Plato's Theory of Forms

Author(s): A. K. Rogers
Source: The Philosophical Review , Nov., 1935, Vol. 44, No. 6 (Nov., 1935), pp. 515-533
Published by: Duke University Press on behalf of Philosophical Review

Stable URL: http://www.jstor.com/stable/2179922

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Duke University Press and are collaborating with JSTOR to digitize, preserve and extend
access to The Philosophical Review

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 Number 6 Whole
Volume XLIV November, 1935 Number 264

THE

PHILOSOPHICAL REVIEW

PLATO'S THEORY OF FORMS

THE theory of Ideas, or forms, lies in the opinion of most

critics at the heart of Plato's philosophy: but what the forms
are, and what exactly is the part they play, is still a matter about
which there is enough difference of opinion to make it incumbent
on a student of the dialogues to exercise some degree of inde-
pendent judgment. The present article does not aim at novelty,
but is an attempt to canvass the more important points of evidence
in order to discover if possible to what extent they lend themselves
to a consistent and plausible picture of Plato's state of mind.
We are fortunate in possessing a firsthand contemporary ac-
count of Plato's doctrine from a thoroughly competent source;
and there would seem to be a certain advantage, therefore, in tak-
ing this as a starting point. The only thing against it would be
some definite presumption that Aristotle has radically misunder-
stood his teacher. This always remains a possibility; indeed it is
even probable that his interpretation will be found colored more or
less by his own philosophical beliefs. But when he says explicitly
that Plato held this or that opinion, even though it may not be an
opinion we should gather naturally from the dialogues, we are
plainly bound to take the statement into serious account; we should
be justified in rejecting it only in case it can be shown to be de-
monstrably inconsistent with something that Plato has himself
to say.
Setting aside for the moment technical complexities, there are
two major points in Aristotle's testimony. Plato, he tells us, iden-
tified the forms with "numbers"; this is perhaps its outstanding
feature. Also he seems to say that such a doctrine does not represent
the origin of the theory in Plato's mind. "Now regarding Ideas",
he writes, "we must first examine the ideal theory itself, not con-

5I5

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
5i6 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

necting it in any way with the nature of numbers, but treating it

in the form in which it was originally understood by those who
first maintained the existence of the Ideas."' There is one way to
evade the apparent implication of this statement; we might sup-
pose, with Burnet, that Aristotle is intending to contrast, not two
stages in Plato's thinking, but the Platonic theory with an earlier
and non-Platonic one. Such an interpretation is however pretty
plainly influenced by the need to get around a difficulty which
Aristotle offers to Burnet's personal views; there is nothing in the
passage itself to demand it, and the natural impression one gets
from Aristotle's whole treatment of the matter, to say nothing of an
explicit statement elsewhere,2 is that Plato was the originator
of the doctrine of Ideas. And now of this original sense attached
to "forms" Aristotle has a simple and intelligible account to give;
a form, namely, is a universal-anything, that is, which is the
object of a definition. "To each thing there answers an entity
which has the same name and exists apart from the substances,
and so also in the case of all other groups there is a one over many,
whether the many are in this world or are eternal."
I shall assume then, as a working hypothesis, that Aristotle un-
derstood the original doctrine of the form to have taken its start
from logical definition, and that the object of any such defini-
tion in common or universal terms has therefore a presumptive
title to be called a form. There is the less need to question this in
that it is supported by a large mass of evidence from the dialogues.
Plato employs the terms ecoq and Be'a with a wide variety of
shades of meaning; but most frequently they stand for a class
of objects or for the common nature which makes classes possible.4
It is unnecessary to quote the fairly numerous statements where
the form appears as the outcome of an act of generalization leading
to a definition ;5 but I may call attention to two passages where
the bearing on an ultimate philosophy is particularly clear. Every

1M Io78b 9. 'M io86a ii.

'A 99ob 6. See also A 987b 30; M 1o78b 30; and cf. Parm. 135ab.
4E.g., Crat. 386e, 424cd; Laches I9QId; Phaedo 7ga; Phaedr. 263b, 265a,
27id; Symp. 205b; Rep. 357c, 477ce; Theaet. I48d, I57C, I78a, i8id; Soph.
235d; Phileb. 32b, 44e; Polit. 258c, 262b, 285a, 307d; Tim. 37e, 57c, 69c, 83c;
Laws 63oe, 700a, 963c.
6Euthyphro 6d; Laches IgId; Meno 72b f, 75a; Phaedo 74b; Phaedr.
24gb, 265d, 273e; Rep. 5o7b; Pouit. 285b; Phileb. 25a; Tim. 83c; Parmen.
I32ac.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 No. 6.] PLATO'S THEORY OF FORMS 5I7

existing object has, he tells us in the Seventh Letter, three things

which are the necessary means by which knowledge of that ob-
ject is acquired-the name, the definition, and the image or phe-
nomenal appearance; then comes knowledge itself as a fourth
thing, and as a fifth we must postulate the object itself-the real-
ity which is cognizable and true." Here the special inference to
be drawn is that a knowledge of reality lies directly in the line of
progress that starts from the attempt to render the meaning of
the name precise through definition. The same conclusion follows
from the well known passage in the Parmenides.7 It is true of
course that this is put in the mouth of Socrates and might just
conceivably be dissociated from Plato's own opinions. But even
granting that Plato would have been likely to insert a fragment
of literal biography in a dialogue with the general characteristics
of the Parmenides, the probability would still be strong that he does
not himself reject the implications of the passage. We may fairly
gather, then, that for Plato the most decisive examples of the
form are connected with logical and ethical concepts; that he
had at first reason to be more hesitant about the ideal reality of
"natural kinds" like man or fire though he came to see that such
a conclusion could hardly be avoided; and that while there is still
more difficulty in believing that things like hair and mud have
likewise their corresponding forms this too is a natural outcome
to which his principle might be supposed to lead.
Before going further, there is one distinction which I think is
of some importance for the formulation of his doctrine. I have
just noticed incidentally that a difference might be made between
a group of objects that possess some common feature and this
universal element itself which justifies us in giving them a common
name. Plato often uses elboq in the former or logical sense where
no metaphysical assumption is or need be involved; when how-
ever his words do carry such a natural implication it is the latter
emphasis he has in mind.8 This is bound to follow from the r6le
which the Idea plays. Plato leaves us in no doubt as to why the
defining and universalizing intellect gives us access to a more ulti-
mate realm of being than do sensation and empirical opinion; the

6 342a. Cf. Laws 895d. 1 T3ob f.

8 Cf. Euthyphro 5d, 6d; Meno 72c f.; Phaedo 74a, T02b, To3e;
437e, 476a f.; Parmen. i2gb f.; Soph. 254c f.; 258c.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 5I8 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

reason is that nothing is deserving to be called "real" which is

not eternally fixed and stable, alike in itself and in the reflex of
itself that constitutes knowledge.9 Knowledge is genuinely know-
ledge only when it is absolutely certain and unchanging-this is
one of the major premises of Plato's thought; and it cannot be
certain unless it is attached to an unchanging object. Such an object
cannot belong to the world of phenomena within which sense and
opinion move. The only thing that persists is the eternal essence
whose presence in particular changing things enables the mind to
grasp them firmly and rescue them from the empirical flux;
and the essence, consequently, arrived at not through the bodily
senses but through the power of reason inherent in the soul, is the
goal at which all true thinking aims. It is this verifiable difference
between sensation and opinion on the one hand and rational know-
ledge on the other which establishes decisively the existence of the
forms. "If reason and true opinion are two distinct kinds, most
certainly these self-subsisting forms do exist, imperceptible by
our senses, and objects of reason only.-Now these two kinds must
be declared to be two, because they have come into existence sepa-
rately and are unlike in condition."'10
Meanwhile the special point I started out to make is a somewhat
different one. Among the forms that for Plato belong to the in-
telligible world it is possible, and for certain purposes almost ne-
cessary, to make a further distinction. This is the distinction be-
tween forms where the emphasis is on the common nature of a
more or less well defined group of individuals, and those that
stand primarily for some single abstract characteristic, like beauty
or equality or greatness, in which indeed a number of objects may
participate, but without thereby constituting a generally recognized
class or species for the ordinary purposes of human thinking.'1
It is seldom we have occasion to speak of "equal things" or "great
things", or even of good and beautiful things, unless our interest
lies in equality or greatness rather than in any "class". And in
the large it is in the second connection that the metaphysical status

9Meno 98a; Crat. 439g f.; Phaedo 78de, 79d; Rep. 479a, 485ab, 490b, 500c,
524e, 585c; Theaet. i83ab; Parmen. I35bc; Phileb. 59bc; Tim. 52a; Epis.
VII, 342C.
10 Tim. 5id. Cf. Meno 98ab; Rep. 477e f.
" Cf. Met. g9ob i6.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 No. 6.] PLATO'S THEORY OF FORMS 5I9

of the Idea is most unequivocal. This is by no means universally

the case. In the Philebus, for example,'2 man and ox appear in
company with beauty and the good as typical Ideas; indeed we
shall have to consider later the possibility that such natural kinds
came to be for Plato forms in a special and distinctive sense. But
taking the dialogues as a whole I think that with qualifications the
statement just made will stand. At least the ability to univer-
salize objects of discourse for purposes of identification does not
seem necessarily to give them an ideal status; empirically defin-
able class-terms may be ultimately real, but then again to all ap-
pearance they may not. It is difficult to suppose Plato would have
allowed a place in the "heaven of Ideas" to forms of the changing
visible object, of doubtful things, of madness, of evil, of imita-
tion ;13 whereas the same doubt seldom arises when he is talking
about the abstracter characteristics which logical analysis uncovers.
And now it follows that we are called on to distinguish two
steps in the intellectual procedure which philosophy presupposes.
In the first instance Plato will be found engaged with such a logical
analysis-with breaking up, that is, the complexity of thought
into its constituents. This is of course the first essential if we are
to have anything like exact scientific definition or explanation; and
there is a side of his thinking, therefore, where the ideal world
may seem to take the form of a multitude of such simple, self-
identical and unchanging entities. This would help explain why
natural species such as man are presumed in the Parmenides to
carry less immediate conviction than the abstracter forms of
beauty or similarity. To stop here would however clearly be im-
possible. For analysis is not as yet definition and as a matter of fact
might appear to leave us with a host of indefinables; for con-
crete thinking it is merely a preliminary stage. When Plato tells us
that the mass of men can never be expected to rise from the per-
ception of beautiful things to the Idea of beauty, he is taking too
low a view of the powers of the average mind if he means by
beauty-in-itself only the abstract character implied in calling ob-
jects beautiful. One who has had no practice in analysis may find

13 Phaedo 79a; Phaedr. 263b, 265a; Lysis 2i6d; Rep. 445c, 476a; Soph.
235d.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
520 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

it a little hard at first to avoid confusing instances of beauty with

the quality they embody, but Socrates usually has no great trouble
in making even rather stupid hearers recognize the difference;
such a direct jump from the use of a common term to the percep-
tion that the term points to some identity in the things to which
it is applied is well within the reach of a man of very moderate
intelligence. But after the feat has been performed we still have
taken only a short step. "Beauty" does not define beautiful things
in the sense of interpreting them rationally; at the most it can
be held in a certain sense to explain why they are beautiful, and
even this it does after what Plato himself calls a "safe and un-
tutored" fashion.14
If we are genuinely to enter the field of dialectic, then, logical
analysis will need supplementing; it sets problems which are in no
way answered merely by translating the characteristics it brings
to light to a higher realm of being through the recognition that they
are now freed from the taint of relativity and change, or by em-
ploying such abstract forms to characterize a vague class of
objects which participate in them. Philosophy is interested in the
definition of the forms themselves. Now it may be that some of
them are really indefinable; and as a matter of fact this very
well may seem to be the case with the logical "categories"-being,
identity, unity and the like. Accordingly in treating these Plato in
practice accepts them for most purposes as ultimates immediately
open to the soul's perception, which dialectic has to presuppose.'5
This is however to anticipate a little; what needs here to be ob-
served is that in other cases, and those the ones that come closest
to his primary interest, Plato does aim at a definition of the form.'6
The earlier dialogues in large part are concerned with defining vir-
tues, which evidently does not mean the mere easy recognition of
an abstract identity among virtuous acts but assumes the possibility
of some more illuminating description of that wherein the essence
of the virtues-temperance, justice, holiness-consists; and the
ultimate quest of Plato's whole philosophy is to define the nature
of the Good. What then is the direction such a program takes?
First a word about the more general conception of method.

14 Phaedo 105c. Cf. iood.

15 Theaet. i85c-e. 16 Cf. Laws 964a.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
No. 6.] PLATO'S THEORY OF FORMS 52I

Dialectic in its simplest form is the way of question and answer

which we can fairly assume goes back to Socrates. Here the dis-
putants aim to find some statement or hypothesis on which both of
them for the moment can agree. This is examined to see what con-
sequences follow from it, and so long as no contradictions develop
it is allowed to stand; if however it does give rise to inconsistencies
it is abandoned and a new hypothesis is set up in its place. So far it
represents primarily a formal condition of fruitful argument. But
for the philosopher who aims at definitive knowledge it is some-
thing more.17 The mere fact that the disputants agree does not
show that their opinion is correct; even the mathematician finds
himself starting from hypotheses which, while their general ac-
ceptance makes his special science possible, are nevertheless taken
for granted rather than demonstrated. It is the business of an ul-
timate philosophy to "do away with hypotheses in order to es-
tablish them". They must be derived, that is, rather than as-
sumed; and so unless we are to be committed to an infinite regress
we must sooner or later arrive at principles which themselves stand
in no need of further evidence.
In a purely formal way this might perhaps be interpreted as a
process of abstraction which leads us in the end to the most
abstract form of all-pure Being; that is the course which phi-
losophy has sometimes taken. But it is not Plato's way. Somehow
reality is for him a system-a universal which includes complexity
and unity alike; he is explicit in rejecting the Eleatic One inter-
preted as just Being itself into which no diversity enters. The
blanket objection to Eleaticism is that it wipes out forthwith
the whole phenomenal world, and so leaves no place for science
in the sense in which science aims to throw light on the actual
facts of the universe and of man's experience. The business of a
true philosophy is to explain appearances, not to explain them
away; and it can only do this in terms of inclusiveness rather than
of exclusion18 And for the same reason Plato is bound to reject
a conception of the form which would regard this as a self-sub-
sistent entity alongside a mere conglomerate of other forms.

17 Cf. Meno 98a; Rep. 435d, 504b f., 5Ioc f., 53I-4-
18 Phaedo 99c; Phaedr. 265d-266b, 268de, 27iab; Rep. 476a, 53id; T
2oga f.; Soph. 253d; Polit. 285b, 304c; Phileb. I5bd, i6c ff., 62a-c; Tim.
3ocd, 4ib, 69c, 92c.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
522 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

"The complete separation of each thing from all is the utterly final
obliteration of all discourse; for our power of discourse is derived
from the interweaving of the forms with one another."19
Before going on it will be well to call attention to the existence
here of two separable problems which in the interest of clearness
ought not to be confused. The first thing which the notion of com-
munity or "participation" is likely to suggest is the relation between
universals and particulars; here the ultimate question is an onto-
logical one and concerns the nature of the connection between two
realms of being-the ideal realm and the world of phenomenal
existence. The second and related question is one of epistemology.
On its more superficial side this has to do with the possibility of
knowing objects, and it makes its appearance in the first instance
in connection with the logical difficulties raised by the eristics, with
their insistence that only identical propositions can avoid self-
contradiction. Plato does not take this seriously enough to dwell
upon it; the difficulty would apparently be met, in part, by the
conception of the individual as a meeting-place of various forms-a
doctrine logically unexceptionable whatever its metaphysical basis
may be thought to be. The real trouble comes when we turn to the
forms themselves and ask whether these may be supposed to par-
ticipate in one another.20 The solution of the more fundamental
problem is found in the Sophist, where in Plato's theory of predica-
tion the separation between ontology and a theory of knowledge
becomes explicit; here the "is" of existence is definitely distin-
guished from the "is" of predication, and "is not" is made equiva-
lent, not to "does not exist" but to "is other than".
Coming closer now to the notion of system in its concreter
application, there is one general consideration which it will be
well to notice at the start, since it will need to be borne in mind
constantly in any attempt to interpret Plato's meaning in detail. It
has appeared that Aristotle identifies Platonic forms with num-
bers; and whatever conclusion we arrive at about this identifica-
tion, it at any rate is true that in a certain large though rather in-
definite sense number, or mathematics, has an important part to
play in Plato's actual procedure.21 Even in the earlier group of
"9Soph. 259e. Cf. 249c; 25ie f.; 257d ff.
20 Phileb. I4C f .; Soph. 25ia-d.
2 Cf. Meno 82 ff.; Gorg. 5o8a; Rep. 522C ff., 546, 6o2d-603a; Phileb.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
No. 6.] PLATO'S THEORY OF FORMS 523

dialogues we find the new mathematics drawn on to give concrete-

ness to the method of dialectic and to lend true scientific character
to a philosophic education; and in the more technical context of
the later writings the influence of the mathematical ideal is con-
stantly making its appearance, expressly or by implication.
However, the interpretation of this role is not altogether easy.
We might perhaps be tempted to think of it as the reduction of all
exact knowledge to mathematical formulae in the shape of "laws";
but this, in the modern sense of such a claim at any rate, goes
considerably farther than the evidence warrants. At the other ex-
treme it would be possible to hold that all Plato really means is to
recommend exact science-which meant in his own day mathe-
matics-as a disciplinary study to train the philosopher in the
habits of intellectual precision which belong to true or dialectic
thinking. Plato does recommend it for this purpose; or again, at a
somewhat lower level, it may be spoken of as a cultural attainment
of which the educated man ought for his own credit not to be in
ignorance, as an intellectual game appropriate to persons of in-
telligence, or, in the form of astronomy especially, as a require-
ment for genuine piety toward the gods.22 But this hardly by itself
does justice to all that his words frequently imply. In the Republic,
for example, where such an interpretation is more than once sug-
gested,23 the mathematical sciences seem likewise to be regarded
as a stage leading to the vision of the Idea of the Good, the princi-
ples which bring us into contact with reality being just the principles
by which the hypotheses of mathematics are themselves estab-
lished or explained.24 Similarly in the Timaeus we are told that
Time "moving according to number" is created by God to serve as
a movable image of Eternity; living creatures attain to reason
through their participation in number; it is number that supplies
the means of research into the nature of the universe; the elements
are generated by means of forms and numbers.l2l5 At any rate we
cannot rule out the possibility that mathematics is constitutive for

25a-e, 55d ff.; Laws 8I7e ff., 967a-e; Tim. 35b f., 43d. See also Epin. 9)76e,
9goa fif.
22 Laws 82obc, 822bc.
23 Rep. 503e, 524d-525e, 526b-e, 527d, 530c, 534d, 536d. Cf. Lawzs 747b.
533c. Cf. 5II; 53Ic. See Epinomis g9oc.
2 37d, 39c, 47a-c, 53b, 54d.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 524 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

philosophy prior to a more detailed consideration of Plato's actual

reasoning and results.
It is of some importance to start from the right angle. Now we
are pretty safe in saying that Plato's primary interest lay in ethics,
and in ethics as a political or social rather than as a narrowly indi-
vidual affair: he himself tells us this is so in the Seventh Letter; his
two largest works-one of them also his latest-are political
treatises; the Academy to which he devoted a large share of his
life was a school for statesmen; and everywhere the idea of the
"good" crops up as the goal of the philosopher.
On the negative side this carries two conclusions, one of them
perhaps a little more obvious than the other. At least it is clear
that Plato is not first of all interested in the physical sciences, as
was the case with the earlier Greek philosophers. In only one late
dialogue does he give these any explicit treatment, and he makes it
plain that he does not consider them capable of definitive results;
they are interesting as intellectual diversions,26 but conclusions here
are probable or plausible rather than certain and necessary, and so
are not science at all in the stricter sense.
It cannot be set down as quite so clear that metaphysics is like-
wise subordinate in Plato's mind to an ethical or political
philosophy; indeed in one sense this cannot be the case, since ethics
is itself in intellectual terms dependent on an ultimate theory of
reality. It is this last conviction which explains one of the most
distinctive features of Plato's political system-the place, namely,
that philosophy and philosophers hold in the state. Plato maintains
consistently that the only hope for the creation and maintenance
of a rational society lies in the few, and ultimately perhaps in the
one man, in whom genuine wisdom resides; and he means by this
more than the familiar demand for "intelligence" in rulers as a
compound merely of a sound education, clear thinking, social dis-
interestedness and a wide experience with affairs. What is called
for is precisely the sort of undebatable knowledge which, fostered
if not constituted by the mathematical disciplines, is able to relate
statesmanship to the ultimate principles of order in the universe.27

a Tim. 59c.
27 Theaet. I73e; Laws 965cd, 966b-968a. Cf. the attitude toward the
"atheist" in the Laws.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
No. 6.] PLATO'S THEORY OF FORMS 525

Nevertheless, in practical terms of interest a

we seem justified in calling ethics fundamental. Both as an indi-
vidual and as a member of society, "reality" translates itself for
the philosopher into a source alike of happiness and duty: of duty,
because it constrains him to bring his vision of the good down from
the heavens into the world of men as a pattern for remodelling this
world; of happiness, because only in "knowledge" does the true or
rational soul attain the fulfilment of its essential nature.
It is of course impossible to make anything like a concrete
summary of Plato's ethics in a few paragraphs; but about what I
shall need to say there will be no great difference of opinion. At
least this will be true of the practical pattern of the good life which
he draws for us, and from which most of his influence upon
modern thought and culture has derived. This takes the form of
an impressive rendering of the historic Greek ideal of moderation,
proportion, harmony; and it has a permanent appeal in so far as
such qualities do actually enter into goodness. So in terms of indi-
vidual living it upholds the claims of an orderly and seemly and
aesthetically pleasing satisfaction of desire under the control of
mind or reason, in opposition alike to the chaotic and tumultuous
gratification of particular desires and to an arid asceticism. In the
same way on the social side it sets up the ideal of a state in which
personal ambitions and the clash of warring factions have given
place to a harmonious community of interest, and where each man
by confining himself voluntarily and uncomplainingly to the specific
function for which his nature fits him is ready to subordinate his
private happiness to the demands of a single organized community
governed by a common reason. Such an ideal, it is possible to ar-
gue, is likely to minimize, and in Plato's hands does minimize,
certain requirements of which man's subsequent experience has
increasingly revealed the need. It accepts too easily the assumption
that human beings have determinate and determinable natures
which can be fitted into a universal scheme of fixed social relation-
ships without prejudice to a reasonable demand for personal good,
and so without the inevitable appearance of a conflict between indi-
viduals and classes. It places too much faith in the power of a
strictly regulated education and of custom guided from above to
keep men from questioning whether in reality the limitations thus

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
526 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

imposed are genuinely justified in terms of their own satisfaction

with the outcome. In particular it fails to give its proper place to
the need for experimentalism if man is to discover his possibilities
of good as a changing creature in a changing world, and badly
overestimates the ability even of the wisest man to canalize human
life through pure scientific reason in anticipation of such a growing
experience. Nevertheless, as an ideal standard, rational harmony
has unquestionably an important r6le to play as a constituent, if
not as an inclusive definition, of the good or satisfying life.
It is not however with the content of Plato's ideal that I am here
concerned but with its underlying principles. And concretely the
issue turns on two alternatives. A justification of the Platonic
notion of the good as a human concept might be put, and would
most commonly be put in modern thought, in terms of the standard
of human happiness itself; empirically man finds his truest satis-
faction in a rational control of the desires rather than in excess
and an anarchical freedom, and in submitting to a common benefit
rather than in subordinating society to personal or class ambitions.
And Plato does in fact argue that such an orderly and ordered ex-
perience is the only sure guarantee of human happinesss.
But it is not such concrete happiness that makes the ideal good;
happiness is a consequence rather than an ultimate norm. As a
standard, the content of happiness is of necessity dependent on a
process of discovery that is bound to be tentative and groping and,
possibly or even probably, irreducible to a single common measure.
But for Plato such an experiential goal will unavoidably fall short.
Ethical good, if it is to constitute science, must find its warrant
in some rational source binding on every rational being alike; ex-
perience or experiment may give rise to right opinion, but in the
nature of the case never to real knowledge. In its ultimate interpre-
tation, therefore, the Good is not a specifically human concept; it
is a higher category from which man and all his empirical notions
of the good will need to be derived. Human teleology in its ordinary
meaning is for Plato never constitutive, and he is inclined to treat
it with something like distaste. The true "end" even of the most
reputable sense-the sense of sight-is not vulgar utility, but the
service which sight performs in revealing to the mind the order and

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 No. 6.] PLATO'S THEORY OF FORMS 527

beauty of the heavenly bodies in their cour

implies that the mathematical relationships embodied in the uni-
verse are themselves good, and not first rendered good by some
human purpose they may serve.29 Only in one field is a "form"
actually constituted by a human use. In the case of articles of
human workmanship, like a bed, the "essence" of the ideal bed is
definable in terms of the service it performs. But while such forms
of artificial objects would seem to be called for by Plato's premises,
and are occasionally referred to,30 it is only very occasionally, and
they evidently lie outside his main interest.
It is not enough for the philosopher, accordingly, that he should
have an acquaintance with the good such as grows out even of the
most intelligent "experience"; he must be able also to show why
the virtues are good by showing how they are the necessary out-
come of ultimate principles which approve themselves to pure in-
tellect. Reason in ethics means not an adjustment of the empirical
capacities of human nature to the conditions of its environment,
to be tested by any man's felt satisfaction with the outcome; rea-
son is a distinctive part of man's nature of a sort to impose its com-
mands authoritatively on other and lower parts, and in it alone
or, more strictly, in the real object which the exercise of mind
discloses, the final nature of the good is located.31 What is this
object, then, from which human good derives?
Up to a point Plato's answer is explicit; it is order, harmony,
proportion.32 Order is not justified in terms of human life, but
human life is justified in terms of order. One consequence follow-
ing from this is the tone of disparagement that is apt to crop, up in
Plato's attitude toward the "merely" human. Not only can indi-
vidual happiness enter no claims against the common good, but the
common life itself is relatively insignificant when set alongside the
universal cosmic order, so that possibly indeed we ought to look
at the whole human race as just a plaything of the gods ;33 what
Tim. 47; Rep. 530d.
2 Cf. Philebus 5Ic for the beauties of geometry.
90Crat. 38gb f. (cf. 39ob); Rep. 596-7; Phaedo 78e; Epis. VII, 342d.
" Tim. gob-d.
82Gorg. 493c, 504b f., 5o6e, 5o8a; Rep. 403a, 430e, 443de, 444d, 486d,
59Id; Phileb. 26b, 64d ff.; Tim. 3oa, 47d, 87c f.; Laws 689d, 6gicd, go6c;
et passim.
3'Laws 644d, 8o3bc, 9o3bc. Cf. Rep. 604c.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
528 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

lends significance to man and to the state alike is the embodiment

in them of the ideal principles of balance and proportion. Only on
such terms can we justify the displacement of intelligent experi-
mentation by "science" as a disinterested intellectual discipline for
which desire is not a datum but a recalcitrant, and on the whole
a rather deplorable, occasion for the exercise of rational authority.
Meanwhile if we were to ask why order itself is good-as Plato
fails to do-there are two answers possible. Plato himself assumes
its goodness as self-evident; it is a premise which for the phi-
losophic mind requires no proof. But even on his own terms there
is a little more that might be said. Measure is recognized as good,
and evokes our human love, primarily because for the thinker it has
an inherent attractiveness and charm; it is an object of aesthetic
appreciation, possessing as such that emotion-compelling beauty
which is never far apart from the conception of goodness in
Plato's pages.34 And here of course the modern man very likely
would go on to raise a further query: Why, he might ask, is beauty
good? Is it not just for the reason that it gives rise to this peculiar
sense of satisfaction? But Plato does not raise this question.
Beauty likewise for him is good as such, self-evidently good; its
ideal goodness is no doubt the source of satisfaction, but it is the
goodness of beauty which explains the satisfaction and not the
satisfaction that gives us the right to speak of beauty as a value.
Now in resolving ultimate goodness into order or measure we
are once more, it might seem, brought within hailing distance of
Aristotle and the "number" theory. However, at best a good
deal will still remain to be done before such an identification can
be found acceptable. Certainly we are still at a long remove from
the "science" of mathematics in any technical sense. Whatever
Plato's ultimate ambition, he does not even make a start toward
deriving the principles of ethics or of statesmanship from such
a science in its modern acceptation; in the existing state of mathe-
matical theory he could not have done this even if by any chance
the thing is possible. His actual procedure comes much closer to
the practices of ordinary human debate. He starts with the abstract
principles implied in order-stability, subordination, organic.
unity-which as has been said he takes as in themselves self-evi-

"Cf. Phileb. 64c-e.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
No. 6.] PLATO'S THEORY OF FORMS 529

dently good. Next he postulates as well, not indeed the actual

variety of human nature, which because it fails to reveal at once
an orderly pattern is aesthetically unpleasing and therefore to be
kept outside the ideal picture, but certain large empirical distinc-
tions of abstract human type. The task of the philosopher-states-
man is then to arrange these types in the relationships which are
most satisfying to the order-lover's sense of proper subordination,
and to devise ways, first for bringing into existence an actual State
that shall in so far as possible embody the ideal form which alone
has perfect being, and then for guaranteeing its persistence without
change, since by hypothesis any alteration must of necessity be for
the worse. It is because he thus works, not upward from the actual
but downward from an ideal pattern revealing itself to mind in
independence of phenomena, that none but the philosopher, again,
has the slightest chance of discovering the true form of the State.
We are left accordingly with two general problems: What more
precisely is the Good-in-itself which lends reality to all human
good? and, What is the specific nature of the intellectual process
through which genuine "knowledge" is attainable? The two ques-
tions are connected very closely, and it will be convenient to begin
with the second.
The point with which I am mainly here concerned will come
into view if we try to determine more exactly the relationship be-
tween the two large interpretations or emphases that have made
their appearance in the previous discussion. To the modern phi-
losopher, influenced more especially by Hegel, it probably will seem
that the most intelligible account of knowledge is in terms of
system. What constitutes a virtue, for example, whether political
or individual, is the possibility of so defining it that it may have a
place in a harmonious order of experience that meets the intellec-
tual demand for consistency, interdependence, stability and
rational completeness. Pointing in the same direction is the import-
ance for scientific method of classification or division, which aims
at a definite number of specific and qualitatively limited sub-divi-
sions.36 On the other hand it would be possible to suppose that
Plato is after something more than this-a method of deduction,
namely, resting on mathematical science in the stricter sense. The

35 Cf. Rep. 557c. 36 Cf. Phileb. i6d if.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 530 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

difference may be illustrated in terms of aesthetics. For ordinary

purposes it may be enough to stop with the direct aesthetic per-
ception of harmonious relationships. But conceivably such a
harmony may be capable of translation into mathematical terms;
and if this is the case its mathematical expression will be the final
desideratum of the theorist.
That this latter program Plato fails to carry through in practice
is by itself no proof that it may not in his mind have represented
an ideal; nor is there lacking a certain amount of evidence in the
dialogues to support such a supposition. The central place of
mathematics in Plato's system of education creates perhaps some
presumption that he actually was looking in this direction. Again,
while it doubtless would be unsafe to allow too much weight to his
occasional attempts to derive various aspects of the social order
from the principles of number-the size of the ideal state for
example, or the laws of birth-contro137 -at the same time it is
difficult to avoid altogether the impression that Plato takes this a
little more seriously than critics have been ready to allow.3'8 And in
general it is number which, as has appeared in previous references,
makes possible exact or scientific knowledge; in the Timaeus
knowledge seems even to be identified literally with those simple
and undeviating motions of the soul which copy the mathematical
precision of the cosmic system, as against the irregular and dis-
orderly motions that confuse and interfere with these.39 Especially
significant is the part which proportion and measure play in the
whole Platonic ideal, since proportion and the "mean" imply the
possibility at any rate of a numerical expression. So in the Philebus
we find measure and definite quantity pretty thoroughly identified;

"Laws 737e f.; Rep. 546.

' The number 5040 is justified most obviously by its empirical convenience
in terms of the subdivisions it permits; but also it would appear not to be
without a certain intrinsic authority of its own. "My most excellent friends,
be not slack to pay honor, as Nature ordains, to similarity and equality
and identity and congruity in respect of number and of every influence
productive of things fair and good"; each such portion or subdivision
"must be regarded as a sacred gift of God, conformed to the months and
to the revolution of the universe". And again, a propos of fixing the size
of the utensils that citizens may possess: "He [the legislator] must rec-
ognize it as a universal rule that the divisions and variations of numbers
are applicable to all purposes.-The lawgiver must keep all these in view
and charge all the citizens to hold fast, so far as they can, to this or-
ganized numerical system; for in relation to economics, to politics and to all
the arts, no single branch of educational science possesses so great an in-
fluence as the study of numbers." Laws 74ia, 77ib, 747ab.
43b f., 47C.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
No. 6.] PLATO'S THEORY OF FORMS 531

it is the class of the finite-equality and the double and anything

which is a definite number or measure in relation to such a num-
ber or measure-which by being mixed with the infinite of more
and less "puts an end to the differences between opposites and
makes them commensurable and harmonious by the introduction of
number", and in this way gives rise to the many beauties of the
world of nature and the world of man.40
Here however a point is likely to arise which complicates the
issue. It is perhaps worth noticing that the most unequivocal ref er-
ence to the quantitative possibilities in the way of measuring excess
and deficiency occurs in a passage where Socrates for the sake of
argument is adopting the premises of a commonsense hedonism.'41
But a pleasure-calculus is of course something that Plato decisively
rejects. His whole practical philosophy rests not on a quantitative
calculation in this sense, but on a system of values to which the
quantity of pleasure, in the form of intensity at any rate, is irrele-
vant ;42 it is such intrinsic value rather than anything belonging to
pleasure in its own right which in the last resort decides what
pleasures are to be regarded as true and desirable. The same thing
comes to light in his doctrine of political equality. Here the simpler
notion of equality as determined by measure, weight and number
is subordinated to the more difficult principle of a "natural equality
given to things unequal" ;43 before, that is, we can attain an ideal
justice we shall need to settle the relative degrees of rational im-
portance attaching to the various men and classes in the state, and
this involves estimates of value to which mere numerical calcula-
tion seems inadequate.
The nature of the distinction Plato himself formulates in terms
of the presence of a norm; measure involves not simply a relation
between the greater and the less, but their relation to a third
thing the "standard of the mean".44 Now it may be possible,
once more, that the ratio involved in every mean is ideally capable
of being determined numerically; but Plato no more than Aristotle
attempts to show this.45 In terms of practical procedure the mean
will be found assuming some underived principle of comparison;
and such a principle is itself not a ratio but a value. This is implied
40 25a-26b. 41 Prot. 356a-c, 357b.
4' As appears from the Philebus, intensity as suc
4' Laws 757b-d. 4 Polit. 283c-284e.
4 Cf. Rep. 6iga; Laws 728d f., 744de; Phileb. 26b.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
532 THE PHILOSOPHICAL REVIEW [VOL. XLIV.

in the frequent insistence in the earlier dialogues that ethics has to

do not with the way of attaining some de facto and unexamined
end but with the significance of the end itself. As the passage in the
Politicus remarks, the mean is the moderate, the fitting, the oppor-
tune, the needful ;4,6 and if we go on to inquire, Fit for what? we
are introduced almost inevitably to preferences which are human-
istic rather than mathematical. Such a preference may take a
variety of forms. Not infrequently it is hardly more than the Greek
gentleman's instinctive and undebatable sense of what is decorous
and seemly-the dislike of laughter for example, or of any other
immoderate indulgence, as sinning against the precept of "nothing
too much".47 Or it may be the outcome of some empirical end or
purpose.48 Commonly it finds a backing also in a larger principle;
thus the esteem felt for obedience to authority and a contentment
with one's lot, for the "middle state of cheerfulness" as a condition
of the soul, for a blend of high spirit and of gentleness in the ideal
citizen avoiding savagery on the one hand and an undue docility
and supineness on the other,49 is partly, again, an immediate
aesthetic preference, and in part it has its ground in the standard
of an ultimate human satisfaction or welfare. Most of the detailed
specifications of the ideal state have this latter justification.50
More ultimate still, and penetrating every corner of Plato's ideal
state, are the abstracter demands which order as such involves as
Plato thinks of it. Of the two ways in which political order might
be justified-as a strait jacket for restraining the prodigality of
human nature, or as a condition and starting point for enlarged
opportunity and free experiment-it is the first, as has been said,
to which he is inclined; and the resulting principles of stability,
authority, discipline, aesthetic harmony, everywhere set standards
therefore to which the outlines of the political community must
conform. Deserving of some special notice in this connection is
Plato's strong aesthetic preference for simpiicity5 -a preference
reflected in his feeling for the beauty of straight lines and circular
motions, and one which inevitably draws him away from much
sympathy with the tangled life of concrete human beings toward

4 284e, 286d. Cf. Phileb. 66a.

4 Cf. Rep. 388e, 425b; Phileb. 47a, 65d-66a.
4 Polit. 286d f., 293b. ' Laws 792cd; Rep. 4IOC f .
5" Cf. Laws 743c, 9i8bc; Polit. 293d 51 Cf. Phileb. 5Icd; Ti

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms
 No. 6.] PLATO'S THEORY OF FORMS 533

an artificially simplified ideal of essential man. Specifically it takes

shape in the standard of "purity". Purity, conformity to type,
points always toward simplicity of nature unobscured by the com-
plications of the mixed; as such it has an intrinsic beauty for
reflective contemplation which tends to disappear as complexity
increases. Just as a little pure white is fairer than a larger quantity
of white when it is mixed with other colors, so in the humanistic
field the same aesthetic demand is everywhere in evidence. It
influences Plato's judgments on the content of man's happiness; it
is for this reason that a "good" pleasure must be free entirely from
its opposite, pain, while knowledge likewise loses in purity and
goodness in the degree that, through the intrusion of empirical
elements, it departs from its essence-immutability and mathe-
matical exactness.52 It determines his rejection of the "mixed"
types of art in favor of the simpler and more austere ones.53 And
it would not be easy to exaggerate the significance which the same
principle possesses for the pattern of the ideal state. In particular,
it is a main source of "one man, one task" as a fundamental social
principle, and of various communistic features that grow out of
this. Here utility no doubt also plays a part. But it is not merely
in the interests of efficiency that Plato would have us limit each
man strictly to one particular function; quite as important prob-
ably is his dislike of the whole notion of variety and complexity
in human nature as derogating from a purity of type whose ideal
form the mind can contemplate and rejoice in.54 It is this which
makes it so easy for him to sacrifice without regret all the concrete
human joys that arise out of special tastes and personal relation-
ships for the purely general goods that belong to all men alike by
reason solely of their common nature.55

(To be concluded)
A. K. ROGERS
ROCKPORT, MASSACHUSETTS

52Phileb. 52d-53c, 58a ff. For a psychological reason for this exclusion
pain see Philebus 32ab. Of course, as an empiricist, Plato cannot carry out
such a principle consistently; the whole thesis of the Philebus is that in the
"'mixed" life of pleasure and knowledge the good for man consists. But
this is because man belongs to the natural world; ideally we should expect
pleasure to be excluded, as we seem bound to believe it is in the life of God
(22c, 33b).
5 Rep. 397d.
5 Rep. 397e. 5 Rep. 462a-e.

This content downloaded from

86.142.145.236 on Sat, 29 Aug 2020 09:56:55 UTC
All use subject to https://about.jstor.org/terms