0DVV7HUPV*HQHULF([SUHVVLRQVDQG3ODWR

V7KHRU\
RI)RUPV
5RELQ6PLWK
Journal of the History of Philosophy, Volume 16, Number 2, April 1978,
pp. 141-153 (Article)
3XEOLVKHGE\-RKQV+RSNLQV8QLYHUVLW\3UHVV
DOI: 10.1353/hph.2008.0360

For additional information about this article

http://muse.jhu.edu/journals/hph/summary/v016/16.2smith.html

Access provided by Universidad Complutense de Madrid (2 Sep 2015 13:25 GMT)

Mass Terms, Generic Expressions,

and Plato's Theory of Forms
ROBIN

SMITH

I.
A FUNDAMENTAL PURPOSE of Plato's Theory o f Forms is to explain why
things have certain properties; for instance, why some things are beautiful, why
some people are just, why some objects are beds. We can (and Plato often does)
state this purpose in the formal mode also: the Theory explains why certain predicates are true of certain things; for instance, why 'beautiful' is true o f some things,
'just' of some people, 'bed' of some objects. The Theory produces explanations by
supposing that at least for each of a certain group of predicates there exists a
corresponding Form; that the predicate is true of the Form, and indeed supremely
true of it; and that if the predicate is true of anything else besides the Form, it is
because that thing is in a certain relationship with the Form (which Plato most often
designates 'participation'). Stated in terms of properties rather than predicates, this
would be: the Form has the property, and indeed supremely; and if anything else has
the property, it is because that thing partakes of the Form.1 Plato sometimes says
this as: each Form has a certain 'power' (dunamis) which it imparts to its participants (Prt. 330, Prin. 133e). As a sort of corollary, Forms serve to give the meanings
of predicate terms; those terms get their meanings by referring in some way to the
Forms to which they apply primarily. Forms also play several other roles in Plato's
philosophical picture. From the last point, it follows that in order t o know the
meaning of a predicate (or in order to know what a property really is) one must have
knowledge of a Form. Thus Forms are the center o f Plato's epistemology. They
occupy an analogous metaphysical position, since he regards them as real in a way
that other things are not. In a number of places, they play a role as paradigms-paradigms that things in nature imitate and aspire to (Phd. 75, Prin. 132d), or
paradigms that an artisan would make use of (Cra. 389, Rep. 597). Both o f these
conceptions appear in the Timaeus. We may also note that Forms play roles as
objects of aesthetic and ethical admiration.
In order to play these roles, Forms must have certain characteristics. First, they
must be predicatively pre-eminent: a F o r m must be the pre-eminent bearer of its
name, the paramount instance of its quality. This includes the claim that Forms
must be, in the phrase introduced by Vlastos, 2 self-predicative. Next, whatever
Forms are, they are to be distinguished from ordinary physical objects. Plato's view
' Condensed versions of this account of the Theory of Forms appear several times in the Parmenides:
130b2-5, 130e5-131a2, 132al-4.
Gregory Vlastos, "The Third Man Argument in the Parmenides," Philosophical Review 63 (1954):
319-349; reprinted, with "Addendum (1963)," in R. E. Allen, ed. Studies in Plato's Metaphysics
(London, 1965), pp. 231-263.

[141]

142

HISTORY OF P H I L O S O P H Y

here undergoes development as he becomes more aware of his views and their consequences, but at least in the Phaedo and later works we find definite claims distinguishing the Forms from those things perceptible to the senses. We may conveniently refer to this characteristic of Forms as separation (the term is of course
Aristotle's). Both a motive for and a consequence of the separation o f the Forms is
Plato's view that they are eternal and immutable. Finally, especially in later works,
there is considerable emphasis on the unity and the intelligibility o f Forms.
A rather surprising thing about this Theory is the ease with which participants in
the dialogues accept its claims. In the Euthyphro, for instance, when Socrates asks if
" t h e impious isn't always contrary to all the pious, but similar to itself, with whatever is going to be impious having some one Form [mian tina idean] with respect to
its impiety," Euthyphro, who is presumably a philosophical innocent, replies, " B y
all means~ yes, Socrates" (5d2-6). In the same way, in the Hippias Major Hippias
answers without hesitation that the just are just by justice, the wise wise by wisdom,
the good good by the good, and all the beautiful beautiful by the beautiful, as well as
agreeing that all these apparent Forms exist (287cl-d2). Responses of this sort are
altogether typical, especially in the earlier dialogues: only in the Phaedo and later
works, and then only occasionally, is any justification or explanation offered either
for claims that Forms exist or for claims that things have properties by partaking in
Forms. Moreover, when we do find any argument on these matters, it usually
concerns what Forms must be like, granted that they exist. For example, in Phaedo
74a-75b, Socrates argues that The Equal Itself is not an object of the sensible world.
He does not argue that it exists; he takes it for granted that it does. Similarly, there is
no argument in the Republic (596a) that there is a Form of bed; we are just
" a c c u s t o m e d " to suppose a single Form whenever many things share a name.
The role Forms play in explaining predication is presented with a similar lack o f
controversy. In the Phaedo (100c-d), Socrates apparently regards it as so obvious
that the beautiful are beautiful because of the beautiful itself as to be a simpleminded commonplace. More surprising still, perhaps, is the complete absence in the
dialogues of any suggestion that the predicative pre-eminence o f the Forms is
anything in need of explanation or defense. In the Phaedo passage just mentioned,
Socrates takes it as obvious that " i f there is anything else beautiful besides the
beautiful itself," it must be by partaking of that Form (100c4-6). Cebes shows no
surprise at the notion that the beautiful itself is beautiful. Likewise, in Republic X
the Form of bed is clearly taken to be a bed. Consider, for instance, the following:
" T h e n will there not be three beds? One is that which is in n a t u r e . . . " (59765-6).
No argument for this proposition is found. Similarly, the shuttle itself is taken to be
a shuttle at Cratylus 389b. Indeed, even the most famous self-predicative claim of
all, that at Parmenides 132a6-8, passes without a hint of trouble.
I think that the relatively commonplace manner in which Plato introduces language characteristic of the Theory of Forms, especially in the earlier dialogues,
should be taken at face value. This language simply would not have been problematic for any of Plato's contemporaries or for any of those characters portrayed in the
dialogues. This, of course, sets us a problem in exegesis. To most modern scholars,
at any rate, the Theory of Forms has seemed anything but uncontroversial, and
some elements o f it (including self-predication) have often been regarded as paradoxical or absurd. If we are to accept Plato's unconcerned introductions o f Formlanguage as proper representations of contemporary attitudes, then we must find

P L A T O ' S T H E O R Y OF F O R M S

143

some way to interpret that Form-language that makes it at least plausibly uncontroversial to c o m m o n sense. Below, I shall try to provide such an interpretation. Let us
note here a consequence of this approach. If Plato does accept some Form-language
as obvious to c o m m o n sense, then the purpose of Plato's Theory of Forms will be,
not to support such language as a novel way of speaking, but to explain it as a
datum. Plato's philosophical contribution in developing a Theory o f Forms thus lies
in his analysis of the necessary presuppositions he thinks this language makes.
Consequently, another object o f my investigations is to sketch the increasing sophistication with which Plato approached this datum and the changing conclusions
concerning the nature of the Forms (whose existence is never in doubt until the
Parmenides) and of their relationship with sensible particulars.
II.
A survey of the sorts of expression Plato uses to name Forms will be useful in
clarifying both these difficulties. We may most easily classify these by keeping in
mind that there is associated with each Form a predicate term that is true of that
Form and of its participants: let us call this the associated predicate of the Form.
Three sorts of names are c o m m o n :
(1) definite article + associated predicate in the singular, often with 'itself': 'the
equal itself', 'the pious', 'the bed itself';
(2) abstract singular term corresponding to the associated predicate, often with
'itself': 'equality itself', 'piety itself';
(3) ho esti ('that which is') or auto ho esti ('that itself which is') associated
predicate: auto ho esti ison, ho esti klin~. 3
Certain other n a m e s - - p e r h a p s one should say descriptions--do occur, most notably
locutions involving ' F o r m of' (eidos or idea with genitive): ' F o r m o f bed' (idea
klinOs), ' F o r m o f size' (eidos megethous). For the present, however, let us restrict
our attention to the three categories above. We may first observe that virtually every
Form has a name in class (1) or class (3)--indeed, P l a t o ' s usage suggests that either
of these procedures invariably can be used. In one place (Phd. 75dl-3), he specifies
the Forms as those things " t o which we set the seal auto ho esti both in asking our
questions and in answering our answers." Names o f class (2), however, are limited
to those Forms for which there is an appropriate abstract singular term. As we shall
see, this is not an insignificant distinction. We should also note that the use of the
intensive p r o n o u n auto with any of these sorts o f Form-names is frequently a mark
o f Form-language. 4
Form-names o f these three classes are not similarly distributed in all the dialogues.
Names o f class (3) occur only in the Phaedo and later works, s and in many cases the
same Form is named interchangeably by names from each of these classes, sometimes within the same sentence. There is, however, as noted above, an important
restriction on names o f class (2): they can be formed only when an appropriate
abstract singular term is available. In order to see the implications o f this, we must
make a distinction between the associated predicates of Forms. Virtually all of the
Forms mentioned in the earlier dialogues, and most of those mentioned in the later
3 I do not discuss the interpretation of these problematic phrases in this paper.
4 Interestingly, in Aristotle and in later Peripatetic and Neoplatonic writers, we find a profusion of
compounds (about eighty are listed in L & S s.v. auto-) formed from autos and apparently designating
Forms, including autoagathon, autoanthr6pos, autokerkis, automegethos, even autos6krat~s. Since a
number of such terms .appear in the Topics, this sort of terminology is probably Academic.
I assume here that the Cratylus is later than the Phaedo, a dating that is at least possible.

144

HISTORY OF PHILOSOPHY

ones, h a v e a d j e c t i v a l a s s o c i a t e d p r e d i c a t e s : ' j u s t ' , ' p i o u s ' , ' l a r g e ' , ' e q u a l ' , ' g o o d ' .
S o m e F o r m s , h o w e v e r , h a v e a s s o c i a t e d p r e d i c a t e t e r m s t h a t are n o u n s : ' s h u t t l e ' ,
' b e d ' , ' m a n ' , ' o x ' , ' l i v i n g c r e a t u r e ' . W i t h o n e p o s s i b l e e x c e p t i o n , n o n a m e s for these
F o r m s f r o m class (2) o c c u r , o b v i o u s l y b e c a u s e t h e r e a r e no a s s o c i a t e d a b s t r a c t
s i n g u l a r t e r m s f o r these n o u n - p r e d i c a t e s in G r e e k . T h a t in itself m i g h t seem an arbit r a r y m a t t e r , since we c a n d i s c o v e r the requisite t e r m s in English, o r at least invent
t h e m : ' h u m a n i t y ' , ' b e d n e s s ' , a n d so on. H o w e v e r , this d i s t i n c t i o n p a r a l l e l s a m o r e
f u n d a m e n t a l one. E a c h o f these n o u n p r e d i c a t e s is countable: w h a t it applies to, it
a p p l i e s to a c c o r d i n g to discrete units (in Q u i n e ' s p h r a s e , each " d i v i d e s its refere n c e " ) . 6 A c c o r d i n g l y , let us t a k e n o t e o f this d i s t i n c t i o n b e t w e e n c o u n t a b l e a n d
u n c o u n t a b l e p r e d i c a t e s a n d n o t e t h a t F o r m s a s s o c i a t e d with the f o r m e r occur
i n c r e a s i n g l y in later d i a l o g u e s a n d a r e a l m o s t never 7 n a m e d b y a b s t r a c t singular
t e r m s . T h e u n c o u n t a b l e p r e d i c a t e s include m o s t l y a d j e c t i v e s a s c r i b i n g m o r a l , aesthetic, o r r e l a t i o n a l qualities; b u t t h e r e are s o m e n o u n p r e d i c a t e s , n a m e l y , ' w a t e r ' ,
' e a r t h ' , ' a i r ' , a n d ' f i r e ' in the Timaeus a n d ( p o s s i b l y ) ' f i r e ' a n d ' s n o w ' in the
P h a e d o . T h e s e are all mass t e r m s : n o u n s t h a t d o n o t d i v i d e their reference. ~ T h e y
a r e u n c o u n t a b l e in the sense t h a t they reject p l u r a l i z a t i o n a n d n u m e r a l q u a n t i f i c a t i o n ( o n e c a n n o t s a y ' s e v e n w a t e r s ' in the s a m e sense t h a t o n e can say ' s e v e n m e n ' ) .
W e s h o u l d a l s o t a k e n o t e o f the w a y s in w h i c h P l a t o describes t h e r e l a t i o n s h i p o f
F o r m s to sensible p a r t i c u l a r s . In the early d i a l o g u e s , as Ross n o t e s , 9 the terms for
this r e l a t i o n s h i p all suggest ' i m m a n e n c e ' r a t h e r t h a n ' t r a n s c e n d e n c e ' : p a r t i c u l a r s
' p a r t a k e ' o f (metechein, metalambanein), ' p o s s e s s ' (kektbsthai, echein), o r 'receive'
(dechesthat) F o r m s ; F o r m s are said to ' b e in' (eneinai, einai en, engignesthaO or ' b e
p r e s e n t t o ' (paragignesthai, pareinat) p a r t i c u l a r s . H o w e v e r , as we m o v e to the later
d i a l o g u e s , we f i n d a n i n c r e a s i n g use o f w h a t Ross calls the v o c a b u l a r y o f ' t r a n s c e n d e n c e ' : F o r m s a r e ' p a r a d i g m s ' (paradeigmata), w h i c h p a r t i c u l a r s ' r e s e m b l e '
(eoikenaO, ' i m i t a t e ' (mimeisthat), o r 'strive f o r ' (oregesthat). A s we shall see, the
l a n g u a g e o f i m m a n e n c e occurs a l m o s t always with F o r m s w h o s e p r e d i c a t e s are
u n c o u n t a b l e , while F o r m s o f c o u n t a b l e p r e d i c a t e s u s u a l l y a r e d e s c r i b e d with the
language of transcendence.
T h e fact t h a t the F o r m s with w h i c h P l a t o deals in the earlier w o r k s a r e m o s t l y
F o r m s o f u n c o u n t a b l e p r e d i c a t e s suggests a p o s s i b l e a c c o u n t o f w h a t F o r m s might
b e t h a t e x p l a i n s t o s o m e extent t h e c o m m o n p l a c e c h a r a c t e r o f F o r m - l a n g u a g e . A s a
v a r i e t y o f p h i l o s o p h e r s h a v e n o t e d , ~~ t h e r e a r e similarities b e t w e e n the things to
w h i c h m a s s t e r m s p u p o r t to refer (let us call t h e m stuffs) a n d the a b s t r a c t objects
t h a t m i g h t be seen as the d e s i g n a t a o f a b s t r a c t t e r m s . I n d e e d , w h a t e v e r our ont o l o g i c a l s e n t i m e n t s , we o f t e n talk as if we s u p p o s e d c e r t a i n a b s t r a c t t e r m s to refer
I use 'countable' as equivalent to Chomsky's 'count + '. For Quine's expression 'divided reference',
see W. V. Quine, Word and Object (Cambridge, Mass., 1960). pp. 90-100.
7 An exception seems to occur at Parmenides 133d-e, where the same Form is named by autos despotOs
ho esti despot,s and aut~ despoteia, and also by autos doulos ho esti doulos and aut~ douleia. I discuss
this below.
8 1 use the expression 'mass term' in Quine's sense: I do not regard abstract singular terms as mass
terms. Jespersen introduced the expression 'mass word' in a broader sense, so as to include abstract
singular terms among mass terms (Otto Jespersen, The Philosophy of Grammar [London, 1924], pp.
198-201. Quine's use has become standard in the philosophical literature.
9 Sir David Ross, Plato's Theory ofldeas (Oxford, 1951), "Retrospect" (pp. 225-245).
~oTwo diverse examples: Max MOiler, "The Original Intention of Collective and Abstract Terms,"
Mind o.s. 1 (1876):345-351; Quine, Word and Object, pp. 120-121.

P L A T O ' S T H E O R Y OF F O R M S

145

to stuffs (even if abstract ones). We talk about love and hate spreading themselves
through groups of people just as we speak of water spreading over the ground.
There is also considerable similarity in their syntactical behavior. In subject position, mass terms and abstract singular terms both resist pluralization and articles. In
predicate position, however, they are distinct in function. Mass terms function also
as predicates, whereas abstract terms require ' h a s ' or an equivalent to function
predicatively: this object is gold, but it has beauty; this thing is water, but it has
clarity. It is significant here that certain terms seem to have an uncertain position
between mass term and abstract term (a good example is 'heat'). O f course, to be
able to differentiate between mass terms and abstract singular terms on the basis of
their purported referents requires the ability to m a k e ontological distinctions o f
some sophistication. Ordinary usage is content to treat the two classes of terms
similarly in m a n y circumstances, leaving it to philosophers to decide whether the
stuff-language is to be taken seriously in any given instance.
The relevance o f all this to the Theory of Forms is that we can construct an interpretation for that Theory, at least as it appears in the early dialogues, that preserves
the c o m m o n p l a c e character o f Form-language by understanding Forms to be
something like stuffs. We need only note the following points. First, the language of
immanence is all appropriate to describe the relationship of a stuff to an object in
which it is an ingredient. Gold, for instance, is in or present to golden objects, and
golden objects contain, possess, or have a share of gold in them. Second, it is at least
plausible to see each sort of stuff as having a certain character that it imparts to the
thing containing it. Gold imparts its particular c h a r a c t e r - - i t s color, its lustre, its
incorruptibility--to those things that contain it. We might note here that this relation admits o f degrees: the more gold something has in it, the m o r e it exhibits the
character o f gold. Third, it is perfectly natural to say that gold itself has that character which it imparts to golden things, and has it supremely. T h a t there is such a thing
as gold is even m o r e c o m m o n p l a c e . If, then, we can plausibly construe P l a t o ' s language about F o r m s as resembling language a b o u t stuffs in these respects, we shall
b o t h explain its uncontroversial character and have some better idea what motivates
Plato to suppose that Forms exist. 11
There is, in fact, a historical antecedent to P l a t o ' s Theory that provides us with
'stuffs' having m a n y of these characteristics. In numerous cosmological writers of
the fifth and fourth centuries B.C. we find use being made o f primary substances
whose names are expressions like 'the h o t ' , 'the cold', 'the d r y ' , and 'the wet'. Such
cosmological primaries are clearly mass objects, and each of them is identified as the
unique object that has a certain property or power (hotness, wetness, etc.). Any
other object is, for instance, hot because, and only because, it has some of the hot in
i t - - t h a t is, because it partakes o f the hot. O f course, the hot itself is hot; indeed, it
must be, for how else could it m a k e anything else hot? H o w else could the presence
o f gold m a k e things yellow, if gold were not itself yellow? This last claim is an
1~ Thomas G. Rosenmeyer, in "Plato and Mass Words," Transactions and Proceedings o f the
American Philological Association 88 (1957): 88-102, also construes Plato's Form-language in such a

way as to make the existence of Forms commonplace; and his account resembles mine in that he sees
Form-names as "mass words" in many cases. However, although I have profited from his arguments, his
perspectiveis quite different from mine. In particular, he uses the expression 'mass word' in such a way as
to include abstract singular terms.

146

H I S T O R Y OF P H I L O S O P H Y

instance o f what Henry Teloh '2 has called the "causal principle": " a cause must
have the quality which it produces in something else." Teloh finds the root of this
principle in Anaxagoras's view that " i n everything there is a portion o f everything,"
and indeed it does seem to be a natural assumption of many pre-Socratic (and later)
cosmologists. More important, Teloh argues that it is implicit in many places in
Plato's dialogues. On his view, predicative pre-eminence results from Plato's acceptance of the causal principle and from his view that Forms cause things to have
properties. If a Form causes a thing to have a property, then (by the causal principle)
the Form must have that property.
Teloh presents a great deal of circumstantial evidence that Plato does in fact accept the causal principle. This clarifies the somewhat perplexing argument in the
Protagoras concerning the unity o f the virtues (329c-333e, 349b-360e). Socrates
argues that wisdom, temperance, courage, justice and piety, the parts o f virtue, are
not parts "like the parts o f a f a c e . . , but, like the parts o f gold, they do not differ
at all from each other or from the whole except in greatness and smallness" (329d58). T o prove this, Socrates undertakes to show such things as that justice is pious or
that piety is just. In other words, he argues that one virtue has another. Teloh points
out that the argument seems to rest on something like the causal principle. Socrates
argues that anyone who has a certain virtue will necessarily have another: that is, he
argues that having one virtue makes one have another. But then, by the causal principle, the first virtue must have the second, in order to cause it in someone. Thus, if
being just makes one pious--in language used in the Protagoras, if partaking of justice makes one pious--then justice must itself be pious. On a mass-object interpretation of Forms, this thesis of the unity of the virtues amounts to the claim that the
terms 'wisdom', 'temperence', 'courage', 'justice', and 'piety' are all names of the
same thing, namely virtue; that this one object is itself wise, temperate, courageous,
just, and pious (and o f course virtuous); and that it makes whatever it is present to
virtuous in all these ways. It is curiously appropriate o f Socrates to compare the
parts o f virtue to the parts of an undisputable mass object, gold.
Obviously, there are serious obstacles in the way of identifying Forms with stuff,
even with the peculiar sorts of stuffs that we have been talking about. However, I
think that this is as much a strength as a weakness of my interpretation. My view is
that Plato was inclined to take abstract singular terms as referential because they
seemed in many ways to behave like referring expressions in ordinary language. He
did not argue that there were quality-stuffs: rather, he tended to accept some such
view without argument. What he did eventually come to argue for was the distinction between Forms and sensible objects of any sort, including quality-stuffs. In
other words, as Plato came to appreciate clearly what a quality-stuff might be, he
also came to adopt the view that Forms were not like that. We see this in evidence in
Plato's increasing insistence on the separation of Forms from sensible particulars.
Statements suggesting separation constitute part of c o m m o n sense. We distinguish
gold itself from golden objects, and we can even agree that gold itself is not subject
to the same sorts o f changes as ordinary physical objects: in a certain sense, gold is
immutable (it always has the same character). Again, it is possible to see gold itself
as a sort o f paradigm that all golden objects aim at and fall short of: none of them
has just the character of gold itself (of pure gold, we might say), although the more
,2 "Self-Predication or Anaxagorean Causation in Plato," Apeiron 9 (1975): 15-23.

P L A T O ' S T H E O R Y OF F O R M S

147

golden among them come closer. However, the parallel cannot be maintained for
long in the face of serious philosophical analysis. For one thing, the question is
bound to come up eventually whether a quality-stuff has its character in the same
way that its participants have that character. If so, can we distinguish the character
from the stuff? Will we then need to appeal to this character, separate both from the
quality-stuff and the things that partake of it, to explain why things have that character? If so, then we are dealing no longer with a simple mass object but with an abstract object, a n d we must ask serious questions about what such an object can be
like. ,3 Another sort of difficulty, perhaps intensified by these last problems, is the
failure of any simple concept o f participation based on a quality-stuff model to
stand up under analysis. I think that Plato consciously investigates difficulties of
both these sorts in dialogues after the P h a e d o , and I will try to sketch the thread o f
this development below. However, first I want to consider another inadequacy o f
the quality-stuff interpretation: it does not even account for all o f Plato's Formlanguage at the common-sense level.
III.
As we noted above, Forms whose associated predicates are count-nouns do
not have abstract singular terms as names. If we try to analyze such Forms as
quality-stuffs, therefore, we encounter one difficulty at once: there is no convenient
name for the mass-object. This in itself might not seem to be o f any consequence,
since we can easily suppose that the other name o f the Form, the name consisting o f
singular predicate term with definite article, could name a quality-stuff having as its
characteristic power just the property ascribed by that predicate. However, when we
try this out with particular count-noun predicates, it is hard to countenance the appropriate sorts o f quality-stuffs even with the most sympathetic of hearts. We must,
for instance, imagine that the bed itself--a Form whose Platonic credentials are
beyond suspicion--is some sort of reified quality (bedness? bedsteadity?) which
various particulars partake of, that it imparts its proper character to those participants, more so as they partake of it more; and that it somehow has this character
pre-eminently. Since the character in question is the property o f being a bed, we are
to imagine an object that is at once a b e d - - a n d a very fine o n e - - a n d also stuff-like.
This seems to be nonsense. In fact, whatever one takes Forms to be, pre-eminence
seems to be nonsense when count-noun predicates are in question. What could it
possibly mean to say that auto ho esti z 6 i o n is an animal or that a u t o ho esti klin~ is a
bed? But if this fails, so does any of the explanatory power that idioms concerning
participation have. The explanatory force of the Theory o f Forms, according to the
interpretation of the previous section, is dependent on the Causal Principle, which
requires that Forms be pre-eminent in a fairly literal sense. Consequently, in the case
of count-noun Forms, we seem to have either an absurd Theory (if we accept preeminence) or a useless one (if we reject it). Plato, as we have seen, presents us with
,3 Even the status of"simple mass objects" has provoked considerable discussion among recent philosophers, and a central question has been whether mass terms are to be treated as referential. Among the
extensive literature on the subject, we might note: Helen M. Carlwright, "Heraclitus and the Bath
Water," Philosophical Review 74 (1965):466-485; Terence Parsons, "An Analysis of Mass Terms and
Amount Terms," Foundations o f Language 6 (1970):363-388; Henry Laycock, "Some Questions of
Ontology," Philosophical Review 81 (1972):3-42; Julius Moravcsik, "Mass Terms in English," in Hintikka, Moravcsik, and Suppes, eds., Approaches to Natural Language (Dordrecht, 1973), pp. 263-285;
F. J. Pelletier, "On Some Proposals for the Semanticsof Mass Nouns," Journal o f Philosophical Logic 3
(1974) : 87-108.

148

H I S T O R Y OF P H I L O S O P H Y

what look like unquestionable assumptions of pre-eminence for these count-noun

Forms also. Must we then suppose that he has tried to preserve the usefulness of his
Theory while ignoring a patent absurdity'?. Moreover, he presents this with no indication that pre-eminence is a problem, either to understand or to accept. Is this to be
taken at face value, or is Plato so enamored of his Theory that he does not see fit to
defend it?
In order to make sense o f this situation, let us first note that we ordinarily accept
language very much like that with which Plato expresses pre-eminence, even in the
case of count-noun predicates, without ontological qualms. We are perfectly willing
to say that the cat has four feet, that it is a mammal, that it eats meat. We are even
willing to distinguish the cat itself from individual cats, saying, for example, that the
cat itself is carnivorous though some cats might refuse meat. Expressions like 'the
cat', 'the bed', 'the animal', ' m a n ' can all function as what linguists call generic
expressions; '4 roughly, as universally quantified terms in which explicit words indicating quantification ('all', 'every') do not occur. There are, in fact, quite a number
of expressions in English that have generic uses. We have just noted one class:
count-nouns in the singular with the definite article. We also use adjectives in the
singular with the article (only of inanimate objects: " T h e beautiful is precious"),
plural adjectives similarly (only with terms applying to persons: " T h e good die
y o u n g " ) , abstract singular terms with no article ("Justice is impartial"), and plural
nouns with no article ( " C a t s are carnivorous"). Perhaps we should also classify
mass nouns in subject position (without the article) as generic ( " G o l d does not
dissolve in most acids"). 1~
I will be concerned in what follows only with generic expressions that are grammatically singular, since only such expressions function as names of Forms. '+ Moreover, since our concern is Plato's language more than our own, let us take note of
some differences between the various singular expresions used generically (I will call
them generic singulars) in the two languages. The first difference is that in Greek,
adjectives applying to persons can be used as generic singulars: whereas we must say,
for example, " T h e just are pious," Greek syntax permits ho dikaios esti hosios.
Next, while English distinguishes between those generic singulars that require an
article and those that do not, in Greek all adjectives and nouns that can be used generically can be used with the definite article. Thus, the Greek equivalents to 'Fire is
hot', 'Justice is pious', ' M a n is mortal' are (at least superficially) grammatically
similar: to p u r esti thermon, h~ dikaiosun~ esti hosia, ho anthr6pos esti thnEtos. We
might also note that the use o f the indefinite article with predicate nominatives,
required in English, is unnecessary in Greek: we must say " M a n is an animal," but
Greek prefers ho a n t h r f p o s esti z6ion.
It is possible, of course, to regard generic singular expressions as simple alternative forms of universal quantification; that is, they can be seen as equivalent to
locutions containing explicit quantificationai words such as 'every' and 'all'. Under
'+ See Jespersen, Philosophy of Grammar, pp. 203-205, for a brief discussion of generic expressions.
So far as I know, the term is Jespersen's coinage.
,s In English, the two count-nouns 'man' and 'woman' are peculiar in that they are used genericallyin
the singular without an article ("Man is mortal"). Might this oddity have influenced some of what has
been written about the "third man" argument?
'+ I ignore the problematic auta ta isa at Phaedo 74cl; if this is indeed a name of auto to ison, its precise
explanation is a difficulty for any other interpretation no less than for mine.

P L A T O ' S T H E O R Y OF F O R M S

149

some circumstances that is no doubt the correct interpretation. However, in m a n y

cases we distinguish between the generic expression and its explicitly quantified alternative. 17 Suppose that it should happen that every cat alive at some m o m e n t
should have exactly three feet. Under such circumstances it would be true that every
cat has three feet. However, we might still resist the claim that, under those circumstances, the cat would have three feet. To make the distinction clearer, we might say
that, although it would be true in such a case that every cat has three feet, nevertheless, the cat itself would not have three feet. Whatever series of accidents caused
each of the various cats to be missing one o f its normal number of legs would not
cause the cat itself to become a three-legged animal.
This sort of language resembles what Plato says about Forms, including countnoun Forms, in several ways. First, it is quite natural to talk about the cat itself (or
similarly about the bed itself, or man himself, or the animal itself) and apply to it the
same sorts of predicates that one applies to individual cats (or beds or men or animals). Second, it is possible to " s e p a r a t e " the cat itself from individual cats: things
can be true of it that are not true of some, or of any, individual cats. Third, as in the
case of mass terms and abstract singular terms, these expressions have the appearance of referring expressions. This appearance of referentiality is enhanced in Greek
by the absence of those superficial differences of syntax distinguishing various
species of generic singulars from one another: the mass-term case, at least plausibly
referential, is similar to the count-noun case.18
That this appearance is not entirely absent even in cases when we know that it is
not to be taken seriously forms the basis of numerous jokes. It also perhaps provides
the motive by means of which we pass from talking about, for instance, the cat itself
to providing a detailed description o f it. Such a description, in detail, might resemble a description of a specific animal: " T h e cat has four feet, is covered with fur, has
a long tail, eyes with vertical pupils especially adapted to night vision," and so on.
Obviously, this sort of language enhances the apparent referentiality of 'the cat'
here. If we wish to avoid this appearance of referentiality, we might have recourse to
explicit quantificational idioms. However, we might also explain, " I d o n ' t mean any
particular c a t - - I ' m talking about the typical cat, or the perfect cat, or the standard
c a t . " These latter idioms, quite as natural as the rest o f this language, suggest referentiality even more strongly, and they suggest in addition a paradigmatic function,
to which we should turn next.
In talking about the cat, or the bed, or anything else involving this sort of generic
singular expression, we frequently run into difficulties that do not arise, or at least
that do not so obviously arise, in the case o f quality-stuffs. Every cat has a c o l o r - except, of course, for the cat itself; we would all agree that the cat itself is neither
black, nor white, nor tabby, nor any other particular shade or coloration. If we incline to treat the generic expression here as in some way referential and as referring
to some paradigm cat, then we must somehow make sense o f this oddity, which
arises with any count-noun Form: such Forms are somehow not determinate in the
ways that particulars are. The problems to which this gives rise are familiar, and
,7 For technical discussions of some of these problems, see John Bacon, "Do Generic Descriptions
Denote?" Mind n.s. 82 (1973) : 331-347; and "The Semantics of Generic The," Journal of Philosophical
Logic 2 (1973):323-339.
LsRosenmeyerargues for a point something like this, although along different lines. See "Plato and
Mass Words," pp. 93-94.

150

H I S T O R Y OF P H I L O S O P H Y

they are p r o b a b l y the principal reason most of us reject the idea that there are such
objects. However, the initial presupposition o f these objects does not arise in the
context o f an arcane theory but, as I have tried to illustrate, within ordinary discourse. Similarly, P l a t o ' s treatment o f F o r m s as paradigms can be seen as only an
extension of ordinary discourse. If asked to explain why the cat itself is of no particular color, we might answer that we have in mind a perfect cat, a model cat which all
other cats are copies or imitations of, in some sense. Paradigmatic language
becomes particularly appropriate when it is possible to imagine an individual member of a species actually serving as a paradigm or perfect exemplar, and in m a n y situations we have no difficulty doing just that (consider the " s t a n d a r d s " of dog-and
cat-show judges, which a m o u n t to descriptions that, though they might be satisfied
by no particular animal, nevertheless c o u l d be construed as characterizing a perfect
instance o f the species). I think some of P l a t o ' s language about Forms treated as
paradigms involves just such a view. In fact, the three most extended instances of
such l a n g u a g e - - i n the Cratylus, R e p u b l i c X, and the T i m a e u s - - q u i t e clearly treat
the F o r m as a perfect instance to be imitated by particulars. What makes Plato's
Theory o f Forms a significant intellectual creation is not his acceptance of this
language but the fact that he takes quite seriously its apparent referentiality. The
Theory o f Forms is then largely Plato's attempt to interpret that referentiality by explaining what sorts of things Forms can be.
IV.
I have argued that the apparently referential character of certain expressions
was important in motivating Plato to develop his Theory o f Forms. He takes generic
singulars as clearly denoting and then concerns himself with the analysis of their
denotata. An earlier f o r m of this analysis resembles in some respects the preSocratic cosmology of quality-stuffs, whereas later Plato increasingly recognizes the
inadequacy of such a model and replaces it by construing Forms as paradigms. In
either case, predicative pre-eminence is an essential element in the Theory: immanent Forms must be pre-eminent if they are to cause their characters in particulars
(by the Causal Principle), and paradigmatic Forms must be pre-eminent in order to
be paradigms. I have not argued that predicative pre-eminence itself, and in particular self-predication, meets with Plato's acceptance only because he fails to
realize that generic expressions need not be construed as proper names. Something
rather like this view, however, has been advanced in several recent papers by
Gregory Vlastos,'9 and in fact it was largely reflection on Vlastos's position that led
me to the conclusions o f this paper. Consequently, I should make clear the distinction between his view and my own.
Vlastos claims that Plato's self-predicative language frequently consists of what
he calls " P a u l i n e predications." He explains this term as follows:
A Platonic sentence predicating the adjective, A, of an abstract property, B, may be read as
assigning A either (I) to B in contradistinction to B's instances, or else (II) to B's instances; if
any, in contradistinction to B, and to them necessarily? ~
,9 Principally in "The Unity of the Virtues in the Protagoras," Review of Metaphysics 25 (1972):
412-458; reprinted, together with another previously unpublished work dealing with "Pauline predications" ("An Ambiguity in the Sophist") and several related appendices, in Vlastos's Platonic Studies
(Princeton, 1973), pp. 221-308. His most recent remarks on the subject are "A Note on 'Pauline Predications' in Plato," Phronesis 19 (1974) : 95-101.
20 Platonic Studies, p. 273.

P L A T O ' S T H E O R Y OF F O R M S

151

Reading (I) here is what Vlastos calls " o r d i n a r y predication," whereas reading (II) is
" P a u l i n e predication." It will be recognized that Pauline predication is something
like the generic use of abstract singular terms and other names for Forms, and that is
part of Vlastos's intent. He thinks that we can explain P l a t o ' s self-predications, in
m a n y troublesome instances, as Pauline predications. Now, I have already argued
that in m a n y cases generic expressions, especially generic singulars, do not seem
equivalent to their explicitly quantified alternatives; and Vlastos similarly does not
understand a Pauline predication as nothing other than a quantificational alternative. Instead, a Pauline predication (which apparently he means to limit to statements in Plato's dialogues a b o u t Forms) must contain a reference to a F o r m and
must include the modal operator 'necessarily'. Thus, 'Justice is pious', read as a
Pauline predication, would be: 'Justice is such that anyone who has this property is
necessarily pious', z~ Since this resembles my own position, and since m y position
was in fact developed in response to Vlastos's, let me explain the points in which
they differ. First, Vlastos thinks that this device saves Plato f r o m what he regards as
the absurdity of self-predication. I have tried to argue that self-predication, at least
as a part o f predicative pre-eminence, is essential to P l a t o ' s theory and hence something Plato cannot dispense with, however absurd it m a y appear to be. Beyond this,
actual self-predications in Plato that could be translated as Pauline predications are
rare. 22 Plato does not say " T h e bed itself is, of course, a b e d " ; instead, he simply includes it, without c o m m e n t , a m o n g a group o f other beds. Thus, the textual evidence that Plato accepted self-predications because he had a Pauline reading of
them in mind when he uttered them is missing, since we do not find the requisite selfpredications. Beyond that, it is by no means clear to me that " T h e bed itself is a
b e d " has anything like the same c o m m o n p l a c e character as " T h e bed itself is a piece
of f u r n i t u r e . " Such expressions, if indeed to be treated as Pauline predications,
seem to f o r m part o f the jargon of a specialized theory rather than ordinary language. In that case, however, Plato could come to accept them only after developing
the Theory; as a result, this interpretation fails to account for the ordinariness o f
Form-language in the dialogues.
There is a more serious objection to Vlastos's interpretation. He insists that Pauline predications must be read as containing an implicit reference to a F o r m : we are
not to omit the phrase "Justice is such t h a t . . . " in the example above. The reason
for this insistence is obvious: otherwise, these statements are only illusory F o r m language, since t h e strict quantificational reading eliminates even the apparent
reference to a Form. Yet if Vlastos retains the reference to the F o r m , it is without
any real theoretical point. At least, on m y interpretation, the F o r m s have a genuine
reason for predicative pre-eminence. T h a t is what allows them to do their job.
Vlastos, b y contrast, eliminates any real predicative pre-eminence, and his remarks
suggest that the Forms do not do much work in any event.
For all its transcendentalism, Plato's metaphysics can only fix the interrelations of a multitude
of celestial Forms by charting those of their earthly shadows. If there is irony in this fact--the
z, "Note on 'Pauline Predications'," p. 98.
z2The only exceptions I can find are in larotagoras 330c-d, where justice is called just and piety pious.
However, Pauline readings are not encouraged by such questions as "that thing [pragma] which you just
mentioned, justice, is that thing itself [auto touto] just or unjust?" (330c4-5).

152

HISTORY OF PHILOSOPHY

interrelations of these Forms begin to look like shadows of their shadows--the irony is
generated by Plato's metaphysics, not by the Pauline reading of a class of Plato's sentences. ~3
I have a l r e a d y r e m a r k e d t h a t the class o f sentences r e f e r r e d to here is n o t really in
evidence in P l a t o . A s for t h e " i r o n y , " I suspect t h a t it is as m u c h a c o n s e q u e n c e o f
V l a s t o s ' s style o f i n t e r p r e t a t i o n as a n y t h i n g else. H e earlier 24 p r o p o s e d an i n t e r p r e t a t i o n o f the c a u s a l role F o r m s a r e called o n to p l a y in the P h a e d o t h a t has a similar
effect. T h e r e , he u n d e r s t a n d s P l a t o ' s a p p e a l to p a r t i c i p a t i o n in F o r m s to a m o u n t to
a n a p p e a l to the s a t i s f a c t i o n o f s o m e d e f i n i t i o n : a c l a i m o f the t y p e ' F t h i n g s are F i n
virtue o f p a r t i c i p a t i n g in the F itself' s h o u l d be i n t e r p r e t e d as ' F things are F in
v i r t u e o f s a t i s f y i n g the l o g i c a l l y necessary a n d sufficient c o n d i t i o n s o f b e i n g F ' . A s
far as I c a n see, this e l i m i n a t e s t h e q u e s t i o n w h a t sort o f t h i n g a F o r m m i g h t be by
m a k i n g it u n i m p o r t a n t . V l a s t o s explains:
When I want to know what makes this figure a square rather than a pentagon, what answers
my question is not the existence as such of the Form, S q u a r e . . . but the logical content of its
definition . . . . And the fact that this logical function is performed by a celestial Form, rather
than by a nominalistic flatus vocis, in no way alters the strictly non-causal import of the
formula " F in virtue of satisfying the definition. ''2s
I n o t h e r w o r d s , all t h a t m a t t e r s a b o u t a F o r m is its d e f i n i t i o n . But here, too, an
i m p o r t a n t a s p e c t o f the T h e o r y o f F o r m s t u r n s o u t n o t to need t h e F o r m s at all. T h e
net result o f these t w o p r o p o s a l s o f Vlastos is t h a t the F o r m s b e c o m e mere
e p i p h e n o m e n a . I f I m a y b o r r o w a line used b y R. E. A l l e n in a r e l a t e d c o n t e x t , this
i n t e r p r e t a t i o n leaves the F o r m as " a g h o s t l y w r a i t h h o v e r i n g a b o u t the verges o f
existence, p o w e r l e s s even to g i b b e r . ''26 I f possible, we o u g h t to a c c o r d the F o r m s
m o r e t h a n an h o n o r a r y place. 27
V.
I w o u l d like to c o n c l u d e this p a p e r with a n e x a m i n a t i o n o f the d e v e l o p m e n t
o f the T h e o r y o f F o r m s t h r o u g h o u t P l a t o ' s c a r e e r in o r d e r to s h o w (as I w o u l d hope)
t h a t this d e v e l o p m e n t c o n f i r m s m y analysis; h o w e v e r , such a n u n d e r t a k i n g w o u l d be
far t o o l o n g f o r this c o n t e x t . I n s t e a d , I will n o t e t h r e e p a s s a g e s t h a t I t h i n k lend
s u p p o r t to m y view. First, at P h a e d o 103b, S o c r a t e s r e s p o n d s to a n o b j e c t i o n conc e r n i n g his c l a i m t h a t n o t h i n g - - t h a t is, n o F o r m m c a n b e c o m e its o p p o s i t e : H o w
d o e s S o c r a t e s s q u a r e this with the g e n e r a t i o n f r o m o p p o s i t e s p o s i t e d earlier (70e)?
S o c r a t e s replies:
You have not understood the difference between what was just said and was said then. Then,
we said that the opposite thing [pragma] came to be from its opposite; but now, that the
a3 "Note on 'Pauline Predications'," p. 101.
24,, Reasons and Causes in the Phaedo," Philosophical Review 78 (1969) : 291-325; reprinted in Platonic Studies, pp. 76-110.
~s Platonic Studies, p. 92.
26"Participation and Predication in Plato's Middle Dialogues," in Allen, Studies in Plato's Metaphysics, p. 54.
~7 Since writing this paper, I have become aware of two other articles concerning Vlastos's uses of
Pauline predication: Jerry S. Clegg, "Self-Predication and Linguistic Reference in Plato's Theory of the
Forms," Phronesis 18 (1973):26-73; and J. W. Forrester, "Some Perils of Paulinity," Phronesis
20 (1975): 11-21.

P L A T O ' S T H E O R Y OF F O R M S

153

opposite itself will never become opposite to itself, neither that which is in us nor that which is
in nature. (103bl-5).
It seems to me that Plato is expressly recognizing the distinction between a generic
use treated as a universal quantification and a syntactically similar statement a b o u t a
Form. " T h e large becomes small," for instance, can be read either as " L a r g e things
become small," which Socrates accepts, or as " T h e large itself (i.e., the Form)
becomes small," which he rejects. We should expect such a recognition as Plato
becomes m o r e familiar with the notion of a separate F o r m .
The next passage is in the Parmenides. The multitude of puzzles in this work
concerning the nature o f Forms and the nature of participation would m a k e interesting analysis, but I will pass them by except for one odd point. At 133d-e,
Parmenides is arguing that Forms can have relations only to Forms, not to
particulars, and he is illustrating this with the relationship between the slave itself
(himself?) and the master himself (itself?). These are named, appropriately, autos
doulos ho esti doulos and autos despotOs ho esti despotOs at 133d8-el. However, at
e3-4 we find 'mastership itself' (auto despoteia) and 'slavery itself' (auto douleia) in
their stead. These are contrasted, in fact, with mastership and slavery 'in us', and he
speaks of the ' p o w e r ' (dunamis) of the members of each pair. This is to my knowledge the only instance in which Forms of countable predicates are n a m e d by abstract
singular terms, perhaps because it is the only case in which Plato has available the
requisite abstract terms. It is striking that in this passage, in which Forms are clearly
being treated as separate (and after the introduction of a paradigmatic concept of
Forms at 132d), we find the language of immanence. I suggest that Plato is still not
clear on the notion of separation, even if he is clear a b o u t some of its difficulties.
Finally, I would like to call attention to the theory o f primary bodies in the
Timaeus as an indication of the level of sophistication Plato finally obtained concerning the relation o f Forms to sensible particulars. There, Plato deals with precisely those cosmological primaries that I suggested earlier might have been the
implicit model for the earliest stage o f the Theory of Forms: fire, air, water, earth.
However, he explicitly distinguishes the sensible quality (morphO, pathos, dunamis)
associated with these elements f r o m the things themselves, which are nonsensible
Forms. Since he recognizes--perhaps m u c h like some c o n t e m p o r a r y p h i l o s o p h e r s - that mass terms are m o r e appropriately treated as predicates than as referring
expressions, he is led to postulate a 'receptacle' of which these predicates can be
true. The F o r m is a transcendent paradigm that the qualities appearing in the receptacle imitate in some way. It appears that Plato has at last rejected all vestiges of an
i m m a n e n t conception of Forms. I would suggest that his reasons m a y be not
unrelated to those that have led some c o n t e m p o r a r y philosophers to discount the
apparent referentiality of mass terms and abstract nouns alike. 28

Kansas State University

28A number of people have provided me with criticisms and suggestions concerning earlier versions of
this paper, including Charles M. Young, G. N. Georgacarakos, and several anonymous referees.