaken
in turning the principle of good things into another good thing.
This argument, too, is essentially ontological.
Here, for his moral-philosophic purposes, that is, of showing
that the idea of the good is irrelevant to practice, Aristotle takes
Plato's mythological way of putting things literally. In Plato it is
said that the numbers "strive" for the one, which makes the one
look like an entity. But, as Gadamer's exposition of the Philebus
shows, the one and the two are principles in things, which, nevertheless, are to be distinguished from these things in our thinkingsomething that Aristotle, were he pursuing questions of
ontology, would be the last to deny. Thus they are "for themselves," or christon, in a sense, but that does not mean at all
that they are separate entities alongside other entities.
And whatever the case may be here, the very presence of these
arguments in his ethics shows that there, at least, Aristotle never
loses sight of the ontological dimension of the question of the
good. His very attempt to expel such considerations from ethics
shows "negatively" or "indirectly" that they are important to
him. Thus the question of just what the ariston tn pantn^ the

xxviii

TRANSLATOR'S

INTRODUCTION

good or best of and for all things, might bethe universal/ ontological structure of measuredness in Platoremains present in
Aristode, despite his stated purpose of limiting himself in his ethics to consideration of the ariston tn praktn, that is, what is
good or best in practical matters.
Aristode, of course, is a consummate phenomenologist who
wishes above all to avoid running distinct things together. But is
it just his "descriptive caution" that leads him to try to keep ontological and practical theory separate? Gadamer suggests that
there is something else fundamental in Aristotle's way of inquiring that leads him to put things as he does, namely, his orientation toward life science. Mathematically oriented thinking such
as Plato's would indeed lead to inquiry about the idea of the
good as an abstract structure of good things, much as one might
inquire about the abstract arithmetical structure of what is numbered or the abstract principles of harmony in what is harmonious. But that is not the orientation of Aristotle's questioning,
which gives primacy precisely to the concrete living thing. Consequently, the ariston tn pantn for him must not be a conceptual structure but instead a "mover" of other things, a first reality that is not epeikena tes ousias^ beyond existence as Plato's
idea of the good is, but an existent god. With the postulation of
this god, Aristotle does indeed carry out Socrates' demand that
we understand the universe starting from our moral experience,
that is, that we understand it in terms of what is good, and to
7. Gadamer uses the German allgemein in both its English senses, "general"
and "universal." I have had to rely on intuition, however fallible, in deciding
which to use in the translation, and on occasion I have switched back and forth
in the same passage where it seemed that Gadamer had both the more ordinary
"general" and the more terminological "universal" in mind. One interesting
point: the German allgemeinliterally "common to all"tends to sustain Aristotelian analogical thinking: the form, or eidos, is taken to be to koinon, or what
all particulars have in common.

xxviii T R A N S L A T O R ' S

INTRODUCTION

this extent he too fuses the practical and the theoretical. But for
him this fusion does not mean that a discussion of arete should
lead to a discussion of the cosmos, even though the same sort of
structures might be constitutive in both. His concern is not with
abstract structures but with the "this here."
In the final chapter, "The Idea of Practical Philosophy,"
Gadamer addresses the broader issues raised by Aristotle's attempt to exclude any considerations of physics or metaphysics
from practical philosophy. This exclusion raises questions about
the relationship of the special field of theory that we call practical philosophy, that is, ethics and politics, to theory in general,
that is, physics, metaphysics, ontology, and cosmology. How
does the application of moral theory, as Aristotle understands it,
to moral practice differ from the application of scientific theory,
say, in technology or medicine? For the latter may certainly be
called practices also, and Aristotle even seems to take techne as a
model for practice in general. And just what application does
moral theory have in the first place? How does it bear on life as it
is actually lived? Finally, what is the relationship between the
two lives of reason of which Aristotle speaks, the life of practice
and the life of theory? What does Aristotle mean when he calls
the life of practical reason a second {deuteron) best? Does deuteron mean "another," or does it mean "inferior"? Or is its
meaning somewhere in between?
The first of these questions is introduced directly by Aristotle's
critique of Plato's idea of the good, for a good such as Plato
speaks of has, in Aristotle's view, precisely no application to life
as it is lived. In medicine all things aim at the specific good of
health, at a good, but an idea of the good in general is of no use
here whatsoever. And so it is with all practices. The good,
whichever specific good it might be, must be limited to the specific conditions of particular human practices. Here Aristotle

xxviii

TRANSLATOR'S

INTRODUCTION

sees a fundamental confusion in Plato, in that he fails to distinguish between the use of "good" in ontology and its use in human practice. (As Gadamer pointed out to me during our discussion of this chapter, according to Aristotle we would have to
clearly differentiate between "good" snow insofar as it approaches the ontological good of pure whitenessperfect snow
as snow, as what it means to be snowand "good" snow in
practice, snow that is "good" for skiing, for instance. The ontological sense is wholly irrelevant when there is a question of
what ought to be done, for example, of whether one ought to go
skiing given the snow conditions.)
So the question comes down to this: What role could there be
for practical philosophy, if, as it would seem, each of the specialists would be best equipped to say what is good in their respective technical fields? Put another way, how does doing moral
philosophy contribute to arete, excellence in practice?
With that Gadamer has arrived at a crucial issue in his own
thinking, which, perhaps more than anything else, is aimed at
pointing out the mistake in making modern scientific technical
reasoning the paradigm for all reasoning. The relationship of
moral theory to practice is not at all the modern relationship of
theory to practice in which an objective, neutral theory can be
applied generally to particular problems. In distinction to producing something {techne, poiesis), doing the right thing is not
simply an application of general rules, and thus there is some8. Here Gadamer's thinking is diametrically opposed to the current conception of "moral reasoning" in American philosophy, and, of course, to any attempt to found that conception on Platonic-Aristotelian thought. W. Frankena,
for instance, writes in his Ethics (Englewood Cliffs, 1973, p. 2): "In this [Socrates*] pattern of moral reasoning [in the Crito] one determines what one should
do in a particular situation by reference to certain general principles or rules,
which one takes as premises from which to deduce a particular conclusion by a

TRANSLATOR'S INTRODUCTION

xxiii

thing misleading about Aristotle's constant allusions to the

technai, medicine, military strategy, and so forth. In the first
place, as Aristotle recognizes, we need an "appropriate principle" {oikeia arche) for moral reasoning, which is not to be confused with mathematical deductive reasoning. For in moral reasoning I always find myself in a particular situation, and the task
is not to subsume this particular case under a universal rule
which I could know apart from the situation I am in, but to define from within my situation what the general rule is of which
this situation is an instance. The particular virtues and virtue in
general, as finding the "mean between the extremes," are not
universal principles that I apply to a situation, but universalizations of what I am doing when I do what is right. Hence it should
be noted carefully that, although Aristode does indeed speak of
the "practical syllogism," he in fact uses this syllogistic reasoning only to exemplify the technical choice of the right means to
an end. That he knows, despite his (and Plato's) frequent adversions to the technai, that reasoning here is not to be equated with
moral reasonableness is clearly established by his sharp separation of phronesis from demotes [Nicomachean Ethics 1124a23
ff.). The latter, cleverness, deals with means to an end, and for
just that reason, he says, it is to be distinguished from the
former, moral reasoning. (In essence, Gadamer finds, Kant
makes precisely the same point in distinguishing moral from
technical imperatives.)
But how would this generalization of what I am doing when I
do what is right contribute to my doing what is right? How is
this awareness useful? How would it strengthen arete? The nkind of practical syllogism, as Aristotle called it. One takes general principles and
applies them to individual situations." From Gadamer's point of view, he is completely mistaken.

xxviii

TRANSLATOR'S

INTRODUCTION

swer, Gadamer suggests, is to be found in Aristotle's illustration

of the archer who is able to hit the mark better precisely because
he has chosen a more clearly defined spot on the target at which
to aim. When we have learned to fix our sights on precisely what
we are doing when we choose what is good, namely, hitting the
mean between the extremes, we will choose better. Hence, although moral philosophy enchances ethical training, it is no substitute for it. Just as one must know how to handle a bow and arrow before one can improve one's aim, one must already be
behaving morally before moral philosophy will be of any use.
And finally, what is the relationship between the two ideal
lives, that of theria and that of praxis. Both Plato and Aristotle
leave no doubt that the highest, most divine lifethe Hfe that
for a god would be teleon (complete), hikanon (sufficient), and
thus haireton (to be desired or chosen) {Philebus 20d)would be
the life of pure theria. And if one takes that life as the standard
against which all else is to be measured, as Plato and Aristotle
clearly do, then the life of practice does indeed seem a second
best: less complete, insufficient, hence less desirable.
But one must be careful here. As both the Philebus and Aristotle's ethical treatises make clear, human beings are not gods,
and thus the life that is best for them has to be a life that combines theria and praxis. Plato portrays human life metaphorically as a "mixture" of reason and pleasure, an image that
Aristotle conceptualizes when he speaks of human nature as
syntheton (composite) {Nicomachean Ethics 1177b29). The
point that both wish to makewhich is the fundamental point
of agreement between them concerning the goodis that, in distinction to gods, human beings are always under way toward the
divine, or, as Gadamer puts it, their best life is philosophia^ not
sophia, that is, striving for wisdom, not wisdom itself. Human
beings are finite, not absolutenever absolved from the "rem-

xxviii T R A N S L A T O R ' S

INTRODUCTION

nant of earth" (Goethe) in them that inevitably involves them in

the task of living well here in the practical world. Seen this way,
the life of praxis is a second best not in the negative sense, but in
the sense of one of two "best-nesses" that together constitute the
good human life, the life of reason in its combined theoretical
and practical functions.
Thus, in both Plato and Aristotle, the good emerges as that toward which we are striving, that for the sake of which {hou
heneka)^ that at which we aim {to telos). But not only we. The
whole universe is to be understood as striving for perfection.
Thus, in the end, despite his announced separation of practical
and physical-metaphysical theory, Aristotle, like Plato, conforms
to Socrates' requirement that the universe be understood in
terms of our own moral experience, and he too fuses moral theory and ontology: the universe, and not only human experience,
is to be thought of in relationship to the good. This cosmichuman good is represented in Plato only quasi-conceptually in
his doctrine of the ideal numbers. Aristotle, physikos that he remains and not Pythagorean mathematician, sees the good actualized in the self-mediation of the physei onta with their eide, the
things of nature with their forms. Plato, he argues, separates
these forms, and with them the good itself, from the things in
which they are realized. This separation, as we will learn here, is
a necessity for one who would think mathematically, as Plato
does. But indisputably it creates problems (aporiai), problems of
how the ideas are to be mediated with the appearances, for
which Plato provides only metaphorical solutions. Aristotle begins the translation of these metaphors into the context of the investigation of natural things and into the language of concepts.
One should keep in mind that the text of The Idea of the
Goody as it first appeared in German, was essentially an expanded version of a presentation for a"quite select audience (Die

xxviii

TRANSLATOR'S

INTRODUCTION

Heidelberger Akademie der Wissenschafaten). Its "occasion," in

other words, was very different from the "occasion" for the publication of this translation, which, it is hoped, will reach a wide
range of English-speaking readers with a variety of interests. I
have sought to augment the text both in this introduction and in
a number of notes, for often things are omitted which could be
expected to be self-evident to its original, narrower audience but
are not so for us.
The material for these notes derives in part from my long years
of apprenticeship with Professor Gadamer and in part from detailed discussions we had concerning different passages in the
text.' These discussions provided me with vivid confirmation of
Gadamer's thesis, that any written word, like lines in the script
of a play, is somehow only dead ink on the page that need to be
brought to life by saying it aloud in dialogue. Passages that at
first were obscure came to life and became clear in what he had
to say about them. I have recorded what I could of these discussions here, but as records, they too are petrified speech, and the
train of thought they were meant to elucidate will, I am afraid,
sometimes break down. Gadamer has provided me with a number of elaborations of the text that he wished to have inserted in
the translation; translations of these are preceded by and conclude with an asterisk. He also made occasional changes in the
wording of the original text which I have noted when they are
significant. It should be noted too that for the sake of clarity,
readability, and style the translation could not be word for word
in a number of cases.
A word of caution is in order here, though. It would be inappropriate to expect one continuous, systematically conclusive
9. I am very much indebted to the Translations Division of the National Endowment for the Humanities for making these discussions with Professor Gadamer possible, and indeed this entire project.

xxviii T R A N S L A T O R ' S

INTRODUCTION

line of argument here. For philosophy as Gadamer understands

it does not proceed more geometrico^ as our modern logic, it
seems, expects it to.^ We cannot ascertain an indubitably secured starting point; we cannot know ahead of time where a
philosophical discussion will lead us, and we will never be able
to say that we have arrived at a definitive conclusion. The starting point is always an open question that is never closed completely. Thus we approach it first with one line of inquiry, then
another, all the while aiming to shed light on it from a variety of
perspectives, but fully acknowledging that the insights gained in
this way will of necessity be inconclusive and unsystematic.
Gadamer is serious when he distinguishes the human enterprise
of philosophia from divine sophia, and this book is representative of the sensitivity to human finitude and the inevitable discursivity of human thought that leads him, just as it did Plato and
Aristotle, to insist on this distinction.

10. As Gadamer pointed out to me, C. Perleman's The New Rhetoric (South
Bend, 1981) argues convincingly that the Cartesian methodological model,
which provides the foundation for much contemporary logic and analytical philosophy in the English-speaking world, in fact falsifies language and discursive
reasoning as we actually experience it. S. Toulmin's An Introduction to Reasoning (2nd ed., Englewood Cliffs, 1984) represents a major advance in emancipation from Cartesianism insofar as it acknowledges the contextuality of reasoning,
that is, the differences in the occasions for it, and recognizes its dialogical, openended character. Even so, he has not quite succeeded in establishing "the hermeneutical priority of the question*' (WM 344) over the answer-proposition, or
"claim" as he calls it. All in all, it appears that, paradoxically, modern physics
has long since transcended the Cartesian method, whereas the human sciences
and logic, for which it was least appropriate to begin with, remain mired in it. F.
A. Heyek makes this point well in regard to economics in his Studies in Philosophy, Politics, and Economics (Chicago, 1967).

PREFACE

Hegel, it cannot be denied, did indeed grasp the speculative tendency in both Plato's doctrine of the ideas and Aristotle's substance ontology, since his thinking was so congenial to theirs.
And to that extent he is the first in modern times to break
through the schema of interpretation of Plato's doctrine of the
ideas shaped by Aristotle and further developed in Neoplatonism and the Christian tradition. Nor can one say that Hegel has
remained without any lasting influence on scholarship in the history of philosophy. Such good Aristotelians as Trendelenburg
and Eduard Zeller owe him a great deal. Above all, Hegel was
the first to make the philosophical significance of Plato's "esoteric," "dialectical" dialogues accessible. However, the unitary
effect^ connecting Plato's and Aristotle's logos philosophy
1. Wirkungseinheit. The word is related to Gadamer's concept Wirkungsgeschichte, the history of effects. The point is that far from being opposed to each
other, Plato and Aristotle belong to a continuous line of thought. They have in
mind the same subject matter, or Sache. Hence, rather than stressing the individual contributions of each, Gadamer finds it best to ask just what that subject
matter is which concerns both thinkers, and to explicate it phenomenologically.
One further point regarding Wirkung, or effect: since it is the subject matter and
not the indidivudal thinker that is primary, one must be careful to read "effect"
not as the effect that a thinker has on subsequent thinkers, but as the effect that
the subject matter, as it is passed on by tradition, has on an individual. It is not
the thinking subject that comes first, but the tradition that makes his or her
thou^t possible. Accordingly, Gadamer's wirkungsgeschichtliches Bewusstsein
might best be translated not as "historically effect/W," but as "historically effected/consciousness." TRANSLATOR.

PREFACE

which did not remain hidden from Hegelwas underestimated,

it seems to me, in the period following him and continues to be
until this day.
There are various reasons why. To be sure, there was a concealed, unacknowledged Hegelianism behind the neo-Kantian
interpretations of Plato in Cohen and Natorp and in their successors Cassirer, N. Hartmann, Hnigswald, and Stenzel. But given
their particular mentality, it was exclusively Plato, and not in the
least Aristode, who could sustain this generation of scholars, in
their critical-idealistic purposes. A complete elaboration of Hegel's insights was totally obstructed, on the one hand, by the
dogmatic overlay superimposed on Aristotle by the neo-Thomism prevailing in the Catholic camp, and on the other, by the
hereditary feud between modern natural science and the Aristotelian teleological understanding of nature or, in fact, any idealistic philosophy of nature. Furthermore, when it interpreted the
"idea" as the "natural law," thereby bringing together Plato and
Galileo, the neo-Kantian interpretation of Plato, especially that
of Natorp, proceeded all too provocatively with the Greek text
while remaining insensitive to historical differences. If one starts
from this idealistic neo-Kantian interpretation of Plato, then
Aristotle's critique of Plato can appear only as an absurd misunderstanding. This fact further contributed to the failure to recognize the unitary effect in Plato and Aristotle, thereby blocking a
full incorporation of the Greek heritage into our own philosophical thought. Such trivial and naive juxtapositions as "Plato, the
idealist," versus "Aristotle, the realist," gained universal currency, although they actually only confirmed the truly abysmal
depth of prejudice in any idealism of consciousness.^ In addition.
2. The goal of neo-Kantianism was to develop the "transcendental" forms
that consciousness applies in constituting its objects. Neo-Kantians seized upon

PREFACE

the schema for which Hegel provided the inspiration which construed Greek thought as not yet able to conceive of the absolute
as spirit, life, and self-consciousness, did not promote a proper
evaluation of the fundamental significance of Greek thought for
modern philosophy.
Nicolai Hartmann's dissociation of himself from neo-Kantian
idealism stimulated me to try to penetrate Aristotle's thought,
and the French and English researchof Robin, Taylor, Ross,
Hardie, and, above all, the incomparable Hicksproved most
helpful in my endeavors. At that time, however, I fell far short of
seeing the unity in the logos philosophy, which started with
Socrates' questioning and then quickly deteriorated in the postAristotelian period, but which, nevertheless, permanently determined the entire conceptual apparatus of Western thought. Encountering Heidegger turned out to be decisive for me at that
stage. Heidegger had worked his way through both the CatholicPlato's idealism precisely because it appeared to them that Plato had set out to
uncover these same forms of consciousness. Plato's "ideas" were thus understood epistemologically rather than ontically, that is, as transcendental forms of
thought rather than the forms of things in themselves. Aristode's insistence that
the forms inhere in the thingshis doctrine of the enhylon eidosand on the
primacy of the tode ti thus seemed to them a regression to a naive realism. When
Gadamer refers to Neo-Kantianism as a mere "idealism of consciousness," he is
implicitly contrasting it with Fichte's and Hegel's idealism, which goes beyond
any epistemic-ontic dichotomy, and which, like Aristotle, whom Hegel favors (cf.
Logik II, "The Logic of the Concept"), views the forms of thought at one and the
same time as the forms of what is, that is, being. Heidegger's break with transcendental philosophy, be it either neo-Kantian or Husserlian, exposed the onesidedness of any idealism that founds itself exclusively on our consciousness of
reality. The question to be asked is not how consciousness construes reality, but
how reality, being, presents itself in our awareness of it. The key here is language,
in which that which is assumes its form for us. Language is the medium (Gadamer: Mitte) in which consciousness and world are joined. Hence it, not consciousness, is the "condition of the possibility" (Kant) of anything's being what it
is. TRANSLATOR.

PREFACE

Aristotelian and neo-Kantian traditions, and in appropriating

Husserl's minutely detailed art of conceptualization, he had
steeled the endurance and power of intuition, which are indispensable for doing philosophy with Aristotle. Here, then, was an
advocate of Aristotle who, in his directness and the freshness of
his phenomenological insights, far surpassed all the traditional
shadings of Aristotelianism, who surpassed Thomism and, yes,
even Hegelianism. To this day hardly anything has been made
public of this event, but it has had its effect on academic teaching, and my own path was defined beginning there. By the time
I published my first book in 1931 [Platos dialektische Ethik]^ the
convergence, at least in the area of practical philosophy, of the
aim of Plato's thinking^ with Aristotle's conceptual distinctions
had become evident to me.
Even at that time one thing already directed me beyond the
narrower context of problems in practical philosophy, namely,
the methodological problem that our tradition has preserved
two such disparate things as Plato's dialogical compositions and
Aristotle's working drafts. Since we possess neither an authentic
theoretical elaboration of Plato's teachings nor any of the writings that Aristotle published, we must constantly play off two
dissimilar things against each other. The art of phenomenological description, a little of which I was able to learn from Husserl
and Heidegger, helped me in my first attempts to master this
3. Plato's Denkintention. There is no single felicitous translation here. A
range of things are meant: what Plato and Aristotle had in mind, what their
thought was aiming at, their intent or purpose, and so forth. Denkintention is related to the equally difficult Sache des Denkens, or subject matter of thought, the
thing under consideration. The point is that Plato and Aristode had corroborative insights into the same truth, though they chose different ways to articulate
and communicate those insights. Hence our task is to get past the differences in
their forms of presentation and to get at what they were both after, so to speak.
TRANSLATOR.

PREFACE

methodological difficulty. Neither

textual analysis of'^ the dialogues' mimetic form of communication nor of the protocol
form of Aristotle's papers can chain the authenticity of a descriptive phenomenological exposition based on the text*a phenomenological exposition of their subject matter itself.*
In the meantime almost half a century has gone by. In many
respects the edge has been taken off the methodological problemon the Aristotelian side by Werner Jaeger and his school,
and more recently, on the Platonic side by the Tbingen scholars, who received their impetus above all from Robin. The philosophical stimuli I received from Heidegger led me more and
more into the realm of dialectic, Plato's as well as Hegel's. Decades of teaching were devoted to elaborating and testing what I
have called here the Platonic-Aristotelian unitary effect. But in
the background was the continuous challenge posed for me by
the path Heidegger's own thought took, and especially by his interpretation of Plato as the decisive step toward "metaphysical
thought's" obliviousness to being (Sein). My elaboration and
projection of a philosophical hermeneutics in Wahrheit und
Methode bear witness to my efforts to withstand this challenge
theoretically. The following studies too, it is hoped, will serve to
keep alive both Platonic dialogue and the speculative dimension
common to Plato, Aristotle, and Hegel, as partners in the ongoing discussion which is philosophy. They belong together with a
number of other small building blocks I have gathered in the
meantime.^
I have lectured from these studies twice in meetings of the
Heidelberg Academy of Sciences (in 1974 and 1976). What I offer here is a further elaboration of my presentations, to which I
hope to add still other studies. It is unlikely that these will ever
4. See the list of works in n. 3 of the Translator's Introduction.

PREFACE

turn out to be a truly unified work, just as the essay which I offer
here did not, but is instead a series of reflections and observations on the question posed.
It will be noticed that I have referred to recent scholarship
only sparingly. For one thing, I do not feel qualified to take a
comprehensive stand on it. For another, the presuppositions of
my own interpretation are all too different from those of other
research. I ask that the reader take what follows as an attempt to
read the classic Greek thinkers the other way round as it were
that is, not from the perspective of the assumed superiority of
modernity, which believes itself beyond the ancient philosophers
because it possesses an infinitely refined logic, but instead with
the conviction that philosophy is a human experience that remains the same and that characterizes the human being as such,
and that there is no progress in it, but only participation. That
these things still hold, even for a civilization like ours that is
molded by science, sounds hard to believe, but to me it seems
true nonetheless.

T H E Q U E S T I O N AT

ISSUE

If one surveys the last fifty years of research on ancient philosophyand it has been more than fifty years since Werner Jaeger's book on Aristotle gave new and significant impulses to
scholarship in this fieldone finds oneself more and more embarrassed by the results of that scholarship. In Werner Jaeger a
simple schema still prevailed which gave us the outlines of Aristotle's development from Platonist to critic of Plato's doctrine of
the ideas and, finally, to empiricist. Even at that time, to be sure,
one could have doubted that this construction was universally
valid. But, starting from a literary-historical interpretation of
Aristotle's Metaphysics^ Jaeger extrapolated a line of development backwards and forwards in Aristotle's divergence from the
doctrine of the ideas, and it could be said in his favor that his
construction was at least unequivocal, not to mention the fact
that his analyses exposed the artificiahty in the editing of the Aristotelian corpus up to that time. Even then, of course, it was
noted that Jaeger's construction yielded a "proto-physics" with
far less tangible contours than his "proto-metaphysics," and that
the proto-physics was, if seen from the perspective of literary history, not supported convincingly, given the state in which the
Physics books have come down to us. But most of all [his claim
to have found] a "development" in Aristotle's ethicswhich
Jaeger, with a certain drastic superficiality, managed to fit into
his construction by availing himself of only parts of the Eudem-

T H E Q U E S T I O N AT ISSUE

ian Ethicssoon met with well-founded objections. For Jaeger's

placement of the Protrepticus in this context seemed especially
problematical. In the meantime we have J. Dring to compare
with.^ Today it is an established fact that in the entire traditional
Aristotelian works we never get back to a point where Aristotle
was not a critic of Plato's doctrine of the ideas, but also that we
never arrive at a point where he really ceased to be a Platonist. If
we start with that fact, then just what it meant to be a "Platonist" once again becomes questionable.
It is inevitable that this difficulty with Aristotle would have a
reciprocal effect on our understanding of Plato. Now that our
confidence that we can discern developmental phases in Aristotle
has pretty well vanished, the question forces itself upon us
whether the same thing does not hold for Plato. Is there sufficient foundation for the prevailing historical-genetic way of
viewing Plato's writings? These days the dominant view assumes
that the dogmatic doctrine of the ideaswhich Plato was supposed to have taught at the beginnings and which, with Neoplatonic hues, has shaped our understanding of Platonic philosophy as a theory of two worldswas later taken back, or at the
very least diluted, by Plato himself in his own self-criticism and
revision of his teachings. Even today many scholars cling to the
belief that Plato's Parmenides bears witness to such selfcriticism.
It is more or less fatal for this theory, however, that the ancient tradition never reports such a change in views in either
Plato or Aristotleaside from a single observation in the Metaphysics, Mu 4, 1078b 10 which makes the number theory appear
to be a late form of the doctrine of the ideas. Aristotle cites the
Phaedo just as readily as he does the Parmenides or the Timaeus
1. J. Dring, Aristotle's Protrepticus. An Attempt
(Gteborg, 1961).

at

Reconstruction

9 T H E Q U E S T I O N AT ISSUE

and seems not to have noticed at all that Plato himself had ever
placed his dogmatic theory of the ideas in question. In truth it is
almost absurdly obtrusive to the modern reader that the late
Plato of the Parmenides seems every bit the equal of Aristotle in
criticizing the doctrine of the ideas. Even the famous "third
man" argument is, as is well known, not only found in the critique of the ideas in Aristorie's Metaphysics^ but in the Parmenides as well. Certainly the worst of all hypotheses is to assume
that Aristotle ignored Plato's self-criticism and cold-bloodedly
repeated Plato's critical arguments in his own critique of Plato.
The picture looks even worse with regard to the "development" in Aristotle's ethics. Aristotle's presumed evolution from
a "politics of the ideas" (in the Protrepticus) through a still hesitant distancing of himself from Plato in the Eudemian Ethics to
the "mature" and self-confident position of the Nicomachean
Ethics is an arbitrary and contradictory construction of Jaeger's.
It is particularly unconvincing if one also brings Plato's late dialogues to bear on the issue, for the Philebus and the dialogue on
the statesman would be so far in advance of the supposedly
Platonizing beginnings of Aristotle's ethics that one can properly
ask onself just who is criticizing whom here. The development
schemathe postulation of the ideas apart by themselves, then
participation of the appearances in the ideas, then dialectic of
idea and appearance, and then, at the end, the equating of idea
and numberslowly begins to come apart.
Did Plato at first really underestimate the problem in the participation of the appearances in the ideas? Did he teach that the
ideas were apart for themselves until one day he recognized that
the problem of participation entailed in the postulation of such
ideas for themselves was altogether insoluble? Or do both postulations belong together: the ideas being for themselves, the socalled chorismos (separation), and the difficulty, to which one is
thereby exposed, concerning participation, or methexis^ as it is

10

T H E Q U E S T I O N AT ISSUE

called? Could it be that chrismos and methexis go together even

from the start? At the end, in his so-called self-critique, does
Plato not have precisely these two aspects of the subject matter
in mind? And in the Parmenides is it not precisely his intent to
fend off the oversimplifications in any dogmatic conception of a
doctrine of the ideas that would like to spare itself the trouble of
dialectic?^ Might it have been Plato's actual intent in the Parmenides to make us so acutely conscious of the ontological problem
in the relationship between idea and appearance that the very inappropriateness of the solutions discussed demonstrates the dogmatism implied in the question itself? In any case, it is striking
that throughout the dialogues the terminology used for the relationship between idea and appearance is extremely free: parousia (presence), symploke (interweaving), koinonia (coupling),
methexis (participation), mimesis (imitation), and mixis (mixture) are all found alongside each other. Both the Parmenides
and Aristotle's critique finally single out methexis from these expressions. Plato coins this new word, so it seems, for the "participation" of the particular in the universal, and the problems it
entails are unfolded in the Parmenides in particular. Moreover,
that [it was Plato's own word] can virtually be deduced from
Aristotle's observation that Plato followed Pythagorean philosophy, save that where the Pythagoreans spoke of the mimesis of
the things in relationship to numbersthat is, of the visible exemplification of pure numerical relationships in the order of the
heavens and in the theory of musical harmonyPlato, he says,
merely uses another word, namely, methexis,^
2. The critique of the all too youthful Socrates at Farmemdes 135d may have
been aimed at making just this point.
3. That the word as such existed in the Ionic form, metoche, which Aristotle
uses, is proved by the appearance of it in Herodotus 1.144. But our concern here
is with its conceptual application.

11 T H E Q U E S T I O N A T I S S U E

With this new word, it seems to me, Plato wants to bring out
the logical connection of the many to the one, the thing "in common," a connection that was not implied in mimesis and in the
Pythagorean relationship of number and being conceived of as
the "approximation [of number] to being" (J. Klein)/ And beyond that: if we pay attention to Plato's set of synonyms as such,
we will have to take methexis as well as mimesis more "objectively," *that is, not as "acts" of subjectivity, our ways of conceiving of things, but as real relationships*. Mimesis refers to the
existence of what is imitated or represented, while methexis refers to coexistence with something. Of course, like the Latin
participatio and the German Teilhabe^ the word methexis evokes
the image of parts. That it does is shown by the early usage of
metechein.^ That the part belongs to the whole is precisely what
the new word underscores. Even in what is perhaps the earliest
allusion to the ideas, in the Euthyphro, the question is formulated in such a way that to hosion (what is pious), for instance,
could be a morion (part) of to dikaion (what is just).^ In the first
place that means that where one of them is, the other is too: the
part is present "in the whole." However Plato is obviously fully
aware of the paradox in a participation or taking part {Teilhabe)
that does not take a part, but participates in the wholeas the
day participates in the light of the sun. That he is, is shown by
the use of that very image in the Parmenides and is indirectly
confirmed by the set of synonyms I listed above. Indeed, as I
J. Klein, "Die griechische Logistik," in Quellen und Studien zur Geschichte der Mathematik, vol. 3, no. 1 (1934), and vol. 3, no. 2 (1936).
5. The predominant meaning is plainly to take part in something. But taking
part in something along with other things, links all the participants to each other.
6. Metoche obviously became terminological later in the academy and precisely in such usage as this where one speaks of the participation of a species in a
genus.

12

T H E Q U E S T I O N AT ISSUE

have shown elsewhere/ the aporia (impenetrable problem) of the

whole and the parts always lurks behind the dialectic of idea and
appearance, unity and multiplicity.
Aristotle's account, however, implies something else, which he
conceals. The change from mimesis to methexis is not the harmless terminological variation it sounds like it is in Aristotle. On
the contrary, it actually reflects the decisive turn which Plato
takes: his distinguishing between aisthesis (perception) and 0^sis (intellection). That is to say, it reflects the step toward mathematics' first comprehension of itself as an "eidetic" science. As
long as such a self-understanding had not been achievedand
the Pythagoreans had evidently not achieved itthe numbers
could actually appear to be the existing paradigms that the appearances strove to approximate. The "torturing" of strings (Republic 531b) expresses this striving quite aptly, but the idea of
approximation has something ridiculous about it if one thinks of
the 'pure' relationships as such. To say that these are christon
(separate) is to say nothing else than that pure relationships do
exist; there is precisely no such thing as a ratio of more or less
1:2.
But if so, mimesis becomes an inappropriate expression. To be
sure, it continues to make sense in a certain metaphorical way if
one describes the world of appearances as a mimesis of pure
mathematical relationships, that is, as mere approximations.
Thus it remains possible for Plato to use the expression mimesis^
together with the pair of concepts copy and paradigm, from the
Phaedo until the Timaeus. Methexis^ on the other hand, describes things starting from the other side, the being of the pure
relationships, and in so doing, it leaves the ontological status of
7. Compare my "Zur Vorgeschichte der Metaphysik/* in IJtn die Begriffswelt der Vorsokratiker, pp. 364-90.

13 T H E Q U E S T I O N A T I S S U E

what participates undefined. In the Philebus it is even possible-^

for Plato to formulate the ontological status of what participates
as genesis eis ousian (coming into being) (26d). Moreover, the
new expression methexis fits better in the context of the longstanding Eleatic problem of the one, the whole, and being, which
Plato continues to explicate on his own. Therefore Plato accepts
as part of the bargain that the dialectic of the whole and its parts
will inevitably color the relationship of multiplicity and unity:
parts, like members {ta mele te kai hama mere) (14e), belong to
the whole whose parts and members they are. That may not say
a great deal, but [even so, we must ask] what bearing it has on
the relationship of the many to the one, that is, on the participation in the idea. The Philebus raises this question, and whatever
its solution is supposed to be, a dogmatically rigid conception of
the chrismos is ruled out by the fact that "the many," which are
not being but genesis (becoming), belong to being as parts and
members. This fact does not keep the holon-meros (whole-part)
dialectic in Plato from playing a large role elsewherein the
Sophist and the Parmenides besides the Philebus. Using this dialectic, Plato is able to lay bare the multiplicity in the logos of being [the statement of what a thing "is"]. As a matter of fact,
even in Parmenides' didactic poem, the whole issue of the logos,
the theme of the "multiplicity" of words (names) for the one Being, is completely obscured. And only Plato's Sophist casts light
8. The logos ousias^ or saying of what a thing is in substance, is examined by
Aristotle in the Categories. The significant point for Gadamer is that being is to
be understood starting from the ways in which we speak of something, the ways
something may be said to be (Aristotle: ps legetai). Ontology thus becomes inseparable from the philosophical investigation of language. Plato takes the crucial turn here in the Phaedo, when his Socrates abandons all naturalistic accounts
of reality and "flees" to the logoi, to our ways of saying things. In what follows,
Gadamer will have much to say about this turn to the logoi, which is at the core
of the "unitary effect" of Platonic-Aristotdian philosophy, TRANSLATOR.

14

THE Q U E S T I O N AT ISSUE

into the darkness here, one step's worth at least, by criticizing

Parmenides and demonstrating the interweaving of the highest
genera.'
From what we can extract from the ancient reports about
Plato's students, the same liberality in interpreting the relationship between idea and appearance seems to have prevailed in
Plato's school.^ We know from Alexander" that Euduxus, for
instance, explicitly taught that the ideas were immanent in appearances, and that to make this point he used the concept of
mixis (mixture), which, as a matter of fact, we often find in
Plato's dialogues too. Could it be that Plato's liberality ultimately extended so far that he not only let different theories
stand about the relationship of the ideas to the numbers and to
things but even accepted Aristotle's disputing of the independent
being of the ideas as well? For it is certain surely, and not seriously doubted by anyone today, that from early on Aristotle was
critical of Plato's doctrine of the ideas but nonetheless was, and
remained, a Platonist into his late works.
If one compares Aristotle with the teachings of earlier or later
Greek thinkers, one cannot doubt that all in all he must be
counted as part of the eidos (form) philosophy that Plato established by introducing the ideas and dialectic. Aristotle himself
leaves no doubt in the matter: in his critical overview in Metaphysics^ Alpha, he finds that the Pythagoreans and Plato were
the first to go beyond the explanatory schema of the
9. For specifics here, see my "Zur Vorgeschichte de Metaphysik."
10. H. J. Krmer has attempted to reconstruct the discussion of this problem
by means of a developmental history in his "Aristoteles und die akademische
Eidslehre. Zur Geschichte der Universalienprobleme im Piatonismus," in Archiv fr die Geschichte der Philosophie 55 (1973). (See n. 13 below.)
11. Compare Dring, Aristotle's Protrepticus^ pp. 244 (with bibliography)
and 253.

15 T H E Q U E S T I O N A T I S S U E

"physikoi"hyle (matter) and both en he kinesis (whence motion comes)and he grants Plato and the Pythagoreans their
concept of the ti estin (what something is). (For the Pythagoreans, see 987a20, and for Plato, 988alO and, above all, 988a35.)
Unlike Aristotle, neither the atomists nor Anaxagoras nor the
Stoic school, and perhaps not even a * mathematically oriented'^
man like Strato in the Peripatetic school, can be understood
starting with the legomena (things we say). In other words, unlike Aristotle, they cannot be interpreted as successors of Plato's
"flight into the logoi (ways of saying things).'' In opposition to
the privileged ontological status that Plato accords the idea,
Aristotle emphatically asserts that the primary reality is the particular individual, the tode ti (this-something), but even so he remains within the framework of Plato's orientation toward the
logoi. His "primary" substance in no way excludes the eidos.^^
On the contrary, there is an obvious and indissoluble connection
between that "secondary" substancethe eidos that answers
the question ti estinand the primary substance of any given
"this."^^
12. In Amicus Plato^ Magis Arnica Veritas," Gadamer argues similarly that
what separates Plato from Aristotle is not at all Plato's emphasis on the eidos and
the idea, for Aristode emphasizes these no less than Plato. The difference (aside
from their divergent means of presentation) is that Plato approaches the eidos
through mathematics, Aristotle, in contrast, through life science. For Plato the
eidos, as an answer to ti estin ("what is it?"), is to be understood in distinction
from its appearancesas mathematical circularity, for example, is to be understood in distinction from circular things. Aristotle, on the other hand, sees the
eidos as that which a living thing (tode ti) actualizes as it grows from lack {steresis) to fulfillment. Thus, for Aristotle, the eidos is one predicatealbeit the
principal oneamong many others that may be said of some thing. And as such,
it, unlike circularity, cannot be thought of in separation from the subject of
which it is predicated, TRANSLATOR.
13. It seems risky to me to build an interpretation as extensively upon the
shifting meanings of "primary being" in the Categories and the late books of the

16

T H E Q U E S T I O N AT ISSUE

Though I am using the concepts of the (disputed) Categories

for my explication here, I am relying in essence not only on
these, but just as much on the central books of the Metaphysics^
particularly Zeta 6. It is evident too that in Aristotle's eyes there
is no contradiction here with his critical demarcation of his own
thought from Plato's. The common problem, basic to both
Aristotle's and Plato's investigations, is how the logos ousias (the
statement of being, of what a thing is) is possible. For my part I
would assert that the locution chrismos was never intended to
call into question the fact that what is encountered in appearances is always to be thought of in reference to what is invariant
in it. The complete separation of a world of the ideas from the
world of appearances would be a crass absurdity. If Parmenides,
in the dialogue of the same name, consciously pushes us in the
direction of that complete separation, he does so, it seems to me,
precisely in order to reduce such an understanding of the chrismos to absurdity (see Parmenides 133b ff).
But what, then, is Aristotle's stereotypical objection that Plato
hypostasized the universal, his "chrismos" objection, supposed
to mean? Does it really apply to Plato? Just what is the being-forthemselves of the ideas supposed to express in Plato? Does it really imply the opening up of a second world, supposedly separated from our world of appearances by an ontological hiatus?
Whatever the answer, things certainly do not stand as the caricature in the Parmenides suggests, namely, that this other world of
ideas that always are exists only for gods who always are, and
our sensory world of fleeting appearances only for mortal human beings. That the ideas are ideas of appearances and that
they do not constitute a world existing for itself are expressed
Metaphysics as H. J. Krmer has done in his admirable and perceptive essay
"Aristoteles und die akademische Eidoslehre."

17 T H E Q U E S T I O N A T I S S U E

negatively by Plato in this, the harshest aporia of the Parmenides

(133b). Aristotle himself says explicitly that there is a basic reason for postulating the ideas: in view of the ever shifting tides of
appearances, everything hinges on knowledge of their ideas if
there is to be any knowledge at all (Metaphysics 987a32 ff.).
Surely the entire doctrine of the ideas rests upon an obvious assumption: one cannot take the chrismos to mean that the connection presupposed [between the ideas and appearances] is now
to be severed.
One need only keep in mind what Plato had in view and the
historical motivation that led him to carry out this separation of
the ideas from appearances. Before him lay the entire expanse of
the mathematical sciences. There is hardly a better characterization of the fact that Euclidean geometry refers to pure spatial
relationshipsand not to the sensory images of a circle or triangle that we draw as illustrationsthan its requirement that
mathematical constructs be separated from the sensory world.
And one cannot even say that the distinction here is all too selfevident. Pythagorean mathematics was certainly genuine mathematics, and its theorems and proofs obviously did not refer to
the figures produced to illustrate them. But plainly it had no appropriate understanding of how its true objectscircle, triangle,
numberdiffer from sense perceptions. This lack of understanding is portrayed explicitly in Plato's Theaetetus^ and corresponding to it there existed at this time the practice of mathematical pseudo-proof, which adduced the way things seem to the
eyes, for example, the coincidence of a straight line with a very
slightly curved segment of a circle, in support of its 'mathematical' argumentation. Only the ontological divorce of the noetic
from the sensory, which is to say, Plato's chrismos, cleared
things up sufficiently in this regard for mathematicians to be able
to say what they were dealing with, and to make evident that

18

T H E Q U E S T I O N AT ISSUE

what they were doing was in any case not some sort of physics.
This distinction between mathematics and physics is a fundamental truth. Moreover, it is no coincidence that modern mathematical natural science can make a lot more sense of Plato's
treatment of the world of mathematics as an intelligible reality
for itself and as a prefiguration of nature than it can of Aristotle's derivation of the world of mathematical objects by abstraction {aphairesis) from physical appearances. Aristotle's 'solution' (Physicsy Beta 2) suppresses the actual problem of the
being of what is mathematical, which is to say, that very beingfor-itself which has such seminal relationships to the being of appearances as those discovered by modern mathematical physics
and anticipated by Plato in the Timaeus.^^
Things are similar in regard to moral phenomena. The distinction between justice itself and what is considered [dokei) to be
just is anything but an empty conceptual abstraction. On the
contrary, it is the truth of our practical consciousness itself, the
truth as Plato saw it graphically before his own eyes in the person of Socrates: true and just human behavior cannot be based
on the conventional concepts and standards to which public
opinion clings. Rather, such behavior must take as its standard
only those norms that transcend any question of public acceptance, and even the question of whether they can be, or are ever
found to be, fully realized, and that thereby display themselves
to our moral consciousness as incontestably and unalterably true
and right. This severance of the noetic from the sensory, of true
insight from mere points of viewthis chrismos, in other
wordsis the truth of moral consciousness as such. Again, it is
no coincidence that this insight of Plato's was dignified anew
14. Compare my "Dialetic and Sophism in Plato's Seventh Letter"^ and "Idea
and Reality in Plato's Timaeus.''

19 T H E Q U E S T I O N A T I S S U E

when the concern was to give morality a transcendental foundation. Kant's rigor is exceeded only by the rigor with which Plato,
in the dialogue on the true state, compels his Socrates to separate
the true essence of morality from what is held to be socially acceptable {Republic^ book 2) and to display this separation with
the example of a man who is held by all to be unjust and therefore put to death by use of every conceivable torture (361c ff.).
And finally, what if one now ventures out upon the entire vast
sea of the things we say and looks for a fixed orientation within
the ups and downs of speaking and reflection, amidst the very
unsteadiness, in other words, which had been cultivated at just
that time into the new art of speaking and arguing that Plato derogatorily labels sophism?^^ Once again, only the divorce of the
illusory way an argument sounds from its real meaning and of
the apparent cogency of what is said from the consequential
logic inherent in the subject matter can provide that orientation.
The so-called epistemological excursus of the Seventh Letter
(which I have already treated elsewhere^^) makes completely
clear, I think, what this divorce actually means. It is intended to
expose the weakness of sense experience, a weakness that threatens to prevent our reaching any mutual understanding.^^
The chrismos is not a doctrine that must first be overcome.
15. That the label was discriminatory, at least in the eyes of high society, is
demonstrated by the Protagoras 317b.
16. See "Dialectic and Sophism in Plato's Seventh Letter."
17. In his "Dialectic and Sophism in Plato's Seventh letter^"* Gadamer shows
how the "means" of languagename, image, concept, ideacan all assert
themselves instead of the thing that they were intended to bring to light. Thus
language, as that which "lets things be" (Heidegger), conceals as well as discloses: Sophism is quick to seize upon this potential of language to obfuscate and
confuse, for its concern is not with the truth but with manipulation. The "new
paideia," to which Gadamer often refers, was essentially instruction in the techniques of manipulation by means of language, TRANSLATOR.

20

THE Q U E S T I O N AT ISSUE

Rather, from beginning to end, it is an essential component of

true dialectic. Dialectic, consequently, is not to be advanced as
evidence against the chrismos and is not a remedy for it. That
Aristotle focuses his critique on the issue of the chrismos even
so must be understood in reference to the point he wishes to
make in opposition to Plato, that is, in reference to his emphasis
on physics and the ontological primacy of a given particular.
And in any case, his attack on the chrismos cannot represent a
total break with the fundamental orientation of Plato's thought
toward the logoi. The question of just what Aristotle's critique
meant within Plato's school and among Platonists will always
have to be raised anew, it seems, and in order to raise it a comprehensive yaboration of the underlying accord between Aristotle's and Plato's doctrines of the logos tou eidous (statement of
the form) and of what these doctrines have in common, would
be required here. Only in this way could the ground be cleared
for a meaningful articulation of Aristotle's divergent doctrine.
The following investigation concentrates on the problem of
the idea of the good. The idea of the good is not just any one idea
among all the others; rather, in Plato's view too, it occupies a
preeminent place. I shall attempt to work out what its special
status is and to shed new light on the importance of that special
status for the fundamental problem in Platonismwhich is just
as much the fundamental problem in Aristotle *that is, the role
of the eidetic entities, or forms'^. Here we are in a privileged position with regard to the indirect tradition, since Aristotle deals
at length in all his ethical treatises with the question of the good,
or the megiston mathema (most important insight) as Plato had
designated it. So perhaps in the mirror of Aristotle's critique we
can also acquire the key to why the ideas were introduced in the
first place. There is a common theme in Plato and Aristotle here.
This fact, which alone makes possible any evaluation of Aris-

21 T H E Q U E S T I O N A T I S S U E

totle's critique of Plato, has been neglected in modern scholarship until now because of the dominant "historical" interpretive
schema deriving from Hegel '^that construes its subject matter in
terms ofantithetical relationships. Scholarship, it seems to me,
finds itself at a loss in regard to Aristotle's critique of Plato as a
consequence of this neglect. I see the beginnings of a better understanding of this critique in Anglo-American research, for example, in Cherniss's The Riddle of the Platonic Academy
which in my views gets somewhat bogged down, nonetheless, in
arguing against the traditionally accepted authority of Aristotle's
accountsand in Lee's Phronesis}^ Each rightly relies on, and
emphasizes, the dialectical, propaedeutical character of the critical introductory books in Aristorie's three ethical treatises. But
to do that is not enough. The task is to get back to the common
ground upon which both Plato and Aristotle base their talk of
the eidos.
The question about the good and, in particular, about the
good in the sense of arete, the "best-ness" of the citizen of the
polis (city-state), dominates Plato's writings from the very start.
And even if we leave aside the consensus that has been reached
today regarding the chronology of Plato's dialogues by and
large, there can be no doubt that in these writings the doctrine of
the ideas does not occur in the same way from beginning to end.
Naturally that does not mean that Plato came to this theory only
later on. It is time that we finally abandoned such a naive chron18. Cf. H. F. Cherniss, The Riddle of the Platonic Academy (Berkeley, 1945),
and E. N. Lee, "Exegesis and Argument," in Phronesis, vol. 1 (1973). Only after
I had completed this manuscript did two books appear that show that I am not
alone in my contention: A. Blum, Theorizing (London, 1974), and above all, J.
Findlay's thorough and speculative work, Plato. The Written and Vnivritten Doctrines (London, 1974), my review of which has appeared in the meantime in the
Philosophische Rundshau, vol. 24 (1977).

22

T H E Q U E S T I O N AT ISSUE

ological ordering of Plato's dialogical fiction, which in the end

was tantamount to a veritable game of running back and forth
from one home base to another. Instead, we shall seek structural
similarities among groups of dialogues, so that along the way we
may clarify both Plato's intentions as an author as well as the implicit content of the dialogues.
Consequently, we shall distinguish among the different types
of dialogue in Plato's works, *and once these have been identified,* we will be able to establish a structural chronology based
upon them. The "aporetic" dialogues, in which Socrates refutes
his interlocutors without finally giving an answer to the question
posed, represent one clearly defined type of Socratic discussion.
(Given Xehophon's apologetical efforts to bring out the positive
side of Socrates' art of refutation, there is no real correlation between this type and what the "para-tradition" in Xenophon tells
us, nor between this type of dialogue and the pseudo-Platonic(?)
Clitopho. Still, these very apologetical tendencies in Xenophon
do provide a certain indirect confirmation of the historical
Socrates' "negativity" [just by denying it].) The new role that
Socrates plays in the discussion of the just state [in the Republic],
for instance, should, and must, be perceived as a clearly accentuated change. Socrates talks virtually the whole night long, and
there is certainly no lack of bold positivity in his Utopia. Furthermore, the affinity in the content and themes of the discussions in
the "negative" [or aporetic] dialogues unifies them as a group.
All the refutations which they contain of the preconceptions
about arete that the interlocutors bring with them to the discussionbe they young people, their fathers, or the celebrated
sophist teachers of the timehave a common character: no answer is found to the questions posed, namely, what this or that
arete actually is, or whether arete can be taught or not, in view of
the unclarity and deceptiveness of whatever particular thing is
taken to be arete [piety, courage, and so forth].

23 T H E Q U E S T I O N A T I S S U E

In contrast, the inclusiveness of the question posed in the Republic seems intended to counter the conventionality of the conceptions of arete found in the earUer discussions. Here the inquiry is about the sympasa arete (all-encompassing virtue) and
dikaiosyne (justice), and consequently about all aretai (virtues).
And at the end the idea of the good "beyond" all these is introduced. The knowledge sought is not there, and perhaps is not
even attainable, as long as one does not consciously look beyond
what is generally accepted to be knowledge. *In assessing just
how far we have gone beyond the Socratic dialogues here*, it is
of particular significance that Plato's brothers are the ones who
make the transition with Socrates in the discussion in book 2 of
the Republic.
The key notion in what is generally accepted to be knowledge
is techne (art). For Plato too, that goes without saying. Thus, in
the Apology^ Socrates finds among the handworkers, at least,
real knowledge of the special things they deal with. To be sure,
like other "wise" men, handworkers, too, fail us when the concern is the most important thing of all {ta megista) {Apology
22d), for which all human will to know ultimately strives.
Knowledge of the good is exactly what is not asked about in the
technai (arts) and by the technites (artisan-handworker). That it
is not is the standard argument used later by Aristotle in his critique of Plato {Nicomachean Ethics 1097a5 ff., henceforth EN;
Eudemian Ethics 1218b2, henceforth EE; Magna Moralia
1182b25 ff., henceforth MM). But this fact, far from weighing
against Plato, is precisely what gets Plato's Socrates started on
his way. His awareness of it is the basis of his superior "ignorance." Knowledge of the good would seem to be different in
kind from all familiar human knowledge. Hence, if measured
against such a concept of specialized expertise, it could indeed be
called ignorance. The anthrpine sophia (human wisdom) that is
aware of such ignorance must inquire beyond, and see beyond,

24

T H E Q U E S T I O N AT ISSUE

all the widespread presumed knowledge that Plato later will call
''doxa'' (belief, opinion). That the good can be caught sight of
only in this apoblepein prosthis looking at it in seeing past all
elseis not merely suggested by the negative result of the Socratic discussions. It is stated explicitly in the first dialogue in
which the acceptance of the ideas is actually proposed, the
Phaedo. The Phaedo thus stands as the noteworthy link between
the elenchtic or refutational dialogues, which must be attributed
to the early Plato, and his work on the ideal state.
The Phaedo has been singled out as the dialogue in which
Plato first introduces us to the doctrine of the ideas. The Marburg school in particular availed itself of Plato's introduction of
eidos as the best "hypothesis," for a somewhat forced assimilation of Plato to Kant. To be sure, Natorp's interpretation was
not lacking in insight into the exceptional role of the good. For
him the good was the principle of self-preservation. He saw
hypothesizing the eidos as a procedure for knowing this principle, and in this way he came to identify "idea" with "natural
law." What he had in mind, accordingly, was natural science. In
its ascending hypotheses the latter does indeed move ever closer
to the true order of the universe, and it is carried out in ongoing
determination of its object. For Natorp the "thing in itself" is
nothing more than the infinite, "unending task."
Today that interpretation sounds like false modernism. If, instead, one looks at the Phaedo as an interim stage on the way to
the idea of the good, another structural parallel emerges. After
all, the Phaedo too is a dialogue full of refutations, and in their
own way the interlocutors in the Phaedo represent a position
that is to be taken seriously, even if they defend it only halfheartedly, namely, "scientifically" founded materialism. They are Pythagoreans, of course, but Pythagoreans of a later generation
that is entirely at home in mathematics and science and that ab-

25 T H E Q U E S T I O N A T I S S U E

jures the contemporary sophistic argumentation as such. Thus

they have become Socrates' friends. It turns out that for them the
rehgious background of Pythagoreanism has faded away entirely. The "inner plot" of the dialogue consists precisely in the
demonstration to these Pythagoreans of the true consequences of
their own thinkingconsequences that point to the "idea."
Thus at one and the same time their understanding of themselves
as "materialists" is refuted, and their mathematical idealism
confirmed.^'
Aristotle introduces Plato precisely as a Pythagorean {Metaphysics^ Alpha 6), but it is not only in retrospect from Aristotle
that it becomes clear that these Pythagoreans in the Phaedo are,
in a certain sense, very close to Plato's thinking. The further we
penetrate into the problems surrounding Plato's development of
a doctrine of an archeproblems that first appear in the late dialogues, in particular in the Philebusthe more evident it becomes that the Pythagorean position represents the real link
between the insoluble problems * Plato displays* in this new
techne-based paideia and his later dialectic. It is no accident that
the question of the "invisible" psyche (soul) is the vehicle first
chosen to exemplify the "invisible" noetic dimension of mathematics and the ideas. Seeking the koinon past agathon (the good
common to all things), Socrates turned to Anaxagoras's treatise
in the expectation of finding it in the latter's concept of nous (intellection, mind). And when, perplexed and disappointed, he finally sees himself forced to hypothesize the eidos in order to
get clear about the true sense of ''aitia'' (cause) in all things
(lOOa-c), he takes the approach of mathematical illustrations.
But in the end it becomes fully clear that only looking to what is
good (apoblepein pros ton agathon)^ or to what is better or best,
19. Cf. my study, "The Proofs of Immortality in Plato's Phaedo."*

26

T H E Q U E S T I O N AT ISSUE

respectively, promises him real knowledge, or, as we would say,

"understanding"understanding, namely, of the universe as
well as the polis and the psyche. To this extent the work on the
polis [the Republic] (which is every bit as much on the psyche)
and then, later, the Timaeus^ adhere to, and articulate, the program projected in the Phaedo's flight into the logoi (statements).
It is not in the least excessive to see in Aristotle's textbooks the
execution of just this project, particularly in his Physics and his
Politics^ which contains his ethics. We have here an ideal of science which plainly did not impede the birth and astonishing development of the Hellenistic specialized sciences. For modern
natural science, on the other hand, that Aristotelian ideal represented the antagonist: it was dogmatic, teleological anthropomorphism that had to be overcome. All attempts to renew this
ideal of a teleological, homogeneous science, beginning with
Leibniz and continuing through the Romantic philosophy of nature to thinkers like Whitehead, have been unable to assert themselves against the inexorable, step-by-step progress of the modern experimental sciences. Precisely the failure of these attempts,
however, indirectly confirms that reason has need of unitya
need that underlies the Socratic requirement. And it is from just
this need that the claim of universality in the doctrine of the
ideas derives.^
20. The point to which Gadamer often returns in this work is that for Plato
and Aristotle the good could function simultaneously as an ontological-theoretical, as well as a practical, concept, indeed, that these two realms remained continuous. As the Phaedo makes clear, natural processes were to be grasped theoretically as purposive, as tending toward what is good. Aristode's teleology
merely extends this way of thinking. Consequently, for both Plato and Aristotle,
human practicemaking choices for the sake of what is good, better, or best,
respectivelyoccurs within the setting of the cosmos. Modern science, which rejects teleology as anthropomorphism, leads to a severance of theoretical science
from practice and of theoretical reason from practical reason. Reason as such,
which seeks unity in its enterprises, naturally rebels once that happens.

27 T H E Q U E S T I O N A T I S S U E

In this regard the idea of the good assumes a special significance within Plato's new noetic orientation. The Republic focuses on it, and it provides the basis for determining the order of
the polis and the psyche. Socrates' partners in the discussion
would be content if he were to speak about the good in the same
way as he had spoken earlier about dikaiosyne (justice) and sophrosyne (temperance) {Republic 506d). But whenever the conversation turns to this highest and ultimate subject, the speaker
in Plato begs off, saying that it would be unnecessary at that moment to go into it and perhaps even beyond his abilitiesjust as
in the Timaeus at 48a, for instance. And as a matter of fact, the
famous epekeina tes ousias (beyond all being) lends the idea of
the good a transcendence that distinguishes it from all other noetic objects, which is to say all other ideas.
That Plato uses only the word idea^ and never eidos^ for the
agathon^ smdy has something to do with that transcendence.
There is no denying, of course, that these words, idea and eidos^
were interchangeable in the Greek of that time and in the language usage of the philosophers too. Still, that Plato never
speaks of the eidos tou agathou (form of the good) indicates that
the idea of the good has a character all its own. Eidos always refers only to the object, as accords with its being neuter. In following the natural tendency of our thinking to objectify, the
Though Gadamer does not pursue the thought here, one might add that the
contemporary pseudo-scientific forms of "moral reasoning," utilitarianism in
particular, also confirm reason's need for unity insofar as they aim to carry the
scientific method over into the realm of practice, and thereby reestablish the
unity of reason. But here the result is the evisceration of moral thought, A.
Maclntyre, in After Virtue (Notre Dame, 1980), shows convincingly why such
attempts inevitably miscarry and end in "emotivism." In so doing, he corroborates Gadamer's argument that the methodology of modern natural science is
misplaced in many fields of the human quest for truthin aesthetic interpretation, politics, history, and, not least of all, moral philosophy, TRANSLATOR.

28

T H E Q U E S T I O N AT ISSUE

feminine form, idea^ can certainly designate an object too, as do

doxa and episteme. But in idea^ taken as a "view of something,"
the viewing or looking is more pronounced than in eidos^ taken
as "how something looks." Consequently idea tou agathou (idea
of the good) implies not so much the "view of the good" as a
"looking to the good," as numerous turns of speech such as
apoblepein pros show.
In any case, in the Republic Socrates treats the idea of the
good as something that is difficult to grasp and that can be observed only in its effects. Like the sun, which by granting warmth
and light, gives everything visible its being and perceptibility, the
good is present for us only in the gifts that it bestows, gnosis kai
aletheia^ insight and truth. It would seem futile to want to grasp
the good directly and know it like some mathema (learned insight), for the very nature of it appears to preclude this possibility. To begin with, its ineffability, its being arreton (unsayable),
should be interpreted as soberly as possible. To say that is by no
means to deny that the religious background of Greek thought is
discernible here. Thanks to the expositions of Gerhard Krger,
we know that it is.^^ But one should be aware that Plotinus takes
a new step when he also calls the "One" epekeina noeseos (beyond all thinking), and when he takes all being and all thinking
too as a pointer into transcendence. In the context of the Republic^ in contrast, the good is presented as the unifying one of the
many. In other words, the good is articulated precisely in respect
to the inner duality and dialectical function of the one which
Plotinus's double "beyond" [beyond being {ousia) and beyond
thought {noesis)] is specifically intended to exclude.^^
21, G. Krger, Einsicht und Leidenschaft. Das Wesen des platonischen
Denkens (Frankfurt, 1948).
22. Gadamer's argument, which he elaborates in "Plato's Unwritten Dialectic," is that Plato is distorted in the Plotinian reading of him insofar as the ines-"

29 T H E Q U E S T I O N A T I S S U E

Now there seems to be a simple way to account for this privileged status and incomprehensibility of the idea of the good,
which distinguishes it from the remaining ideas. The idea of the
good, it is said, is precisely what comes "first" {to proton), and is
thus removed from any derivation. Thus it is what one was later
to call a principle. And, after all, this solution has generally been
the one resorted to ever since Aristotle, who was the first to introduce the concept of the arche (first principle). Furthermore,
wherever Aristotle introduces Plato's philosophy as an extension
of Pythagorean teachings, he does in fact treat it as analogous to
the Pythagorean doctrine of the peras (limit), and the apeiron
(unlimited, indefinite). He presents it in the doctrine of the hen
(one) and the ahoristos dyas (indeterminate two)the two principles of either the ideas or ideal numbers, as the case may be. It
seems consistent to grant the idea of the good in Plato the same
special place that Aristotle assigned to those two principles in his
account, for it would explain why any speaking of the idea of the
good, which is to say, all dialectical treatment of it, could never
take "the good itself" as its object directly, and why, consecapable concomitance (Heidegger: Gleichurspriinglichkeit) of unity and indeterminate duality is lost from view. It is far more accurate, he contends, to see
Plato's number doctrine as an extension of both the Pythagorean concepts: the
peras and the apeiron too. For Plato, whatever is unified or delimited as "one"
finds itself within the indeterminate duality {ahoristos dyas) of unlimited extremes, for .example, large and small, hot and cold, pleasure and pain, and so
forth. Hence what is determined as one, as a thing, is inseparable from the indeterminacy in which it is embedded.
This point has important consequences for Plato's ethical thought as Gadamer
understands it: human beings are never purely good, which is to say never purely
harmonized, integral, unified, but always struggling for (diamachesthai) unity
against those passions and drives that threaten them with dissolution and loss of
integrity. It is in this sense that Plato's ethics may be said to be "dialectical": reasonableness (nous, phronesis) is always in dialectical tension with the immediate
desire for what is "sweet" (hedone). TRANSLATOR.

30

THE Q U E S T I O N AT ISSUE

quently, whenever Plato fixes his sights on it, he does so in the

language of analogy.
Convincing corroboration of this point comes to light in the
Philebus, Of course there the concern at first is not the idea of
the good but the question of the good in human life. But in the
end, any talk about the universal idea of the good always takes
as its point of departure this human question: What is the good
for us? That this is the case emerges clearly in Socrates' selfportrayal in the Phaedo. The particular question raised in the
Philebus is the extent to which the passion of our drives and our
consciousness while thinking can be harmoniously balanced in
life. Putting the question this way implies turning from the realm
of the ideal to what is best in realitywhich appears to run directly counter to the Republic's project of constructing the ideal
state. But there, too, this turn to the practical is not altogether
unheard of. To start with, the Republic also addresses the good
in human life, for it opens with the very same question as the
Philebus^ namely, whether the good is hedone (pleasure), as the
mass of people {hoi polloi) believe, or phronesis (reason), that is,
whether it consists in satisfaction of one's vital drives or insight
into the good {Republic 505b). Of course, here in the Philebus
this antithesis is no longer stated as a clear-cut either-or. Harmonizing and balancing the two sides has been elevated to the
central concern.
This shift, we can say, makes human practice the theme, and
alongside of this consideration of the practical, a physics can
now be placed, of which the Philebus also gives preliminary indications and upon which the mythical narrative of the Timaeus
discourses. After all, both these dialogues are directed to the
realm of genesis (becoming), and in a way that contrasts fundamentally with Plato's sharp separation of being and becoming.
Nevertheless, in the Timaeus especially, this separation provides

31 T H E Q U E S T I O N A T I S S U E

the framework that sustains the dialogue, just as it does in the

Republic. Thus the actual question in the Timaeus remains the
same, namely, how we are to think of chorismos and methexis in
combination with each other.
The relationship between the two comes into focus even more
clearly in the Philebus than it does in the ideal celestial mathematics and terrestrial physics of the Timaeus, In the Philebus the
idea of the good has precisely the function of providing practical
orientation for the right and just life insofar as this life is a mixture of pleasure and knowing and insofar as the mixing of this
life, which is described at the conclusion, is explicitly said to be
regulated by the idea of measure, measuredness, rationalityor
however else one might epitomize all the structural determinations of the beautiful in whose appearances alone the good itself
is to be comprehended. From Aristotle, and above all from the
subsequent accounts of philosophers' teachings, it would appear
to follow that the good can simply be called the one, and the one
the good. The fact that it can does not seriously conflict with
what the dialogues say. For this "one" is certainly not Plotinus's
"One," the sole existent and "trans-existent" entity. Rather, it is
that which on any given occasion provides what is multiple with
the unity of whatever consists in itself. As the unity of what is
unitary, the idea of the good would seem to be presupposed by
anything ordered, enduring, and consistent. That means, however, that it is presupposed as the unity of many. When we make
this fact our starting point, we get our first sense of the paradigmatic function of the arithmos (number) structure, a structure
that plays a decisive role in the indirect tradition's account of
Plato. The number, after all, is also unity and multiplicity simultaneously. That it is both is illustrated in particular by the fact
that for the Greeks "one" was not an arithmos, that is, not a
sum, not a unity of many. Rather, it was the constitutive element

32

THE Q U E S T I O N AT ISSUE

of the numbers. The smallest arithmos is the "two," and thus

any number must already be precisely a multiplicity and a unity
at the same time." In any event, the idea of the good does not
seem to be a supreme mathema to which every other sort of
knowledge (techne), all practical knowing and all physics alike,
would be subordinate, and which could be known and learned
as these are.

23. Compare my work on "Plato's Unwritten Dialectic."

111

s o C R A T I C K N O W1 N G
A N D N O T - K N O W1 N G

Our preliminary review of the special place of the good in Plato's

thinking about the ideas has shown us that Plato himself saw
that knowledge of the good cannot be understood using techne
as a model, althoughor better said, precisely because
Socrates continually uses this techne model in his critique and
refutation of the views of his partners in the discussion. If one
brings Aristotle's illuminating analyses of the modes of knowing
{hexeis tou aletheuein) {EN, book 4) to bear on this insight, and
in particular his differentiation between technical and practical
knowledge, the end result is not even surprising: we see how
close the knowledge of the good sought by Socrates is to Aristotle's phronesis. In treating phronesis, Aristotle explicitly distinguishes practical knowledge from both theoretical and technical
knowledge.^ He characterizes it as a different kind of knowing
1. Plato's efforts to distinguish Socratic moral reasoning from the new
techne of sophism are of particular interest to Gadamer insofar as in WM he attempts to show that there are forms of reasoning and of giving justification that
are quite different from those of natural science and technology. In this book he
shows that technical or scientific expertise, if carried over into the moral sphere,
becomes an amoral, if not immoral, art of achieving success in the pursuit of
pleasure. Again, modern utilitarianism (see chapter 1, n. 20) and post-utilitarian contractualism, however well-intentioned, would open the gates to sophism
if Gadamer is right. For these are attempts to carry scientific, methodical reasoning over into ethics. J. Rawls, for instance, in A Theory of Justice (Cambridge,
1972) p. 121, goes so far as to speak of a "moral geometry." TRANSLATOR.
33

34

SOCRATIC KNOWING AND

NOT-KNOWING

{alio eidos gnses) (EN 1141b33,1142a30; EE 1246b36). The

virtue of practical knowledge, of phronesis, appears as the epitome of everything that Socrates' exemplary life displays. The
picture of Socrates that Plato draws in the Apology also shows
him to be far removed from any merely "theoretical" concern.
Nevertheless, elsewhere in Plato things look quite different,
and he forces upon his Socrates figure far more complex roles
the antagonist in his dramas, for instance, to sophist paideia,
and, in addition, the enthusiast for myth who tells us of bizarre
realms lying somewhere between religious fables and philosophical thought, and, above all, the dialectician who broadens his
own art of leading a discussion into a paradigm for all cognition
and truth. Here Socrates becomes a mythical figure in whom
knowledge of the good ultimately coalesces with knowledge of
the true and knowledge of being in a highest theria as it were.
Our task originates here: it is to raise this mythical unity of
knowledge of the good, the true, and the real to the level of conceptual thinking, and in so doing, to make comprehensible what
Aristotle shares with Plato even when he critically separates himself from him.
It is characteristic and significant that Plato himself consistently delimits knowledge of the good from all other knowledge
only in a negative way. In the Socratic dialogues he elaborates
the difference by sovereignly reducing the discussion to an aporia. Whoever thinks he knows what arete is, is refuted. And it is
always taken for granted that it is the standard of the technai by
which his "knowledge" is to be measured. It turns out that each
one of those questioned lacks knowledge of the good. Of course,
when Plato's Socrates, be it in the Phaedo or the Republic^ goes
on to speak of passing beyond everything else knowable in moving toward knowledge of the good, and when he singles out the
good because of its transcendence, he does tie the argument into

35 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

the Socratic question about the virtues and the resultant question
about the good. But, imperceptibly, Plato's Socrates proceeds
here to a theoretical knowledge of the good and of being. What
he calls dialectic is a sort of meta-science that opens up behind
the mathematical disciplines described here as pure theoretical
sciences.
Nevertheless, this science is not called "dialectic" unthinkingly. The heritage of Socrates and his art of dialogue lives on in
it. Accordingly, Plato often applies the word phronesiswhich
for Aristotle characterizes the virtue of practical reasonablenessin a wide sense. And he can also use it as synonymous
with both techne and episteme. This usage is never meant to imply that knowledge of the good is really the kind of knowledge
that techne is. [Rather, it shows that] the knowledge of the handworker plays such a paradigmatic role in any kind of knowing at
all that language usage conforms to it. Besides, there is indeed
something that practical reason and technical know-how have in
common: in a certain sense it is true of anyone who has a science
or art that his knowledge is based on grounds, reasons. That
holds for the doctor (an example of which Aristotle is particularly fond), for the mathematician, and for anyone at all who
claims to have knowledge as opposed to mere opinion.
Still, an essential difference between technical-theoretical reasonableness and practical reasonableness becomes discernible
here at once. When he who knows is required to give reasons in
any other case but practical matters, he can draw upon a general
knowledge that he has learned. It is exactly this recourse to general knowledge that characterizes techne or episteme. Hence, in
regard to these Plato speaks of a mathema (an insight to be
learned). But things look very different in respect to the exercise
of practical reason. Here one cannot rely upon previously acquired general knowledge, and yet one still claims to reach a

36

SOCRATIC KNOWING AND

NOT-KNOWING

judgment by one's own weighing of the pros and cons and to decide reasonably in each case. Whoever deliberates with himself
and with others about what would be the right thing to do in a
particular practical situation is plainly prepared to support his
decision with nothing other than good reasons,^ and he who always behaves this reasonably possesses the virtue of reasonableness, of "well-advised-ness." {Euboulia was a political slogan of
the new paideia of that time.) Now it strikes me as significant
2. Gadamer is raising the issue here of moral argumentation, that is, how I
can justijpy to myself and others the choices that I make at the end of my moral
deliberations. Giving justificationlogon didonaimeans giving grounds, or
reasons, in support of my decision. In a society such as that in which Socrates
finds himself, a society, namely, in which "no one does what is just voluntarily,"
the mere patterning of one's behavior on paragons of virtue no longer suffices.
(See "Plato and the Poets" and "Plato's Educational State.") And appeals to traditional morality become empty rationalizations of self-interest. Consequently,
one must take a stand and be able to hold to itto justify it. The issue then becomes what sort of justification this will be, for it is clear that giving reasons, or
grounds, in scientific accounts of things is quite different. What the difference is
emerges clearly in regard to teaching and learning. Scientific knowledge and technical know-how can be taught. What is right, just, or good, on the other hand,
cannot.
With that we come to a critical question for Gadamer. On the one hand, he argues that we are dependent on tradition for justifying our moral choices and
places great emphasis on Sitten, or moral customs (see WM 11). The authority of
these alone, and not autonomous intellection, founds our morality. On the other
hand, he emphasizes here that Socrates, as an individual, knows a good that transcends the merely "conventional" morality to which hoi polloi, or the "many"
(Heidegger: das Man), uncritically subscribe. This apparent contradiction could
be resolved, I suggest, if one were to distinguish between traditional, as opposed
to conventional, morality. The latter, though its sophist advocates often appeal
to traditional authorities (Homer and the like) for justification, consists in fact of
nothing more than prudential accommodations coerced and acceded to, respectively, by unequal individuals whose sense of solidarity and community with
each other has vanished. What is convened on here is not what is sittlich, but
what is convenient for the most powerful. Here, might indeed makes right, and
no real justification is possible (cf. Republic, books 1 and 2). TRANSLATOR.

37 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

that Plato holds fast to this characteristic of practical knowledge,

and that he distances himself from technical knowledge. Dialectic is not general and teachable knowledge, even if Plato often
follows customary language usage and also speaks of it as techne
or episteme. It is not in the least surprising, however, that he can
call dialectic "phronesis" too. Dialectic is not something that
one can simply learn. It is more than that. It is "reasonableness.''
Plainly, in calling dialectic "phronesis," Plato is again following a language usage in which it is perfectly natural to give that
name to the arete proper to human dealings. That fact comes out
most clearly in the Meno, 88b ff.^ This passage is of particular interest because Socrates, here in the midst of one of his didactic
speeches, is astonishingly conciUatory. He does not insist at all
on a complete equation of virtue with knowledge. He leaves
open the question of whether other things besides phronesis pertain to arete: " . . . phronesis ara phamen areten einai etoi sympasan e meros tV (we say then that phronesis is arete, be it either
the whole of it or a part). Obviously his point is only that, whatever the case may be, phronesis plays a role. Aristotle himself
says exactly that in book 6 and the Nicomachean Ethics
(1144bl7ff.).
From this observation one may conclude that Aristotle remains true to the actual language usage ot phronesis in his ethics,
as in fact he does generally. It is not the case, as Natorp believed,
that Aristotle restricted a ceremonious artificial word of Plato's
to the ethical realm. Quite the reverse: Plato in fact widened the
3. The way in which the arete of phronesai (being reasonable) is distinguished from the so-called aretai of the psyche at Republic 518e provides a nice
illustration of the main meaning that is to be heard in phronesis and which
Xenophon always has in mind. Compare, too, the Symposium 209a: Phronesin
te kai ten alien areten (phronesis and the other virtue) and similar passages.

38

SOCRATIC KNOWING AND

NOT-KNOWING

customary usage, whose proximity to practice must have always

been sensed/ to include dialectical knowing, and he did so in order to ceremoniously exalt dialectic. In other words, he took
what was called practical reasonableness and expanded it to include the theoretical disposition of the dialectician.
Conversely, if on occasion Aristotle himself follows Plato's
widened language usage, one should not overburden this circumstance by basing hypotheses about Aristotle's "development" on
it, as Jaeger, Walzer, and others once tried to do. In truth, it
demonstrates only that Aristotle continues to live in the same
word of language as Plato's. Nor does it in any way mean that in
such cases Aristotle forgot the proper sense of phronesis
whose meaning he himself analyzedor that he was consciously
"Platonizing." By the same token, however, the use ot phronesis
in Plato himself does indicate that he was aiming at something
common to both practical and theoretical knowing that transcends the distinction between them. Precisely this may have
been his motive for broadening the usage of phronesis to include
the highest form of knowing: he wants to assign to the true dialectician not a mere skill, but real reasonableness. At the same
time, this distinction implies that for Plato the dialectician does
not possess some superior art, which he employs in se//-justification,^ but that, instead, he seeks real justification. Hence he does
not possess an art that he uses whenever he so desires. Dialectic
4. Compare ShnelPs recent exhaustive investigation of Homer's use of phronesis in Glotta 55 (1977): 32-64.
5. Rechthaben; the allusion here is to the sophist's desire always "to be in the
right." In "Dialectic and Sophism in Plato's Seventh Letter,'" Gadamer points out
that for Plato self-assertiveness and obduracy are among the chief obstacles in the
pursuit of the truth: precisely because a belief is mine, I tend to cling to it rather
than permit myself to be shown that it is false. Sophism only reinforces this propensity insofar as it gives me the means of confounding any counterargument to
my contention, TRANSLATOR.

39 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

is not SO much a technethat is, an ability and knowledgeas

a way of being/ It is a disposition, or hexis in Aristotle's sense of
the word, that distinguishes the genuine philosopher from the
sophist. To be sure, it will take some doing to defend this distinction to the advocates of the new paideia. The pivotal reason why
dialectic is only dialecticthat is, a process of giving and receiving justification, and not knowledge like that of a handworker or
knowledge in the so-called sciencesis plainly that talk which
confuses and confounds does not constitute a threat in both
these other realms in the same way that it threatens inquiry into
the good. It may certainly be said of techne, and episteme too,
that they know, that is, that they have reasons for proceeding in
this way or that or for holding this or that to be true. Socrates'
questioning of the handworkers in the Apology shows this
knowledge to be characteristic of each of them in their specialties. Once the handworkers have learned their trade, they know
why they do what they do and how they do it, and in any event,
their knowledge is displayed in the fact that they know how to
teach their art. Within their competencies they are not to be
shaken by the sophistic arts of rhetoric and argumentation, and
a figure such as Hippias, who passes himself off as some sort of
6. Something of the early Heidegger can be detected here, in particular his
contrasting of authenticity and inauthenticity (cf. Sein und Zeit (Tbingen,
1960] henceforth SZ). Philosophers not only think differently, they exist differentlydifferently from das Man (everybody), or, as Plato would say, hoi polloi
(the many). They are differently disposed insofar as they hold to what is true. The
decisive passage in Plato is Theatetus 174a ff., where the otherworldliness of the
philosopher is portrayed from the viewpoint of hoi polloi: he appears hopelessly
inept and is mocked by maidservants (and sophists). In his lectures at Boston
College, Gadamer pointed out that in fact Thales would not have fallen into the
well, as the maidservant presumed (174b), but would have climbed down in
order to view the stars without the interference of peripheral light. A contemporary reader of Plato would have understood just whom Plato is portraying as ignorant herenot Thales, but self-proclaimed "practical" people, TRANSLATOR.

40

SOCRATIC KNOWING AND

NOT-KNOWING

expert at everything, is not likely to have impressed any

handworker.
The knowledge of the doctor or the mathematician is also secure in the same way. We know that the new paideia tried to create confusion with its techniques here too. In the case of the
medical art, after all, it can play upon a general suspicion that
arises again and again, given the limits of medical skills and the
impossibility of medicine's proving that it is responsible for its
successes. Are we really dealing with a science here? The tract
Peri Technes {On Techne) which comes from the era of sophism,
gives us a good idea of this vulnerability. Something like it exists
even in the knowledge of the mathematician, who, as we can
also easily understand, is vexed on occasion by the intransigence
of everyday experience. We know that to a large extent Protagoras devoted his rhetorical arts to discrediting the mathematician.^ Since the subject matter in mathematics tends not to be
grasped as readily as the subject matter of handwork, it is all the
easier to understand why he did so. It may also be true that the
mathematician himself is incapable of exposing the false impression that such deliberately confusing arts of argumentation give.
However, the Theatetus dialogue, which speaks to just this predicament, confirms at the same time how a real mathematician
would respond. Theodorus and Theatetus remain immune to
such arguments and simply steer clear of them (see Theatetus
164e, 169c). They see what they are dealing withnumbers or
figuresas if these were there before their eyes. Knowing is the
immediacy of having something before one's eyes in this way,
and therefore it can be called "aisthesis" (perception).
The situation is very different, however, where there is no specialized knowing and no speciaHsts, and where, nevertheless,
7. H. Diels, Fragmente der Vorsokratiker, vol. 2 (Berlin, 1907), p. 266.

41 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

each one of us must have the right to his own opinion. I am referring to the question of the good in personal, as well as social, life.
Here everybody talks to everybody else, and each seeks to convince the other of the veracity of his opinion, especially when it
comes to political decisions. A debate about the good is always
going on, and as we saw, everyone would maintain that he was
doing nothing other than advancing reasons or grounds. The
claim made by the new paideia is founded on this circumstance,
in particular the claim it makes for its rhetoric, which is really
the new art. We see this foundation most clearly in the way that
Protagoras defends this claim for his rhetoric in Plato's dialogue
of the same name. He maintains that he educates people to be
good and, indeed, that he does this apparently by nothing other
than his rhetorical and dialectical art; he leaves aside any and
every specialized knowledge {Protagoras 318d). In opposition to
Protagoras's claim, Plato advances the true dialectical art of giving justification, which submits assertions about the good to
question and answer. Plato has a way of making both clear and
convincing that in order to keep oneself from being confounded
in such matters, something else is required besides a technique of
speaking and disputation and besides mere acuity. To be sure,
the ability to differentiate things according to their genera and
thereby expose confusions must be called an art. But more is required than mere acuity to do this.
And true dialectic entails still more. It provides practice in
holding undisconcertedly to what lies before one's eyes as right,
and in not allowing anything to convince one that it is not. Plato
can also call this true dialectic "phronesis," and with good reason. Here, in the question of the good, there is no body of
knowledge at one's disposal. Nor can one person defer to the authority of another. One has to ask oneself, and in so doing, one
necessarily finds oneself in discussion either with oneself or with

42

SOCRATIC KNOWING AND

NOT-KNOWING

Others. For the task is to differentiate one thing from another, to

give preference to one thing over another. Later, when Plato
characterizes this art of differentiating as dihairesis (division),
thus making it almost identical with dialectic, he still has in mind
less a method than the practical task of differentiating where
confusion is especially threatening and prevalent. Such differentiation is not a scientific method in the logical sense. To this extent Aristotle's questioning of the value of dihairesis in demonstration is justified: his objection is that it yields no logically
cogent conclusions. One must know ahead of time in which species the thing under consideration falls. And, as a matter of fact,
dialectic is not demonstration or proof in the scientific sense of a
proof (apodeixis)y which cogently deduces things from presuppositions. On the contrary, the dialectical art of differentiation
presupposes antecedent familiarity with the subject matter and a
continuing preview of, and prospect toward, the thing under discussion. Aristotle was right.
But his objection is not a valid objection to Plato. It seems to
me that the insight which guides Plato is that such an ability to
differentiate dialectically is exactly the same sort of thing as that
abiUty to give justification which characterized the man Socrates
in his holding undisconcertedly to what he had recognized to be
good. Here we really have knowledge (and not doxa) insofar as
someone, knowing his ignorance so well, is completely willing to
give justification. After all, Socrates' guiding theme is arete. In a
certain sense arete is something that one always knows already
and always must know already. To use the fashionable word today, arete requires self-understanding,^ and Socrates proves to
8.
wrote
9.
ergon

I attempted to make a convincing case for this point at the time that I
Platos dialektische Ethik.
Here again Gadamer has in mind that unity of logos (word, reason) and
(deed) that figures so prominendy in all his interpretations of Plato: any

43 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

his partners that this is what they lack. Plato gives self-understanding a more general meaning: wherever the concern is
knowledge that cannot be acquired by any learning, but instead
only through examination of oneself and of the knowledge one
believes one has, we are dealing with dialectic. Only in dialoguewith oneself or with otherscan one get beyond the
mere prejudices of prevailing conventions. And only the person
who is really guided by such pre-knowledge of the good will be
able to hold to it unerringly, Plato has various metaphorical
ways of expressing this fact. For example, he says [that such a
person can hold to the good] because the good "resides" in him,
or because it is "related [syggenes) to him." Thus, unlike the unfortunate victims of Socrates' refutational arts who do not understand themselves, the true dialectician does not allow himself
to be artfully misled past the truth. But the converse holds too:
whenever someone who knows how to give justification nevertheless goes astray, as Socrates himself occasionally does in
Plato's dialogues, he finds his way back and then knows how to
articulate what he intended to say better than before. That occurs, for instance, after Protagoras's famous repudiation of a
conversio falsa which Socrates commits (350c ff.).
question about arete and the good is not merely a question for intellectual inquiry but also an existential question of how I am to understand and lead my
lifea question of what I am to do.
One must be careful, however, not to conflate this element of "existential"
concern in Gadamer's reading of Plato with the subjectivism of modern existentialism. Neither Plato nor Gadamer begin with the individual subject, but rather
with the language we speak and the traditions and customs that constitute our
world and from which we as individuals derive. Hence gnthi s'auton, Socrates'
"know thyself," really means, know the logoi, our ways of saying things, which
lend our world its significances and values. We come to know the logoi and what
we are to do not in solitudinous "existing towards death" but in dialogue with
others and with ourselves. Wir sind ein Gesprch: we are a discussion (cf.
Phaedrus 23Od on Socrates* passion for discussion), TRANSLATOR.

44

SOCRATIC KNOWING AND

NOT-KNOWING

There are places where Socrates with vacuous finesse seems to

abandon himself to making formal distinctions, and thereby to
disavow his true intentions and to deny that he knows better
for instance, in his first ers (love) speech in the Phaedrus. But an
inner voice, whether he calls it his daimonion or not, will hold
him back and compel him to retract what he has said in a second
speech. And this retraction is anything but a blind act of submission to a traditional religious rule.^ Socrates covered his head
when he gave the untrue panegyric (Phaedrus 257a), and only in
the retraction is he once again entirely himself. Only Socrates' retraction places the experience of ers within the grand and broad
setting of Plato's rendering of his master's charisma. Socrates uncovers the frue distinctions, as befits one who knows what ought
to be known and given justificationas befits, in other words,
the dialectician. What the latter is, is expressed mythically in his
second speech. There he not only properly identifies the essence
of ers as a god-given gift. Put another way, he not only makes
the necessary distinction between "good" and "bad" madness
a distinction that is essential if human beings are to have a true
understanding of themselves and what they experience in love.
In making this distinction, he shows himself to be a real dialectician, who, in the very process of giving justification, unveils at
the same time the nature of dialectic as such. The passion for the
beautiful is consumed here in the passion for the true. We will return to this point subsequently.
Thus, in Plato's dialectic the concern is still that Doric harmony of logos (word) and ergon (deed) that gives Socrates' refutational enterprise its particular ethos (character)." This har10. That is, that one should recant one's blasphemy, in this case, of the god
Eros (Phaedrus 242e). TRANSLATOR.
11. For a programmatic exposition of the relationship of these two, see my
Logos and Ergon in Plato's Lysis,"

45 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

monyor the lack of it in his partnersnot only provides the

background for Socrates' (often logically unsatisfactory) art of
argumentation in the elenchtic dialogues. Even in the so-called
dialectical dialogues of Plato's later period it plays a far greater
role than is generally acknowledged.^^ For example, one will only
be able to understand the Theaetetus fully once one has properly
evaluated the complete paradox in Theaetetus's giving a sensualistic answer to the question "What is knowledge?" For Theaetetus is a brilliant mathematician." Also, the task of differentiating
between the philosopher and the sophist, which the Sophist dialogue proposes, can only be carried out successfully by someone
who is really (onts) a philosopher and no mere artist at refutation. From the very start of the Sophist the search for the essence
of the sophist is guided by the vision of the true essence of the
philosopher and dialectician. The Stranger from Elea is no mere
artist at refutation, and hence it is quite in the order of things
that the essence of the philosopher and dialectician must also be
cleared up if one is to arrive at a clear grasp of the sophist.^^ In
the Phaedrus the task was to distinguish between good and bad
madness. Similarly the task here is to distinguish the true dialectician from the false one, the sophist. Even here the differentia-

12. Sprague, for example, shows an understanding of the logical fallacy at

Theaetetus 163 ff. insofar as she quite rightly acknowledges the positive qualities
that Plato assigns his Theaetetus figure. However, the pervasive dependence of
logos on ergon extends much further into the content of the dialogue than she
perceives. See R. Sprague, Plato's Use of Fallacy (London, 1962).
13. What Theaetetus is in deeda mathematicianis a far cry from the position he takes intellectually when he proposes that knowledge is sense perception (aisthesis). TRANSLATOR.
14. Some observations are to be found in my "Plato und Heidegger" (in Der
Idealismus und seine gegenwart. Festschrift fr Werner Marx (Hamburg, 1976,
pp. 166-75) concerning the limited success with which the "sophist" can be
grasped and the limitations under which, in Plato's view, any comprehension of
him is possible at all.

46

SOCRATIC KNOWING AND

NOT-KNOWING

tion by dihairesis is still aimed at differentiating the good from

the bad. That this is so becomes fully explicit once again in the
PhilebuSy in which the dihairesis is wholly concerned with the
good in human Hfe.^^ It makes good sense to call the dialectical
virtue "phronesis."
Let us return from this perspective on things, to which we
were led by Plato's use of the word phronesis, to the actual
thema probandum^ namely, that in Plato's eyes knowledge in
arete has something special about it that sets it apart from the
technai. On occasion the difference emerges explicitly, for example, in the Meno at 74b, where it is declared to be particularly
difficult ''mian areten labein kata pantn^' (to take one virtue
above all), evidently because conventional morality only knows
many different conventional aretai (see 71e ff.). But above all, the
difference between arete and techne is demonstrated by the role
that the problem of teaching arete plays in Plato. The Protagoras
is of decisive significance in this regard. That arete can be taught
ought to follow as cogently from the character of knowing in
arete as the fact that techne can be taught indisputably follows
from knowing in techne. But as a matter of fact, it does not. That
is common knowledge, for which Plato can find support as early
as in Theognis. In a certain sense the problem at the core of all
education is, after all, the fact that unlike the technai, arete is not
teachable. Traditional ethical and moral customs are based not
so much on teaching and learning as on taking someone as an
example and emulating that example. Xenophon says explicitly
that Socrates never promised to be a teacher of arete, rather that
by the example he settoi phaneros einai toioutos (himself appearing to be of such character)he inspired followers
15. J. StenzePs "Socrates," in the Realenzyklopdie der Klassischen Altertumswissenschaften, vol. 3, col. 856 ff., has already brought this element of dialegein
(sorting out, selecting) to our attention.

47 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

mimoumenoi (emulators) {Memorabilia Alpha 2). Ethical norms

are passed on with an accepted self-evidentness that social traditions lend them, and Protagoras comports himself quite in keeping with this traditional thinking when he names "all Atheneans" as teachers of political virtue (Protagoras 327e). Of
course, he leaves unsaid that just these Atheneans can also be
teachers of injustice, and Socrates' and Plato's lives are defined
in their entirety by precisely this fact that they can. Plato's, Seventh Letter (325f.) and his interpretation of Socrates' life in the
Apology fully concur on this: only the demand for justification
lays bare the empty presumption in the usual moral and political
self-understanding. ^^
But Plato goes even further by contrasting Socrates with the
great sophists in his dramas. In this way he shows that precisely
by claiming falsely that it teaches by means of a new knowledge,
the new ideal of paideia secures this conventional morality. For
the new ideal of paideia was precisely conscious learning. And
just that is what the great sophists of the time, of whom Protagoras is representative, claim they can bring about. They promise
that by instructing the student in the new art of speech making
and argumentation, they will educate him to be a just citizen of
the statein the ancient sense of taking active part in public affairs. As the Protagoras shows, sophistic paideia does not claim
at all that, by teaching its art, it is putting new norms in the place
of those traditional moral norms whose decisive force is perpetuated through education using paragons of virtue. By its art it
claims to impart civic virtuethe same civic virtue as always
16. Compare "Plato's Educational State" and "Plato and the Poets." The
problem is that the communal solidarity upon which traditional morality is
based has been destroyed. As a consequence, what one uncritically assumes to be
right turns out to be nothing more than a rationalization of self-interest (see n. 2
above), TRANSUVTOR.

48

SOCRATIC KNOWING AND

NOT-KNOWING

was and always will bein a new way. Knowledge about arete
is always sustained and transmitted by all. But the sophist does
what everybody does with "art," and that is what he means
when he praises his new "art" as the perfection of education. We
see this claim equally in Plato's characterizations of both Protagoras and Gorgias. And a lot of what we know of both these men
from other sourcesfor example, that they were held in high esteem by societyaccords with Plato's portrayal of them.
But now the Socratic question overturns the whole sophist
claim. Plato shows in the dialogues named for these two sophiststo which group the Republic^ book 1, the so-called "Thrasymachus," also belongswhat ill fate this paideia's new claim
to be knowledge actually bodes. Theirs is a technical mentality,
which passes itself off as arete without really being arete at all.
This pretense is exposed by displaying its radical consequences
in radical immoralists of the cut of a Thrasymachus or a Callicles. In the Protagoras the unmasking is not as explicit, but certainly the way is cleared for it. For there is no doubt that there
Protagoras ought to be forced into a radical hedonism as the true
consequence of his concept of knowledge. Precisely by decking
himself out in another garb and thereby evading this radical consequence, he makes clear negatively that it is a conclusion he
would have to draw. In truth, his pragmatic knowledge and his
art are incapable of establishing any other norms or defending
them. Then, in book 6 of the Republic (493 ff.), this point is
stated explicitly: the so-called sophists are actually the hirelings
of public opinion. Every one of them teaches nothing other than
the opinions that people form when they get together. And that
is what they call wisdom. The new paideia appears to accommodate itself to the traditional system of norms. But this appearance, we see, is false and only obscures the fact that the
traditional concepts of arete cannot be given sufficient justi-

49 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

fication and conceals that this world of norms has itself become
equivocal.
Thus the purpose of Socrates' pointing to the paradox in the
knowledge of virtue is precisely to show that the traditional
world of norms has come to need justification, but that it is incapable of being justified: what this new paideia is really claiming
is that it is a techne of success. When it purports to impart a
knowledge founded in [what for it is] the self-evident continuing
validity of the world of norms, its apparent knowledge proves
false, and its sureness of being able to justify itself illusory. As we
saw, its pretense is exposed in the Protagoras in the caricature of
an art of living that would amount to technical knowledge
{metretike techne) of how to get the greatest amount of pleasure
possible. But most of all, its pretense is exposed in the paradox
that all arete, despite its claim to be knowledge, is unteachable.
One now sees how skillfully Plato's Protagoras is composed,
and what a powerful statement its dramatic setting makes just by
itself. Plato's idea of confronting the famous sophists of the time
(Protagoras, Gorgias, Hippias, Thrasymachus) with his Socrates
appears to be an invention all his own, and it serves the purpose
of defending Socrates against the fateful equation of him with
the sophists, an equation that was the reason for his tragic condemnation. Xenophon reports only one other comparable pairing with a sophistwith Hippias (Memorabilia^ Delta 4)but
it is set up quite differently. The sophist knows Socrates well. He
complains that Socrates is forever saying the same things, and
that one can never get a positive answer out of him (the theme of
the Clitopho), Therefore Hippias insists that Scorates himself finally say something positive about to dikaion (what is just). But
what Xenophon has Socrates say in response could hardly be
based on authentic remembrances. Xenophon's apologetic motif
is all too trivial and transparent. Socrates may have argued be-

50

SOCRATIC KNOWING AND

NOT-KNOWING

fore the court that he himself never committed an injustice, and

Plato also has him say that. However, w^hat Xenophon has
Socrates say about to dikaion itself, namely, that it is to nomimon (what is lawful), and about the agraphoi nomoi (unwritten
laws) and in justification of the prohibition of incest and about
the godsall that is colorless. Moreover, Xenophon has the
sophist interlocutor agree in the same colorless way in which all
the partners in the Memorabilia do. The other conversation in
Xenophon between Socrates and a sophist (Antiphon) is totally
superficial (Alpha 6). In any case, these naively apologetic confrontations bear no similarity to Plato's great sophist dialogues.
In the Protagoras, in contrast, Plato displays the falsity of sophistic pseudo-knowledge and of the claim to teach it by confronting this "knowledge" and claim with Socrates' claim to
know. And the conclusion with which the dialogue ends says
a great deal: Socrates forces the sophist to agree that arete is
knowledge, but for his part he disputes that it can be taught. But
if Socrates really took arete to be knowledge similar in character
to the knowledge of techne, he would have to maintain that it
can be taught. What sort of knowledge, then, is this knowledge
that he has in mind that is evidently unteachable? The reader is
meant to put this question to himself. In any event, he has to see
clearly that the knowledge and justification for it that Socrates
seeks has nothing to do with sophistic techne thinking. That it
does not is obvious from the start. The doubt about the teachability of arete dominates the discussion from beginning to end.
Even in the opening scene it lurks in the background. Thus the
logical point of the comedy-like ending that Plato invents for the
Protagoras dialogue is most of all this: knowledge in arete can
have the character of neither knowledge in techne nor the
knowledge of this new paideia, which boasts of being techne.
The whole series of the Socratic discussions whose conclusions
are negative could be advanced to demonstrate the inadequacy

51 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

of the techne concept for attaining a clear concept of knowledge

of the good and of the nature of arete. The Meno is particularly
crucial in this regard, for the exposition here is advanced one
step further. At first the dialogue deals with the same problem as
the Protagoras^ namely, teachability. And in essence it reiterates
the paradox in which the claim that arete can be given justification gets caught: if it can be given justification, then it can also
be taught. Once again, the claim that arete is knowledge founders on the facts of our moral and political experience. The sons
of great men, who have had the best education and upbringing
thinkable, are often grave disappointments. Hence something
other than knowledge must play the decisive role here, something that Plato calls theia moiradivine dispensation.
And now we have arrived at a truly crucial test for the traditional interpretation of Plato. Socrates' own demand that justification be given, which he pursues relentlessly, seems weakened
when Plato substitutes "divine dispensation"the latter appears to provide only half an answer to the problem. Subtle interpreters of Plato see in this divine dispensation an indication
that Socrates himself is the only true teacher of arete. It is certainly correct to say that the end of the failed discussion with
Meno points to Socrates as the actual and only teacher. But one
has already forfeited the truth of this insight if, at the same time,
one misses the general point implied here. Plato's concern is not
to sanctify this charismatic Socrates, even if in Plato's eyes he
certainly was charismatic. Rather, he is much more concerned
with overcoming the false conception of learning and knowing
that prevails in the young Meno, as it does in his teacher Gorgias. It is to this end that he adverts to divine dispensation.^^
The whole discussion with Meno is devoted to this task. One
17. The allusion to divine dispensation at Republic 492e is also aimed polemically against sophistic paideia.

52

SOCRATIC KNOWING AND

NOT-KNOWING

need only ask oneself Plato's question: Who was this Socrates
and what was his knowledge? After all, he had declared that precisely knowing that one does not know is the real human wisdom. His teaching could never be different from what it always
is, namely, demonstrating that his partner does not know, and
by doing that, making it urgent that one know and give justification. For someone who has come to seek and question on his
own, the pretentious assumptions that Meno, for example, has
learned from the likes of Gorgias and advances himself are
empty. And emptier still is a sophistry that would argue someone
out of seeking and questioning altogethersuch a sophistry,
that is, as Meno produces with blind acuity. The significance of
the Meno is'that here Plato expressly thematizes the aporia (perplexity) in which the other Socratic dialogues tend to end.
Like these other dialogues, the Meno begins with a series of
failed attempts to define arete that disclose sometimes more,
sometimes less, clearly that the sole reality behind moral conventions is the pursuit of power. The last answer that Meno ventures virtually says as much. He appropriates the poet's Hne:
charein te kaloisi kai dynasthai (to delight in the beautiful and
have power) in such a way that arete would mean nothing else
but having the power to acquire the beautiful thing that one desires (77b). But Plato takes a new step here. He shows that reaching the aporia in which Meno's attempts to determine the nature
of arete end is the precondition for raising the question of arete
in the first place. But here, raising the question means questioning oneself. The knowledge in question can only be called forth.
All cognition is re-cognition. And in this sense it is remembrance
of something familiar and known.
The conversation with Meno makes this fact clear e contrario.
Meno appears on the scene as one who wants to acquire the new
wisdom as cheaply as possibly, and he bolts when he is about to

53 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

be forced to place himself in question. Thus he is just the right

foil for allowing us to see what knowing and recognizing actually are. The doctrine of anamnesis (recollection) brings out the
true sense of the Socratic question. As one who himself only "reminds," Socrates is a teacher. And in portraying Socrates' deeds,
Plato at the same time reminds us that knowledge is recollection,
knowing again.
The idea of recollection is introduced here as a myth, which is
to say, apparently not as an argument per se but as a sort of religious truth. But one has to view the myth of anamnesis in the
light of the question we are raising. Is it a myth at all? Certainly
this doctrine is introduced in the Meno like a mythwith references to verses from Pindar and the Pythagorean doctrine of the
transmigration of souls. But the authorities upon whom Socrates
relies already sound odd. For here we find priests and priestesses
who are able to give justification! In the context of Greek religion there is something absurd about that. For Greek religion
was not a religion of scripture and orthodoxy but of individual
awe and piety and of regular public honoring of the divine.
Moreover, the thesis that seeking and learning is recollection is
then demonstrated quite soberly with no reference at all to religion. The famous lesson that Socrates gives Meno's slave is far
removed from a proof of the religious doctrine of the preexistence of souls. Of course, in every step of this lesson Socrates
carefully adheres to the premise that the slave is not taught anything but instead grasps each of the steps himself, the negative
ones as well as the positive. In other words, the slave displays a
kind of knowledge without ever having "learned" mathematics.
But given the lengths to which Socrates goes it is all the more
striking that the conclusion drawn at the end is not viewed as
validly provedthe conclusion, namely, that there was a time
before human beings existed at which the soul already -knew

54

SOCRATIC KNOWING AND

NOT-KNOWING

things, and that the soul is consequently immortal. On the contrary, any such claim that it has been proved is explicitly retracted (86b). The only thing accepted is the practical certainty
that we are better off holding firmly to the belief that one can indeed seek the truth, and that one should not allow oneself to be
misled in this search by sophistic objections. And it is accepted
logi kai ergbi (in word and deed) (86c). Hence the mythical horizons within which Plato places this certaintyand not without ironic ceremoniousnessserve essentially only to display
and explicate the capacity of the human mind to place things in
question.
The Phaedo demonstrates fully and convincingly that we are
not dealing' with a religious truth here. There the anamnesis
theme is taken up anew, and once again it is explicated quite
unmythologically. The way in which the doctrine of preexistence
is "proved" hereby the "prior knowledge" that underlies all
knowledgeeven has a comical side to it. To be sure, it is made
clear here that as religious heritage what this preexistence proof
demonstrates with its pseudo-stringency, is worthy of solemn respect. But this comical aspect of the argument makes clear that
what is 'proved' hardly lends itself to a rational legitimation in a
style such as this. In particular, the sharpening of the argument
after Simmias's objection that knowledge could, after all, be
given to one at birth makes the discrepancy between the mythical claim and the logical concepts with which the argument proceeds especially palpable. Obviously it is with this discrepancy in
mind that Plato has Socrates now venture the following argument (Phaedo 76d): since knowledge cannot be attained after
birth, it must derive from a "previous" Ufeunless, that is, it is
acquired at the moment of birth. But after all, as the initial ignorance of the newborn shows, it is not present at birth. So at one
and the same time, it would be acquired and losta pretty piece

55 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

of nonsense, it would appear. And with that Simmias's objection

seems to have been disposed of. Or, in the final analysis, is this a
hint that we should examine in earnest the concomitance of
knowing and not-knowing? For if we do, we might perceive in
this concomitance an intrinsic interweaving of cognition and recognition that splits apart into a mythical prior life and a subsequent recollection only in mythological thinking. ^^ Whatever the
case may be, we must abstract from Plato's mythical mode of
presentation if we want to understand what he is getting at. And
that requirement holds in regard to the Meno as well as the
Phaedo. Let us, then, attempt to conceptualize some of the
things he has in mind.
After he has put the false solutions behind him, the slave in the
Meno recognizes that a square constructed on the diagonal has
the double area he seeks. That he does so implies that he already
knows what "double" meanshe must know Greek (82b)
and that he keeps his attention focused on doubling and what is
double. Accordingly, we have a real seeking here. The slave has
enough of an idea of what is sought to recognize that his first at18. One of the essential tenets of Gadamer's thought underlies this interpretation, namely, that human beings never have insights that are fully clear and distinct, but only partial insights within persistent obscurity. Thus any aletheia, or
truth, that they know is embedded in lethe, or forgetfulness. This concomitance
of knowing and not-knowing (which Heidegger would call. Gleichursprnglichkeit) has far-reaching consequences. For one thing, it renders the project of Cartesian methodology incapable of execution: since there is no certain starting
point, no certain conclusions can be drawn. For another, it makes systematic
unity and conclusiveness unachievable: we always find ourselves in media res
under way in the middle of things whose beginning and end are beyond the horizons of our knowing. Gadamer finds this principle of human finitude throughout
Plato and, in particular, in his doctrine of ideal numbers, the one (unit) and the
indeterminate two. For Plato, Gadamer maintains, any unitary thing we know is
given to us within the indeterminacy that surrounds it. Consequently, our inquiry
will remain inconclusive (unabschliessbar). See "Plato's Unwritten Dialectic."
TRANSLATOR.

56

SOCRATIC KNOWING AND

NOT-KNOWING

tempts to solve the problem by doubling are wrong and to recognize the true solution when it is presented to him. As we know,
he does not find it on his own. Socrates has to show it to him
(85e). That fact is of no concern, however. The point is that he
himself recognizes it as the solution he seeks.
It should be noted that we are dealing with a mathematical insight here, that is, not with a result of empirical generalization.
The slave already knows enough of mathematics to accept without question that the problem put to him is eidetic-universal and
to grasp it as such without giving it a second thought. The entire
path along which the slave is guided to his eidetic insight proceeds through eidetic terrain. Even his first mistaken attempts at
solving the problem are meant to be eidetic. They are wrong only
mathematically. For him, unlike his master, the insight that his
proposals are false is not anything that might cripple him. Instead, it actually makes the right insight possiblean insight
that would require only sufficient repetition of the exercise to be
stabilized in him as genuine mathematical knowledge (85c ff.).
Here, however, this mathematical example stands for everything that Plato would call real knowledge or insight. One always has aletheis doxai (true beliefs) in oneself concerning what
ones does not know {Meno 85c). Indeed, just this fact emerged in
the mathematical lesson: the refutation of false assumptions is
needed in order for these to be recognized as false, but that entails that one always already^' has some idea of what the true as19. Immer schon. This common tum of speech has special importance in
Heidegger's work and also in Gadamer's. It underscores the fact that I actually
never was, and never will be, in the state of unprejudiced objectivity which the
Enlightenment considers prerequisite for valid knowing. Put another way, I am
never in an "original position" (Rawls); rather, I can understand what I encounter within my world only because of the pre-knowledge that I "always already"
have. Implied here is Heidegger's and Gadamer's theory of the circularity of understanding. Gadamer extends Heidegger's line of thought in arguing for the in-

57 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

sumption is. Thus, what is displayed here is the nature of seeking

and learning {zeteitiy manthanein) (8Id). Seeking and learning
presuppose that one knows what one does not know, and to
learn that, one must be refuted. Knowing what one does not
know is not simply ignorance. It always implies a prior knowledge which guides all one's seeking and questioning. Cognition
is always re-cognition.
Plainly that holds especially in regard to arete. And though the
Meno too does not say so explicitly either, Meno's renewed evasion of the issue at 86c makes clear even so that the question of
what arete is would necessarily lead us to knowledge of the good
(see Meno 87 bd). Knowledge of the good is always with us in
our practical life. Whenever we choose one thing in preference to
another, we believe ourselves capable of justifying our choice,
and hence knowledge of the good is always already involved.
Socrates' recapitulation of the doctrine of anamnesis in the
Phaedo is no less instructive. In a masterful analysis he unfolds
the argument that shows why all knowing is recollection, and
leads us through it step by step. He begins with clear instances of
our being reminded of something. A lyre reminds us of a beloved
friend. A friend reminds us of his friend. Even the picture of a
friend also reminds us of the other friend. And, yes, the picture
of a friend also reminds us of the friend himself. We are being led
along very artfully here, and the final step is astonishing. In this
last case we would not say that we are reminded of the friend,
but instead that we recognize him in the picture. Exactly because
recognition emerges here as a kind of recollection, or being redispensability of tradition and authority for any understanding of our human
world. His point is that the "condition of the possibility" (Kant) of my understanding my world is not so much consciousnesses interpretive acts or performances as it is consciousness of and recollection of, what is always already pregiven in the traditional authority of language and customs (Sitten), TRANSLATOR.

58

SOCRATIC KNOWING AND

NOT-KNOWING

minded of something, Socrates succeeds in establishing what he

set out to demonstrate. In this way recognition is set apart from
all learning.
It could be important that this example is not a genuine instance of being reminded. After all, seeing the lyre of his friend
does not remind the lover of someone he had forgotten! The
friend, in fact, is so close and present that the lover is made to
think of him by all sorts of things. It is as if he sees all things
and hence the lyrein the light of his passion. That is significant. The assimilation of this being reminded to being reminded
of something forgotten is completely contrived. This fact is made
clear negatively by the insertion of malista mentoi (better yet) at
73e. In truth, what we have here is far more a matter of mneme
(remembering) than of anamnesis (recollecting, being reminded
of something).^^ And properly speaking, knowing, or cognition,
too, is not being reminded of something forgotten. Rather, it is a
new revelation about something already known. When I recognize something as something, I view something I know in the
light of what I take it to be. I interpret it in regard to something
which, for its part, is also known to me and present to mind, "^es
physes hapases syggenos ouses" (since the whole of nature of
akin) {Meno 81d).
We can see that this phenomenon of prior understanding
applies above all to our self-understanding in arete and to the
question about the good. After all Meno wanted to evade just
this supposition [that we already know the good] and by trying
to evade it, he induced Socrates to advance the theory of anamnesis. But there can be no doubt that even in the Meno Plato intends anamnesis to have a much broader sense which should
20. With regard to the relationships here, compare the extensive excursus in J.
Klein, A Commentary on Plato's Meno (Chapel Hill, 1965), pp. 108-72. Klein
is right in bringing in the Philehus,

59 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

hold for every sort of real knowing. The dialectical art of making
distinctions allows us to distinguish the good from the bad or, as
we might say with moral reserve, to distinguish the right thing to
do from everything which would not be right. But in its full extent this art has to be applicable to knowing anything worth
knowing. In the end, the structure of anamnesis proves to be coextensive with all possible questioning. Questioning is seeking,
and as such it is governed by what is sought. One can only seek
when one knows what one is looking for. Only then, only with
what is known in view, can one exclude the irrelevant, narrow
the inquiry down, and recognize anything. That is what the
Meno teaches us.
Another illustration, albeit negative, of what Plato has in mind
is the failure of Socrates' sophist interlocutors when they want to
do the questioning themselves. The questioner seems to them to
play a superior role, to which, accordingly, one should aspire.
But questioning is not a technique of role playing. The questioner is always one who simultaneously questions himself. The
question is posed for him just as it is for the other person. What
we have here is the dialectic of dialogue, and its logical structure
is simultaneous synopsis (seeing things as together one) and
dihairesis (division, or differentiation). Both recognition of what
one knows oneself to bethat is, recognition of how one understands oneselfand recognition of everything one knows are always at one and the same time synoran eis hen eidos (seeing together as one form) and kata gene dihairesthai (separating
according to species), which is to say, differentiation. We always
find ourselves in dialectical tension with the prejudices which
take us in and parade themselves as knowledge but which really
mistake the particularity and partiality of a given view for the
whole truth. That holds for both the person asked and the person asking. Plato's most abstract way of expressing this phenom-

60

SOCRATIC KNOWING AND

NOT-KNOWING

enon is to say that we confuse ta metechonta (things which take

part [in the truth] with to auto ([truth] itself) {Republic 476d).
The beginnings of this idea are to be found in the Meno, although it is only later, above all in the Republic^ that we are explicitly compelled to draw this conclusion.
Socrates' statement that arete is knowledge thus proves to be a
provocation. For unlike knowledge, arete is not teachable, although it cannot be denied that it can and needs to be justified.
Now it is obvious that Aristotle openly accepts this positive side
of the Socratic paradox of knowledge in virtue. Arete is not
logos (reasoning), says Aristotle, but exists meta logou (along
with reasoning). I will attempt to make credible that in truth this
way of putting things is in complete agreement with what Plato
and Socrates intended to say and is implicit in the total "intellectualization" of arete articulated in Plato's dialogues. The fact
that Aristotlein half agreeing with "Socrates" (EN 1144b
ff.)takes the paradoxical equation of virtue and knowledge in
Plato and Socrates literally and "corrects" their mistakes, can, in
my view, not be advanced as a counterargument. Aristotle has a
way of taking statements not as they were intended, but literally,
and then demonstrating their one-sidedness. Aristotle's use of dialectic consists in balancing off the one-sidedness of one person's
opinion against the one-sidedness of someone else's. And while
often doing violence in the process, he thereby succeeds in better
articulating his own position and also in conceptualizing previously unquestioned presuppositions. What Aristotle gains from
his critique here is the concept of ethos (habituation), which, Hke
every conceptual advance, becomes the source of new questions.
These we must raise later under the heading of practical
philosophy.
Here, a general hermeneutical insight must be our guide, the
clear outlines of which emerge precisely in discussions like the

61 S O C R A T I C K N O W I N G A N D

NOT-KNOWING

Socratic ones, but which ultimately applies to any "discussion of

the soul with itself" that we call thinking. Statements like "Virtue is knowledge" do not come out of the blue, rather, their
sense is determined in part by what they answer or respond to.
Socrates is responding to the confusion in the moral tradition, a
tradition for which sophism promises the new foundation of a
bogus techne, a new "knowledge." Plato and Aristotle seem to
have been in complete agreement in their common efforts to fend
off this sophistic technical conception of knowledge. In linking
logos to ethos, Aristotle only further formulates what Plato had
in mind.
Is not the prevailing view absurd that Plato underestimated
the role of habituation and character molding as these are implied in Aristotle's concept of ethos? After all, Plato fabricates an
entire ideal city, the tone of which is set, for all intents and purposes, by a super-ethos, a formidable habituation in virtue. To
be sure, this ethos component comes to light only in the quasi-mythical exposition of this new super-ethos and not in conceptual form. The rule of the philosopher-king remains an
enormous provocation insofar as it makes it seem that pure
knowledge, theria, is supposed to provide the answer to the human question about the just life, the question of the good. But
exactly what, one must ask, is the point of Plato's Utopia? Why
did he write it? It appears to me that the Utopia of the Republic^
to the extent that it answers the question about the good, aims at
the mature conceptual clarification which the Nicomachean Ethics offers in its considered analysis of the relationship between
ethos and logos. In any case, Aristotle's ethics itself presupposes
the Socratic-Platonic turn to the logos and rests upon the foundation of it.
Therefore it seems justified to me to go ahead with my attempt
to get behind the web of polemical-critical relationships that

62

SOCRATIC KNOWING AND

NOT-KNOWING

bind Plato and Aristorie together and, for once, to read the
moral-philosophic paradoxes of Plato's writings with an eye to
what he and Aristotle have in conimon. The next step in this direction will be an exploration of the relationships between the
Protagoras and book 4 of the Republic.

111

THE

POLIS

AND

K N O W L E D G E OF
THE G O O D

The unity and multiplicity of the so-called cardinal virtues became a problem in the Protagoras^ and Socrates demonstrated
therealbeit often with sophistic meansthat these cardinal
virtues are reducible to being knowledgeable. As we emphasized
above, the sophistry of these means presupposes in principle that
Plato is secure in what he is aiming at: the sophistry is there to
serve his purposes. Being able to give justification, being responsible for what one does, is essential to the ethical disposition, and
that implies that the whole of one's ethical consciousness and
ethical being are at stake here. Consequently, giving justification
cannot be limited to any single moral phenomenon, tendency to
behave, or special skill. Obviously Socrates has this fact in mind
when he plays his game of refutational dialectic and juxtaposes
particular virtues with the good. In essence, however, his doing
so amounts to the same thing as Aristotle's rejection in the
Nicomachean Ethics of the separation of the aretai from each
other and his assignment of the very same unifying function to
the logos (rational principle) of phronesis (EN 1144b). The
problem figuring in many of Socrates' numerous elenchoi (refutations) becomes especially easy to get hold of in the Protagoras,
insofar as the question concerning the unity of the many virtues
63

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 64

is posed there explicitly from the beginning. In particular, the

superficiality of conventional moral judgments becomes clear
when Protagoras argues for the privileged status of andreia
(courage) in distinction to the indissoluble unity of the rest of the
virtues. Andreia appears to be a special trait required of soldiers.
As such, in Aristotle's eyes, it would only be a physike arete (natural excellence) (1144b and elsewhere). That is exactly what
Plato has in mind. In the Protagoras Socrates shows that andreia
too is knowledge. Plato pursues this line of thought from the
Protagoras and Laches through the Republic and into his late
work the Laws, in which he gives special attention to andreia
in criticism, I think, of just that unyielding partiality to Sparta
of which ht might well have been accused by readers passing superficial judgment on the Republic.
The content of the corresponding book in the Republic (4)
fully concurs with the Socratic aims of the Protagoras. In a certain sense even the theme that provides the framework for the
Republic, the ideal polis, already gives a preliminary answer to
the question raised in the Protagoras about the unity of the virtues. That the politeia (state) now provides the framework tells
us from the start that the issue is unity in multiplicity: the multiplicity of the political classes is ordered to provide unity and concord, as is the multiplicity of the parts of the soul. In book 4 the
order of the state is first discerned in *the structure of* its
classes, and then the order of the virtues in the soul is discerned
in that order of the state. The virtues here are the traditional cardinal virtues. I will call them the so-called Platonic virtues because in truth they are not Platonic but traditional. This fact is
now generally recognized. One need only closely examine the
definitions in book 4 to see that all four of the traditional virtues
are artfully stylized in order to emphasize the element of knowl-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

65

edge that they contain. In other words, they are reinterpreted

Socratically.
Here, couragewhich in the Protagoras still provided the
point of stiffest resistance to conventional moralityis not only
understood as it was there, namely, as knowledge of what is dangerous and what is not. In the Republic Plato outdoes this paradoxical and provocative formulation with an even greater paradox. What he really has in mind is thereby made clear: courage is
demonstrated not so much in response to fear of an obvious
threat as in response to the hidden danger in what is charmingly
pleasant [hedu). And in the political realm the latter danger is the
danger of flattery, which is more to be feared than the overt
threat of an enemy. The universal meaning of courage, to which
Plato is pointing, becomes plain if one places courage in a more
general and comprehensive frame of reference, one that includes
civil courage. Above all, courage is needed in response to the
danger of conformismcourage, that is, which does not allow
itself to be misled but "knows."
Then, in book 4 of the Republic^ one can see how in the end all
four cardinal virtues very nearly fuse in the knowledge that they
all are. This fusion is obviously what Plato wants to display. He
wants to show that the old norms, the traditional aretai, having
been established on a new basis, have become something different. For now justification of what is good in them is required,
and merely choosing a paragon and imitating it no longer suffices. All the aretai are phronesis. The question that Socrates
raises in the Protagoras concerning the unity of the aretai
whether they are more like parts of a clump of gold or more like
parts of a facegives us a clear profile of Plato's new understanding of arete. Both comparisons are inappropriate, for both
tacitly presuppose a conventional understanding of arete ori-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 66

ented toward external appearances. Hence both of the alternatives are misleading. Arete is not to be thought of at all as a unity
or multiplicity of ways of behaving primarily presented to an observer. Rather, it is self-knowledge, phronesis. In the end our behavior attains its unity when our actions are undertaken in regard to the good.^
To be sure, Plato's Republic confronts us with a difficulty: it
introduces the question of knowledge of the good only in a second stage of the argument, as if it were an afterthought. Once
the aretai have all been shown to have the character of knowledge, the analogy between the harmony of the classes in the polis
and the harmony of the soulits "health"would seem to suffice as an answer to the question about the definition of justice.
With the conclusion drawn in book 4 the goal has been reached.
As is common knowledge, it is only after this apparent conclusion that the question about the megiston mathema (the greatest
insight), the idea of the good, arises, leading us further along
winding paths. It is striking that this question about the good
does not follow, as one might have expected, from an attempt to
1. The thought here can be traced to Kierkegaard too (cf. WM 91). In rising
above the aesthetic "stage on life's way" to the ethical stage, I move beyond a life
that consists in discontinuous "great moments" of aesthetic exhilaration to a life
that maintains continuity in temporal transienceto a life, in other words, in
which unity and integrity are established within what would otherwise disintegrate in the flow of time. A question worth pursuing would be the extent to
which Heidegger's reception of Kierkegaard in SZ opened up new possibilities
for understanding Plato, say in Gadamer's PDE, It is striking that in ethical matters at least, Kierkegaard and Plato emerge in Gadamer's writings as astonishingly close. That by no means says that Plato is an "existentialist" (see ch. 2, n.
9). Of interest to Gadamer is the transition from the aesthetic, to the ethical stage
on life's way, not the transition from the ethical to the religious stage, in which
the existential decision made in anxiety supersedes dutiful and consistent adherence to what ought to be done (Plato: to deon; Gadamer: das Tunliche).
TRANSLATOR.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

67

establish what the unity is in the multipUdty of the aretai. Nor is

the question about the good based on the foundation of that
unity when this question is raised in regard to the class of the
guardians of the ideal state. And in any case one might find such
a foundation for the question of the good missing in what results
from book 4, which [merely] defines [the virtues] (435c ff.).
In book 6 (504a) Plato does remind us of the achievement of
book 4, the definition of the four virtuesin particular, the definition of dikaiosyne (justice). The insufficiency of the provisional sketch {hypographe) (504d) given in book 4 is articulated
nicely here by contrasting learning (manthanonti) and practicing
(gymnazomenoi). As early as 503e an omission is also explicidy
acknowledged, namely, the failure to say that the guardians
must have practice at the various sciences too, so that their "nature" might also be capable of holding fast to the knowledge of
what is most important of allthe good. But this line of thought
is not really derived from the problems found in arete. On the
contrary, knowledge of the good is simply treated here as something supreme that, as no one would question, is recognized to
be indispensable: the crowning accomplishment in the selection
and education of the guardians of the ideal state must be knowledge of the good. Thus the connection between founding the
unity of arete in knowledge of the good and the propaedeutic
function of the sciences in attaining this knowledge remains
more or less obscure.
The education of the guardians, who are to be guided through
the various disciplines of the mathematical sciences to dialecticwhich is to say, the art of distinguishingis education by
and to theria (theoretical thinking). This emphasis on theory
would appear to lead us far away from the "element of knowledge" contained in every arete. Indeed, the theoretical concerns
in which the guardians are to be educated end up in virtual con-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 68

flict with the task of poHtical leadership for which the guardians
are chosen {Republic 519d ff.). Left to themselves, those who
have been freed from the cave of murky sense experience and
practical routine, those who have been set free for theria, it
could be objected, cannot possibly feel any impetus to return to
the cave of politics, in which all knowledge is inexact and where
things always go wrong.
Of course Socrates sees no difficulty in defending his position
against this objection {Republic 519c): in his ideal state there is
no private sphere for the individual at all, and hence no question
of the happiness of the individual either. Accordingly, he rejects
the question of whether those dedicated to theria would not be
done an injustice if they were forced away from the higher fulfillment they find in theria and, for a time at least, were constrained to pursue the ugly business of politics. Not their own
happiness is at issue but the happiness of the whole. Indeed, one
even trusts that these ideal guardians of the ideal state will not
feel externally coerced but will submit understandingly to the
political task assigned to them (520d).
Nonetheless, one must ask oneself if this is really supposed to
be an answer to the question of how an existence devoted to
theria sees itself in this world of appearances, the world of social power structures? In Plato's state in the clouds, of course, all
problems that would otherwise confuse and distort political and
social life are solved ideally. Everyone does what he is supposed
to do, and consequently everything is ordered in such a fashion
that the whole prospers and flourishes. Just like all the other
classes, those who have knowledge and who have been brought
up to be leaders of the polis and have been educated in science
have careers in the ideal state defined for them in advance, which
they must follow. Knowing himself to be almighty, the poet who
invents this ideal state is not bothered by the fact that the guard-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

69

ians, who have knowledge and who find their fulfillment in

knowing and in dedication to theoria, cannot really be expected
to submit docilely to the onmiscient providence of the founder of
the state. For Plato each person belongs to the whole.^
Thus Plato denies that there is any conflict here. But what does
that mean? Does not this very denial prove to the thinking reader
that the conflict is real? Indeed it does. To my way of thinking,
there appears to be no doubt that Plato has this conflict in mind
in all of its sharpness and that he displays it negatively by the
very impossibility of his Utopian "solution"the conflict, that
is, between knowledge of the truth, to which the theoretical life
is dedicated, and actual political life. The choice of an apolitical
theoretical life seems fully justified to Plato. The grand excursus
in the Theaetetus says that it is in plain language, but the point is
proven here e contrario insofar as in Plato's ideal city those initiated into theoretical studies must acknowledge the political duty
they incur as a condition of their having been selected for such
initiation (520a ff.). That they must, fits with Plato's testimony
about himself in the Seventh Letter. We see there that Plato had
recognized that not only in his father city, but in all cities, the entire being of the state was thoroughly and incurably ruined
"unless a reform of quite incredible proportions were to occur"and that he had turned away from politics completely,
''epainesas ten alethinen philosophian" (giving himself over to
true philosophy) and placing the theoretical search for the truth
above everything else. The Seventh Letter even makes explicit
references to the ideal of rule by the philosopherevidently as
negative legitimation for retreating into the private realm.
Still the question is: Does autobiographical interpretation also
2. Keep in mind that the true sphrosyne is already introduced in book 4
{Republic
ff.) as arete common to all.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 70

suffice for understanding Plato's purpose in writing a political

Utopia and for an exhaustive interpretation of its meaning? This
work does indeed constitute a most extreme affront to, and rejection of, Athens. But should we say that it asserts in any way
that philosophy and politics are absolutely irreconcilable? Did
Plato wish to characterize the conflict between a theoretical and
a political existence as irresolvable?
One thing is clear in any event: this ideal state cannot be actualized. All the preconditions for itfrom the sharing of women
and children to the rule by philosophers to the exodus of all
those older than ten years from the city to be reorganizedall
these things demonstrate its impossibility. Glaucon hesitates visibly at 484b when he can find no other answer to the question of
who would be the right leaders of the poHs than ''the philosophers." And he remains a hesitant partner to the end: see 541a,
"eiper pote gignoito" (if it were to come about sometime), and
592ab, where this is even more pronounced. But what is the
whole point of Plato's invention? That we see its absurdity? Is it
meant to highlight the impossibility of the ideal? Are we supposed to read this political Utopia only negatively and be convinced by it only of the irreconcilability of theoretical and civic
life? If so, a great expenditure of intelligence and wit has been
wasted. For a blind man would see that such a state is impossible, and precisely its /mpossibility is underscored by the clumsy
and circuitous demonstration of its possibility. Does Plato seek
nothing more than to show that the conflict between theria and
poHtics is irresolvable?^
On the contrary. Surely one must read the whole book as one
grand dialectical myth. On occasion Plato himself virtually says
that dialectic is its principle. (See 497e: 'Hounantion e nun'' [the
3. Such is the opinion of Leo Strauss and Allan Bloom.

T H E P O L / 5 A N D K N O W L E D G E OF T H E G O O D

71

opposite of what is now].) Surely one must take all the institutions and structures in this model city as dialectical metaphors.
Of course, reading dialectically does not simply mean taking the
opposite of what is said, to be the true belief. Here, reading
dialectically means relating these Utopian demands in each instance to their opposite, in order to find, somewhere in between,
what is really meantthat is, in order to recognize what the circumstances are, and how they could be made better. Per se, the
institutions of this model city are not meant to embody ideas for
reform. Rather, they should make truly bad conditions and the
dangers for the continued existence of a city visible e contrario.
For example, the total elimination of the family is intended to
display the ruinous role of family politics, nepotism, and the idea
of dynastic power in the so-called democracy of Athens at that
time (and not only there).
Indubitably, one must read the argument for the rule of philosophers just as dialectically as everything else that is said about
this splendid state in the clouds. This argument is not meant to
specify a way to actualize the ideal city. But it is not intended either solely as a negative demonstration of its impossibility.
Rather it uncovers somethingand not only the obvious fact
that no polis would let itself be governed by such philosophers.
Is the paradox of the philosopher-king not also meant to give us
the positive insight that both aiming at the good and knowing reality pertain to the political actions of the true statesman as well
as to the true theoretical life? In support of this thesis one could
appeal to biographical factsI mean Plato's repeated attempts
with Dionysius II in Syracuse. Plato certainly had no intention of
proposing communal women and children to this tyrant, nor
later to his friend Dion, to whom he made very reasonable
recommendationsfor instance, that of a general amnesty. But
the same point could be deduced directly from what Plato says

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 72

using this Utopia. Making the ideal city possible in reality is of so

little concern to him that even the question of whether one
should make the philosophers rulers or educate the rulers to be
philosophers can be left open.
The sole issue is what the paradox of the philosopher-ruler
means, that is, what it uncovers about rulers and ruling generally. On the one side, we have the law inherent in all power, according to which power never aims at anything but the increase
of itself. On the other side, in opposition to this law, we have the
individual who gives himself fully to knowledge and for whom
power is of no interest. He knows something better, something
higher, to which he wishes to devote himself. When Plato opines
that such a person is more suited than anyone else to direct public affairs, he thereby exposes what seductiveness there is in having power: power wants only itself. The education of the guardians has the purpose of making them immune to this seduction.
The point "in between" to which Plato directs us here is a state
so arranged that the exercise of the power of government will be
carried out as a public office and not exploited as a chance to advance one's own interests. My contention is that there is more
significance here than the merely negative insight into the incompatibility of philosophy and politics. And it is right when one
sees the institution of modern professional officialdom and the
ideal of honesty in public officials [foreshadowed] in Plato's requirement [of immunity to the seductiveness of power]
That we are meant to read this projection of an ideal city
dialectically is confirmed most of all by the way in which Socrates paints in the transition from this ideal construct to historical reality with its cycle of constitutionsthe famous miscal4. Cf. Hegel on the principle of modern states: "Thus essentially Plato's requirement is met here" (Smtliche Werke, ed. Glockner [Stuttgart, 1927], vol.
14, p. 155).

THE P O L / 5 A N D K N O W L E D G E O F T H E G O O D

73

culation of the marriage number (546d). The mystifying, yet ingenious, thing about this invention of Plato's, it seems to me, Ues
in the fact that this comical shortcoming of a comical institution
symbolically displays why no system of human social order,
however wisely planned or thought out, can endure. What can
only be brought about by an artfully contrived institution will in
the end be done in by its own artificiality. This is the insight
Plato gives us here. The successful calculation of mating, which
insures the continuance of the ideally ordered polis, fails not because of malevolence or external forces, but because of its own
complexity. That is a true statement concerning something we
all know to be the reality of any humanly planned economy: be
the rationality of the planning ever so highly developed, in the
execution of it there is always the power of coincidence, and
above all, there is always human shortcoming. Because we are
human beings, not because we planned mistakenly, even an ideal
self-sustaining organization in full accord with the plan for it
will nevertheless go under in the rolling seas of historical life. To
say this is not at all to deny the task of reason to shape action
reasonably. Book 8 of the Republic undertakes to show that wisdom and reason are not only at home in the game of utopianism,
but that in our dealings with "real" historical life too, foresight
and insight are attainable within certain limits. The doctrine of
the cycle of state constitutions presented in book 9this brilliant example of intellectual penetration of the course of historyconfirms that human reason is not restricted to the realm
of Utopia and strict ideal order. On the contrary, it is fully capable of expanding into the historical world of vague regularities.
The disorder of human things is never complete chaos. Ultimately this disorder represents the periphery of a sensibly ordered universe that under any circumstances would have its periphery. That * there is order, albeit finite in human events'^ is

THE P O L / 5 A N D K N O W L E D G E O F T H E G O O D 74

brought out above all by the fact that in Plato's dialogical fiction
the Timaeus follows the Republic. To display the republic that is
really supposed to come into being, the entire grand project of
the demiurge's ordering of the world, which the Timaeus portrays, is necessary.^
The question that guides our inquiry here is how Plato unites
his concern with the Socratic question about arete and the good
with his scientific program. If we are to find an answer to it, the
requirement that we read dialectically must also be taken to
heart in regard to the assertions in the Republic concerning scientific knowing/ In respect to interpreting the allegory of the
cave, reading dialectically entails that we abandon all attempts
at an exact interpretation of this wonderful and many-layered
metaphor regarding its bearing on the theory of scientific knowing. Instead, we must focus on only one point, namely, what
function the allegory has within the course of the discussion.
Here there is no ambiguity: it is intended to dispel the illusion
that dedication to philosophy and the theoretical life is wholly irreconcilable with the demands of political practice in society and
the state. The theme is the blinding by the brightness that befalls
5. The demiurge, it will be noted, is always confronted with intractable
anange (necessity). Hence he can create order, but never complete order. "Demythologized," what we have here is the principle of unity or order within indeterminancy, the one and the indeterminate two (see ch. 1, n. 22). TRANSLATOR.
6. The German appears here in adjectival form: wissenschaftstheoretisch.
Wissenschaftstheorie (translated here as "theory of scientific knowing") is the
theory of science in a broad sense of science that would include not only the natural, but the social and human, sciences, which is to say, any body of knowledge.
One of its major concerns is methodology. Of particular importance in this work
will be the method appropriate to the "science" of practical philosophy, and how
it differs from method in episteme and techne respectively. Gadamer maintains
that Plato and Aristode are alike in arguing that the method of practical philosophy must be rigorously distinguished from any sophistic "technical" method.
TRANSLATOR.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

75

those accustomed to the dark, and conversely, the blinding of

those who leave the brightness and enter the dark. The allegory
is supposed to explain why those caught up in the practical life
consider the theoretical life worthless (515d and 515e). The
story is intended to enlighten us regarding this putative worthlessness of the theoretical human being in practice. One must not
only get used to the light; one must also get used to the dark.
Those who return from the light of the true day to the twilight of
the cave are also blinded at first by the contrast in brightness.
That they are does not imply that they are really blinded or are
incapable of getting oriented there. Plato tells us that their bUndness passes quickly (517a).
Evidently a common objection made to Plato was that philosophy makes one unfit for real life. Plato's desire to defend philosophy against this criticism completely dominates the Gorgias
(see 485a), and this motif is often heard elsewhere in Plato. It
was generally believed that philosophy had a place only in the
years of youth, and that it ought not to be continued for long:
one must seek entry into political life early enough and give up
philosophy. In the Republic Adeimantus appoints himself the
advocate of this general view (476d). In response, Plato's program for education of the guardians, which lasts decades, develops the idea of the absolute primacy of the theoretical life. Only
unwillingly, and only for a limited time, he says, would someone
consent to interrupting this life by taking a political office.
One must perceive the deliberate provocation in what Plato is
saying when he expressly proposes that after ten years have already been devoted to the study of the sciences, twice as much
time then be spent on the exercise of dialectic as on gymnastics.
At thirty years of age, the future guardians are to be schooled in
dialectic for five years. Then they should assume subordinate political offices for fifteen years. Only when they are fifty are they

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 76

considered mature enough for the task of political leadershipa

task at which they would take turns for a relatively short time.
They should continue to spend the larger portion of their time in
dedication to a life of study.
It is from this perspective that we must understand the ascent
from the cave to the true day and the vision of the true sun.
Herein lies its dialectical function within the context of the discussion. And in comparison, interest in the relevance of the allegory to a theory of scientific knowing remains entirely in the
background for the time being. Even the fundamental distinction
in Plato between doxa (belief) and episteme (science, knowledge) (476b ff.) is not introduced at first in the context of epistemological theory. Rather, within the framework of the Republic,
its function is to prepare us for the paradox of the philosopherking.
The way in which the allegory of the cave is introduced also
leaves the epistemological aspect of the subject matter in the
background for the time being. At first it seems that the knowledge to be preferred to all other knowledge is knowledge of the
good in the practical, political realm, and this alone is taken to
be the knowledge characteristic of those emancipated individuals called to be leadersknowledge, that is, tou dikaiou (of the
just) (517d) and kaln te kai dikain kai agathn peri (concerning things beautiful, just, and good) (520c).
To be sure, it becomes clear subsequently that the way traversed in getting used to the light that the allegory describesa
way which begins with shadows, passes on to reflections, images, and nocturnal stars, and then ends with the sunis a way
to the sciences, through the sciences, and beyond them. Nonetheless, nothing is said either here or later about the application
of such theory to human practice. The allegory deals exclusively
with the superiority of those who know the good over those who

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

77

remain caught in moral, political conventions. This fact is displayed particularly well when, in first introducing the paradox
of the philosopher-king, Plato places him alongside the eroticist,
who loves everything beautiful, and the spectacle-seeker. All
three are alike in the universality of their passion. Glaucon is so
little mindful of the sciences when the philosopher is spoken of
that he even interchanges the spectacle-seeker's curiosity with
thirst for knowledge (475d). That conflation, of course, is seriously misleading. Whoever is drawn to spectacles, whoever is
swayed back and forth in indiscriminate curiosity about everything there is to see, has in truth no similarity whatever to the
philosopher. Indiscriminate passion for novelty constitutes the
extreme opposite of philosophy, for philosophy has to do with
the just and the unjust, the good and the bad, or, as Plato puts it,
with discriminating between the beautiful and the ugly.
To be sure, in what follows it is shown that such a sense for
the just, the good, and so forth, implies a fundamental distinction between what is known and what is believedthe distinction, namely, between the "one beautiful thing" itself and everything which only participates in it {ta metechonta). To this
extent, that sense of the just and the good is indeed philosophy,
which is to say, that it passes beyond the question of the conventional "just and good." This notwithstanding, the allegory of the
cave is, as we saw, applied to nothing other than the life of the
polis. It is stated expressly that those who have returned will
have to deal with the shadows and images of the dikaion, which
is to say, with what Plato calls ta tn anthrpn (human things)
(517c) or ta anthrpeia (that which is of human concern) (517d).
Accordingly, one ought not to take the description of the cave
and the superior insight of those who have been led upward to
the true sun to imply that those who have thus been liberated
are, by virtue of their comprehensive knowledge of all true

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 78

things, better prepared for the upcoming competitionmore

experienced, more provident. That is not the point here. On the
contrary, the cave-dweller knows full well how things tend to go
in social and political life and what practices promise to be successful there. What he does not know and what he does not even
ask about is the good, for the sake of which all these practices
are to be carried out.
Hence the contrast between theoretical knowledge and political practice that is portrayed hereso that it might ultimately
be transcendedis not the contrast between theory and practice
in the modern sense. What we mean when we oppose theory to
practice in our language usage today has to do entirely with the
realm that'Aristotle calls techne and nothing to do with what is
under consideration in Plato's projection of a statenothing to
do, that is, with the ideal of theria and its relationship to political reality. But just like Aristotle, Plato, of course, knows full
well the problem of theory and practice in the realm of techne,
which is to say, on the one hand, in any realm in which the concern is to apply general rules, but, on the other, in the realm of
general experience having to do with the relationship between
ends and means in practical or poHtical action. Here we are dealing with a knowledge of rules, knowledge which, per se, contains
nothing to ensure that it is applied correctly. Hence, as is well
known, Aristotle rightly observes that the practitioner, for example, the healer, can be more successful than the specialist, for
example, the scientifically educated doctor {Metaphysics^ Alpha,
981al2 ff.). Plato is no less aware of this fact. He explicitly acknowledges the importance of practical experience in his program of education {Republic 484d, 538e), and more generally,
he sees the hermeneutical problem which is entailed in every use
of rules, that is, in their correct application.
One need only compare the thorough discussion of this prob-

THE P O L / 5 A N D K N O W L E D G E O F T H E G O O D

79

lern in the Phaedrus (268 ff.). Therewith an eye to the knowledge of rules in rhetoricmedicine and the arts of writing
drama and composing music are cited, in all of which there is
also "general" knowledge. That a natural ability remains indispensable is not disputed (269d), but besides having that ability,
education in the art is necessary. Moreover, as the Phaedrus
shows by means of Socrates' ironic dissimulation, in the case of
the art of oratory that means that the true art presupposes both
dialectical knowledge of the subject matter and dialectical
knowledge of "souls." Only he who has this double knowledge
is a true rhetorician. Here Plato specifically adds that the rhetorician must also be able to apply all this knowledge correctly in
practice: ''dei de tauta hikans noesanta meta tauta theomenon
auta en tais praxesin onta te kai prattonmena'' (it is necessary
that once he knows these things sufficiently, he see them actualized in practice and being done) (271e). Clearly this necessity
pertains to every techne. Moreover, it would seem to me that the
art of measuring what is "fitting," spoken of in the Statesman^
also accords with this argument. We shall return to this point
later.
In any case, this hermeneutical problem of concretizing [a general rule] has no bearing on the relationship between politics and
philosophy or on the tension between the political and the theoretical ideals of life. Certainly, application of a techne always
presupposes practice and experience. But the procedure remains
one and the same, whether it be the practice of the educated specialist who knows the reasons for the practical measures he takes
or only the action of an experienced man. On the other hand,
when Plato expatiates in his allegory on the relationship between
the political practitioner and the person who returns to the cave,
his concern is to draw a distinction of a very different sort.
Whether the political practitioner in this or that field is at the

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 80

same time an experienced specialist, for example, a military

strategist, seaman, or the like, has no bearing at all on the issue
at stake. The good that the one who returns to the cave has seen
outside is nothing that those chained in the cave want to know
anything about.
Their obliviousness can be stated in simple terms. Plato himself alludes to the fact that the art of the navigator per se does
not insure that the work he has done well will lead to the good.
Agamemnon's navigator might well have had doubts whether he
had done something good for his master by bringing the voyage
to a safe conclusion {Republic 60 Iff.). In techne the process of
production is subordinated to utility, and this subordination sets
a limit to techne which, as Aristotle points out {EN Zeta 4), excludes it from making any claim to be an arete.
The way in which Plato organized the ideal city in book 4 already made clear that real knowledge, which one could also call
Sophia (wisdom), is distinct from specialized knowledge of any
subject matter (428b ff.). In book 4 this specialized knowledge is
depicted in such a way that the city as a whole could be "well advised" {euboulos) by it, and likewise the soul: "'echnti ten hyper
hapases tes psyches prometheian" (it has care of the entire soul)
(441e). Here the good is still circumscribed simply as knowledge
of what is advantageous {sympheron) for everyone and for the
whole. It can scarcely be overlooked that here in book 4 the
word agathon is avoided.^
7. Plato's reservation of the actual question about the good for book 6
would seem to be fundamental to the composition of the Republic as a whole.
Whether a four-book Republic ever existed or not, the effortless and fortuitous
introduction of the question of the sharing of women and children, which leads
to the extended discussion, may surely be assumed to be as deliberate as the
avoidance of the word agathon in book 4.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

81

That it is accords quite well with the fact that in book 6 Plato
sets the question of knowledge about the good in total opposition to a Hfe led in doxa, that is, in mere conventions, and that to
this end he even places knowledge of the good in analogy to
knowledge of one's own advantage (505d). There we read that
as far as the just and the beautiful are concerned, many might
content themselves with an appearance, with what is currently
accepted, ta dokounta. With regard to the good, on the other
handand that means even with regard to the benefit that one
hopes to have from somethingthe consensus of others is of no
importance to an individual. Only the real advantage counts [ta
onto). It is instructive that here the rationality in the relationship
of means to ends suffices to illustrate the knowledge involved in
knowing the goodsuffices, that is, to establish irrefragably
that it transcends all conventions. Nobody contents himself with
merely conventional concepts when the issue is the utility of the
means to be chosen.
Now one could also view the so-called technai (the knowledge
of the handworker and the so-called sciences) in the same way,
that is, as knowledge of the right means, and hence as knowledge
of a relative good. Evidently that was the reason why techne
knowledge was paradigmatic from early on for the Socratic art
of persuasion. But this knowledge is not the knowledge that is of
ultimate importance to human beings as human beings, for it
fails to provide an ultimate justification. In other words, it
knows nothing of "the good itself."
Plato has various names for the knowledge of the good,
knowledge which by giving justification sets itself apart from all
technai and epistemai (sciences). For example, he speaks of it as
a dynamis tou dialegesthai (ability to distinguish dialectically)
(532d), a methodos (method), and an episteme (333c). In so do-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 82

ing, he orders it among these modes of knowing and opens the

way for the misunderstanding that knowledge of the good is a
highest, teachable knowledge that is to be attained at the end of
a long course of education leading through all the mathematical
disciplines. Here the bond between knowledge and action seems
to tear apart completelythe bond, that is, between those two
things which, in the Socratic question, were so closely connected
that arete itself appeared to be knowledge. But what is the significance of "good" in the mathematical sciences? To be sure, in the
discussion of different disciplines, its usefulness for the handwork of war is mentioned repeatedly. But here Socrates is all too
obviously responding to the trivial expectations that his partners
have in regard to any knowledge in the service of the polis. It is
for Glaucon's sake when Socrates says that one must take care
that such education is not useless for warriors (5lid), or when,
in appending a sequel to his program of education, he starts with
numbers and arithmetic and points out himself their indispensability for the techne of waging war. It becomes clear, nevertheless, that this reference to war and warriors does in fact make
some sort of sense within [the discussion concerning] the education of the guardians of the ideal city when Plato refers back to a
fundamental line of argument for him first stated in book 2: the
close connection between self-control and having power. He first
opened up the political dimension as such with that theme, and I
have elaborated on this connection in my "Plato and the Poets."
At this point in the dialogue, however, this connection is no
longer the issue. Hence, when it is said in the summary at the beginning of book 7, for instance, that guardians trained in dialectic are "the best in philosophy just as they are the best suited for
war" (543a, and similarly 525b), "best" is always meant in the
conventional sense and has an almost ironic overtone. The issue
is actually something else. From the beginning, Socrates pro-

T H E P O L / 5 A N D K N O W L E D G E OF T H E G O O D

83

claims his program to be an entirely new kind of education

(SlSb), in which the concern is not so much with learning something as with turning "the whole soul" around (521c). For instance, when Glaucon emphasizes in regard to astronomy that it
is useful for the art of waging war, Socrates takes Glaucon's argument as an occasion to criticize his worry about the masses of
people to whom the sciences could appear useless (527d). As a
matter of fact, in what follows Plato is emphatic in always justifying exercises in the mathematical sciences by their single essential preparatory function: they ready us for knowledge of the
idea of the good. (Plato says this explicitly at 526e.) Particularly
in respect to music and astronomy, this preparatory function entails a turn quite unexpected by Socrates' partners, the turn,
namely, away from the audible and visible to the purely mathematical-arithmological. It is difficult to see how this turn is still
supposed to bear on the Socratic question [about arete] and on
the precondition for asking itknowledge of one's own
ignorance.
Then, in a concluding summary, it is asserted that the preparatory curriculum through the mathematical disciplines is designed
to lead what is best in the soul {to beltiston en psychei) to a vision of what is best in reality {pros ten tou aristou en tois oust
thean) (532c). This assertion, made with recourse once again to
the myth of the cave, also readies us for the final transition to dialectic, and it is said of dialectic here, in similar fashion, that it
pulls the soul up out of the mire (533d). But just what this assertion is supposed to mean is so obscure for [anyone schooled in]
the ways of thinking prevailing at the time that Glaucon calls it
"difficult." He would like to be instructed in the skill of dialectic
in the way one is instructed in a new science having its special
objects and methods. This expectation scarcely speaks for his
understanding of the matter. And indeed such expectations in re-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 84

gard to Plato's thought are never fulfilled. In the realm of dialectic, there is no differentiation that would correspond to differentiation in the mathematical sciences.
Thus this curriculum of education, insofar as it leads through
the sciences up to dialectical knowledge of the good, leaves us
with a peculiarly ambiguous result. The deduction and summary
with which the previous arguments conclude would appear to be
the crowning piece in the theoretical ascent to dialectic. Yet, in
fact, they are more than that. The issue is now the good itself,
which would correspond to the sun in the allegory, and now the
question of what this is, is finally supposed to be answered without a metaphor (oud' eikona) (533a). It turns out, however, that
this particular question literally dissipates in the universality of
[inquiry concerning] everything that truly is: the dialectician is
characterized as one who strives to find what any and every existent reality truly is {auto ge hekastou peri ho estin hekaston)
(533b), one who grasps the logos (concept, definition) of being of each thing {ton logon hekaston lambanonta tes ousias)
(534b). This characterization is given in order to distinguish dialectic from the mathematical sciences, which are to be termed
mere dianoia (understanding), and in this regard it is entirely
accurate.
But then the good too is said to be an objectsupposedly in
"just the same way" {hsauts) [as the other realities]: like the
ousia hekastou (being of each), one must separate the idea tou
agathou (idea of the good) from everything else, and, as if in battle, one must endure tests of mettle and make one's way undistractedly through every challenge aptti ti logi (with a logic
that cannot be overthrown). If not, one will recognize neither the
good itself nor anything else which is good.
We must pause here. This "in just the same way" is a source of
no small difficulty. Certainly one can comprehend that in the

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

85

case of the good too, it is the procedure of the dialecticiangiving justificationthat alone can prevent our being confused by
false similarities, being guided by mere conventions, or being seduced by flattery. In the imagery of the Republic, book 2, the dialectician is like a "philosophical dog" [in his faithfulness to his
task]. But it is astonishing even so that the idea of the good appears here merely ordered alongside the other ideas. This equation suggests that it is just one idea among others. At the very
most, one could say only that the dialectical differentiation of
this idea from all others is especially difficult to carry out because of the particularly strong interference of interests and preferences in this case. And perhaps one is also supposed to detect,
if one listens carefully, that cognition of the good, which is either
to be won or lost here, is more important than anything else for
one's whole life. That it is, is brought out negatively by the juxtaposition of "life here" with Hades (534c).^ But no retraction of
the likening of the good to the other ideas is implied thereby. To
say that one can only know everything else good if one knows
the idea of the good is to say practically nothing, for this way of
putting things holds just as much for all the other ideas also. At
the beginning, too, the question about the good was introduced
with the same schematic formulanamely, that it is the thing
by virtue of which everything else {kai dikaia kai talla [what is
8. "And it is the same in regard to the good: if someone is unable to define
the idea of the good by giving a reasonable account of it, abstracting it from all
else, and fails to get through to it, getting through all tests and seeking to defend
it against all these with a logic not to be overthrown (aptti ti logi), as if in
battleif all this be true of him, you would not say that he has recognized either
the good itself or any other good. Rather, you would say that if he has attained
to any image of it at all, he has done so through opinion and not through science,
and that in dreaming and slumbering this life away, he will land in Hades before
ever awakening here and sink into the deepest sleep" (Republic 534c). (Text provided by translator.)

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 86

just as well as everything else]) becomes useful and good (503a).

But how are we to reconcile the manifold of true reality with the
unity of the true good? And at the expense of what? Should the
reality that is split apart into the multiplicity of ideas be sacrificed, so that, by suspending all "hypotheses" of ideas (tas
hypotheseis anhairousa) (533e), one ascends to a good that
would no longer be the good of each particular {hekastou)} Or
should the "separate" (apheln) good be sacrificed, so that the
good would then be in everything good, and so that, like all
ideas, it would have its true being only in what participates in it?
The text here calls for no such radical interpretation. Since we
are faced with an explicit "in just the same way," we cannot
treat the matter as if we were dealing with some new, ultimate
step leading from the multiplicity of ideas to the "principle" of
the one and the good. On the other hand, one would certainly
not want to say that the idea of the good is comprehended "in
just the same way" as all the other ideas. Is the idea of the good
only one of the ideas? And if so, what happens then to the "one,
[and] good"?
Hence we must ask ourselves: In the end, is it only Plato's
mythical way of speaking that involves us in this apparent dilemma? Can one develop from Plato's metaphor a better way of
putting the question? Let us reexamine the metaphor and our interpretation of it. Back when Plato introduced the analogy between the sun and the idea of the good, things looked quite different. There, the good was not ousia but was explicitly said to
be "beyond" ousia, exceeding it in majesty and power (509b).
There, we found no "in just the same way," but instead a surprising new step. And it would still be a new step even if it were
to signify only a new insight into what can be called "true,"
what thereby constitutes the being of any given thing.
For, initially, one can interpret the comparison of the good

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

87

with the sun at 508e in the following way: what gives aletheia
(truth) to what is known and gives the capacity to know it to the
one who knows it is supposed to be the idea of the goodjust as
visible things are visible and the eye can see, thanks to the sun
that sheds light. Thus, to begin with, the idea of the good is the
cause of knowledge and of truth (epistemes or gnses, and
aletheias). Obviously the point is the analogy with seeing and the
visible, and their dependence on light. Just as seeing and light are
sunlike, so too, knowing and truth are to be called "good-like,"
even if they do not yet count as the good itself, that is, as ton
agathou hexis: the character the good has about it, what it is.
And to this extent the analogy is indeed most expressive. One
can completely set aside the question of the cause of the light, the
question, that is, about the sun. True beingthe noumena (objects of intellection), the ontos onta (things that really are), the
eide (forms)appears in thinking in the same way that light
connects the visible with seeing: the good makes thinking what it
is. The capacity {dynamis) for something is, after all, always defined by what it is a capacity for, and by what it effects {eph'
hoite esti kai ho apergazetai) (477d). Thus, what lights up
(katalampei) as aletheia te kai to on (truth and being) (508d) allows thinking to be thinking, that is, allows it to noun echein
(have reason, be right, avoir raison) in that nice double sense of
both getting an insight and being capable of reasoning. With this
formulation it seems clear that the whole realm of the noeta
(things thought of) has been opened up.
The metaphor of light accomplishes all this, and it is significant that in the scene in the Parmenides in which the young
Socrates is supposed to clarify the participation of the many particulars in the idea, he takes refuge in the same marvelous metaphor. * Plato's own recourse to the metaphor of light is plainly
behind Socrates' answer in the Parmenides that the idea is like

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 88

the day. Socrates wants to say that just as things are only visible
in the one light of day that floods around them, so too the idea is
visible only to the extent that it emerges in thinking. This means,
however, that the idea is only visible to the extent that it allows
the beings which appear to be thought of as what they are. Thus
we have a threefold methexis (participation) here: (1) the methexis of the individual in the idea, (2) the methexis of the soul in
the idea, but ultimately, (3) the methexis of the ideas in each
otherfor any thinking of something is both a delimiting-from
and a combining-with (cf. the Sophist on dihairesis and synagdge). My thesis, then, is this: these three kinds of methexis are
nothing but aspects of one and the same relationship. To think
of "this here" as "what" it is, is always at the same time to differentiate the "what" from the "this." But to differentiate the
"what" is always to differentiate one "what" from another
"what": the impurity in the appearing "this" is in truth the existence of some other pure thing in it.' This circumstance provides
the background for the dialectic in the Parmenides*
There, of course, Socrates shows himself incapable of keeping
a firm hold on the sense of the metaphor of light and of elevating
it to the conceptual level. He is still young. In truth, Aristotle is
the first who could. He did so in availing himself of the distinction between poiein (doing or making something) and paschein
(suffering something), or poietikon (active) and pathetikon (passive), to conceptualize the structure of nous (intellect). In Aristotle too, nous enables, "makes" {poiei), thinking, hs hexis tis
(as a kind of condition), just as light * "makes potential colors
9. The impurity of fire, for instance, is coals or ashes, which to someone who
might use them for fertilizer are something else pure. It is not by accident, Gadamer points out, that one says reiner Schmutzpure filth. Thus there are indeed ideas for "mud" and "hair" though Socrates is reluctant to admit it. TRANSLATOR.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

89

into actual colors"^ {De anima 430al4

The principle of
the good, then, would seem to have the function of a principle
only in epistemological matters.
But if we follow Plato's text, we see that Socrates takes a decisive step beyond this interpretation of the allegory. And when he
does, the special status of the idea of the good again becomes a
problem. For now the "transcendence" of the good emerges with
all its ontological implications. And not without a dramatic
mise-en-scene: as the sun bestows upon things not only visibility
but also their coming into existence, their growth, and their nurture, without it itself being "becoming" (genesis)^ so too the idea
of the good, without having "being" itself, is said to give being,
to einai kai ten ousian (being and reality) (509b), to what is
known in thinking. The comparison forces us to take this ascent
beyond being in such a way that the good becomes the "cause"
of the being of the many ideas. But of course the question remains: "cause" in what sense? The word "cause" {aitia)^ which
we know as the fourth genus of the Philebus^ is not used here
either for the sun or for the idea of the good. On the contrary,
the text moves in the semantic field of dynamis (power): parechein (allowing), pareinai (standing by), proseinai (being present). The rendering of the good that Socrates gives (511b) makes
unequivocally clear that here the good is interpreted as tou
10. "But since, as in the whole of nature, to something which serves as matter
for each kind (and this is potentially [dynamei\ all the members of the kind) there
corresponds something else which is the cause or agent (poietikon) because it
makes (poien) them all, the two being related to one another as art to its material,
of necessity these differences must be found also in the soul. And to the one [passive part of] intellect, which answers to this description because it becomes all
things, corresponds the other [active part of intellect] because it makes all things
(panta poien), like a sort of definite quality {hs hexis tis) such as light. For in a
manner light,, too, converts [poiet) colours which are potential into actual 'colours'" (De Anima 430al0-16; translated by R. D. Hicks).

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 90

pantos arche^ the "starting point (principle) of everything."

Consequentlyand just as unequivocallyit is the good from
which the descent begins and from which the multiplicity of the
ideas that make up the noetic realm originates. It is explicitly
stated in regard to the mathematical-dianoetic realm (5lid) that
the multiple ideas are called noetic because they exist meta
arches (by virtue of the starting point), which is to say, because
they derive their intelligibility from the arche. There would be no
problem here in saying that this mathematical realm of entia
rationis derives not only its intelligibility but also its being from
the arche. The interrelationships here in the dianoetical realm of
the mathematical, at least, have been splendidly explained in the
investigations of Konrad Gaiser.^^
Now the realm of dialectic, of course, is first reached when we
pass beyond the dianoetical, and only then have we taken the full
measure of what truly is: ''autou ge hekastou peri ho hekaston"
(concerning each thing itself, what it is in itself) (533b). Hence
we can say that wherever we imprint the seal "it is"wherever,
in other words, we make use of words and speech and say sentenceswe dream about being {oneirottousi men peri to on)
(ibid.). But for the time being we just dream. Only the dialectician awakens from the hold that the dream of the life-world has
on us. And "waking up" is precisely what he does: he suspends
the hypotheses in which our linguistic interpretation of the
world is set down, and in thought alone, he inquires what lies behind them.^^
11. K. Gaiser, Platos ungeschriebene Lehre (Stuttgart, 1963).
12. It is here that Plato moves beyond our world of things as they are originally given to a "metaphysical" world (Heidegger). What Gadamer emphasizes is
that this transition occurs primarily within Plato's approach to language. Given
the ambivalences of language as it is ordinarily spoken, it is all too susceptible to
sophistic manipulation. Consequently, Plato finds himself forced to seek secure

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

91

It is common knowledge that Plato discusses the hypothesis of

the eidos in the Phaedo. The hypothesis of the eidos, it is argued
there, should not be involved in dialectical dispute prematurely,
that is, without prior testing and scrutiny. In the Phaedo, however, it also remains uncontested that when all is said and done
dialectic must be given a final justification. In his quest for an ultimate starting point for that justification, the dialectician suspends, one after another, the hypotheses set down in our discourse. And the same thing holds here in the Republic. Thus, in
both the Phaedo and the Republic we are led by dialectic up to
the arche that is the one. And just as Aristorie did, we may discover in this one the duality of one and two. But there are reasons for assuming that this first thing, or principle, does not
serve as the foundation for a system of ideas that could be deduced from it. Apart from the paradigmatic case of the sequential mathematical disciplines, the concern is always with limited
realms wrested from the apeiron (indefinite), within which the
dialectician attains his insight.
In the Philebus Socrates gives the examples of musical tones
and of letters and phonemes. These, certainly, are examples of
the one's unfolding itself into the many. For letters, like phonemes, have systematic character: "oM<i an hen auto kath' hauto
univocal ideas behind our ordinary "linguistic interpretation of the world."
Gadamer finds here the beginnings of formal logic as we know it today, that is,
the ideal of a univocal sign system that would allow us to transcend the confusions induced by the unclarity of ordinary language (cf. WM 383 ff.). Still, insofar as Plato remained oriented toward Socratic discussion, and insofar as he recognizes our finitude and the consequent impossibility of systematizing language,
heunlike modern logiciansremains true to our original experience of language and the world that it constitutes. Put another way, he is far more "original" (ursprnglich) than Heidegger makes him out to be. Compare Heidegger,
Platons Lehre der Wahrheit (Bern, 1954) and Gadamer, "Plato's Unwritten Dialectic." TRANSLATOR.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 92

aneu panton autn mathoi" ([we could] not learn any one [of
them] by itself apart from all of them) (18c). However, as accords with their specific nature, they constitute a deHmited
realm even so, and thus they serve as an example for any techne
(16c). Here, [knowledge of] the systematic structure (desmos
[bond]) is tantamount to mastery of writing or of making music
respectively. In regard to their structure, certainly, these two are
the same as dialectic: ascent to the primary and descent from the
primary belong together: ^'touton ton desmon au logizamenos
hs onta hena kai panta tauta hen ps poiounta^' (and we conceive of this, their bond, in turn as being one and somehow making all of them one and one into all of these) (18d). In each case
the reference is plainly to a relative unity to which each tone or
phoneme, respectively, belongs as such. Compare the parallel execution of dihairein (division), which departs from the one voice
(phone mia) (17c) and, conversely, at 18b, from indeterminate
voice [phone aperios), (The last approach seems easiest to Philebus, evidently because he follows the path of experience [18d].)
Structurally, dialectic here corresponds closely to dialectic as it is
portrayed at the conclusion of book 6 of the Republic (511bc).
Of course, in the Republic the only arche spoken of is a single
one, and there is no talk of relative first principles such as the
principle of either voice or tone in the Philebus. That there is not
would seem to signify that the examples in the Philebusinsofar as they involve something like tones or lettersmust belong
to what the Republic calls dianoia. Here, at 511c, the distinction
between dianoia and dialectic still sets dialectic sharply apart
from the "so-called technai," which do not justify their presuppositions. Only in the Philebus are technai such as music or the
science of letters described in such a way that they themselves
could be called dialectical. Does that not mean that Plato was indeed fully aware that the ideal of dialectical derivation of all
things from a single arche could never be carried out completely?

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

93

An astonishing statement of Aristotle's would also seem to

support this thesis. At the beginning of his ethics (EN 1095a32)
he says that Plato considered it problematical whether one
should work up toward the archai (first principles) or down
from them. The plural is striking in itself. To be sure, it is characteristic of Aristotle's own use of language. But the aporia as such
seems indicative of the fact that Plato did not envision a unified,
fully elaborated deductive system. Otherwise it would not have
been such an aporia at all. Furthermore, this aporia does indeed
correlate with the examples of a two-sided procedure given in
the description of the structure of techne in the Philebus, In
"Plato's Unwritten Dialectic," I have attempted to clear up how
the imperfectibility impHed in Plato's doctrine of two principles an imperfectibility that Aristotle elaborates for usis reflected in Plato himself in the way indeterminate duality
(ahoristos dyas) functions.
In any event, we are far removed here from Socrates' question
about the good. This question arises as a question about arete
(virtue) andgiven all the special forms of areteas a question
about the aretai (virtues). And the aretai are not to be investigated in regard to universal considerations but rather in regard
to the narrower realm of the good in human life. Certainly the
most natural starting point for the question of the good is the
good in human life. The program of the Phaedo addresses the
good in human life, and so does the Republic when the question
of the good is introduced there. And the same is true later when
this question is raised again in the Philebus, But in all these cases
the discussion transcends this narrower realm and leads to a universal ontological inquiry.
That it does so is not surprising if one remembers how universal the context was in which the question of the good is introduced in Plato's Phaedothe context namely of inquiry into the
cause of coming-into-being and pasing-away in general: ''holds

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 94

gar dei peri geneses kai phthoras ten aitian diapragmateusasthai'' (for it is appropriate to treat as a whole the cause of generation and destruction) (95e). Socrates advances his own example
of knowledge of the good only as an illustrative introduction to
the general question of what knowledge is. As early as the
Phaedo a teleological cosmology is postulated, which, to be sure,
is not worked out. Similarly, in the Seventh Letter^ we find the
extension of knowledge about arete to knowledge about the
whole of reality (344b). And, in the final analysis, the Timaeus is
the mythical exposition of the unelaborated postulate of the
Phaedoeven if, strictly speaking, the Socratic question is no
longer mentioned there at all. One sees that Aristotle is extending a Platonic line of thought in his teleological physics and
metaphysics. But how the widening of the agathon (good) to the
arche tn pantn (principle of all things) is supposed to follow
from the structure of dialectic is still obscure.
It seems to me that real clarification of this problem can be
achieved only if one analyzes the actual procedure of dialectic, a
procedure that is specified in the Republic only as a general program. We must find justification in this procedure of dialectic for
the disquieting "in just the same way" that places the good
alongside the other ideas. Let us remember that the underlying
principle of Plato's Utopian state (even in the first outlines of it)
was to educate the guardians, in whose hands the power of government lay, to be immune to the seduction of power: in the end
education in science was to be education by science. It is striking
that in our passage, with its confusing "in just the same way,"
grasping the good is portrayed as the breakthrough that brings
victory in battle. What battle? Against what enemy?^^
13. The image Plato useswhich, of course, was suggested in particular by
the warrior-guardian analogy of his Utopiain fact belongs to one self-con-

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

95

Obviously the concern is not only the abuse of power. Or, better said, even the abuse of power, which the constitution of a
state is intended to prevent, derives from another fault. And
Plato maintains that this other fault is a lack of dialectic, a lack
of the art of differentiating. That, of course, sounds absurd. As if
the passions which carry us away were forms of thinking, and all
thinking were not overpowered precisely by the force of them!
The Protagoras pushes this absurdity to the extreme: there, succumbing to the passions is said to be mere ignorance (352 ff.).
This assertion notwithstanding, Plato's intellectualization of
courage in book 4 of the Republic did disclose something important. He demonstrates convincingly that in regard to courage as
a qualification for the warrior-guardians, the concern is that
they hold fast to the right doxa (belief) about danger and not let
themselves be dissuaded from that beUef by anythingnot even
the seductive power contained in hedone (pleasure). That may
indeed be expressed too "intellectually," and Aristotle, who always takes Plato word for word, accordingly discounts the Socratic position all too much when he comes to its aid with the argument that to some, a man might actually appear courageous
only because they themselves overestimate the danger he faces
(EN 1116b3 ff.). What Socrates has in mind is substantiated insofar as the issue is real political courage and not just physical
courage (430b). For in political courage the concern is not so
much the physical anxiety as such that overpowers us, as the rationalizations into which fear seduces us. And in the case of
hedone the concern is even clearer: the seductive power of pertained semantic field: the expressions for logical operations are taken in large
part from the language of wrestling and similar forms of fighting; so too, aptti
(not to be overthrown) in the passage here (cf. n. 8 above), and frequently diamachesthai (to fight hard, or contend), for instance, in the Gorgias at 503a and
elsewhere.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 96

suasion that emanates from hedone. To be sure, I cannot prove it

in regard to this particular passage, but it seems clear to me that
Plato is thinking of the flattery that legitimates the powerful in
their abuse of their power. One need only recall the Gorgias, for
instance, where the sophistic rhetorical art is portrayed as an art
of flattery. We encounter something analogous here in the Republic when sophistic wisdom is likened to the art of comporting
oneself properly toward a dangerous beast (493b). And in the
depiction of how a youth (similar to Alcibiades) is lured away
from dedication to philosophy and ruined, there is explicit reference to flattery, namely anticipatory flattery: ''prokolakeuontes
ten mellousan autou dynamin'^ (flattering in advance a power he
will yet have himself) (494c).
In any case it is true that whoever finds himself in the possession of power must fight his way through {diamachesthai) temptation in order not to succumb to the seduction of power. And
anyone can see that such resistance is really an accomplishment
of reason {logos) and, yes, even a question of reasonableness
{phronesis). In Plato's language reason and reasonableness are
called dialectic, for his language usage is taken from [Socrates']
guiding of a discussion through all its episodes and gathering
and holding together what is sought and actually meant through
all the unsteadiness and errancy that pervade any discussion
be it with others or with oneself. Used this way, "dialectic" is, as
we saw, not simply an art that can be learned. One succeeds in
holding fast to what one sees before one's eyes as right not only
logi (in discourse) but also ergot (in deed). Hence the guardians
must be monimoi, steadfaststeadfast specifically in the sciences, just as in war and in everything else that is considered
right {en nomimois [in lawful obligations]) (537d). The unerring
way in which Socrates lived his life in Plato's portrayal of him
right up to the final temptation of the escape offered to him in

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

97

the Critoand, most of all, the long discussion as Socrates is

taking leave of his friends in the Phaedo are the best illustrations
df this steadfastness. An essential point that gives Plato's dialectic as a whole its underlying meaning is that he demands justification in logos from the persona of Socrates. For this
reasonand not for reasons of aesthetics and taste^^it is vital to read Plato's dialogues not as theoretical treatises but as mimesis (imitation) of real discussions played out between the partners and drawing them all into a game in which they all have
something at stake.
In regard to the passage we are analyzing, it is indeed striking
that the portrayal of the dialectician, who has the idea of the
good before his eyes, puts particular emphasis on the struggle to
withstand "tests" (elenchoi): ''hsper en machei dia panton
elenchn diexin" (as if in battle, getting through all tests)
(534c). Neither in reference to the mathematical studies that lead
to "pure" thinking nor in reference to the universal requirement
of giving justification for "being" was there any mention at all of
tests. With what sovereignty are the peri tauta deinoi (the "experts" about these things) repudiated at 525d, for instance, when
the doctrine of numbers is introduced! Plainly, the temptation to
rationalize and the seductive sophistry of the passions play a special role in the matter of the goodand, one might certainly
add, in the matter of the aretai, even the "intellectual" aretai,
too. It is the role of what Kant calls a "natural dialectic" (Foundations of a Metaphysics of Morals^ at the end of part
14. How astonishing that Gerhard Mller can hold that reading Plato's
Socrates' mimesis as mimesis robs Plato's thinking of its seriousness {Gttingische gelehrte Anzeigen, vol. 229 (1975), pp. 157 ff.). What a degraded, aesthetical and mistaken concept of mimesis, poetry, and myth!
15. Kant figures prominently in Gadamer's ethical thought. In "ber die
Mglichkeit einer philosophischen Ethik," in KS, vol. 1, pp. 179-91 {henceforth

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 98

Thus one should take note that the portrayal of the dialectician here contains not only the word elenchos (test, refutation),
which we know from the Socratic discussions, but a series of
other words that we find in the later dialogues in the context of
characterizing dialectic: dihorisasthai, aphelein^ diexienai^ to define thoroughly, to abstract, to run [get] through. "To define
thoroughly" means to mark off one thing from another, hence,
to differentiate. And that implies removing the thing meant from
everything which is not meant {abstrahere) and getting through
all differentiations until the end, that is, until an understanding is
reached with others as well as oneself. The prefixes are of particular importance. The dia (through, thoroughly) implies at the
same time'an "asunder," hence a differentiation. And the apo
(off, away) implies at the same time a "to," that is, a seeing together of what has been taken away. This vocabulary already anticipates the analyses of dialectic undertaken in the Sophist using
the highest generabeing, identity, difference, and so forth.
Here in the Republic the exposition still has a purely dialogical,
indeed even military, timbre [hsper en machei [as in a battle]).
That does not happen by chance for the danger lurks in logos
(discourse) itself.
Confusing something is the counterpart to distinguishing
something, and the wrong separation is the counterpart to the
right one. In confusing something and separating falsely, one
MP), Gadamer attempts to develop an ethical theory in which Kant and Aristotle, far from being in conflict, complement each other. The application of
practical reason in hitting the mean between the extremes in a particular situation can only succeed if reason is secured against the seductive influence of the
desire for gratification and the "flattery" of the senses. For when reason is not so
secured, it degenerates into rationalization. Gadamer reads Kant's Foundations
as devoted primarily to this preparatory task of purifying the rational sense of
duty from the subversive influence of the senses, TRANSLATOR.

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

99

gives oneself over to the play of opinions being bandied about.

The counterpart to dialectic's getting through (dia panton elenchn [through all tests]) is the dazzling art of the forceful answer
{kata doxan [counter to what someone believes]), that is, the art
of contradicting (antilogike techne). In a word, the counterpart
to dialectic is sophism. Sophism is a constantly threatening
dangersomething that can always happen to the logoi (assertions) "in us" {''tn logon autn athanaton ti kai ageron pathos
en hemin^ (an experience in us, immortal and unaging, of the
statements themselves) {Philebus 15d). Even in mathematics
there are pseudo-proofs or pseudo-arguments which, as we
know, Protagoras advanced in great number against mathematicians. Of course, Plato would say that someone who can really
do mathematics is out of range of such rhetorical arts, even if the
mathematician escapes simply by withdrawing from verbal disputes {psiloi logoi [vacuous arguments]) (Theaetetus 156a) and
from Protagoras's entourage, as does Theodorus in the Theaetetus. On the other hand, where the subject is of the greatest importance, that is, in the realm of dialectic, one is threatened
steadily by the danger of sophism. In the excursus of the Seventh
Letter^ this threat is explained in detail. But even where the
"sophist" is firmly pinned down in a definition, as he is in the
Sophist dialogue, his intrinsic proximity to the true dialectician
and philosopher emerges. For merely clearing up how something
has been mistaken for something it is not does not really put an
end to the abuse of logos. This abuse is a moral matter.
Using the exemplary case of the Lysfs, I have already elaborated in greater detail how Plato's early elenchtic dialogues are
based entirely upon the correlation of logos (word) and ergon
(deed) (see "Logos and Ergow in Plato's Lysis''). As a matter of
fact, the connection between the logical and ethical aspects of
true dialectic runs through the whole of Plato's work. And even

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 100

Aristotle confirms this correlation when he finds the difference

between sophistry and philosophy to be purely the prohairesis
tou biou (choice of the life [one leads]) (Metaphysics 1004b24;
Sophistical Refutations 169b24, 171b8).
Thus it is not in the least remarkable that, in our passage,
holding steadfast in the face of all confusion is emphasized especially in regard to the idea of the good. That it would be is already inherent in the whole Utopian construction of the ideal
state's constitution and of the education of the guardians which
that constitution specifies. Hence, in the discussion of the educational program for these guardians, neither the content of the
mathematical sciences nor the particular set of problems in the
doctrine of ideas are under consideration as themes of dialectic.
The sole interest is in turning the soul around toward noetic
reality.
To be sure, when dialectic is raised out of, and beyond, mathematical dianoia in the Republic, we get a first hint that in the
truly noetic realmwhich has a presuppositionless {anhypotheton) arche, *that is, a principle preceding all hypothesis*the concern will be the relationship of the ideas to each
other. (And in an elaboration of that relationship, the inner
unity in the structure of the mathematical disciplines would become transparent.) But this hint given at the end of book 6 is not
pursued in what comes afterward, that is, in the interpretation
and exposition of the allegory of the cave as a curriculum of education leading through mathematics and dialectic. And the "hyperbolic" position of the good, represented by the simile of the
sun, is not really explained, as our detailed examination of the
summary at 534b-c has shown. When it comes right down to it,
the "ascent" to the noetic dimension is the single theme. Accordingly, the problem of participation, of methexis, which is so
sharply disputed later on in the Parmenides, enters in here only

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

101

insofar as it is said to be characteristic of the philosopher that he

sees the beautiful itself and does not mix it up with the many
beautiful things that "participate" in it. Here, methexis appears
to imply only the differentiation of "it itself" from those things
that participate in it. In other words, its reference is limited to
what the Phaedo called the "simple" hypothesis of the eidos.
Significantly, when the "idea" is introduced explicitly in the
Phaedoperhaps for the first time in Plato's writingsit occurs for the stated purpose of putting a stop to the anti-logical
arts of inducing confusion. The interpretation of the Phaedo's
"hypothesis of the eidos'' in recent Plato scholarship as a
wholeand not only in the Marburg school's radicalization of
the questionhas kept its attention focused all too much on science. As a result, the ideas are taken to be the "tranquil realm of
laws" (Hegel) that one approximates by critical testing of the hypotheses one postulates. But Plato did not have the process of
scientific investigation in mind at allrather, the sophistic
abuse of that new marvel, an art of arguing able to confound any
assertion: Socrates had been so confounded in his studies that in
the end he no longer understood anything, not even what causes
a human being to grow (96c). One has to read this text precisely:
the hypothesis of the eidos is not to be tested against all-decisive
"experience." The test of "experience" would be a complete absurdity for the postulation of an idea. What constitutes being a
horse can never be confirmed or refuted by a single empirical
horse. Rather, the test here relates to the immanent consistency
of what the eidos comprises: one should not go a single step further until it is clear what such a postulation says in each case,
and what it does not. Accordingly, one should note that it is not
the hypothesis that gets tested against its consequences, but the
presumed consequences that get tested against the hypothesis:
everything that the hypothesis does not include should be ruled

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D 102

out. Above all, that means that the individual thing which participates in an eidos "counts" in an argument only in regard to the
eidos in which it participates, that is, only in regard to its essential, eidetic content. All logical confusion has its origin in not
keeping the eidos separate from what only participates in it. For
if one fails to keep these separate, one easily gets entangled in
contradictions, such as saying that the number two "comes into
being" both by addition and by division. A proper testing of the
hypothesis of the eidos repudiates as sophistic all adulteration
introduced here by the concept of "coming into being." (A good
illustration of this function of the hypothesis. Republic 525d,
has been treated above.)
The portrayal [of this procedure] in the Phaedo is obviously
intended to be of the highest formal generality: the illustration
given is the noetic example of the number. Even so, what is said
here is in complete accord with the demand made of the philosopher in the Republic^ that he differentiate the eidos from everything that participates in it. Of course, one must be clear that this
procedure of hypothesizing the eidos is only the preliminary precondition of all argumentation, only the first step which provides an initial foothold on the shaky ground of the logoi. No
knowledge is attained by this procedure yet. As early as the
Phaedoy no doubt is left about that. And to this extent there
never was such a thing as Plato's "Eleaticism": the schema of
"development" in Plato proposed by Stenzelfrom arete to
dihairesiswhich, per se, contains many correct observations
must be qualified accordingly. In the Phaedo only the application of the procedure of hypothesis to the immortality of the soul
and the comparison of the soul to snow, which disappears when
the fire of the sun warms it, is supposed to yield anything like
knowledge.
Nevertheless, this first step on the path of dialectic lays the

THE P O L / 5 A N D K N O W L E D G E OF T H E G O O D

103

foundation. It is the step into the noetic realm as such, a step that
is presupposed whenever one is serious about giving justification. When Plato describes dialectic in the Republic^ its differentiating is played out entirely in the noetic sphere (511c). We
will see later in our analysis of the Philebus that there, too, this
first step of reflection, the step into the noetic, is taken explicitly.
There, Protarchus names the contradictions that result from the
correlativity of the one and the many and, similarly, of the large
and the smallcontradictions that are introduced in the Republic (523a ff.) expressly as the "call" to awaken to thinking. In the
Philebus (14d) Protarchus has to be told by Socrates that these
contradictions are trite and overworked. Matters only become
serious once noetic unities are under consideration, and when
these are said to be one and many at the same time. And in what
follows, the dialectical nature of all science is founded on this
noetic basis. Even in the very words he uses here, Plato is
opposing true dialectic to the art of confounding someone: for
example, in the Phaedo, lOle, ''phyroio'' (would confuse), and
in the Philebus^ 15de, ""symphyron'' (confusing).

IV

THE DIALECTIC OF THE

I N TH E

GOOD

PHILEBUS

Of all Plato's dialogues, the Philebus has long been considered

the most important source we possess for the mysterious Platonic doctrine of the ideal numbers. Of course, this aspect of the
dialogue should not concern us, given the context of our present
investigation. Our question is the reverse one: what relationship,
if any, does this doctrine, known to us above all from Aristotle,
have to the Socratic question about arete and the human
good'^? Put another way, what was Plato thinking of when he
took up the principles of the one and the indeterminate two in
his famous public lecture "On the Good," and then nevertheless
came to speak in it of human virtues?^ The Philebus must contain
an answer to this question, for in no other dialogue of Plato's is
the theory of dialectic so tightly interwoven with the dialectic of
human practice. The "transitions" in this dialogue pose a longstanding and famous problem. But of primary importance here is
the fact that the Socratic question about the good in human life
is actually posed again, and yet in such a way that, simultaneously, the universal nature of the good is always kept in view.
Similarly, the nature of dialectic is treated in statements of the
utmost generality and universalityalbeit without any mention
of the good per se in these reflections which are specifically said
1. See Aristotle's gloss in the Magna Moralia, cited below, ch. 5, n. 3.
104

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

105

to concern the principles of dialectic. Instead, the universal question about the good is woven completely into the plot of the discussion, that is, the dispute about the respective importance to
human life of hedone and phronesis.
Thus here we find together again all those things that in the
Republic got dispersed over the wide-ranging considerations of
Socrates' long discussion of the true state: the Socratic question
about the good, the doctrine of the ideas and their dialectic^ and
the doctrine of the uppermost principle, that is, the good. And all
this occurs in a discussion between Socrates and quite young
people, whom he must introduce to these things ab ovo.
And indeed, here in the Philebus the question that we found
posed in the Republicwhether pleasure or thinking is the
highest good (505b)is made the theme of a dramatic confrontation. The advocates of pleasure are no longer dismissed from
the start as they were in the Republic. After all, in defending
their position they can summon in support a truly important
trait of life that pervades all living things. Behind these advocates
stands, we suspect, the figure of Plato's great friend, the mathematician and scholar Euduxus, to whom Aristotle later makes
respectful reference in the same context {N, Kappa 2). Those
who advocate "thinking" will have to justify their claim that it is
supreme against this universal principle of life that pervades even
the human being, who is distinct by virtue of memory, deliberation, and the like. At first it seems as if two irreconcilable basic
attitudes were being pitted against each other here. For just as it
steers the behavior of any living thing, the pleasure principle has
a kind of obvious predominance, unlimited and overpowering,
in the human being too. That one should argue for this principle
in [rational] statement and answer would seem to be self-contradictory, and hence it is entirely consistent that those who do advocate it do indeed resist^giving justification of their position in

T H E D I A L E C T I C O F T H E G O O D I N T H E PHILEBUS 106

this way. The most visible indication of this reluctance is the fact
that Philebus, in whose honor the dialogue is named, withdraws
from the discussion entirely.
Thus there is good reason for why the old Socratic question of
the good in human life leads precisely at this point to thematizing the dialectical principle of giving justification. Resistance to
the demand that justification be given is part and parcel of the
hedonist position. Philebus is consistent when he does not oppose this demand with a logical argument but, instead, dogmatically insists on the unconditional priority of hedone: "That's
what I believe and that's what I always will believe" (dokei kai
doxai) (12a). He bids adieu to the whole thing in order not to do
any harm to his "goddess" Pleasure by accepting any uncomfortable assumptions about her. And when Socrates gives a demonstration of the principle of differentiation, illustrating it with examples of particular technai [grammar and harmony], Philebus
cannot see what that could possibly have to do with his contention, of which he is so completely certain (18a-b). The allusion
is the same at 22c, where we find, "and not your nous (reason)
either!" In this phrasing, "your reason" (ho sos nous) (22c), we
detecteven in the negative Philebus's absolute partiality for
his goddess Hedone. Philebus's intransigence is made all the more
evident by the skirmishing in which Socrates defends the official
cult name. Aphrodite, against Philebus['s attempts to rename her]
(12c). Philebus wants to consecrate Hedone artificially. Socrates,
in contrast, stays with the cult name for Aphroditethat is to
say, he recognizes her as a member of the Olympian family of
gods. In substance, his doing so establishes the merely partial vaHdity and limits of hedone's claim to be the dominant power in
the world. After that Philebus allows himself to open his mouth
just once more, and then solely to reinforce that no limits can be

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

107

set to the desire for pleasure (27eas in Nietzsche's "for all

pleasure wants eternity'').
Protarchus, who takes over Philebus's part, is now introduced
step by step to the process^ of giving justification. As his initiation proceeds, the structure of dialectic is simultaneously clarified. At first he resists any differentiating among pleasures: as
pleasure, he says, they are all one and the same thing (12b and
13c). Socrates counters with the example of the genus color and
points out that to insist on the mere unity of the genus is to say
nothing at all. On the contrary, at the moment when one wants
to say anything about a genusfor instance, that it is good
one must differentiate. Precisely in order to escape the sheer irrationality {alogia) of such all-inclusive arguments that run everything together, Socrates develops the dialectic of the one and the
manya dialectic implicit in the postulation of any such ideal
generic unit. He sees an abuse of this dialectic precisely here,
where the principle of hedone is defended by means of such an
all-inclusive reduction. When he goes on to demonstrate what
for him is the productive application of differentiation, using
convincing examples, his partner does accede, but then tries one
last objection: Is there, he asks, really a need for such subtle dif2. Bewegung, The word figures prominently in the Hegelian tradition and
means primarily motion or movement, not in the spatial but in the logical-developmental sense. Hegel wishes to trace the "movement" of the "concept" or subject matter (Sache) of thought, as it unfolds itself. Gadamer finds this same attention to the self-unfolding of the subject matter in Plato's dialogical art, the only
difference being that Plato stays much closer to the discursive character of our
thinking and speaking. For Gadamer, every discussion, if we are willing to submit ourselves to it, takes its own course, and the task is thus to let the thing under
discussion display itself apart from any arbitrary intrusions (Hegel: Einflle) in
the process. See "Hegel and the Dialectic of the Ancient Philosophers," in HD
p p . 5 - 3 4 . TRANSLATOR.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 108

ferentiations in the types of pleasure and thinking, when the issue is as existentially important as the question of the good?
Protarchus is speaking quite Socratically here: one should not remain hidden to oneself, he says (19c). But paradoxically, he says
this in order to evade Socrates' demand that he pursue the
dialectica highly ingenious, ironic twisting by Plato of existential seriousness and the dialectical game. Socrates is surprisingly
willing to go along with seeking an alternative approach to
resolving the dispute. As he so often does, he introduces an argument quite on the sly, which, when all is said and done, will serve
to support his thesis, which here is to prove the priority of nous
over hedone. The argument is the doctrine of the four genera
(23b ff.). From this point on, his partner goes along with him
ever more readily, no longer allowing himself to be misled even
by Philebus's final intervention (28b). Something quite surprising now happens: when the doctrine of the four genera is applied
to the two contestants, hedone and nous, hedone is assigned in
toto to [the first genus,] the apeiron (indeterminate), and nous to
the fourth, the aitia (cause). With this conclusion the argument
might be considered setried. But instead, Socrates asks where
and in what way {en hoi te kai dia ti pathos) these two, hedone
and phronesis, show up in the visible realm (31b). And from this
moment on, no more persuasion is needed to involve his partner
in more and more subtle, and further and further differentiated,
analysis of the most varied forms in which hedone appears (and
later, phronesis too).
This transition is the least obtrusive, yet perhaps also the most
important of all in this dialogue, which is so rich in transitions.
The transition here to the concrete manifold of experience takes
place automatically, so to speak. And the application of the procedure of differentiation, which Protarchus had shied away from

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

109

before, now takes place automatically too, and with his willing
assistance (32c).
It is remarkable how this "guiding of the soul" [psychagge)
is combined with highly theoretical expositions of the principles
of dialectic. Socratesin his typically secretive, oracular way
supports his argument with appeals to obscure sources of knowledge and even vague dreams. Plainly he makes no claim that
what he says is authoritative. Rather, he sets the listener free
once again to recognize himself in what is said.^ In this way
Socrates does indeed get his partner to enter into the dialectical
movement voluntarily.
After all, dialectic, as the art of differentiating rightly, is really
not some kind of secret art reserved for philosophers. Whoever is
confronted with a choice must decide. Being confronted with
choices, however, is the unalterable circumstance of human beings. Their having to make choices removes them from the realm
of the rest of living things, which unquestioningly follow their
animal desires {therion erotes) (67b) wherever theselike forces
of naturemay drive them. To be a human being means always
to be confronted with choices. As Aristotle puts it, human beings
"have" prohairesis (choice). They must choose. Having to
choose, however, entails wanting to know, that is, to know what
is best, to know what is good. And that means knowing reasons
why, knowing grounds, and using grounds to differentiate.
Socrates' partners in the discussion experience this: they learn
3. Gadamer maintains that Plato's myths are a sort of mirror in which we are
meant to recognize ourselves. (See ch. 2. n. 9, on self-knowledge and the Socratic
gnthi s*auton.) Their validity, accordingly, is not dependent upon any authorities to whom they might appeal, but upon their pertinence and accuracy in portraying the phenomena of human existence, and their efficacy in helping one to
achieve self-knowledge, TRANSLATOR.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 110

that concern for a life of justice and rectitude necessarily leads to

giving justification for the good.
Thus, the opening scene, in which the two ideals pleasure and
knowledge are opposed to each other, is already pervaded by a
latent contradiction. The way in which all living things blindly
submit to the immediacy of the pleasure principledriven as
they are by the hidden power of the life urgeis not the way for
human beings to fulfill their potential for leading their own life.
Hence, right from the start, the exposition of the question is such
that a contradiction must ultimately emerge, namely that here
one of the options supposedly open to choice can in fact not be
chosen at all. The blindness of the life urge, which prevails in everything, exists completely apart from any choice. The other
"choice" or option balancing the blind life urge is choosing
itselffor which one has already decided *as soon as one begins
weighing these two against each other*. And this choice presupposes knowledge. What makes human beings human beings is
the fact that they must ask about the good and must give preference to one thing over another {prohairein) in conscious, deliberate decision. In other words, they must give themselves justification {Apology 38a).
Protarchus outgrows his immature dependence on Philebus
and his partisan support of him. He develops into Socrates' candid partner. At the beginning he just goes along with the differentiating. At the end his thirst to know is such that he will not let
go of Socrates. Socrates leads the discussion through the differentiation of various kinds of pleasure and knowledge. The task
thenafter it has been settled that neither one of these two by
itself can constitute the good lifeis to "mix" the proper portion of each correctly in the right life. That, of course, is a metaphor. Life is treated like a potion in which various ingredients
are to be mixed to achieve blended, full-bodied tastiness. Se-

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

111

leering the right ingredients and combining them well obviously

presupposes the good as the standard to which one refers. That
is to say, one mixes with a view to the [harmoniousness and consonance] of the whole. The important thing is that we be able to
gather from the metaphor what makes the mixture "good."
We should not let ourselves be led astray by the metaphor. We
are, of course, not dealing here with a real mixture of substances
that are found separately from each other. Such talk is merely
the language of a simile, as is expressly stated at 59e. Both ingredients exist solely in logos (reasoning). Would pleasure be anything at all without consciousness, without our being aware of
it? And pure absorption in knowinginsofar as it is a total loss
of oneself to what is thoughtwould fall just as short of being
life. Even Aristotle's god "enjoys" his vision. Hence, here in
Plato, self-knowledge is concealed on both sides of this abstract
antithesis, that is, self-knowledge that alone, obviously, can
make both sides desirable in the first place. Given the hiddenness
of this self-knowledge, we also see why only a joining of both
sides, of knowing and perception on the one hand and pleasure
on the other, can display the sole thing in concrete human life
that is desirablethe human good.
From the beginning, something is evident here which is ever
more firmly established as the discussion proceeds, namely, that
the good life must consist in a third thing, a mixed genus in
which hedone and "limitless drives," on the one hand, and reason {nous) as the source of all measuring and measured restraint,
on the other, are both found. Certainly we should not take this
metaphor to mean that the mixture is real and can be brought
about properly by some kind of art (techne) of living. Unlike an
artisan who is smart at selecting his materials, we do not stand at
a distance from the components of our life, our drives and our
intelligence. On the contrary, we ourselves are both of these.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 112

Plato knows how to indicate this fact. He finds an ingenious way

to momentarily suspend his techne analogy of making a mixture.
He has Socrates insist that the question be put to both manifoldsof pleasure and of knowingto what extent each will
accept the other side. In truth they are both only abstract aspects
of the one life we really live, the life that is both; and our task is
to inquire how this one life unfolds itself (63 a).
There is an implication here, however: put in the language of
the likeness, if the good life is to be the most beautiful and the
most free of inner discord, what is good about it "good'' in the
same sense in human beings and in the cosmos (en fanthrpi
kai toi panti pephyken agathon)must be comprehended, and
also the idea (idea) of that good (64a). Even if one grasps the
mixture metaphor in this way as an image of real human selfknowledge, the question about the good, about its normative
role in this mixture, remains. Indeed, it becomes more acute than
ever. Consequently, we cannot escape having to articulate conceptually here what we call good in regard to this mixture, which
is to say, what we call good in regard to our concrete human
existence.
We now see the actual meaning of that mysterious doctrine of
the four genera. It is obviously an extension of the Pythagorean
doctrine of opposites, the peras (limit) and the apeiron (limitless,
indefinite), and it is introduced as such. But there is something
new: Plato, namely, is not simply a Pythagorean. On the contrary, he explicitly distinguishes the noetic world of numbers and
mathematical relationships from what is given in the reality of
concrete appearances. The latter he calls ^genesis^ (becoming).
Genesis does not identify still another form of ideal being *but
rather the "real" being of what comes to be*^. There "is" the reality of the genus of things mixed from the peras and the apeiron
[the reality of genesis] just as there "is" [the eidetic reality of ]

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

113

the peras and apeiron themselves (and just as there "is" necessarily a cause for the third, mixed genus). Not only hedone and lype
(pain) appear in this mixed genus of the "real" (31c). In a decisive passage it is emphasized that the good is to be sought here
too (61b). At issue, to be sure, is solely the good in human life.
The good in human life, however, is just as much the good in the
state and in the cosmos too. That the good is the same in all three
is confirmed when Plato reminds us of hygeia (health) and harmonia (harmony) and points out their cosmic relevance (for example, at 31c). The doctrine of the four genera thus proves to be
both the ontological preparation for, and prerequisite of, the
debate in the Philebus. Only when the mixture is no longer
thought of as a diminution and clouding of the pure, true, and
unmixed, but as a genus of its own, can it be the place where we
see how the being of the good and the true is constituted. In this
way we arrive at the metaphor of the potion of life: the way is
prepared for it ontologically by the differentiation of the four
genera of being.
This doctrine has far-reaching consequences for any appropri^:
ate understanding of Plato's dialectic and of the problems of
chorismos (separation) and methexis (participation). If limit and
determinacy do not exist apart, for themselves, then neither does
the entire noetic realm of the ideasany more than do the ingredients of this potion of life that is supposed to be mixed. That the
noetic world of numbers and pure relationships belongs together
with their dialectical opposite, the apeiron, implies that they are
only abstracted aspects of this third thing called the "mixed" {ex
amphoin symmisgomenon, meikton [combined from both,
mixed]) (23d, 25b). It is established expressly at 27b that our
Hfethis life, mixed from pleasure and knowledgebelongs in
the third genus. But that it does is, after all, virtually self-evident.
It was indeed difficult for Protarchus to grasp this third genus

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 114

and precisely because it is ubiquitous: "Its numerousness {plethos) startled you" (26c). Plato is pointing to a self-evident truth,
namely, the obvious fact that the particular participates in the
universal. After the confusions of a dialectic of making one many
and vice versa, a dialectic that ended in vacuity, the third genus
of the mixed now appears as the reality, or being, that has come
into being (gegenemene ousia) (27b). The fourth genus, the
"cause" of the mixture, makes clear that this third genus is a genus of its own and is not to be derived from the eidetic opposition of peras and apeiron, but is instead a special kind of being.
The doctrine of the four aspects of being developed here is a
universal ontological doctrine, which is to say, that it extends far
beyond the particular occasion for which it was introduced
herenamely, the question of the good in human Hfeand embraces the whole cosmos and its constitution. We may go even
further: nowhere in the entirety of Plato's dialogues are we as
close as we are here to Aristotle's parallel account of the two
principles, the one and indeterminate duality. If one starts here
in Plato, even something like a physicsthat is, an eidetic science
of what, in its essence, coming-into-being isno longer seems
completely impossible. Coming-into-being, becoming, is, after
all, becoming being. It is being that has come to be. Even so, the
Socratic question about the good in hiiman life is included here
too. Physics and ethics can still appear here, undifferentiated
from each other, as mere applications of the basic ontological
structure of the good. And the mode of discourse that is used
here to describe both these ways in which the good appears,
could, if viewed in relation to Aristotle's technique of conceptualization, be termed_inythical, A world whose origination and
determinate order are caused and executed by a master craftsman who possesses reaspn, or a human life whose ingredients are
knowledgeably and expertly combined into a blended potion by
an ideal drink-mixerthese are mythical metaphors. And it

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

115

seems to me that Aristotle's physics and ethics achieve a translation of them into concepts.
If the good is accepted as the cause of any mixture being
goodand ultimately this means, as the cause of everything real
being good (64d)the famous "beyond all being" {epeikena tes
ousias) takes on a new meaning. The good is no longer the one.
On the contrary, it is explicitly conceived of according to the
ideal of mixture and as having three aspects (syntrisi [in three together]). The dynamis (power) of the good has taken refuge in
the physis (nature) of the beautiful: measure and measuredness
constitute what beauty and arete are everywhere {^^metriotes gar
kai symmetria, kallos depou kai arete pantachou symbainei
gignesthaV) (64e).
We are far removed here from some esoteric, abstract, dialectical doctrine. It is stated expressly that all human beings know
what is meant (64d). For this reason there is no terminological
precision in this description whatsoever. Beauty, symmetry or
measuredness, and truth {aletheia) are named as the three structural components of the good, which appears as the beautiful.
Thus, in the intrinsic connection between the good and the
beautiful, which is brought out so emphatically here, we can see
an indication that "the good," which is at the same time "the
beautiful," does not exist somewhere apart for itself and in itself,
somewhere "beyond." Rather, it exists in everything that we recognize as a beautiful mixture. What is viewed from the perspective of the Republic (or the Symposium) as the pure unmixed
good or beautiful "beyond being" is here determined to be the
structure of "the mixed" itself. In each case it would seem to be
found only in what is concretely good and beautiful. And precisely the unity and integration of the appearance itself would
thus appear to constitute its being good. This thesis, it seems to
me, does not represent a change in Plato's teaching, a change
that would have led him to abandon the doctrine of ideas or the

T H E D I A L E C T I C O F T H E G O O D I N T H E PHILEBUS 116

transcendence of the good. It is still true that the good must be

separated out of everything that appears good and seen in distinction from it. But it is in everything and is seen in distinction
from everything only because it is in everything and shines forth
from it.
The Phaedrus already points us decisively in this direction
when its grand myth about the divine gift of ers (love) singles
out beauty as the only idea that preserves something of the
former lustrousness of an idea even after our plunge into this
earthly world. Beauty lights up here in our world. It shines forth
most of all, and it, most of all, stimulates love in us {ekphanestaton [esti] kai erasmitaton) {Phaedrus 250c). Thus it awakens in
the lover the longing and passion for what is higher. To say that,
of course, is not to provide a conceptual answer to the problem
of the participation of the particular in the universal. But it is
nontheless significant that beauty is singled out because it
"shines forth." For after all, that means that it is in the visible.
Actually the beautiful, as the thing which is loved, is surpassingly
"pure" beauty, It stands fully visible in its lustrousness. Being
beautiful (kalos) means in the first place being presentable and
refers to what can be seen in public. (Cf. the Philebus 65e, on the
ugly (aischron), which "we keep hidden out of sight" (aphanizontes kryptomen).) Thus, if we start with the Phaedrus, we can
understand what is meant when it is said that the good's capacity
to be, its potential (dynamis)^ is displayed in the beautiful. Of itself and according to its own nature {tauten esche moiran [having this destiny]) (250d), the good is appearance {erscheinen)^
Ughting up {aufscheinen)^ shining forth {herausscheinen^ or in
Greek, ekphainesthai)^^
4. This Heideggerian play on words is meant seriously. The root verb,
scheinen^ corresponds to phainesthai in Greek, whence the noun phainomenon.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

117

Thus, out of the rushing flood of the audible, the numerically

determinate relationship of the "pure tones" comes to the
foreharmony in music. Or, in their function as letters, determinate lines of script emerge from the whole of the visible, and
phonemes from the rest of what is to be heard, each as the special articulation of meaning that it is. Similarly, that which lives
is brought forth as an organic body (Philebus 29d), and the entire universe as a harmonic order. These are raised out of the limitless flowing away of mere genesis and are raised up into ousia
(being). Its having been raised to ousia constitutes the intelligibility (woMs), or dis-concealedness (aletheia)^ of the cosmic order. In
all these instances the eidetic-ideal can be discerned, picked out,
because it is contained in them {heuresein gar enousan [for (it is)
to be found existing therein]) (16d).
That the good has sheltered itself in the beautiful thus means
nothing less than that it is to be found only in the beautiful. Measure, symmetry, and openness to view [aletheia) characterize the
beautiful. To this extent the beautiful is at the same time the
good, which provides everything that is with its true being
with the being, namely, that we have called here eidetic-ideal.
The dynamis of the good had already been spoken of in the
Republic^ and it was also said there that the good provides everything with its aletheia. Here the good "appears" precisely as the
beautiful. It is in no way separable from that which it is in each
instance. That it is not is vouched for by Plato's language. The
language of the Philebus^ to be sure, might indeed be nonterminological. Still, I would call to mind how the puzzling nature of
dynamis is characterized in the Republicnot as existing apart
Phainomena are not "mere appearances," but the ways in which a thing really
displays itself. Compare SZ, pp. 28-31, on phenomenology as the art of letting a
thing show up, forth, and so forth, in discursive exposition, TRANSLATOR.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 118

for itself, but "in that in which it exists and which it effects"
{eph' hoi te esti kai hoi apergazetai) {Republic 477c-d). That
holds universally. Hence, one must look for the dynamis of the
good in the manifold of what the dynamis of the good brings
aboutas, for instance, the dynamis of seeing consists in the
manifold of sights and nothing else. In conceptual language, that
means that we are dealing here with the inseparability of the one
from the many. True reality, or "being," is one but nevertheless
in all the many things. And that means that it is separated from
itself "which, however, seems to be the most impossible thing of
all" {Philebus 15b). Since it is one and the same in many things
that are separate from each other, it is simultaneously in them
completely, and hence it is separate from itself. This circumstance is the seemingly nonsensical state of affairs with which
Socrates is confronted in the Parmenides. He seeks a way out by
referring to the light of day, which is everywhere at once and yet
one and the same, not separated from itself (131b). In the Parmenides^ of course, he does not succeed in keeping a proper hold
on his thesis. He is still too young. But in the Philebus, Socrates
characterizes precisely this problem as the source of all perplexity {aporia) if it is not properly allowed for, and the way to all felicitous advance {euporia) if it is {Philebus 15c).
And in fact the felicitous, good way of reaching an understanding, which the discussion in the Philebus traverses, gets
completely beyond the danger that Socrates had warned against
at the beginning, namely, the eristic tricks of sophistic pseudodialectic. This sophistic dialectic is not real thinking, for in pursuing it one succumbs to the blind desire for success in contentious argument: ^hyph' hedones enthousiai te kai panta kinei
logon asmenos"" (enthused with pleasure he delightedly sets
every sort of argument in motion) (15e).
To be sure,-the way toward reaching an understanding and

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

119

making sense of things that the discussion pursuesthe path of

first differentiating and then finally coming to a decision by
weighing what is to be allowed and what rejecteddoes not fit
all that well with the theoretical reflection that Socrates initiates
via this path. The examples that he uses, music and letters, have
to do with differentiations of voice or, better said, articulations
that constitute a system. All singing and all speaking are based
upon the intelligible lawfulness of the respective musical and linguistic uses of voice. Insight into this lawfulness is techne. But
what is produced on the basis of this techne or skillthe musical mimesis (imitation) or the speech content, say in persuasion,
instruction, or poetrycould surely be brought about by some
other, quite different techne too. For example, one might think
here of the true rhetoric in the Phaedrus^ as opposed to mere
"grammar," this general skill which, in Plato, hovers in a remarkable way somewhere in between the articulation of letters
and the articulation of the spoken voice. Grammar is presupposed in any discussion (and in any dialogue that is given literary
form), and ultimately in any thinking at all that differentiates. In
this sense it also plays a valid role, of course, in the search for the
right life portrayed in the Philebus. But those in search of the
right life are not on the good dialectical path to reaching a genuine understanding about the good just because they have
learned to speak or write. The path one takes in search of the
truth is plainly dialectic in a different sense from the dialectic of
the arts of speaking or writing.
The two senses of dialectic may have something in common,
namely, the division of a one into a determinate manifold that is
itself eidetic-ideal. In the language of logic we would speak here
of the unit of the genus and the manifold of species and subspecies in which this generic unit is specified. Also, one might grant
that the number has a general paradigmatic function [in both

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 120

cases]. After all, the art of differentiating only reaches its goal
when one finds no more specifiable unitstones, phonemes,
and so forth. So one might see in all dialectical division of a one
into many a certain approximation to Plato's coordination of
idea and number.
Differentiation takes place here within the noetic one, and it
is the principle of number that the Philebus introduces in this
context as the truly illuminating Promethean fire. Here, the Pythagorean heritage, the identification of being with number, is
explicated on the new level of noetic being. In this way the multiplicity that the one contains receives numerical determination. It
is many but not indefinitely many, rather so and so many. The
numerical determination of what constitutes tones and tonal relationships in music, this ancient Pythagorean inheritance, has
its correlate in the ideality of language and writing, both of
which articulate the whole of human phonemes and, by doing
so, put them at our disposal.^
Hence it is an eidetic-ideal structure, a relationship of ideas,
that underlies the knowledge and skill in which any techne consists. This ideality certainly does not eliminate techne's relationship to the production of something in perceptible reality, for example, tones and music or articulated speech and what is fixed in
writing. But the thing produced in this way retains a special kind
of ideality itself. It is a world of signs and indices that directs us
to the ideal. Accordingly we are not dealing here with just any
particular handcraft among many others, one that Plato would
say is less a techne than a mere routine (tribe). Rather, we have
here two arts which were later called "free" (liberales) because
5. Even within the book of aporiai at Metaphysics, Beta 4,999b30, the ideality of letters becomes clear, and this ideality entails that they have an ideal universality v^^hich takes multiple forms in the individual instances to which they are
applied^

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

121

each in its own way is subordinate to no particular aims, and because each is so comprehensive.
Now the mixing of the potion of life also has an inclusive, universal aspect. The ingredients, which one after the other, are
found acceptable for the mixture, have something to do with
number insofar as the examination and testing of them is supposed to be comprehensive and exhaustive, *that is, is supposed
to include just the right number of them's-. Nevertheless, the
thought here cannot be that one learns how to live in the right
way, and is finally capable of it, in the manner in which one
learns how to sing, speak, or write.
Or should we say only that one leams about the right and just
life in the way one learns about giving justification for something
and in the end is finally able to give it? But that is just what the
discussion [in the Philebus] teaches us: dialectic is not a techne
that one learns like writing, not something that others (illiterates) cannot do. Thinking, to be sure, is an art, but an art that is
practiced by everyone and that one is never finished learning.
And how to live is just as little an art that one could ever be finished learning. Futhermore, right thoughts about Ufe and the
idea of the right and just lifethe highest thing that one could
learn {megiston mathema)only become visible in general
outlines *and not in regard to specifics (cf. Aristotle, EN
1098a21)'^. Aristotle knows that the theoretical reflections that
he calls "ethics" have to be of use in Hfe as it is actually lived.
Similarly, it is clear to Plato and to the reader of the Philebus
that what results from this dialogue, the ideal of a life harmonized rightly, isprecisely as the result of dialoguea logos
(statement in words), which directs us to an ergon (deed), to
choosing what is right in the moment of choice.
This is not the place to pursue the intrinsic connections between the dialectic of the one and the many and the doctrine of

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 122

the ideal numbers. Of interest to us here is only the fact that human life, just as all other being, belongs to the mixed genus, and
that what is called "being-good" appears in the reality of what is
mixed. That it does so must mean that everything that exists has
reality only in its concrete determinacy. And that means precisely that it is set in, and surrounded by, the unlimitedly variablegenesis. Similarly the conduct of human life that is
guided by practical reason, also has the good in it only insofar as
the good is concretized in the actual doing of it, that is, in giving
preference to one thing over another (prohairesis). That would
mean too that any deed, to the extent that it is decision, always
includes a component of uncertainty, for it must move in an element that exceeds all determinacy and delimitation, which is
therefore called ^'apeiron" (indefinite).
Consequently, human life is eo ipso dialectical.^ It is one and
many at the same time. At every moment it is itself and, exactly
for that reason, separated from itself, just as the "what-it-is" {ti
estin) of every existent thing ultimately exists in such a way that
it is in all that participates in it {to metechon). The aporiai for6. The use of the word dialectic in Gadamer shifts. Here the reference is to
the inner tension in human existence between order and disorder, the rational
and the bestial. Our task is to maintain unity of self, integrity, within ever threatening disintegration into boundless chaos. Thus we must be constant in holding
to one thing (Kierkegaard) through the vicissitudes of our life (cf. PD and ch.
1, n. 22) Dialectic in this sense also has to do with dialogue insofar as the Socratic
"art" of leading a discussion is an "art" of keeping it from getting lost in the
indefiniteness of many things that are not important and of holding to the one
thing that isthe one subject matter under discussion. Gadamer shows that
phronesis is the requisite virtue for both these forms of constancy in holding to
one definite thing within threatening indeterminacyin moral practice and in
discussion. And it is the "dialectical" nature of its content in both these applications that distinguishes phronesis from any techne whose content is a systematic
whole that we could ever be finished learning, TRANSLATOR.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

123

mulated in the Philebus are to be taken literally. They are not

contradictions but the path taken by thinking itself.
Though the perspective is somewhat different, this doctrine of
human existence, it seems to me, is also reflected in the Statesman^ a doctrine which so decisively preempts Aristotle's criticisms. In the Statesman a distinction is made between a relative
art of measurement [Messen) and an art of measurement that
takes what is fitting [angemessen) as its measuring rod. The relative art of measurement knows only what is more, relative to
what is less or vice versa. The true art of measurement, for which
what is "fitting" is the measure, knows a "more" that is not only
a "more" relative to a "less" but is really [hos on) more [Statesman 284a). With such knowledge, this art of measurement
brings about what is fitting: "pros ten tou metriou genesih" (in
regard to establishing what is fitting) (284c). Even without going
further into the context in which the distinction between these
two occurs, we can say that, viewed ontologically, "the fitting"
[to metrion) here is what is called "the mixed" in the Philebusnamely, everything that has to do with what is fitting
or appropriate, with the right moment, the obligatory [to
deon)in short, with what is in the middle between the extremes (284ef.). Here one finds precisely the fundamental concepts of Aristotle's ethics. Though we must concede that the Socratic concern for one's own soul is extended by Plato's Socrates
into the realm of the political-utopian and cosmic-universal, we
see, nonetheless, that the Socratic question also lives on even
here where the universal doctrine of ideas and the universal nature of dialectic are under discussion. That, more than anything
else, is what the Philebus teaches us.
One must view what results here in relationship to the "transcendence" of the good that the Republic emphasizes so

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS 124

Strongly. Though it is not easy to grasp just how it happened,

that which makes all good things good found itself expelled
there from the ranks of what exists. It is not one thing alongside
of others; "it itself," auto to agathon (the good itself), withdraws. The good is the being of the ideas generally and not an
idea itself.
I hope to have made credible that this way of putting things is
simply the mythical form in which Plato expresses essentially the
same thing that he says explicitly in the Philebus when he states
that the good "appears" in the beautiful. And I would like to
show that the problem that Plato's mythical question about the
good contains is the same one that Aristotle later singles out as
the problem of analogy or the analogia entisJ The transcendence
of the good precludes thinking of it as an ideaa ti estinthat
would constitute the highest, all-comprehensive genus, so to
speak. But if this is so, neither the being of the good nor the being of any ti estin needs to be mediated with what exists in order
to be Uke what existswhatever form that mediation might
take, for example, specification, diahairesis, or any other form of
dividing up the whole. It cannot be mediated in this way at all; it
appears in existence /mmediately. That is the meaning of the
statement that the good takes refuge in the beautiful. Thus, to
my way of thinking, the result of our investigation of the special
status of the idea of the good in Plato's works would seem to be
that Plato's so-called self-criticism, as one is wont to interpret his
Farmenides^ is a criticism that we would be better off applying to
ourselves. In the final analysis, our wanting to think of the participation of existent things in being as a relationship of existent
things to each other always involves us in a false concretion. In7. Compare E. Frank, Wissen, Wollen, Glauben (Zurich, 1955), pp. 8 6 119.

T H E D I A L E C T I C OF T H E G O O D I N T H E PHILEBUS

125

Stead we would do better to acknowledge from the start that this

participation is the point of departure for all meaningful talk of
the idea and of the universal.
Whoever seeks to conceive of the transcendent nature of the
good cannot think of it as "a good." On the contrary, he will
have to consider the three levels in the order of reality: the soul,
the state, and the worldan order that is explicated mythically
in Plato's dialogues, above all in the Republic and the Timaeus,
It is not their beauty alone that constitutes the unity of unity and
multiplicity which appears visibly in these things. Rather, it is
the dynamis of the good, which holds everything together everywhere and gathers everything together into a unity.

A R I S T O T L E ' S C R I T I Q U E OF
THE IDEA OF THE G O O D

But how, then, do things look with regard to Aristotle's critique

of the good? Is there still any real object for his criticism? And
what about his critique of the doctrine of the ideas? Let this last
question be our guide, and let us scrutinize Aristotle's critique in
light of the insights we have gained in our study of Plato. On
methodological grounds, let us leave aside the reconstruction of
no longer extant textsin Plato's case, the reconstruction of his
forbidding lecture "On the Good." Instead, let us confine ourselves to the three ethical treatises found in the corpus of Aristotle's works. As far as these are concerned, we do not even need
to involve ourselves in the question whether all three are authentic. Especially after Dirlmeier's commentaries on the Magna
Moralia and the Eudemian Ethics, it may be considered certain
that we are dealing here in any case with an authentic legacy of
Aristotelian thought: the concordance of these three texts in
their three critiques of Plato is almost overwhelming.^ And cer1. F. Dirlmeier, translations of, and commentaries on, Aristotle's Ethica
Nicomachia (Darmstadt, 1956), Magna Moralia (Darmstadt, 1958), and Ethica
Eudemia (Berlin, 1962).
2. Recently, the strikingly didactic style of MM has been brought to the fore
by Brink, Theiler, and Dirlmeier, and rightly so. It seems to me that the simplest
hypothesis herea hypothesis that also accounts for certain stylistic peculiarities
and obscurities in the composition of the workis still that we are dealing with
126

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

127

tainly the texts themselves dictate our general methodological

principle: namely, to bring things in common to the fore, to emphasize these things in common over the differences, and to use
each text to elucidate the other two. Moreover, our principle enjoins us not to concern ourselves with attempts to ascribe individual arguments to specific "Platonists" (cf. the research of
Arnim, Gigon, and others). For again, the truly significant thing
is not so much the divergence of these arguments from one another but the common character of the argumentation that persists throughout all variation and that all three of Aristotle's critiques evince. In individual cases it may well be debatable
whether everything that is weighed and subjected to critical refutation in them is actually founded on Plato's own teachings, or
whether some of it derives from later speculative elaborations of
these teachings. But since Aristotle does not seek to give a reliable account anyway, and since his critical aims as such must always be taken into considerationas the works of Cherniss,
more than any others, have convincingly demonstratedour
approach can only be to evaluate Aristotle's assertions starting
with the set of problems implied by the subject matter in Plato
himself, that is, to evaluate Aristotle's assertions from the perspective of those problems. For our purposes we need not concern ourselves with the question of just who it might have been
who developed the individual arguments.^

some sort of copy by someone else, that is, not with the text of the "author," but
with carefully reworked class notes. In particular, the "logical concern" of the
text, of which Dirlmeier correctly takes note, can be easily explained as a consequence of the lecture's skeletal treatment of its topic. And many "deviations" in
the line of thought in MM [from EE and EN], for example, the delayed introduction of eudaimonia, would not be all that surprising if we are dealing with a rendering of a live lecture which was not intended as a text to be read.
3. Is MM 1182a25 ff. an allusion to the lost lecture on the good or, rather, a
polemic against the Republic} Can apedken hekasti (assigned to each [part of

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 128

Also, let us keep away from any hypothesis about the relationship of these three ethical treatises to each other. We find the critique of the idea of the good at the beginning of all three, and in
all three that critique is aimed precisely at the universal ontological claim which Plato makes *for his idea of the good'^. In all
three the decisive argument is that knowledge of such a good can
have no relevance for the philosophy of human practice. In the
Magna Moralia (1182a25 ff.) Plato is criticized precisely for having introduced the question of arete (virtue) at all into his universal ontological doctrine of the agathon (the good)as he did in
his famous lecture "On the Good." According to this argument,
for Aristotle, the Socratic question about arete would be wholly
incompatible with the universal question of the good, to which
Platonic dialectic is addressed. Here Aristotle in fact raises the
very same question that we put to the text of the Republic^
* namely, how the Socratic question about the good and the ascent to dialectic through and past the mathematical sciences relate
to each other. In what follows I intend to show that Aristotle, the
creator of physics and founder of practical philosophy, holds
fast to the Socratic heritage in Plato: the good is the practically
good. On the other hand, as the creator of physics, Aristotle also
fulfills the demand made by Plato's Socrates, that is, that we understand the world starting with the experience of the good. The
good thus appears in Aristotle's physics as well as his practical
the soul]) be anything other than an allusion to book 4 of the Republic} Above
all, the phrase at 1182a27, kai synezeuxen (and coupled [arete and the good]),
speaks for such an understanding of the allusion. In substance the critical observation would then amount to a rejection of Plato's having extended the so-called
four-book Republic, Or perhaps one must read the note as follows: Plato's lecture on the good would have been just fine had he only omitted all references to
the aretai. That would fit well with the account of Aristoxenus. To be sure, I am
of the opinion that in this case no one other than Aristotle himself said this, and
that he was attempting to reverse popular expectations with such sarcasm.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

129

philosophyin his physics as the hou heneka, in his practical

philosophy as the anthrpinon agathon. This double function of
the good can be demonstrated using the text of his critiques of
the idea of the good. Thus one sees that his philosophy of human
practice remains embedded in the whole of his conceptualization
of reality.*
Aristotle's restriction of his inquiry to the concerns of practical philosophy does not silence the question of just what constitutes the common property in all being-good. For it is surely not
just a superficial equivocation that such a diversity of things are
called good. What the Nichomachean Ethics states expressly
ou gar eoike [to agathon] tois ge apo tyches homnumois ([the
good] is indeed not like those things that only have the same
name by chance) (1096b26)holds all the more for the other
two treatises. The Magna Moralia comes close to granting the
argument that what is most good, to malista agathon^ the good
itself, must be an idea, which is to say, the good of all ideas:
alethes men estin iss... (that is perhaps true...) (1183a32). And
the mutilated closing sentence of the critique in the Eudemian
Ethics virtually seems to demand the investigation of the multiple meanings of agathon with an eye to the ariston pantn (the
best of all things), once the ariston tn praktn (the best of the
practical) has been treated (1218a25).^ In any case, the author of
4. To be sure, in emending the text Dirlmeier attempts to eliminate the
posachs (in how many senses ["best of all" is said]) and, with that, the allusion
to the universal problem of the good. But in regard to 1218b4 ff., which in its
content is close to 1217a31 ff., I find that unacceptable. In these passages either
agatha (goods) or a part of the agathon are mentioned, which are not prakton.
Of course I do not know if Aristotle himself wrote skepteon posachs (it remains
to be investigated in how many ways . . .] in this spot after the concluding
statement, which, together with the addition in question reads: "to d* hs telos
agathon anthrpi (esti) kai to ariston tn praktn, skepteon posachs to ariston
pantn^ (the good for human beings [exists] as the goal and the best of prac-

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 130

the Eudemian Ethics also has the broad sense of a comprehensive agathon very much in mind. Thus we can estabHsh that in all
three treatises Aristotle does not limit himself to what for him is
the decisive argument concerning the practical relevance of the
idea of the good. Instead, he finds himself forced beyond the
confines of his theme of practice.^
ticesit remains to be investigated in how many ways "the best of all" [applies]).
The last part could have been appended by an editor who wished, after this summary, to remind us of the larger universal ontological horizons here. But in any
case, in our text {1218a25 ff.), the contrast between to ariston tn praktn (the
best of practices) and to ariston pantn (the best of all) is manifest.
Here, a general observation concerning the status of our texts is in order.
Gigon likes to speak of an editor but leaves open the question of whether this editor was Aristotle himself. He is entirely right insofar as one may not always apply
the yardstick of stylistic consistency appropriate to a literary exposition of doctrineeven to the ethical treatises. But his predicates (negligent, careless, imprecise, and so forth) go too far. One should not allow oneself to be deceived in such
fashion that one mistakes single locutions, often subtly and cleverly formulated,
for a text. To be specific: the composition is indeed often "careless," or better
said, dependent on the kind of care that would be given to it in oral presentation.
5. Things like this occur elsewhere in Aristotle. For example, his definition of
physics ought, strictly speaking, to prohibit any discussion of Eleatic philosophy
whatsoever within the framework of physics, for Eleatic philosophy, after all,
denies the existence of motion altogether. Nevertheless, he inserts his critique of
the Eleatics in his lectures on physics (Alpha 3,4). In the case we are considering
the definition of practical philosophy ought, strictly speaking, to preclude a detailed discussion of Plato's idea of the good. Nonetheless, he takes it up, even if
he constantly points out that it actually belongs in another context. It would
seem that he is somehow uncertain of just where to find an appropriate place for
such things. Ultimately one must say, then, that such uncertainties, which often
occur in Aristotle, reflect the larger uncertainty we call "metaphysics." The stack
of papers that later received this name is, so to speak, a collection of uncertainties
that share this character of being marginal. Once one is clear about this fact, it is
no longer very puzzling that the start made in Book Gamma, which gives the impression that the science to be pursued is a formal ontology, does not quite fit
with what is begun in Books Zeta and Eta (the doctrine of substance and of
dynamis and energeia) and in Book Lambda (the so-called theology), and that the
book of aporiai. Beta, stands by itself in a peculiar way. The fact that Book Epsilon ultimately provides a certain editorial harmonization of what precedes it,
does not contradict what I have said here.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

131

The category argument thus finds a place in all three treatises.

Of course, it is particularly well suited to repudiate the claim
that a universal science of the good is relevant practically. But at
the same time its very usefulness here suggests that the problem
of the good remains indissolubly tied to the problem of being. In
regard to being, there is a genuine question how the various
senses of the word, that of "substance" and those of the other
categories, are related to one another. And the same would 3eem
to hold necessarily for being-good. As a matter of fact, in regard
to both being and the good, Aristotle directs us to the problem of
analogy (analogia). So he is not at all blind to the universal ontological question of the good, despite his critique of Plato in the
three ethical treatises. It is only the practical concern that prevails in his ethics that allows him to dodge the problem in a facile
way, butand this is the pointnot completely.
Aristotle takes up Plato's doctrine and the refutation of it most
extensively in the Eudemian Ethics, Hence it serves our purposes
to fix our sights on that work, not, of course, without bringing in
both the other treatises as well for clarification. Let the results of
our study of Plato be our guide here. It emerged there that in
Plato too, "knowledge of the good" was a special kind of knowing "beyond" the "sciences," a mode of knowing that had a different epistemological structure from techne. Unlike the latter, it
does not attempt to draw inferences from given presuppositions
but consists, instead, in giving justification {logon didonai) of
the highest telos (goal or purpose), that is, justification in the
sense of dialectical exploration and elucidation of what Hegel
calls the "concrete universal." Do we not find the good in everything outside of human praxis (practice), as well as in what is
prakton (of practical concern)? Is it not the immanent measure
(the metrion of the Statesman) in the soul, the state, and the
world? And precisely in regard to the idea of the good, is not talk
of the chrismos (separation) especially misleading?

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 132

We can expect that the critical purposes for which Aristotle

wrote his ethics will prevent this way of seeing things from
surfacing by itself. The opposite clearly holds: he must play
down the "transcendence of the good," which, if he did not,
would set it apart from all ideas. He must put the idea of the
good in the same class as the other ideas. Consequently, he must
be particularly emphatic in insisting that *like other ideas* the
idea of the good exists for itself separately (choriston).
One must keep in mind that the concept choriston has two
aspects in Plato and Aristotle. On the one hand, it refers to a
thing's being separate and, on the other, to its consisting in itself
{In-sich-stehen). If one starts with the latter, Aristotle's ontological deviation from Plato becomes understandable at once. It is
not the ideas that consist in themselves but rather the physei onta
(things which are by nature), and, ultimately, the highest existent
thing, the god. For Plato it is precisely these things that do not
exist for themselves, but rather, only the ideas. The divinelike
the goodis beyond being {epekeina tes ousias)^ in a sense that
prohibits its being called an existent thing. For Plato, eidetic or
noetic constructs, for example, numbers, lines, and so forth, are
to be separated from phenomenal existence, not fused with it as
the Pythagoreans held. For Aristotle, the physei onta are inseparable from their ti estin (what-it-is). That is the meaning of his
doctrine of primary and secondary substance [ousia). But that
means, conversely, that the eidos is not to be separated from its
phenomenal appearance and, thus, that it is an enhylon eidos
(materialized form). And to this extent not only the todi ti (thiswhatever), but also the ti (what) is "separate" from all other
symbebekota (accidents). That, too, is a fundamental departure
from Plato. What methexis is in Platoa being together [koinnia (coupling)] of ideas, for example, of human being and
whitenessis in Aristotle predication that refers to a subject.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

133

the synholon (underlying whole). Still, the question remains: despite these differences,* is it not possible that what Plato truly intended becomes visible in Aristotle's discussion, nevertheless
against the latter's will, as it were?^
6. Gadamer grants that this chapter in particular often needs to be filled out.
The argument as stated here, for instance, is somewhat compressed. As I understand it, the line of thought that Gadamer is pursuing is as follows: Aristode
wishes to include the idea of the good in his general critique of the ideas, which
argues that Plato unnecessarily doubles the world by postulating that the ideas
have a christon reality apart from the things that they inform. For Aristode, being christon is indeed a characteristic of anything that is; for anything that is,
exists for itself and in itself, which is to say, exists as "what it is" {ti estin) apart
from changing accidental predications (the kategoroumena or symbebekota). A
horse is a horse in itself, apart from (christon) being brown or old, here or there,
next to this or smaller than that. But, according to Aristode, Plato is guilty of a
misplaced concretion, so to speak, insofar as he assigns precisely this being christon to the ideas themselves, as if they too were realities, while in fact only a thissomething {tode ti) is real and christon. When Aristode comes to the idea of the
good in Plato, he finds the same mistake that Plato makes with all the ideas:
Plato, he says, treats the good as if it were a thing in itself, and that leads to an
empty abstraction: "We say first, then, that to say there is an idea not only of the
good, but of anything else whatever, is to say something abstract and empty"
{EE 1217b20).
This criticism, of course, accords with Aristotie's overall project, but it must
systematically suppress the fact that Plato himself argues that the good transcends all existence, that is, that it is epekeina tes ousias. As Gadamer has shown,
despite all the differences, Plato's aim is to make just Aristotle's point: the idea of
the good is precisely not another thing alongside things that are good; rather, it is
the structural order in any thing that is good. We call the good, insofar as it exists, the beautifula shining forth in things, an appearance. Aristotle's intended
criticism thus actually reinforces Plato's point.
One should not overlook the fact that, for the purposes of his critique, Aristotle shifts the weight of christon ever so slightly. As Gadamer has shown
above, christon in Plato implied independence from contingencies, as circularity
is independent of the aberrations that may occur in any particular circles we may
draw (see "Dialectic and Sophism in Plato's Seventh Letter^'), Aristode, too,
would not deny that what a thing is in essence is separate, or distinct, from the
variable things that might be said of it.
But since his starting point is the living thing (tode tt)^ not mathematics, he also

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 134

In interpreting the idea of the good, the Eudemian Ethics^ Alpha 8, begins with two interpretive formulations of the idea of
the good, which, in tying into Plato's assertions, are as faithful to
him as one could possibly imagine. The first states that the good
is that which is first {proton)^ the negation (anhairein) of which
also negates everything else [to heteron) that follows from what
is first and is therefore "good" *just as when there is no longer
such a thing as line, for instance, there can no longer be a triangle composed of lines*. Although it does not occur in the dialogues themselves, this formulation has an undeniably Platonic
stamp, as P. Wilpert, in particular, has demonstrated.^ Konrad
Gaiser, whose argument we touched on above, has convincingly
shown that tfie system of the mathematical disciplines is a sequential order of number, point, line, plane, and solid, and he
takes this mathematical system to be a kind of schema for Plato's
overall systematization. It is obvious that the numbers stand or
takes christon to mean "unto itself" in the way that a thing exists self-identically apart from other things. Thus the chra (space) in christon is much more
present in Aristotle's understanding of the word, albeit in a transfigured sense. As
a "physical" thinker, Aristotle has an ear for this chra, or "spatial" separation,
of whatever is christon.
Now if, like Aristotle, one starts with this sense of christon and then applies
the word to the ideas, it does indeed appear that there is a misplaced concretionalmost as if the ideas were said to be things "spatially" apart from the
things that participate in them, and hence Aristode's argument that Plato needlessly doubles the world. Gadamer's point, however, is that thinking of the ideas
as christon in this sense is obviously every bit as much of a mistake in Plato's
eyes too; in fact, it is precisely the mistake in which the young Socrates gets
caught in all his attempts to defend the ideas in the Parmenides. There can be no
doubt that Aristotle knew this. Hence we can only assume that he consciously
slants Plato's thought in order to better articulate his own "physicalist" position;
in fact, he marshals Plato's very own argumentsagain consciouslyto attack
the christon idea (for example, the "third man"), TRANSLATOR.
7. P. Wilpert, Zwei aristotelische Frhschriften ber die Ideenlehre (Regensburg, 1949).

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

135

fall with the oneand, of course, with the two as well. In our
text too, something follows that concurs with this line of argument. Aristotle says {EE 1218al5 ff.) that we may not deduce
the good from the numbers but instead from what everybody acknowledges to be good. Conversely, at the very most, we can
conclude from the being-good of types of order in things (such as
health or harmony of the soul) that the numbers too, on account
of their ordered structure, are good in a certain sense. The numbers are then described as "striving" toward the onea metaphor that Aristotle in his accustomed manner, takes literally {EE
1218a22 ff.). Here too, given my purposes, the concern is not to
establish whether this doctrine is the special teaching of some
Platonist. Even if it were, it would still be a conclusion drawn
from Plato's doctrine of the good and the one, a conclusion that,
in the context of our investigation, has to be of interest to us.
The principle of the proton^ which the Eudemian Ethics takes as
its point of departure, is in any case palpable in the numbers. At
this point a question forces itself upon us: if the numbers have
the central function suggested here, how, exactly, do things
stand with regard to the chrismos of the good? And how do
things stand with regard to the chrismos of the ideas if the ideas
are numbers? After all, are not the numbers in the things? (Again
I refer the reader to the Philebus 16d.) And is not the one, which
resides in all numbers, each being a manifold of ones, actually
"separated from itself" while in them? If one interprets the ideas
8. Instead of agathon^ Aristotle says kalon here (cf. Metaphysics 1078a31 ff.).
But surely he does so solely to reserve agathon here for what is prakton and to
avoid misunderstandings. This, it seems to me, is a minimal terminological differentiation, which Aristotle makes for his own purposes, a modification that accords with the close concatenation of agathon and kalon in Plato's language usage. Hence I would not ground any hypothesis concerning divergent teachings on
the passages bearing on this one that Dirlmeier carefully assembles, and certainly
not any hypothesis concerning Aristotle's "development."

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 136

as numbers, this puzzling and deliberately contradictory formulation of Plato's (cf. Philebus 15b: auten hautes chris^ and
Parmenides 131b) becomes understandable at once: the one-ness
of the one is both for itself and in the numbers.
That brings us to the second formulation, which may count as
good Plato if anything may, and which is solidly documented
precisely in reference to the idea of the good as well as the other
ideas. It is the general formulation for "participation" {methexis), which is also used elsewhere in Plato in speaking of participation in the ideas. Here it is applied to the good, whose presence (parousia) is said to cause all things that are good to be
good (EE 1217b5). Precisely this formulation is used to introduce the idea of the good in the Republic (book 6, 505a). That
the good is the most important subject matter [megiston mathema) because it embraces everything is advanced as a kind of
self-evident argument. It will be recalled, of course, that presence
{parousia)^ participation {methexis)^ and similarity [homiotesY
are always only metaphors, which the young Socrates of the
Parmenides cannot succeed in conceptualizing when the old
Parmenides locks him in his Socratic grip. If one wishes to understand the aims and limits of Aristotle's critique of Plato, one
must constantly keep this fact in mind. Aristotle must have been
aware of it when he repeated the very argument here that, as
Plato himself had shown in the Parmenides, leads to an insoluble
problem {aporia) and an absurdity, that is, the complete separation of the ideas from the appearances.
As I have shown, indications of a substantive answer to the
question of what is there, what is present, when something is
9. Homiotes, incidentally, is the preferred expression in Diogenes Laertius's
account of Alcimus, whose ties with the old academy Gaiser has argued for convincingly {Rheinisches Museum fr Philologie^ Frankfurt, 1975).

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

137

"good," are most likely to be had if, in some vague way, we take
the triad of metrotiy symmetron^ and alethes (measured, symmetrical, and true) that constitutes the beautiful in the Philebus^ to
be what is first (proton) and gives oneness to things. In any case,
the structure of Plato's Republic also implies that somehow the
good is the one: the "oneness" of the ideal city is a unified order
of such a nature that neither strife nor disturbance can occur in
it. That the good is what is first and gives oneness to things is
also implied, incidentally, in Aristotle's critique in the Nicomachean Ethics. Aristotle's express praise there of the Pythagoreans for simply putting the one in the series of good things and
consequently, unlike Plato, not equating the one with the good
{EN 1096b5), obviously presupposes that for his part Plato did
think of the good as the one. Of course, Plato's one is not at all a
Neoplatonic hen (One). On the contrary, in the Parmenides insoluble puzzles are displayed in the concepts of being and oneness in order to establish the dialectical unity of the one and the
In returning to Aristotle's introduction of his Plato critique in
the Eudemian Ethics^ which we discussed above, we can now see
that there Aristotle is striving for the least metaphorical exposition possible of what is meant by the idea of the good. Being primary among all those things that are good and being the cause of
everything other than itself by virtue of its presencethese are
obviously two aspects of Plato's methexis metaphor. First, as the
10. The critique in the Metaphysics, Alpha 6, introduces Plato as a Pythagorean and works him into the doctrine of the arche in the same way as in other
places in the first book of the Metaphysics. In contrast, Alpha 9 exceeds this
framework and, for its part, fits perfectly in the arrangement of the doublet of
this passage in Mu: aistheta mathematike-megethe-ideai-arithmoi (perceptiblemathematical-magnitude-ideas-numbers). How the two chapters fit together,
and how the chrismos critique in Alpha 9 is materially connected with the arche
doctrine in Alpha 6 are, of course, not explained by this observation.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 138

principle of number, the good is primary logically. That, as we

saw, is a genuinely Platonic line of argument, which Aristotle often mentions under the heading of the logical term synanhairein
(suspension or negation of the species by suspension of the genus) (cf. Topics 141b28; Metaphysics 1059b30 and elsewhere).
Second, the good is the highest presence (eminently good) and,
precisely for that reason, the cause of everything good being
goodvia participation. It is striking that Aristorie, in dealing
with this second aspect [of Plato's methexis metaphor], only pursues the chrismos argument incidentally, an argument which,
given Plato's Parmenides with its decisive repudiation of the
chrismos, is not without problems of its own." That is to say,
he places the idea of the good in parallel with the other ideas
* insofar as both are said to be "separate" (christon)*. Aristotle
evidently senses full well that the good, which is the object of his
concern here, does not quite fit among the ideas taken as the genera of species. (Ultimately, this disparity also holds for being
itself, and it will be recalled that the highest determinations supervening upon being in the Sophist are only called "species" of it
in an improper sense.) Hence Aristotle does say of the idea of the
good: ''kai gar christen einai... hsper kai tas alias ideas'" (and
it is indeed separate for itself . . . like the other ideas) {EE
1217b 15). But that is a questionable argument. After all, Plato
singled out the good as above the level of realities {epekeina tes
ousias). Aristotle deliberately ignores that here and is emphatic
11. It might be recalled that the Parmenideswhich constitutes a sort of
conerstone in my argumentdoes not stand by itself. The Sophist, Philebus^ and
the Statesman attach so little value to the chrismos of the ideas that some people, in fact, have even attempted to find in them a turning away from the doctrine
of ideas. Of course, that creates difficulties in regard to both the Philebus and the
Timaeus, and above all it makes the enigma of Aristotle's critique of the ideas
that much more impenetrable.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

139

in equating the idea of the good with the general postulation of

the ideas. Clearly, he wants the full weight of his critique of the
ideas, to which he refers here, to fall on Plato's idea of the good
(cf. 1217b20:
. . to einai idean me monon agathou alia kai
allou otououri*' ([it is empty to assert] that there is an idea, not
only of the good, but also of anything else whatever).
Then he introduces his second and, he says, decisive argument: the idea of the good is useless in practice. It is striking that
this argument is later treated only very briefly (1218b34). In introducing it, he inadvertently lets a peculiarly clumsy sentence go
by: "e/ kai hoti malist' eisin hai ideai kai agathou idea . . ( A n d
however much there are ideas and an idea of the good [they are
perhaps useless in regard to a good Ufe and to practice])
(1217b23).
To be sure, the general, logical argument against the doctrine
of the ideas as a whole is not repeated in what follows. But it is
remarkable that the particular question of the idea of the good,
which is taken up now, is pursued exclusively along the paths of
logic. One must ask oneself if the kinds of argument used here do
not actually, if unintentionally, confirm the special status of the
idea of the good. For Aristotle does not simply bring the chrismos critique to bear on the idea of the good and let it go at that.
Rather, independent arguments are devoted to the question, arguments designed to display the contradictoriness of the idea of
the good (1218b33).
To begin with, we find the category argument, which places
the good in strict parallel with being: ^oude to on hen ti esti peri
ta eiremena . . . " (being is not one thing in all that we have spoken of [and neither is the good]) (1217b33). This argument, of
course, precludes any being-for-itself [christon) of the idea of
the good. But does it not go too far? For it would follow from
this parallel [not only] that no science [is possible for the good.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 140

but also that none] is possible for being as such (1217b34). Obviously, the Nicomachean Ethics, as well as the Magna Moralia,
seeks to avoid this undesired consequence. Hence, when the
Nicomachean Ethics uses the same argument, it speaks only of
the good and rules out that it could be "something universal and
one" {koinon ti katholou kai hen) {EN 1096a28). In general, it is
striking that in the Nicomachean Ethics "being-christon" is
mentioned only once and is used primarily as a synonym for
^koinei kategoroumenon^ (predicable in common) (1096b32
ff.). Here the issue is the koinos logos (common expression).
The kind of argument that follows next is taken from the sciences and based on the category argument. According to it,
knowledge of the good disappears among the particular arts
{technai). Actually, Aristotle thereby touches on the same, familiar difficulty which we uncovered in Plato, that is, the difficulty
which arises when one tries to understand the good taking the
mode of knowing in the arts as a starting point. To be sure,
Aristotle's rejection of a science of the good is meant as a criticism of Platoscholei auto ge to agathon theresai mias (there
can hardly be a sole theorizing about the good itself) {EE
1218al) or en an mia tis episteme there would have to be [but
cannot be] a single science) {EN 1096a30). But in Plato's dialogues too, equating knowledge of the good with the mode of
knowing in the arts tended to miscarry. Accordingly, Aristotle
does not seem so far removed from what Plato had in mind even
when he criticizes him.
There is a difficulty with the second argument {EE 1218al
15). The text of the Eudemian Ethics cannot be right here. The
point is that the good cannot be both something in common and
something for itself {koinon kai choriston). But it seems to me
that the line of thought here is incoherent and cannot be followed out to a logical conclusion. The prteron-hysteron (prior-

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

141

subsequent) argument and the illustration of it with the numbers

is familiar to us from the Nicomachean Ethics. That there are no
ideas of the numbers because there could be no idea of number
as such is specifically cited there as a Platonic doctrine. And
there, Aristotle uses Plato's own reasoning in arguing against the
idea of the good. In view of the privileged position of substance
over the other categories, which are only subsequent to it, the
good, just like the numbers, would have to be excepted from the
postulation of an idea. That is: if the number per se were what is
first [to proton)^ the first number would no longer be the first
number, and that argument ought to hold analogously for the
good per se.
The exposition in the Eudemian Ethics of the same argument
that led Plato to reject the postulation of an idea of number for
the numbers is entirely accurate: the "multiple" can only be
something in common and not something existing for itself.
Hence, there is no idea here. This was obviously the conclusion
Plato drew. But how does Aristotle then proceed to derive an
argument against the idea of the good from Plato's own teaching? The text does not say expHcitly. We must, it would seem, assume (as in 1218al5 and a24) that the good is thought of
here as the one, that is, as the first in the series of the ideal numbers. ^^ Then the argument makes sense, and only then is there no
12. Thus 1218a6: "If, for example, duality is the first in the series of the multiple, the concept common (to all multiples), 'multiplicity,' cannot be separable.
For if what is in common is the ideaif, that is, one were to make *the common'
separate (christon)multiplicity would be prior to duality."
The point is that multiplicity (pollaplasion) is a principle in number and not
another numerical entity. Two, as Gadamer will point out, is the first arithmos,
since arithmos or number, is by definition sum number. Hence there are two
principles in number: the one, or unity of so and so many (which unity is precisely not a number), and multiplicity. Any arithmos is a unity of many. Compare
"Plato's Unwritten Dialectic." TRANSLATOR.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 142

need of a detour by way of the categories. Just as there is no idea

of the numbers, so too there is no idea of the good for itself. The
one of the ideal numbers cannot be an "idea existing for itself" if
it is the first number of a series. That is Aristotle's objection.
But how does the argument, augmented in this way, relate to
what follows? How does it relate to what is common {koinon) to
all virtues? Certainly the issue continues to be whether a universal is to be viewed as something existing for itself. And in some
form or other, something like this question is in fact raised in the
corrupted traditional text: ''ei christon poieseie tis to koinon''
(if one were to make that which is in common separate and for
itself) (EE 1218a9). That is a conditional charged with skepticism. Nonetheless, it introduces an entirely new argument,
namely, the hypostasizing of the good as the thing common to all
virtues. Pure induction leads from dikaiosyne (justice) and
andreia (courage) to a common being-good. We find that argument in the Magna Moralia at 1182b31as the kaf epaggen
koinon (thing in common arrived at by induction)and its logical structure is clear. To start with, the good is nothing other
than the common logos (expression) and now the Platonists proThe argument with the numbers, then, extends the general "third man" argument against any christon idea, and in particular, the idea of the good (see n. 6
above). It is a misplaced concretion to turn the principle of something into another thing like the thing of which it is the principle, and thereby make it part of
the same series of things. That will only necessitate yet another principle governing the new seriesad infinitum.
It now becomes clear why Gadamer treated Plato's hsauts in the Republic so
extensively (cf. ch. 3 above). If Gadamer is right, Plato could not have meant
hsauts in a way that would lead to a strict equation of the good with the other
ideas. For, as the one is a principle of number and not a number itself, so too, the
idea of the good is the principle of the ideas and not an idea itself. (It will be
noted, of course, that here we are moving within eidetic realm and not as before,
between the eidetic and phenomenal realms. Even so, the structure of Aristode's
argument remains the same.) TRANSLATOR.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

143

ceed to call it "itself (auto), by which they must mean "eternaP'

{aidion) and "for itself" {christon). This line of thought appears to end in a conclusion which Aristotle draws: if it is "itself," then it is "for itself," and in that case it is not something
"in common." For what is in common is an attribute not of a
particular individual thing but of all individual things (EE
1218al5 = MM 1182bl3). Now if one considers all those things
that argued for the fact that Plato himself also had in mind no
other presence of the good than its presence in all good things,
the deduction of its being for itself (christon) from its being
eternal (aidion) is, to say the very least, quite misleading. This
deduction is more an expression of what in Aristotle's view is an
unavoidable consequence than of what Plato intended.^^
It seems to me that something similar holds in regard to what
follows. Aristotle now criticizes Plato's inference from the numbers, for the being-good (being-beautiful) of the numbers can
perhaps be deduced from those things that are accepted as good,
for example, justice and health, and the ordered structure that
pertains to these, but not vice versa. Let us forget the aims of
Aristotle's critique for a moment and consider the Platonic argument per se, that the good is found in orders of things and in
numbers. But is it separate from them? Is the unity of an order of
things separate from the order itself? It makes no sense to say
that it is. In the mythology of the Timaeus too, the world soul, to
which the order of the motion of the universe conforms, is spoken of as a harmony composed of pure ratios. But the one is not
spoken of as the good. In our passage (1218a30), to be sure, we
have "the striving of all numbers for the one," '^as if it were a
13. Perhaps the addition that Rassow and Susemihl insert at 1218a 14 is unnecessary, since Aristotelian explications of the meanings of auto, aidion^ and
christon constitute a unity, and the argument is intended to show only the incompatibility of these with whatever is koinon.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 144

good*. That, however, is just a metaphor, and Aristotle, in his

familiar fashion, makes things easy for himself when he takes
this metaphor of the striving of the numbers literally and then
finds that the numbers ought to have soul(s) but do not. Just
what, however, does the metaphor actually stand for? Does it really stand for the being-for-itself of the one?
The numbers are units of ones. The principle of being one is
generative in them. They all follow the principle of + 1. That
they do is obviously the sole meaning of the being of the number
one, just as the sole meaning of the one *or unit'' of the world
order is being that one order. Plato is able to express this fact
only in such fashion that it looks as if the world were striving to
come as close as possible to the pure numerical relationships,
and as if, for their part, the pure numbers and measures that constitute the order of the universe were being "striven for.'' Evidently, the ideal order in the Timaeus is called "world soul" on
account of this movement toward the pure numbers and measures. When it is said that not only all numbers, but all existences, strive for the one and the good, that is to be taken in this
metaphorical sense. But surely Plato would not find Aristotle's
objection that every individual thing strives for its own good, for
its measure (EE 1218a31), an objection. In the mythical language of the TimaeuSy this same idea is expressed when it is said
that the self-moved heavenly bodies follow a course prescribed
for them {''to prostachthen emathen'' [they learned what had
been commanded]) [Timaeus 36e). And what does Socrates do in
the Philebus but raise precisely this issue [of each thing's striving
for its own good] ? But whereas all other living things strive for
their proper good instinctively, human beings are special insofar
as they choose theirs, using reason.
To this extent, the difficulty that Aristotle creates by deliberately taking Plato literally is an artificial one: the dialectical per-

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

145

plexities {aporiai) (1218a33) are contrived. They are designed to

introduce Aristotle's own solution: there can be no "good itself," no "good in itself," except in the functional sense of the
telos (goal) or hou heneka (that for the sake of which). This conclusion emerges at 1218b20. His arguments are based on a literal reading of Plato's metaphorical statements within the framework of his own conceptual apparatus. But for just that reason
they miss what Plato intended.
Thus the actual question before us is as follows. If the idea of
the good is "the good itself," it must be common to all things
that are good. But what does that mean? In the end, does it not
mean that "it itself" is nothing other than what is common to all
things? Such an interpretation of the doctrine of the idea of the
goodnamely, that it means nothing except the koinon (what is
in common)might have been advocated by liberal Platonists,
for instance, and might even accord with what Plato himself intended. But whether it does or not has no bearing whatever on
Aristotle's argument. The important thing for us is that, whatever the case, for Aristotle, too, the question of just what it is
that is in common remains inescapable. The inevitability of this
question is also confirmed by the fact that he gives his own positive answer to itto telos tn anthrpi praktn (the goal of
human actions) (1218b)while expressly mentioning the ou
prakta (what is non-practical), and in fact even the akineta (what
is unmoved) *in order to exclude them from consideration:
"Moreover the good is spoken of in many ways, some of which
are the noble and practical, as well as the nonpractical. The practical good, however, is that for the sake of which. It is not something in the realm of what is unmoved" (1218b4 ff.)*.^^ Hence
14. The reference here is to the stars, which, it will be noted, the Pythagoreans
called "good." TRANSLATOR.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 146

the foundations of ethics are provided by a kind of narrowing of

the question about the good to the prakton (practical). Within
the range of "practical" problems, the question of what is good
and, at the same time, one and unifying is easily answered, and
in all three treatises it receives the same answer: "that for the
sake of which" {to hou heneka). In each case the for-the-sake-ofwhich is the good, and it comprehends within itself that which
contributes to this end, which is to say, the means [to achieve it].
But clearly the best in human action is something different in
each case, that is, not something common to all cases, and least
of all, common to all existent things in general. The concern here
is solely with^the ariston tn praktn (the best of actions). But as
clear as that is, the question of how such a "practical" best relates to the "best of all things" remains meaningful. And, as a
matter of fact, the text that concludes the critique of Plato in the
Eudemian Ethics raises a question aimed in this direction:
skepteofiy posachs to ariston pantn (it remains to examine the
best of all things in general) (1218b26). I have treated the corrupted text above (see n. 4), and I hope that my analysis of the
context of Aristotle's argument has demonstrated convincingly
that the phrase "posachs to ariston pantn" cannot be conjured out of existence. It corresponds to the pollachs (in many
ways) at 1218b4. The question is prepared for therea question that is most likely raised in our passage with the sole intent
of showing that the practical good, although it only has to do
with a narrower realm, can also be called a "best of all." That
Aristotle says this, however, in no way alters the fact that the
more general question is found in the text here as an unelaborated question mark.
Fortunately we do not have to depend exclusively on the
unreliable text of the Eudemian Ethics, Both the other treatises
confirm that an outlook [as it were] into the universal is always

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

147

nearby. To begin with, let us turn to the Magna Moralia. The

first thing that strikes us is that the argument starts out with the
question of arete. But then it proceeds to the question of the
agathon, and it is clear that with this transition Plato's line of
questioning is taken up. Of course, right away his question is restricted to the "good for us," the single and highest object of
"politics." Still, it should be noted that the concept of the agathon and the multiplicity of its meanings continue to be the general, guiding theme. That may be the reason why the expression
for the restricted topic characteristic of the Eudemian Ethics and
the Nicomachean Ethics^ namely, prakton agathon^ never occurs
at all in the Magna Moralia. So one must ask oneself just what
the restriction here to the "good for us" actually signifies. It is introduced in vague distinction to the "good for the god," which
brings no universal ontological or teleological implications to
mind. Then, still another distinction is introduced within this reduced "good for us": ^'palin de kai touto dielein deV (moreover,
we ought to distinguish in this too) (1182b6). The distinction
needed, we are told, is between the two meanings of "good"
here: on the one hand, to be good in the general sense (called
here "fo koinon^' [what is in common]), and on the other, to be
the "idea of the good." These are two renderings of "good." The
first refers to a being-in-the-individual, the second to that in
which the individual participates: "e to ariston en hekastoi tn
ontn e hou talla metaschonta^^ (either the best in each existent
thing or that of which another partakes) (1182b8). The consideration of both senses would seem to have its locus within the
framework of politics, that is, of the "good for us," but, at least
when Aristotle takes up the doctrine of the ideas, this framework
is broken through. "The good" in both these senses is rejected as
untenable.
Even in the Eudemian Ethics we do not find these alternatives

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 148

in the interpretation of the good. There, being-for-itself is always

explicitly associated with "what is in common." Here, in the
Magna Moralia, on the other hand, the good of each individual,
as the koinon en hapasin hyparchon (the common thing occurring in each (MM 1182bll), is divorced in an illuminating way
from the being-for-itself {christon kai auto kath' hauto [separate and itself for itself]) (1182bl3) that the idea of the good
presupposes: the good as the koinon has only the logical universality of a genus. It is treated in regard to each of the two possibilities of arriving at a universaldefinition and induction
and is rejected on the grounds that it can be established as a universal in neither of these ways. Obviously the good, which is discussed here as the "universal" {koinon)at 1182b20 the same
thing is called katholouand whose definition is what is auto
di' hauto haireton (to be chosen for its own sake), is thematized
entirely within the context of politics and is to be understood accordingly, that is, as the good for us. (We know this definition
from Plato's Republic^ book 2 [357b] and from the detailed critique of the idea at EN 1096bl6, where the good is taken to be
the one consideration [mia idea] in reference to which all human
beings^^ choose and give preference. The fact that thinking and
seeing [phronein kai horan] are mentioned here, on the one
hand, and certain delights [hedonai tines] [1096bl7] on the
other, correlates nicely with the PhilebuSy where, it will be recalled, certain delights are explicitly allowed. But our passage
also speaks of "honors." This addition accords with the ancient
doctrine of the three ideals of life, and from Aristotle's point of
view, honors should automatically be joined with the other two
wherever the concern is politics.) Thus in the Magna Moralia
15. Gadamer changed Seienden in the text here to Menschen,

TRANSLATOR.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

149

too, there is no altering the fact that the question about the good
is deliberately restricted.
And in similar fashion, the elucidation of the other, inductively derived, concept of the good stays within these confines.
For it is the different aretai that are discovered [by induction] to
be good.
Thus, this first part of the overall argument does not point beyond itself in the direction of the question on which we have focused our inquiry, that is, the relationship between to agathon
(the good) and to on (being). Nevertheless, that the question has
been narrowed to the specific issue of "the good for us" is noted
explicitly in the summary of the Magna Moralia at 1183a7, and
even more explicitly at the climax of that summary: ""hyper tou
agathou ara, kai hyper tou aristou kai hyper tou hemin aristou^^
(about the good and about the best and about the best for us)
(1183a23). In this way we are kept aware of the universal, general reference of "the good."
We are made even more clearly aware of this universal reference in the discussion of the concept of the idea and what this
concept entails. For the postulation of a good in itself plainly derives from Plato's having founded his thinking generally on the
ideas: for Plato, "what a thing is most" means in each case "it
itself" (malista [most] . . . auto [itself]). We find this line of
thought applied here to the good and, as we mentioned at the beginning, Aristotle even grants that Plato's argument has a certain
validity which only in political matters is unimportant and irrelevant. The restricted form that the rejection of Plato's argument
takes here becomes even more conspicuous when it comes time
to take aim at the more general problems that arise in connection
with the universal doctrine of ideas. When it is said then that, in
the realm of politics, we are concerned with a class of goods for

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 150

which the idea of the good has no real relevance as an arche, the
idea of the good as such is not called into questionany more
than the doctrine of the immortality of the soul, for instance, is
called into question when one says that this doctrine has no
place in mathematics.^^
That the overall concern here is with a particular instance of a
more general set of problems is made fully clear by the far more
universal overtones of Aristotle's way of putting things. When
expressions are used here such as dia ten hautou physin haireton
(to be chosen according to the nature of it) (1182b9)which refers back to en hekasti tn ontn (in each of the existent things)
(1182b8)this could easily be taken in a universal teleological
sense such as is implied, for instance, in the Eudemian Ethics at
1218a30, where the reference to Euduxus's doctrine is obvious.
In the Magna Moralia^ to be sure, such general considerations
are not pursued per se, even though Aristotle's formulations virtually invite us to pursue them, and even though in his critique of
the ideas we find such a universal concept as ta noeta (the noetic), which certainly does not make us think primarily of the
practical. Quite the contrary: the allusion of ta noeta to the numbers is patent (1183a24), an allusion that, nonetheless, is not
stated outright. Instead, the text speaks only of the "idea." This
fact notwithstanding, we are dealing here with the same methodological argument that we find in the Eudemian Ethics at
1218al6, where it is expressly pointed out that numbers are not
generally accepted to be the good. The correspondence to the
Eudemian Ethics thus shows that the Magna Moralia too implicitly points beyond the problems of practical philosophy.
16. Dirlmeier makes a convincing case for the dio (therefore) at 1183b7. For
what follows, the simplest emendation, it seems to me, is: ouk oikeian arch en
einai toutn tagathon (the good is not a proper principle of these).

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

151

The ontological aspect of the problem of the good is all the

more evident in the Nicomachean Ethics. Of course, here too the
main argument, in which everything culminates, is that the idea
of the good is of no practical use (EN 1096b33 ff.). But again,
even the refutation implies acknowledgment of the issue. Aristotle says explicitly "e/ gar kai estin hen ti to koinei kategoroumenon agathon e christon auto kathauto'' (if indeed what is
said to be the good in common is one thing or separate and for
itself). All the logical arguments against the one common good
kata mian idean (corresponding to one idea) (1096b25)the
category argument as well as the argument more concerned with
pursuing Plato's differentiations of the kath* hauto haireton
(that which is to be chosen for its own sake) (1098b90)prove,
as it were, uncomfortably more than they should. For it cannot
be the case either that "good" is a pure equivocation, that is, that
there is one word for completely different things merely coincidentally. Thus Aristotle undercuts his own postion, so to speak.
Even though he banishes any consideration of the common element in the use of the word "good" from ethics, he indicates two
possible answers to the question of what that common element
might be. The first is that the common meaning of "good" being
asked about could be thought of as "derived from one thing,"
that is to say, in such a way that everything (good) "contributes
to" {syntelein) one thing. Although there is some disparity in the
expressions used, there can be no doubt that Aristotle is alluding
here to the logical structure that he explicates initially in the
Metaphysics Gamma 2 and which was later termed the "analogy
of attribution." The multiple senses of good do not mean only
that the means are good if the end or purpose is good, as the expression syntelein might suggest. Syntelein must obviously be
taken in a wider sense to mean "contributing to a common semantic field," so to speak. The pros hen (to one thing) here thus

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 152

fully accords with the aph' henos (from one thing). Accordingly,
in the Metaphysics^ Gamma 2, we find pros hen or pros mian
archen (to one principle), respectively,^^ and the example of
"healthy," which can be said of a human being, a facial color, a
medicine, and so forth. The word "good" could be multivocal in
this way too. That would presuppose something privileged and
primary such as Aristotle's category doctrine specifies [for "to
be"] when it gives priority to "substance" [ousia). The priority
assigned to "substance" is of great significance for Aristotle's
metaphysics, particularly in regard to the theology of the first
mover, which, for its part, is first in the order of substances.
Given this fact, one might expect that here in the Nicomachean
Ethics Aristotle would favor such an attributive relationship
{analogia attributiva) when he sets about grasping the manifold
of agathon. The god or nous (intellect), both of which appear as
examples of substance in the category argument {EN 1096a24),
would then be the summum bonum, the highest good, in which
the attribution terminates. The universal-ontological sense of the
one good would fit perfectly with Aristotle's doctrine of a god.
The 'theological' conclusion of the Eudemian Ethics would also
lead us to expect the same thing.^^
Hence it is all the more surprising that the text continues: "or
rather more according to analogy" (e mallon kaf analogian)
(1096b28) and makes clear by an example that the bare sameness of the relationships, that is, a proportional analogy, is
meant, in which no one thing is given priority over another, and
in which there is consequently no terminus.
This concept of proportional analogy, too, is known to us
17. At the same time, the pros hen is characterized as a special case of the
more general kath* hen legesthai (to speak in one regard).
18. To be sure, MM 1182b9 characterizes the question about the good of the
god as an allotria skepsis (another consideration).

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

153

from Aristotle's metaphysics and in other connections as well. In

distinction from the later tradition, above all Scholasticisma
tradition that is summed up in Cajetan's famous treatise De
nominum analogia {On the Analogy of Names)we find in
Aristotle simply "analogy." It is self-evident that Aristotle would
not consider this universal question more closely in our passage.
The whole issue of the common sense of "good" belongs, after
all, in the Metaphysics and not at all in practical philosophy.
This makes it all the more surprising that, nevertheless, here in
the Nicomachean Ethics, the bare similarity of relationship [or
proportional analogy] is the preferred topic. Would one not have
thought it much more likely that Aristotle would give preference
to the other possibility, the attributive relationship, in order to
do justice to the vague question of a highest good, the aristn
pantn that appeared to be the focus in the Eudemian Ethics at
19. As opposed to Aristotle, the later Scholastic tradition which Cajetan recapitulates distinguishes attributive from proportional analogyto which alone
Aristotle refers when he speaks of "analogy." The proportional analogy, or analogy of relationship, is to be found in Aristotle in Metaphysics Lambda 4. The
point of his argument here is that things that exist do not share any common substance or idea but, rather, only an analogous structure, namely, that of movement from steresis to eidos (privation to form) or dynamis to energeia (potentiality to actuality), and that each carries out this movement in a way peculiar to it
and different from the others. (Cf. Gadamer's translation of, and commentary
on, Aristorie's Metaphysik, Lambda [Frankfurt, 1948].) Or, put another way, as
the eye is actualized in seeing, so, analogously, the ear is actualized in hearing,
but alio alloUy "the one in one way, the other in another." This argument, then, is
in critical response to Plato's hypostasis of the ideas: when we say that the soldier
who is brave is good or the flautist who is actually playing the flute is good {EN
1097b25), there is no single "the good" in itself to be hypostasized here, rather a
parallel structure insofar as each, only analogously to the other, does what is appropriate to him. The issue in practical philosophy thus becomes the proper function (ergon) or activity {energeia) in which human beings find their fulfillment,
and any idea of the "good in itself," should it exist at all, is irrelevant here.
To that extent, this Aristotelian line of thought implies an exclusion from prac-

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 154

Of course, we see elsewhere that Aristotle is often not entirely

decisive, and that in general the carefulness in description and
tical philosophy of any theoretical-metaphysical considerations of the one good
of all things. Still, Aristotle, precisely by setting such considerations of the one
good apart from practical philosophy, confirms that they remain important to
him. As Gadamer will point out, Aristotle, like Plato, wants to understand the
universe, and not only human practice, as purposive, as aiming at what is best.
Therefore, he follows the Platonic Socrates in passing beyond merely materialefficient causal accounts of things.
And it is here that what was later to be called the attributive analogy comes
into play: we say that a thing is white or is larger than something else, and so
forth, but before we can say any of these things we must be able to say what it is.
When we do that, we are not merely articulating a structural parallel, but rather
a structural convergence on one ultimate sense of "is," namely, ousia, the
"what" or substance, which is the end point in which all the categorially differentiated senses of "is" terminate, and the arche on which they all depend. Aristotle
never expressly carries this argument over to our saying "is good," and for the
most part he argues, as we have seen, that when we say a flautist is good or a soldier is good, we are speaking of a proportional analogy, an analogy of relationship. But if, for instance, we speak of a healthy food, a healthy facial color, or a
healthy person, these different senses and uses are in fact not mere parallels.
Rather, as "is" converges on one sense of being, ousia, they converge on one
privileged sense of healthy: the terminus, or end point, from which they all derive
their meaning is ultimately the health of the human being. Hence the question is:
Is there in Arisototle a good in which our diverse uses of the word converge? Can
the uses of "good" be said to have a terminus? Lambda 10 makes clear that they
can. The best of and for all things, to ariston pantn, is the good in which the actualization of the individual goods culminates.
The real difference from Plato here results from Aristotle's conception of predication: if we say "is good," we must say that some thing is good, however incorporeal this thing might bein this instance, the god (ho theos). "Good," even
the highest good, does not exist independently of the subject that is said to be
good. Even good in the highest sense is thus an enhylon eidos. In Plato, on the
other hand, the good is an idea that is to be thought of in abstraction from any
specific thing and to be analyzed in regard to which other ideas it may be combined withjust as point, line, figure, and solid, are to be considered in their
purely mathematical relationships to each other, apart from, say, pointed,
curved, square, or cylindrical things. Compare "Amicus Plato Magis Amica Verit a s . " TRANSLATOR.

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

155

analysis that governs him often hinders him from dedicating

himself to his own constructive aims, let alone those of his interpreters. As I reminded the reader above, that even holds for
Aristotle's metaphysics and still more for the relationships between theoretical and practical philosophy and, within practical
philosophy, the relationship between the ideals of political and
theoretical life, or even the relationship between political science
and political cleverness {demotes). (We shall return later to this
last point.) Sometimes it seems as if Aristotle prefers the analogy
structure because it accords particularly well with his descriptive
caution. He is inclined to think analogically even in theoretical
contexts. Evidently his analogical way of thinking helps him
avoid the hypostasis of the universal, or "what is in common."
Thus alio allou (the one thing in one way, the other in another) is
virtually an anti-Platonic catchword insofar as it underscores
the merely analogical sameness of things which remain distinct'^.
Consider, for instance. Metaphysics^ Lambda, in which Aristotle
treats the purely analogical character of the "causes" in lengthy
expositions (chapters 3 and 4) and sharply demarcates the entire
analysis from any constructive derivation from some single primary reality, and in which, as a consequence, it seems somewhat
surprising that at the end the priority of the first mover nevertheless results. When, in the Nicomachean Ethics^ the pure analogical likeness is given explicit preference, it also sounds as if
Aristotle finds that likeness sufficient for the task of conceptualizing the whole teleological order of the worldthe program of
the Phaedoy which in a certain way is his own. In Plato the three
great realms of ordersoul, state, and worldare similarly
represented as analogous, and the good appe^lrs in them as that
which unifies, gives unity, that is, as the unit, or one.
There is no doubt that in this regard Aristode shared Plato's
view of the world. Even if his own science of the whole is con-

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D 156

structed on the very different basis of his physics, and not a

mathematical-harmonic conception of the good such as Plato's,
he still cannot ignore the Platonic problem as such. That he cannot is audible in our passage in the Nicomachean Ethics. How is
the good, as that which is in common, to be thought of? Perhaps
like being, he says, which we encounter in manifold categoreal
differentiation, and which, as we know, is nonetheless the object
of a highest science. As a matter of fact, it is in the context of
Aristotle's prima philosophia that the "pure analogy" first
comes into play. We have already mentioned the Metaphysics^
Lambda 3 - 4 , and Eta 4 also belongs here: although there is no
mistaking,the anti-Platonic turn it takes, it nonetheless remains
an answer to the Platonic problem of the good and being. It is remarkable how much the Platonic problem of the universal dialectic in beingparticularly as that dialectic occurs in the Sophist and Parmenidesevidences itself in many parts of Aristotle's
Metaphysics^ for instance, in the aporiai of Beta and in Gamma.
When one compares the way in which the question about being
is put there with the context in which the agathon comes up in
Plato's Republic^ Philebus^ and, indirectly, in his Timaeus too,
one is struck by the identity of the problem, and the idea of analogy profers itself. What Aristotle rejects as such in Plato's philosophy is not the structural order of the whole but the derivation
of that structural order from the hen (one) and the ontological
primacy that Plato gives to mathematics as a consequence.
Aristotle saw Speusippus as the prime example of this mathematicizing that a world construed as [proportional] analogies
would represent. But of course Plato's doctrine of number and
unit is behind Speusippus's theory. After all, there is not only
Aristotle's account in the Metaphysics^ Alpha 6, of the two principles, the one and the indeterminate two, from which all numbers just as all existents in general are derived. The same thing is

A R I S T O T L E ' S C R I T I Q U E OF T H E G O O D

157

Stated in the Philebus's doctrine of the peras and apeiron, and

our linking of Aristotle's account with the Philebus is supported
by the testimony of Porphyrins cited above, according to which
Plato's lecture on the goodof which Porphyrins evidently had
Aristotle's original versioncan only be made comprehensible
by starting with the Philebus. The critical point for Aristotle is
that the number structure is said to be the structure of being.
Here he finds that Plato remained all too Pythagorean.
That is not to say, however, that Aristode did not have to raise
the metaphysical question about the good and the best himself.
On the contrary, this question is always implied in his question
about being. At the end of the grand overview of the whole universe that Book Lambda provides, he formulates an alternative
regarding how the nature of the whole might contain the good
and the besteither in the form of a best being or in the order of
the whole. Here he cannot evade the consequence of the ontological primacy he gives to physics and to physei onta (what is by
nature). Plainly, Aristode is concerned with the unified order of
motion in the universe, an order that he defends in opposition to
mathematical-harmonic theories in Speusippus's Pythagorean
style. Put another way, he is concerned to give physics priority
over mathematicsa concern that underlies his well-known objection to the academy, namely, that there philosophy had been
turned completely into mathematics. Hence he alludes polemically to Speusippus's Pythagoreanism in the Nicomachean Ethics^
and in the Metaphysics^ Lambda 10, in an obviously polemical
reference to Speusippus, he rejects a pure [proportional] analogical structure of the universe, citing Homer's well-known "Let
the ruler be one" (1076a4).
With that he decides against a mere order and in favor of a
good existing apart for itself (kechrismenon ti agathon [a separated good])-^although he certainly does not question that each

A R I S T O T L E ' S C R I T I Q U E O F T H E G O O D 158

individual order is good. In his eyes, however, the primary reality that he postulates, that which is first (to proton)^ cannot be
something mathematical; rather, it must be a mover, hos to
kinoun poiei (such as induces movement). When everything is
ordered toward this primary reality, that is indeed "the best of
all" {to ariston pantn)^ and this highest being is at the same
time the fulfillment of what being means. Thus, Aristotle puts
the Platonic heritage, which the question about the good represents, on the ground of physics. And starting there, he develops
his doctrine of being in the conceptual form of an attributive
analogy which has a highest terminus (end point). 'Pure,' proportional analogy cannot accomplish thisit would remain too
close to Platonic ideal mathematics. Hence, we are left with a
paradoxical result: the chrismos (separation) that lives on in
Aristode's theology is not Plato's. On the contrary, the ontology
of the physei onta and the entirety of motion as a whole force
Aristotle to a chrismos of his own, which passes beyond Plato's
mathematically oriented interpretation of the transcendence of
the good.

IV

THE IDEA OF
PRACTICAL PHILOSOPHY

The question now becomes all the more pressing why here, in the
Nicomachean Ethics^ Alpha 4, Aristotle gives preference to the
mere analogical structure in inquiry about the good. The reasonwhich suggests itself at oncecould be that he finds the
consequence unwelcome that would be implied in pros hen
legesthai (saying, to one [end]) and in the complete parallel that
it establishes. In the Metaphysics^ Book Gamma, namely, such
an attributive proportionality justifies the authority of one and
the same science for the entire semantic field of "to be." The examplary case is medicine: ''kathaper kai tn hygieinn hapantn
mia episteme estin'' (if indeed there is also one science for all
matters of health) (1003bll). Where all things converge on one
goal [for example, health], a single science is always conceivable.
Hence one must obviously interpret the program of a formal ontology elaborated in Book Gamma according to this schema
even if the connection with the books on substance and the theology in Lambda remain obscure by so doing. In any case in regard to the primary science of "being as such" {on hei on)^ which
Aristotle is pursuing, the attributive analogy argument makes
good sense. But it would be absurd to claim that it is relevant to
the practical question about the good. For that would mean that
recognizing a good time for surgery or assessing a good sign in
159

160

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

the evening sky promising good weather would be matters for

one and the same science, when, in fact, one is a matter for medical science and the other for meterological science. Hence
Aristotle has good reason to give preference to the pure proportional analogy.
A science of the "good in general" corresponding to the
schema of medicinea schema that is applicable in the science
of being in generalis meaningless for practical philosophy.
Drawing a parallel with the knowledge of being as such does not
work as long as the concern is solely with the structure of human
practice and of practical knowledge. Whether it be in regard to
technical knowledge or to political-practical knowledge, the
good remains restricted [in its application] to the circumstances
of human practice. Practical reason is far removed from any universal teleology. Aristotle isolates practical philosophy for the
decisive reason that what we find to be good in the theoretical
realm"good" here meaning immutable beingis something
quite different from the right thing to do {to deon) at which the
practical rationality of human beings aims.
This fundamental distinction that Aristotle makes between
theoretical and practical knowledge also has consequences for
the field of philosophical inquiry that we would call the theory
of scientific knowing. At issue is the methodological character of
Aristotle's practical philosophy. It is indisputable that practical
philosophy itself is not knowledge of the right thing to do in a
given situationthis, even though the very fact that the teacher
of theoretical instruction in practical philosophy keeps his distance from the more universal questions of ontology, may betray
that he does have a knowledge of the right thing to do after all.
Be that as it may, practical philosophy has the character of
theory. It can be called ''episteme^'' ''techne,'' ''methodos^
(method), "pragmateia'' (pragmatics), or even ''theria'' (the-

161 T H E I D E A OF P R A C T I C A L

PHILOSOPHY

ory), but not "^phronesis^' in a terminological sense. Nevertheless, the question arises whether practical philosophy is teachable in the same sense that every other science or techne may be
said to be. The state of affairs here becomes even more complicated when we observe that Aristotle disputes that Plato's idea of
the good has any practical utility but emphatically asserts that
his own theory of practical philosophy does. He claims explicitly
that such theoretical instruction as his enhances arete itself. That
it does so is emphasized in all three versions of his ethics. And for
just this reason one can see a problem in how practical philosophy is supposed to relate to phronesis.
This problem has been investigated often in recent times. Still,
it probably appears to be a paradox only to someone who has a
modernistic understanding of "theoretical," such as we derive
from the modern conception of science. For such a person things
do indeed appear odd. Plainly, practical philosophy is not a theoretical science in the modern sensea theoretical science, that
is, which might be applied to practice in the way one puts pure
natural science to use in [applied] medical science. Practical philosophy is rather more like knowledge of cures, and accordingly,
Aristotle often draws comparisons with this kind of knowledge.
Any talk of the "application" of theory to practice would presuppose a separation between the theory Aristotle imparts in
such an ethical pragmatics [as his ethics] and lived practice. And
whatever the case, such a separation does not exist here. The
ideal of an objective theory, neutral in regard to all the interests
at stake in any practical application of it, and consequently capable of any application one might wish to make, is neither Platonic nor Aristotelian. Our investigation of Plato's science program in the Republic^ book 7, made clear how secondary the
application of the mathematical sciences is in his eyes. Aristotle
makes the same point by going to the opposite extreme: in the

162

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

field of practical philosophy he expressly avoids importing

purely theoretical considerations. This restriction holds for his
politics and especially for the vast collection of state constitutions that he organized. It is quite clear to him that there is a
practical-political problem in just how these model constitutions
are to be put to reasonable useas there is in any application of
theoretical general knowledge. (This problem is presented in detail in EN, Kappa 10, but it is also implied in EN, Zeta 8.)
The system of norms that Aristorie's ethics establishes makes
plain that Greek thinking about the polis is tied to empirical contingencies. But this fact does not constitute a theoretical deficiencyany more than the fact that Aristotle advances only
those normative insights against contingencies that are sustained
by everyone's unquestioned assent. It is characteristic of him that
he simply accepts slavery as a natural institution. Plato's normative construct, his ideal city, thus comes much closer to the modern concept of theory, and, as we know, in his Utopia Plato explicitly rejects slavery. We have here a methodological contrast
expressed repeatedly in Aristotle's critique. In practical philosophy the concern is not with the idea of the good or with the polis
that displays consummate justiceeven if in his politics Aristotle is not averse to using the ideal model state as a tool in his
thinking. Above all else, his concern is with the right thing to do.
Thus the true arche (starting point), as he says with startling radicality, is the "that," to hoti (EN 1095b6, 1098b2). He means
that we must start with our practice itself and the living awareness that we have in it of what is homologoumenon (agreed
upon) as good. With this idea in mind, Aristotle points out that
he himself only uses arguments and principles that pertain to the
subject matter and then turns this fact critically against Plato
(EE 1217al0 and, similarly, MM 1183bl). In Aristotle's view, if

163 T H E I D E A O F P R A C T I C A L P H I L O S O P H Y

one wants to do practical philosophy, any argument that starts

with the good taken as a concept of mathematical harmony is
foreign to the subject matter. When he himself makes theoretical
assertions about what is good practicallyas he indisputably
does in his pragmaticsthese assertions, he says, are not to be
taken from somewhere else but are to be derived from the realm
of practical experience itself. For thjey will then be based on a
principle proper to the subject matter {oikeia arche).
The statements that Aristotle himself makes in his practical
philosophy are, of course, general. And to this extent they all
have a theoretical character. They are also not meant at all to be
applied to the concrete case of doing the right thing in the way
that technical rules are applied in the procedure of producing
something. The right thing to dowhich one decides on the basis of reasonable, practical deHberationis not simply a case or
instance of a rule. For example, the general structure of arete
that Aristotle elaboratesnamely, to be the mean between two
extremesis not a rule that can be applied. The point of his
doctrine is not that one has to hold to the golden mean, but that
one ought to be aware of what one is actually doing when one
does what is right. Extremes are always recognizable here, and a
clear consensus exists that they should be rejected whereas what
is right as we know, cannot be definitively specified as such.
Nevertheless, Aristotle insists that the theoretical doctrine that
he presents as practical philosophy has to be of use in practice.
He shows how with an illustration {EN 1094a23): practical philosophy, he says, is useful in the way it is useful for an archer to
pick out a definite point on the target at which to take aim. This
way he will score a better hit. This can only mean that one is better able to keep one's aim fixed in the right direction when one
can set one's sights on a specially targeted point instead of on a

164

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

larger object.^ Aristotle avails himself of this splendid image to

say that the theoretical instruction that can be given in practical
philosophy puts in one's hands no rules that one could follow in
order to "hit" what is right in accordance with an art (techni^.
After all, taking aim does not by any means constitute the whole
of archery. One has to have learned how to handle the bow, and
in the same way, whoever wishes to profit from practical philosophy must be trained for it in the right way. Only then is practical philosophy of use in decision making. It assists our concrete,
practical ability to size things up insofar as it makes it easier to
recognize in what direction we must look and to what things we
must pay attention. Plainly the illustration is intended to show
that one does not rely on the theoretical generalities of practical
philosophy in the way that one relies on a rule.^
To be sure, Aristorie models his exposition of the exercise of
1. In Plato we encounter the word skopos as the expression for that at which
one aims (stochazesthai) at Republic 519c, Gorgias 507d, and elsewhere, and often in Aristotle, but never with direct reference to archery. Nevertheless, this semantic origin is implied at least once, namely at N, Zeta 1, 1138b22, where
tightening and loosening {epitenei kai aniesin) must surely be taken in reference
to the bow and not the lyre. In all other respects the semantic field of these two
words is coextensive. Cf. Heraclitus, fragment 51.
2. The point here is relevant to the contemporary Anglo-American debate
concerning rules in rule deontologism (for example, D. Ross), in rule utilitarianism (for example, R. Brandt), and contractualism (for example, J. Rawls). Cf.
in particular the latter's "Two Concepts of Rules," Philosophical Review 64
(1955):3-32). The question that Gadamer raises, starting from Aristotle, is an
ontological-epistemological one: What is the existential status of moral rules,
and how can we be said to "know" them when we apply them? He is at pains to
show that the methodological models of the mathematical sciences and technology are misplaced here. For the "being" of the rule or ethical principle is not like
that of "triangularity," that is, not like the being of something that is always
apart from its instantiations and toward which the latter may be said to "strive"
while nevertheless always falling short. On the contrary, rules in ethics have their
reality only in the tradition of their applications, instantiations, or interpretations. And each of these, far from being a diminution of some ideal rule in itself

165 T H E I D E A OF P R A C T I C A L P H I L O S O P H Y

practical reason entirely on the logic of the theoretical syllogism

used in demostration (apodeixis). But some precise distinctions
are necessary here. Certainly every reasonable conclusion to
which one comes can be depicted using the schema of a syllogism. It is also easy to see that in the sphere of practice the conclusion (Schluss) is not a proposition but a decision {Entschluss).
Still, it is striking that when Aristotle analyzes the exercise of this
practical reason, he does not use decisions that are really practical or moral, but pragmatic, technical decisions instead. In the
technical realm the only concern is with the right choice of
means for pre-given ends or purposes. At the heart of the matter
here is the subsumption of the particular under the universal,
and the structural correspondence with apodeixis is thereby
given. But one should not overlook the fact that making moral
decisions does not quite fit this schema. For in the realm of practice, holding to a principle, for example, to a certain arete, is not
a merely logical act. Practical reasonableness is displayed not
only in knowing how to find the right means but also in holding
to the right ends.

Aristotle's demarcation of phronimos (prudent, reasonable)

apart from its instantiations, is thus to be viewed as an "accretion of reality"
(Gadamer: Zuwachs an Sein) in the rule.
This understanding of the reality of ethical rules requires that we revise our
conception of how we know them. We do not know them as we know mathematical realities, that is "clearly and distinctly," as objects to be intuited purely
and exactly in abstraction from situational contingencies. Rather, we know them
only in a limited way from within the tradition of their applications, in which we
always already find ourselves "under way" (unterwegs). Consequently, the same
measure of exactitude is not to be expected here as in the mathematical sciences
and technology (Cf. EN 1094b24). Indeed, this kind of rigor would be disastrous: summum ius summa injuria. Phronesis, understanding of moral principles,
is thus anything but being a "stickler" for the rules. It is judicious discretion that,
in faithfulness to the tradition, adjusts to the particularities of the given case (Cf.
Gadamer on Aristotle's dikaste phronesis and synesis in WM). TRANSLATOR.

166

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

from deinos (clever) turns on this point (EN, Zeta 13). In this
distinction one sees how very much aware Aristotle is of the difference between technical knowing and practical-moral knowing, and I have attempted to show that here he extends a genuinely Platonic motif. Not the least indication that he does so is
the fact that no real teaching is possible here in the way in which
science and technical knowledge can be taught. On the contrary,
in practical matters the general hermeneutical task which figures
in all instances, that is, of concretizing general knowledge,
always implies the opposite task of generalizing something
concrete.
Practical philosophy by itself can give us no assurance that we
know how to "hit" what is right. Such knowledge remains the
end of practice itself and the virtue of a practical reasonableness
(phronesis)^ which is precisely not mere inventiveness {demotes).
This distinction is important for the theory of the sciences. The
comparison with the archer is found in the introduction to Aristotle's whole course on practical philosophy and politics. At the
start, the governing role played by politics is worked out. Politics
is the highest science or art: ""kuritate kai malista architektonike'' (first of the arts, worthiest and greatest) [EN 1094a26).
At first glance this comment bearing on the theory of scientific
knowing would not seem to accord all that well with the subsequent analysis of phronesis and especially not with what is said
about political phronesis. Ever since Burnet, this discrepancy has
been attributed to Aristotle's having accommodated himself to
Plato's use of language. Viewed purely on the surface of it, this
account is certainly accurate. But Aristotle could not possibly
have spoken in any other way. For we saw, after all, that in practical philosophy we are dealing precisely with philosophy, which
is to say with theory. Its object and, accordingly, what it ultimately is aimed at, is, of course, practice. But that only means

167 T H E I D E A OF P R A C T I C A L

PHILOSOPHY

that its method must be subordinated to the law governing its

object, which entails in turn that its claim to know is defined and
restricted in accordance with that law. What Aristotle calls
(habituation) is of fundamental importance for human
practice. Whoever cannot master his passions is not in a position
to listen to the logos (reason).^ Aristotle particularly impresses
this fact on us one last time at the very end of his lecture. And the
fact that he does, shows us that he was fully aware of the impli3. Auf den logos zu hren. There is a long tradition behind this phrase, extending to Heidegger'sand then Gadamer'sinterpretation of Heraditus (Cf.
Heidegger, Heraklit, part 2, "Logik. Heraklits Lehre vom Logos" [Frankfurt,
1979]). Heidegger finds in Heraditus an explication of the inauthentic existence
of those who are caught in the patterns of their everyday pursuits, and who, unthinking and deaf, remain oblivious to the logos, the reasonable order in the
many evanescent things with which they are absorbed (Heraditus: ta panta).
Note that the word hren is of particular significance here: the logos is not something I see before me and over which I may be said to dispose. It is rather more
like significant sound, or language, by which I am surrounded and within which I
"always already" find myself under way.
Heidegger's concern, of course, is not so much praxis but theria, even if the
latter is to be understood not so much as visual but as aural. Here Gadamer
makes the transition to the praaical, as he did beginning with his first book,
PDE, But as this chapter of the present book confirms, he discovers a continuity
between logos in practice and logos in theria: any individual caught in the unrestrained play of the passions and the immediate, unreflective pursuit of gratification is incapable of the highest human activity, a life of reasonincapable, that
is, of a life of reason in its practical dimension, but also, and importantly, in its
theoretical dimension as well. Sophrosyne, for instance, is a prerequisite of
sophia. And reciprocally, in some sense sophia, or at least philosophia, which liberates the human being from the "flattery" and coercion of the passions, is a prerequisite for sophrosyne. Plato and Aristotle both teach us that.
Thus one sees in Gadamer's reading of Plato and Aristode just how one might
extend Heidegger's thinking about being, and his theory of our obliviousness to
being to include practical philosophy. This extension would lead to an ethics of
arete and not at all the ethics of agape for which, apparently, Werner Marx
claims to have found a basis in Heidegger (Cf. W. Marx, Gibt es auf Erden ein
Mass [Hamburg, 1983]). TRANSLATOR.

168

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

cations of the structure of practice for the theory of scientific

knowing. Clearly a special prerequisite for receiving instruction
in practical philosophy is that the pupil will not fall victim to the
misunderstanding to which technical thinking gives rise
namely, that such instruction could relieve him of his own practical responsibility. To be sure, that is never stated explicitly.
However, this consequence for the theory of scientific knowing,
which is intrinsic to the idea of practical philosophy, is made evident indirectly at the end of the lecture, when Aristotle prepares
for the transition to politics. There he explicitly repudiates political theoreticians who, as sophists, claim authority in the matter
of giving constitutions and laws without actually being in political life themselves. The claim Aristotle makes for his own political philosophy and his insistence that the pupil have the right
preparation ahead of time are in response to this sophistic claim.
A separate investigation would be necessary to establish more
precisely the locus of Aristotle's politics within the theory of scientific knowing. In the treatise called the Politics we find no special methodological reflection, evidently because the beginning
of the Nicomachean Ethics is conceived of as the general methodological introduction to the whole of politics. The conclusion
of the Nicomachean Ethics^ which makes the transition to politics, takes up this theme again expUcitly. Of course, the version
of the Politics that we have is not very well connected to this
transition. Hence one might well question the extent to which
this treatise on the polis is subject to particular requirements of
its own field of inquiryrequirements that are not the general
ones of practical philosophy as such. For it is clear that the concern in this political treatise is lawgiving. The art of lawgiving,
however, is certainly very different from anything in political or
juridical decision making, both of which are confronted with
concrete cases. In the Nicomachean Ethics 1141b the art of law-

169 T H E I D E A O F P R A C T I C A L P H I L O S O P H Y

giving is explicitly distinguished from the other appHcations of

political reason. One methodological observation which could
result from this distinction might be, for example, that one ought
not to change existent laws in favor of better ones without including in one's calculation the fact that any change in the laws
is bad to the extent that it necessarily weakens the authority of
law in general {Politics 1263al2 ff.).
Aristotle's reflections on the method of practical philosophy
thus have important consequences for the theory of scientific
knowing, but this is not the place to pursue the significance of
these for our own situation and its special problems. Aristotle's
separation of practical philosophy from theoretical philosophy
has demonstrated its power to establish and sustain a traditionthis, not least of all, in recent centuries in which his distinction between the two has been challenged by the idea of
modern empirical science. To be sure, his distinction was n
longer able to maintain a real place for itself within the methodological reflections of modern scienceother than in the weak
afterglow of the rhetorical tradition that Vico, for instance, once
again invoked. Nevertheless, the fact that the philological, historical sciences have actually spread within the cosmos of modern science proves that this tradition is not dead. The relevance
of these so-called human sciences to the moral, political, and social circumstances of human life is convincingly demonstrated if
one considers the role played in these by world views, ideologies,
and enduring human value concepts. In this regard I would call
to mind, for instance, the contribution that the historical sciences have made in determining the identity of modern nation
states, and continue to make to the present day. And not least
important of all, certainly, is the tradition of practical philosophy itself. It is a fact worth noting that philosophical ethics has
almost never renounced its right to inject its normative judg-

170

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

ments into life as it is actually lived. To this extent Aristotle lives

on. The theoretical and practical processes of reflection in human reason seem in the end to be indissociably intertwined. One
can properly understand this correlation of the two only if, following Aristotle, one strives to keep the idea of theoretical constructs and of scientific method out of the practical field, and
only if one recognizes that the relationship of theoretical insight
to practical application is not as simple everywhere as it is in the
fields of technology.^
The clearest exposition of this theme, which Aristotle was the
first to take up, seems to me to be Kant's Foundations of a Metaphysics of Morals, There, at the conclusion of the first section on
the so-called "laying of the groundwork," Kant treats the transition from common reasonable moral knowledge to philosophical knowledge. He starts with the convincing thesis that in human nature as such there is a proclivity at work: human beings
tend to resort to a kind of dialectic in order to evade what their
consciousness of duty, for instance, makes morally evident to
them. Kant calls this a tendency to quibble, and it not only
makes the transition to the philosophy of morals necessary, but
essentially always makes that transition by itself. It is obvious
that Kant adheres to the basic tenet of Platonic-Aristotelian phi4. Aristotle's language usage reflects the [unity of the theoretical and practical] in a way that should not be taken as [just] an echo of Plato, for example, at
EN 1177al4-15: ''eite de nous touto eite alio ti, ho de kata physin dokei archein
kai hegeisthai kai ennonian echein peri kaln kai thein'' (whether this is the intellect or something else which we believe by its nature rules and guides us and
envisions things noble and divine). Here it appears completely artificial to attempt to distinguish between the theoretical and practical exercise of reason. And
even if in the Protrepticus (Walzer, fragment 6) they are distinguished, and practical dianoia and theoretical contemplation of the truth are treated separately,
they are nevertheless characterized using the same words, and both are demarcated from all knowledge in poiesis (making something).

171 T H E I D E A OF P R A C T I C A L P H I L O S O P H Y

losophy in his moral philosophy, insofar as he divorces the practical, moral imperative from the technical imperative of cleverness. But that is not all; his exposition of the transition from
common to philosophical knowledge, has, it seems to me, universal validity. Philosophy never really finds it necessary to justify its existence, since whoever would contest it is also engaging
in the process of reflection that one calls philosophy.
In returning to Aristotle, we find the intertwining of the theoretical and practical that we have seen at many sorts of cracks
and splits in the structure of his teachings. For one thing, we see
that according to Aristode the highest possibility of awareness,
which the Greeks called ''nous^ (intellection), is to be attributed
to that theoretical knowing which has attained complete selffulfillmentto sophia (wisdom). But the same highest awareness is to be attributed to practical reason as wellnamely, to
phronesis, which in each instance is conscious of the rightness of
its choice and decision. The definitive juxtaposition of theoretical and practical knowing, and hence of the theoretical and practical virtues of knowing, in no way infringes upon the unity of
reason, which governs us in both these directions [in which our
reasoning might move].
Aristotle's conception of a "practical philosophy" is plainly
the consequence of the critique we treated above of Plato's idea
of the good. Nevertheless, his separation of practical philosophy
from theoretical philosophy in no way implies a lack of coherence or an inconsistency in the content [of his thought]. On the
contrary, it is solely out of methodological and argumentative
caution that Aristotle forbids himself any and every extension
[of his practical thought] into more universal considerations.
Not that such a universal, more theoretical background does not
show through in many places. But Aristotle makes use of it in his
argument only where rt is based in universally accepted, given

172

T H E I D E A OF P R A C T I C A L P H I L O S O P H Y

facts that provide a methodological foundation for theoretical

philosophy too.
Naturally Aristotle does not speak of practical philosophy in
the context of his metaphysics. But insofar as the world of human practice is located within the entirety of what exists, the
whole sphere of human praxis (action) and poiesis (doing) has its
place within the realm of nature. Not only art imitates nature.
Human practice does so too insofar as it aims at nothing other
than the highest fulfillment of human existence itself. The fact
that it does, however, shows that at the same time human existence points beyond itself [to the divine]. And hence Aristotle
tooin following where necessity leads himmust give precedence to the theoretical ideal of life as opposed to practice and
politics.
Within his practical philosophy this precedence is also made
clear by the ontological implications of Aristotle's concepts
something that cannot be pursued here in detail. Let it suffice to
call attention to one point: the paradigm of being that always
isbe it the being of the divine or of the heavenly bodiesremains the ultimate point of reference in treating the practical nature of the human being. Thus on occasion it is said that human
nature per se is "not simple" {EN 1154b21) or that it is "compounded" {EN 1177b28); it does not consist in the pure intellectuality of any divine living thing. Nevertheless, nous (intellection, mind), or logos (reason), is the most important component
in us, and it is our task to develop it above all the others.
In general, human beings are vulnerable to their drives and
feeUngs {pathe)^ which threaten to overwhelm them. Still, the essence of human practice lies in the fact that we do not simply
give ourselves over to the forces of these drives. Instead, we are
capable of cultivating in ourselves a constant disposition {hexis)^
so to speak, which enables us to obey our reason, the logos. This

173 T H E I D E A OF P R A C T I C A L P H I L O S O P H Y

capacity too is human nature. Aristotle explicitly emphasizes

that it is, when he characterizes the realm of praxis and ethos,
which takes shape in habituation and habit, as specifically human. He thereby distinguishes it from everything that is by nature {pephykosin hemin [has come to be in us by nature]) (N
1103a25).^ The teleological framework into which the entire
5. Gadamer has already shown that in regard to habituation and habit,
where many have seen Aristotle departing from Plato's "over-intellectualization"
of practical philosophy, Plato and Aristotle are in fact very much alike. Plato's
entire Republic may be viewed as a program of training, leading not only, and
not even primarily, to insight into what the good is, but to an inculcated disposition (hexis, ethos) to hold to the good in practice (Cf. "Plato's Educational
State").
On this point Gadamer also finds Kant (and Kierkegaard) to be within the
"unitary effect" of Platonic-Aristotelian philosophy. According to Kant, I must
be made immune to the "sophistic" whisperings of my inclinations, for these
might lead me to deviate from what my reason and conscience tell me is obligatory (see MPE and ch. 3 n. 12, above). It is possible that Gadamer is able to discern this "holding to one thing" in Socrates, Plato, and Aristotle precisely in retrospect from Kierkegaard (cf. WM 91). So one might argue that in Gadamer,
Heidegger's use of Kierkegaard for the purposes of an ontological explication of
the being of the human being is reversed: Heidegger's *^existentiales^ concern
again becomes Kierkegaard's "existentielles" concern.
However, while Kierkegaard and the early Heidegger spoke of the existentielle
"decision" of the solitary individual, Gadamer, in returning to the classical
thinkers, speaks of deliberation and choice (prohairesis) founded in a communal
sense of what is fitting {to deon; das Tunliche). "I" am ethical by virtue of participation in the language, customs, and practices (Sitten) which "we" are: "Each
single individual who raises himself from his natural being to a life of spirit, finds
in the language, customs and institutions of his people a pre-given substance,
and, as in learning language, it is his task to make it his own. Thus the single individual is always in the process of cancelling his physical aspect insofar as the
world into which he grows is a world shaped by language and custom" (WM 11).
(See ch. 2, n. 9, above.)
In general, Gadamer's divergence in ethics from existentialism's solitary individual may be said to parallel Heidegger's "turn" away from the fundamental
ontology of 5Z. Just as Heidegger abandons any attempt to build a theory of being on the "acts" of the human subject and, on the contrary, defines human exis-

174

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

practical world is fitted could be similarly displayed in many

passages. Doubtless, for Aristotle too, the structural order that is
to be attributed to this whole of things can be conceived of in reference to the good. That it can may be deduced from the fact
that Aristotle finds his teleological cause {aitia) missing in all his
predecessors. The teleological cause, together with the doctrine
of the eidos, is Platonic inheritance. Thus Aristotle has no problem at all in calling the teleological cause the cause of the "well"
{eu) {Metaphysics 984bll) or even the hou heneka kai tagathon
(that for the sake of which and the good) (!) (983a31), and thus
in using exactly the same word that Plato uses.
One should not view this choice of words as a remnant of
[Aristotle's 'early'] Platonism, and certainly not as a basis for
dating texts. Rather, in these cases too, such language usage confirms that the problem Aristotle treats is one both he and Plato
see.
The strongest confirmation of what they share, however, is to
be found in the way that Aristotle holds fast to the ideal of theoretical life. That, of course, is made clear above all in the arrangement of the proto philosophia, in particular in its so-called
theology. But the lecture on practical philosophy also concludes
with a corresponding discussion of the relationship between the
theoretical and practical ideal of life. The priority of theria is
based on the ontological superiority of its objects, namely, beings that always are. In contrast, the world of praxis belongs to
that reality or being that can be one way but also be another.
Consequently, knowledge of what is to be done in practice must
be placed second to theria. Even so, both dispositions of knowtence starting with being's "presencing" to the human being, so too, Gadamer
defines human choice (prohairesis) in starting with tradition, which surpasses
any individual, TRANSLATOR.

175 T H E I D E A O F P R A C T I C A L

PHILOSOPHY

ing and reason are something supreme. Practical reasonableness,

phronesis, as well as theoretical reasonableness are "best-nesses"
[aretai].^ That which is highest in the human beingwhich
Aristotle likes to call "nous" or the divineis actualized in both
of them/
In this paradoxical doctrine which affirms [the] subordination
[of one thing to another] and nevertheless acknowledges two
forms of one highest thing, we find displayed the same descriptive carefulness in Aristotle's thinking that we discussed in the
context of the problem of analogy. The cautious formula of analogical thinking ["the one thing in one way, the other in another"] indeed proves its worth in this final and highest problem
of practical philosophy. We can establish the priority of theria
on the basis of practical philosophy alone, without having to
bring in the subject matter of theoretical philosophy. Practice itself is the all-inclusive, distinctive characteristic of the human being. Thus, one must understand even theoretical activity as
highest praxis {Politics 1325b). Aristotle remains quite vague in
discussing the relationships here. At the end of his treatment of
phronesis (N, Zeta 13) he argues that the inclusiveness of human practice entails no subordination of theory to practice.
6. W. Schadewald proposed this non-word, which, better than any other expression for arete, brings out the decisive point here.
7. Aristotle often says that everything that is in some way or other participates in the divine. Of course, this does not mean that the alia zia (other living
things) participate in eudaimoniaapart from the "^kata ten epnamian en tei
physei metechei theiou tinos^ (whatever in the realm of nature, which in accord
with what it is named after, takes part in something divine) (EE 1217a24). This
passage must be explained differenriy from all proposals to date. Epnumai
(things named after...) is used in a seminal sense and refers to particular animals
that are to be assigned to a god and can therefore be called divine, such as a stallion (Homer) or a bird (for example, Zeus's eagle) or a fish. (Is Aristotle thinking
here of the sacred dolphin of Apollo?) In explaining the passage EE 1217a25 ff.,
I would go even further than Dirlmeier's telling criticism of Gigon.

176

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

Practical reasonableness, though, is the precondition for engaging in theory and in developing theoretical reasonableness., At
the same time, practical reasonableness is also something
highest. Indeed, it is this same highest thing, nousalbeit in another application which is not reducible to theory but which is
also a beltiste hexis tou aletheuein (a most excellent disposition
of knowing truly).
With that a final and substantive similarity between Plato's
philosophy and Aristotle's practical philosophy comes into view.
It turns on the relationship [of human life] to the divine, a relationship which both take as the starting point for their thinking
on the finite, conditional, and limited nature of the human being.
Aristotle can repeat genuinely Platonic ways of putting things
when he attempts to describe the approximation of the human
being to the divine. What Hegel claimsnamely, that philosophy itself must surpass its character of striving for knowledge
and become wisdommay not be said for Aristotle.
On that account one may not absolutize the priority given to
the ideal of theoretical life over the ideal of practical-political
life; Aristotle knows just as well as Plato that for human beings
precisely this possibility of the theoretical life is limited and conditional. Human beings cannot devote themselves persistently
and uninterruptedly to thought's pure seeing for precisely the
reason that their nature is composite. Hence, viewed from the
perspective of practical philosophy, the relationship of the two
ideals of life is not such that the complete happiness of practical
life would not be something supreme too. To be sure, Aristotle
calls this happiness a deuteros^ that is, a second best. But this too
is something best, that is, a fulfillment of eudaimonia (happiness). The fulfillment in purely theoretical existence is, after all,
not the full bliss of the gods, since it is a limited fulfillment for

177 T H E I D E A OF P R A C T I C A L

PHILOSOPHY

human beings. The happiness of nous is in a certain sense separate (kechrismene)beyond all comparison. And precisely for
this reason the practical happiness of human beings is not second
rank, rather precisely what has been apportioned to them. That
holds even if at times they can also rise beyond themselves to the
divine bliss of theria. Does one not find the same thing in
Plato's Republic in regard to how the philosopher-kings will
carry out their office?
Thus the overall result of our investigation is as follows: in
basing the question about being on the physei onta and not on
the universality of the eidos or mathematical-eidetic configurations, Aristotle did indeed subject Plato's teachings to a radical
critique. But in the end did he not carry out what Plato intended
to doindeed, even go beyond it in fulfilling it? There are basic
truths that the Socratic Plato did not lose sight of any more than
did the Platonic Aristotle: in human actions the good we project
as hou heneka (that for the sake of which) is concretized and
defined only by our practical reasonin the euboulia (welladvised-ness) of phronesis. Furthermore, every existent thing is
"good" when it fulfills its telos (purpose, goal). Still, Plato only
anticipated symbolically in his number doctrine what the good
in such a universal sense actually means. Aristotle found conceptual answers to this question. The artificial expression entelecheia^ which Aristotle introduces, is obviously supposed to make
clear precisely that the telos is not a goal that belongs to some
faraway order of perfection. Rather, in each case the telos is realized in the particular existent itself, and realized in such fashion
that the individual contains the telos. Aristotelian metaphysics
keeps this fact in focus as its constant theme. It thinks of the being of what is as the self-mediation of an existent thing with its
"what-it-is" (ti estin), its eidetic determination. I have tried to

178

T H E I D E A OF P R A C T I C A L

PHILOSOPHY

make credible that such a mediation of being and becoming has

to be presupposed if the postuiation of ideas is to make any sense
at all. The idea of the good and the barely comprehensible doctrine of the one and the two, point to such a mediation even
though it is formulated only metaphorically in Plato's dialoguesin the game played in the Parmenides^ the likeness of
the Philebus^ or the mythos of the Timaeus. In Aristotle's
thought, what Plato intended is transferred to the cautious and
tentative language of philosophical concepts.

8. Vermittlung von Sein und Werden. Such a mediation is, of course, precisely the task that Hegel sets himself in his Logik. One should not overlook the
Hegelian background in Gadamer's analysis of Plato and Aristotle (cf. "Hegel
and Dialectic of the Ancient Philosophers", TRANSLATOR.

INDEX

Aisthesis (perception), 12, 45, 45nl3

Aitia (cause), 25, 9 3 - 9 4 , 108, 114,
137n, 174
Alexander, 14
Alio allou (the one thing in one way,
the other in another, 153n, 155,
175. See also Analogy
Analogy, 29, 66, 124, 131, 155-56,
175; of attribution, 151,152,153n,
158; of proportion, 152,153,153n,
160
Anamnesis (recollection), 4 5 - 5 5 , 5 7 59. See also Pre-knowledge of arete
Anaxagoras, 15; on nous, 25
Apeiron (the indefinite), 9 1 , 1 0 8 , 1 1 2 14,122,157
Apology, 23, 34, 39,110
Arche (first principle), 25, 29, 90, 91,
93, 94, 100, 137n, 150; as to bote
(the "that"), 162. See also Proton
Arete, 22, 23, 37, 42, 52, 128, 147;
versus techne, 2 3 , 3 2 - 3 4 , 4 6 , 4 9 ; as
unteachable, 50, 51, 60 (see also
dialectic and phronesis); unity of in
phronesis, 63-66; and practical
philosophy, 161; as hitting the
mean, 163
Ariston pantn (best of all, highest
good), 129-30, 129n, 146, 153n,
158
Arithmos. See Number doctrine
The Beautiful, 77, 115-16, 124-25,
133n, 137

Becoming. See Genesis

Being, 5, 98, 117, 131, 138, 139-40,
158-59; coming into, 13,114; relationship of to the good, 27, 149,
156, 160; that always is, 171, 174;
question about, 177-78. See also
Substance
Categories, 13n, 15nl3,16
Category argument, 131, 139-42,
151-52. See also Substance
Cave, allegory of, 74, 7 6 - 7 9 , 1 0 0
Chrismos (separation), 9 - 1 0 , 12,
16-20, 31, 113, 131-32, 135,
138-39, 142-43; in moral phenomena, 18-19; in logic, 19; in
Aristotle, 133n, 158
Clitopho, 22, 49
Courage (andreia), 64-65, 95,142
Crito, 97
De Anima on active and passive nous,
89
"Developmental" interpretation of
Plato and Aristotle, 7 - 9 , 38, 103,
135n, 138n
Dialectic, 5, 20, 83, 84, 92, 109; as
unteachable, 37; as phronesis not
techne, 39, 121; as holding steadfastly to what is right, 4 1 - 4 3 , 9 6 97; as the art of distinguishing
where confusion threatens, 4 4 - 4 5 ,
95, 98-99; versus dianoia, 90-92,
100; in the arts of music and letters,
119-21
179

180

Dianota (understanding), 90, 92,100

Dihairesis (division, differentiation),
42, 46, 88,102,119,124; as differentiating right from wrong, 42; and
synopsis (seeing together as one), 59
Dirlmeier, F., 126,129n, 150n
Doctrine of the four. See Genera
Dynatnis (power), 89,116-18,125
Effect, unitary, 1, 2, 5,173n. See also
Wirkung
Elencohos: as refutation 45,63; as test
to be withstood, 84, 85n, 9 7 - 9 8
Enhylon eidos (materialized form),
3n, 133, 154n
Entelecheia (internal telic principle),
177
Eros (love), 44, 77,116
h h o s (habituation), 44, 60-61, 167,
173,173n
Euboulia (well-advisedness), 36, 80
Eudemian Ethics, 7 - 8 , 9 , 2 3 , 3 4 , 1 2 6 ,
129, 130, 131-46,147,150,153
Euduxus, 14,105
Euthyphro: on piety as a "part" of dikaiosyne, 11
Gaiser, Konrad, 90,134
Galileo, 2
Genera, doctrine of the four, 108,
112ff.
Genesis (becoming), 30, 112, 114,
117, 122,178
The Good, transcendence of, 27, 28,
34, 86,89,115-16,123-25,133n,
157; knowledge of, 67ff., 81-82,
111-12, 131, 140, 164-65, 16870 (see also Megiston mathema);
dynatnis of, 8 9 , 1 1 6 - 1 8 , 1 2 5
Gorgias, 48, 51, 52
Gorgiasy 75, 95n; on flattery, 96
Hartmann, Nicolai, 2, 3
Health (hygeia) analogy, 66,143; as a
parallel to the good, 113,152,154n
Hedone (pleasure): withstanding the
seductive influence of, 65, 95-96;

INDEX

as the counterpart to nous and

phronesis, 105ff.
Hegel, Georg W.F., 1 - 3 , 5, 21, 101,
107n, 131,178n
Hegelianism, 4
Heidegger, Martin, 3 - 5 , 19nl7, 36n,
39n, 55n, 56n, 66n, 90nl2, 116n,
167n, 173n
Hexis (disposition), 33, 172; as a way
of being, 39, 176. See also Logos,
logos and ergon, logos and ethos
Hippias, 39, 49
Homer, 137
Hou heneka (that for the sake of
which), 129,145-46,174, 177
Husserl, Edmund, 4
Hypothesis of the eidos, 9 1 , 1 0 0 - 0 2
Indeterminate duality (the ahoristos
dyas), 29, 93,104. See also Number
doctrine
Jaeger, Werner, 7 - 9 , 38
Justice (dikaiosyne), 11, 23, 27, 6 6 67, 142-43
Justification, giving of {logon didonai), 36n, 49, 53, 60, 65, 81, 85,
103-07, 110, 131; in moral matters, 41-44. See also Dialectic
Kant, Immanuel, 19, 97, 97nl5,173n
Laws, 64
Leibniz, Gottfried W., 26
Logoi (discourse), 20, 102; "flight"
into, 13n, 15, 26
Logos: Plato's and Aristotle's logos
(language) philosophy, 1; logos
(statement) of being, 13; logos
(word and reasoning) and ergon
(deed), 42n9, 43, 46, 54, 96-97,
99, 121 (see also Prohairesis); logos
(word, reasoning) and ethos (habituation), 60-61; logos as reason, 63,
101, 111, 167, 172 (see also Nous;
Phronesis)
Lysis, 99

INDEX

Magna Moralia, 23, 126, 128, 129,

142,147-50
Mathematics, 53, 56, 82, 134, 163;
and the chrismos, 17-18; Pythagorean understanding of, 24; its immunity to sophism, 40, 45, 99; as
preparatory for dialectic, 53,55,83,
84, 89, 97, 99, 100 (see also dianoia); verus knowledge of the good,
82; versus physics, 156-58, 177
(see also Physei onta); inapplicability to ethics, 163
Measurement, 111, 115; as the art of
finding what is fitting, 79,123,131
Megiston mathema (highest insight),
20, 23, 35, 66,121,136
MenOy 46, 60; on the unteachability of
arete and phronesis, 37, 51-54; on
anamnesis, 53, 55, 58
Metaphysics, 7 - 9 , 14, 16, 17, 25,
156; on philosophy as a prohairesis,
100; on ideality of letters, 120n; as
a collection of marginal observations, 130n; on the arche, 137n; on
analogy, 151,153n, 155; on the god
as the highest, separate good,
157-58
Methexis (participation), 6 0 , 8 8 , 1 0 0 01, 122, 132, 136-38; and the
chrismos, 9 - 1 3 , 31, 113; of parts
in the whole, 11; in the beautiful, 77
Mimesis (imitation), 10; and methexis,
11-12; of a character model^ 4 6 47, 97, 97nl4 (see also Ethos;
Hexis; Logos, logos and ergon)
Mixture metaphor, 10,14,113ff.
Natorp, P., 2, 24, 37
Neo-Kantianism (Marburg), 2 - 4 , 24,
29, 101
Neoplatonism, 1, 8,137
Neo-Thomism, 2 - 4
Nicomachean Ethics, 23, 34, 37, 63,
80, 105, 136, 129, 141, 151-59,
169; on the archai, 93, 162; on application, 162; on the archer, 163-

181

64, 166; on law giving, 168; on

phronesis, 175
Noesis (intellection), 27, 28, 87, 100,
102-03, 112, 120, 153; in distinction from aisthesis, 12,17
Nous (reason), 25, 106, 108, 111,
171, 175-77; active and passive,
88-89, 89n. See also Noesis
Number doctrine, 8, 3 1 - 3 2 , 104,
119-20, 121-22, 134, 135, 141,
143, 156, 177. See also The One;
Indeterminate duality
"On the Good^ 104,126,128
The One, 31, 91, 137, 155-56; and
the many, 28, 9 1 - 9 2 , 107, 121,
122,137; and the two, 29, 91,104,
135,137,156,178
Ousia. See Being; Substance
Paideia, sophistic, 19nl7, 36, 3 9 - 4 1 ,
47-50,51n, 61
Parmenides, 8 - 9 , 13, 16-17, 100,
136-38, 156; on methexis and the
chrismos, 10, 124, 134n6; on the
metaphor of sunli^t, 11, 87-88,
100,118
Participation. See Methexis
Peras (limit), 1 1 2 - 1 4
Phaedo, 8,12,13n, 2 4 - 2 6 , 2 9 , 3 4 ; on
anamnesis, 4 4 - 4 5 , 57, 58; on the
ontological significance of the good,
26n, 9 3 - 9 4 ; on the hypothesis of
the eidos, 9 1 , 1 0 1 - 0 3
Phaedrus: on distinguishing between
good and bad madness, 4 4 - 4 5 ; on
the techne of rhetoric, 79, 119; on
the beautiful (kalon), 116
Phenomenological description, 45
Philebus, 9, 13, 2 9 - 3 1 , 9 1 - 9 3 , 103,
104-25,135,136,144,148; on the
good in human life, 46; on the ugly
(aischron), 116
Phronesis (reasonableness), 33, 35,
96, 105, 165, 175; as unreachable,
37-39; unity of the virtues in, 6 3 66

182

Physei onta (that which is by nature),

157-58,177
Physics, 7,18, 26,130n
Pindar, 53
Plotinus: his step beyond Plato, 28; his
doctrine of the One, 31,137
Politics, 26,169,175
Pre-knowledge of arete, 42
Prohairesis (choice), 100, 109-10,
121-22,173n
Protagoras, 4 0 - 4 1 , 4 3 , 47-48, 99
Protagoras, 19nl7,41, 63-66,95; on
the unteachability of arete, 5 0 - 5 1
Proton (what is first or primary), 29,
135, 138, 158
Protrepticus, 8,170n
Pythagoreanism, 14, 15, 17, 24, 25,
53, 120,132,137, 157
Republic, 23, 26, 28, 30, 48, 60, 6 3 103, 105, 128; on knowledge of the
good, 34, 67ff., 80, 85; as a dialectical myth, 70-73; on the human
good, 77, 93, 148; on the dynamis
of the good, 117-18
Recollection. See Anamnesis
Self-knowledge, 42, 43n, 47, 58-59,
66,109n, 111. See also Anamnesis
Seventh Letter, 69; on the sophistic
abuse of the logos, 19, 99; on arete
and the whole of reality, 94
Sophism, 19,25,45,168; as the counterpart to true dialectic, 99,118. See
also Paideia
Sophist, 13, 88; on differentiating between philosopher and sophist, 45,
99; on being, 98,138,156
Sophistical Refutations, 100
Speusippus and Pythagoreanism, 15657
Statesman, 79; on measure as a prefiguration of Aristotle on the good,
123,131

INDEX

Strato, 15
Substance (ousia), 15, 130n, 141; priority of over Ae other categories,
131,152,154n
Sunlight metaphor: in the Parmenides,
11, 87, 118; in the Republic, 28,
76-77, 86-87
Syllogism, practical and theoretical,
165
Symposium, 37n, 115
Temperance (sophrosyne), 27, 69n,
167n
Techne (art), 79; versus arete, phronesis and knowledge of the good,
23, 32, 35, 37, 37n3, 4 6 - 4 9 , 80,
111-12, 131, 165; versus political
practice, 7 8 - 8 0 ; of music and of
letters, 119
Theaetetus, 17, 40, 99; on theria,
39n, 69; on aisthesis, 45, 45nl3
Theognis, 46
Theria: and politics, 67-69; and
praxis, 161,169,171-7^
Theory of scientific knowing, 74,
74n6, 160, 166,168-69
Timaeus, 8,18, 74, 125; on the world
soul, 1 4 3 - 4 4
Topics, 138
The Two, 32, 135, 178. See also Indeterminate duality; Number doctrine; The One
Vico and the rhetorical tradition, 169
Whitehead, A. N., 26
Wissenschaftstheorie. See Theory of
scientific knowing
Wirkung (effect). In; Wirkungsgeschichte (history of effects). In, Wirkungseinheit (unitary effect). In, 2,
5,13n, 173n
Xenophon, 22, 37n, 46, 4 9 - 5 0