Semantics and Self-Predication in Plato

Semantics and Self-Predication in PlatoAuthor(s): John MalcolmSource: Phronesis, Vol. 26, No. 3 (1981), pp. 286-294Published by: BRILLStable URL: http://www.jstor.org/stable/4182130 .

Accessed: 09/09/2013 20:47

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

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

.

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Phronesis.

http://www.jstor.org

This content downloaded from 128.111.121.42 on Mon, 9 Sep 2013 20:47:40 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=bap

http://www.jstor.org/stable/4182130?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


DISCUSSION NOTES

Semantics and Self-Predication in Plato JOHN MALCOLM

It is crucial to the notorious Third Man argument (Parmenides 132a 1 -b2, 132d 1- I 33a6), where an infinite regress is alleged to ensue from the adoption of Plato's Theory of Forms, that the Form and its instances be grouped together in the same class and this by virtue of the so-called Self-Predication of the Forms. In other words, a given Form, e.g., Beauty, or the Beautiful, is itself beautiful and this fact allows this Form to be included along with the beautiful particulars in the class of beautiful things. In general, not only is a certain predicate "F" true of a group of particulars, but the Form, F-ness, in virtue of which this is so, is F as well - assuring the truth of the formulation "F-ness is F."

Some scholars have convinced themselves that an examination of Plato's concepts of eponymy and homonymy is sufficient to disallow self-predication' and so free Plato from the questionable clutches of the Third Man. I propose to challenge this contention and shall maintain that Plato's use of eponymy and homonymy is inconclusive as regards self- predication and hence the evidence for or against the latter must be sought elsewhere. I hope, in passing, to cast new light on an interesting passage in Plato (Parmenides 133d) and, finally, to offer some considerations on the relation between predicate adjectives and Platonic Forms.

Let me begin with eponymy. Plato's account of the eponymous relation between particulars and Forms, whereby the particulars get their name from the Form, is given at Parmenides 130e- 13 la. The passage reads: "'There seem to you, as you say, to be certain Forms from which the other things that partake of them get their names ('ras E'nWvv1ics ... LoXeLv). For example, those partaking of Likeness (6Iot6T-q-Tos) become like (bou[ot), of Greatness great, of Beauty and Justice beautiful and just." (One may compare Phaedo 102b and 102c.) Eponymy, here, is the relation of a particular to a Form. Socrates, for example, gets the derivative designation "just" by partaking of the Form named "Jus- tice."

But these passages have been taken to give us much more. Bestor (PS, pp. 38-40; CPE, p. 190) assumes that, by virtue of the eponymous relation as it is presented by Plato, the same general word "F" has a primary referent, or Form, of which it is the proper name and also secondary referents, the many particulars, which are named after the Form.2 Since the application of the term "F" to the Form is not of the secondary or "named after" variety, we do not require another Form for a corresponding primary referent and hence the infinite regress proposed by the Third Man does not obtain. Allen, PP, p. 170, distinguishes "F" as a common name, when applied to particulars, from "F" as a proper name when applied to the Form. With respect to the use of "F" as the proper name of a Form, he maintains, p. 170, "When 'F' is used in primary designation it is a synonym of 'the F itself' and 'F-ness,' therefore to say that F-ness is F is to state an identity. It follows that it is invalid to infer self-predication from Plato's apparently self-predicative language." For both thinkers self-predication is blocked because "F" when applied to a Form is a proper name ( = "the F") and so "F-ness is F" is really equivalent to "F-ness is the F" or, indeed, "The F is the F" (cf. Bestor, PS, p. 58) and this gives us no license to

286



classify the Form along with the particulars as entities univocally possessing the common name,3 or description, "F."

There can be no doubt that what the particulars get from their eponymous relation to the Form is a common name - "F" - e.g., "just", "beautiful" or whatever. But, with respect to the corresponding name of the Form, the Bestor/Allen position goes beyond the evidence on two counts.

Firstly, it is certainly the case that the name of the Form, given in the paradigm formulation of eponymy, is a proper name, but this has no logical connection whatever to self-predication, for the proper name of the Form from which the particulars get their name is the abstract form of the name, i.e., "F-ness" (e.g., "Justice"), conceivably also the substantivized adjectival, "the F" (e.g., "the Just"),4 but not "F" (e.g., "'just"). We have in effect the equivalence of Aristotelian paronymy5 where what we get is "F-ness" naming the Form and "F" naming the particulars. There is no basis here for a statement such as "F-ness is F."

In the second place, even if Bestor and Allen were right and the eponymous relation gave us "F" as the proper name of the Form, this would only show that eponymy could not be used as grounds for self-predication, for there "F-ness is F" would be an identity statement. But the evidence for self-predication has not generally6 been held to derive from semantic considerations such as Plato's theory of predication, as illustrated by eponymy and homonymy, but from ontological concerns, such as Degrees of Reality and Model/Copy,7 and from contexts such as Phaedo lOOc and Symposium 210e-21 lb where the Form is perfectly F and the particulars imperfectly F. The fact that the Forms do not obtain "F" as a common name from the eponymous relation does not preclude that they may receive it from other sources.

So far I have dealt only with eponymy. I now turn to consider homonymy. Things are homonymous when they, literally, "have the same name." Bestor, PS, p. 46, CPE, pp. 199-200, does not distinguish homonymy and eponymy. Allen, PP, p. 169, takes eponymy as a more precise rendition of homonymy, but I do not think this is so. In the standard formulation of eponymy the Form is named "F-ness" and the particular "'F." Now this, I submit, is not a case of homonymy, for "Justice" and "just," to take a particular instance, cannot by any stretch of the imagination be looked upon as being the same name. Eponymy, therefore, is to be distinguished from homonymy in that Justice and Socrates are related eponymously, but not homonymously, when Justice is called "Justice" and Socrates is called "just". To continue with the same specific example, cases of homonymy, in contrast, are to be found (1) where each homonym is called "justice" or (2) when one is called "the Just" and the other(s) "just."8

The first sort of homonymy is found at Parmenides 133d which gives us both homonymy and eponymy. It reads: "Those Forms, then, which are what they are with respect to one another, have their being in relation to themselves and not in relation to the likenesses in us (Trap' ijLv) ... by participating in which we receive each of the names ('Cxa a E'rovotat6plleOa). These things in us, having the same name (4lvvFta) as those [i.e., the Forms], are relative to themselves and not to the Forms."

We can distinguish two relations here: (1) that of eponymy between individuals and the characteristics in our world. For example, Aesop is a slave and gets this designation from the slavery "among us"; (2) that of homonymy between the characteristics of terrestial slavery or masterhood and the Forms of the same. In this passage immanent characteristics are homonymous with Forms,9 e.g., slavery in us with Slavery as such. Particulars are eponymous with (from) these immanent characteristics, as is also the case

287



at Phaedo 103b. The use of eponymy here differs from that at Parmenides 130e-131a where particulars are clearly eponymous with respect to Forms. The use of homonymy at Parmenides 1 33d is very straightforward, each homonym being called by the same name, "F-ness"/"f-ness," but is not a case of Form/particular homonymy and hence does not bear directly on our prime concern.

Turning from homonymy between Forms and immanent chacteristics to Form/particular homonymy, we may distinguish two possibilities. The first is where both Form and particular share a common name. We would have, for example, "Socrates is just" and "Justice is just", with the latter an instance of the self-predication of the Forms. There is no context in Plato where homonymy must be taken in this way, though a passage such as Timaeus 52a would not preclude it if we had other evidence for self-predication. But, since self-predication is not presupposed by this paper, I shall leave this possibility. It obviously cannot be a type of homonymy which would support the Bestor/Allen thesis.

The second sort of Form/particular homonymy, and that to which one must appeal to undo the Third Man, is where the Form has a proper name and the particular a common name. This is found at Phaedo 78e10 where the homonymy is between auto to ison (the Equal itself), auto to kalon (the Beautiful itself) and the many beautiful and equal things. But, though we are closer to what Bestor and Allen need than we were with eponymy, when taken as equivalent to paronymy, we still do not, despite Allen's contention, PP, p. 169, have the name "F" applied homonymously. "F" is the common name of the particulars, but its counterpart, as a proper name of the Form, is "the F" and this vital difference makes the lack of connection with self-predication readily apparent.

To summarize: The claim to cripple the Tertius Homo by homonymy involves three assumptions:

(1) Homonymy is between the particular having the common name "F" and the Form when referred to by its proper name."1

(2) Since the particular is named "F," the Form, if it is to have the same name, must also be named "F."

(3) Since "F" is the proper name of the Form, when we say "F-ness is F" we are not predicating F-ness of itself because "F" is not a common predicate but a proper name. "F-ness is F" is really a type of identity statement. (Since F-ness is not being classified along with the particulars described as F, we do not get the Third Man regress.)

I shall grant the first assumption. I mentioned another possibility of Form-particular homonymy where each shared a common name (p. 287 above) but this very possibility is now being challenged.

The second assumption is very questionable. There seems no reason to take the proper name of the Form to be "F" rather than "the F." One can surely allow this amount of latitude to the interpretation of "having the same name". This objection applies even more strongly to those who equate homonymy with eponymy, where F-ness is the prevailing name of the Form.

The third assumption is clearly unacceptable. Even if we were to grant the second, which we need not, and admit that when we use the proper name of the Form we call it "F," it does not follow that whenever we call it "F" we must be using its proper name.12

One may try to rescue the second assumption by reminding us that Plato does in fact often use the adjective without the article to properly name the Form.13 But this will not, in fact, salvage Assumption Two since Plato does not have to do this and, indeed, does not do so in the homonymy passage given (Phaedo 78e). Furthermore, and here we proceed to the third assumption, this use of"F" in certain contexts to stand in for "'the F" is in no

288



way evidence that whenever "F" is so used it is the proper name of a Form. When, for example, Socrates remarks at Phaedo lOOc, "If anything else is beautiful besides the Beautiful itself" (or at Symposium 210e-21 lb contrasts that thing beautiful in its very nature with the particulars which are beautiful with certain qualifications and ugly with others) we must extract the use of "beautiful" from these contexts. Only if we get the best reading by interpreting it as a proper name are we entitled to do so. It is not the case that an analysis of homonymy can give guidance as to how we are to take these passages. For, unless we assume that any use of"F" when applied to the Form must be a proper name, and that is the very question at issue, we cannot appeal to this particular sort of homonymy (or, afortiori, to eponymy) to "break the back of Parmenides' Third Man."''4

Our first conclusion, then, is that neither the analysis of the Platonic concepts of homonymy and eponymy nor the fact that Plato sometimes uses "F" for "the F" as a proper name of a Form can determine whether or not he is committed to self-predication.

But does the latter tendency, that of treating "just," for example, as equivalent to "the Just," help make him vulnerable to a criticism often levelled against him - that, for Plato, predicate adjectives, when used as descriptions of individuals, properly name Forms? That is to say, in "Socrates isjust," "Socrates" names Socrates and "just" properly names Justice. 5

One may begin by noting that the Form does not have to be introduced to provide a nominatum for "just" in the sentence above, since, if Socrates is homonymous, or eponymous, with the Form then he gets the common name "F" in virtue of this relation.16 But one may grant that 'just" is a common name or description of Socrates and still maintain that it is also a proper name of Justice.17 All we can get, however, from the occurrence of the relevant type of Form/particular homonymy in Plato is the bare fact that the same word is used to express the proper name of a Form, when it is called the F(or F), as is used to describe an individual as F. We cannot therefore rely on this datum alone to conclude that predicate adjectives, when used as common names of individuals,'8 are also to be understood as proper names of Forms. Before attributing this muddle to Plato, we must, as in the case of self-predication, not content ourselves with discussing his terminology, but must consider certain key passages where he might be taken to confuse the common and the proper naming function of a word and so come to be regarded as having thought that, in order to be a common name of Socrates, "just" must, at the same time and in the same context, also be the proper name of Justice. I shall consider the two passages which are the most likely candidates for such a confusion. If we do not find it in these, we are not liable to find it elsewhere in Plato.

The first key passage is Parmenides 147e:

"You apply each of the names to something do you not? "I do...." "If you utter the same name once or several times, is it not necessary that you always mean the same thing by it?" "Certainly." "Isn't the name 'other' the name of something?" "Yes." "Therefore, whenever you utter it, either once or several times, you do not apply it to anything else or name anything else than that of which it is the name?" "Certainly not." "Whenever then we say that the Others are other than the One and the One is other than the Others, uttering the term 'other' twice, we do not, for all this, apply it to something else, but always to that nature of which it is the name?"

There is no doubt that the word "other" ('iTEpOV) in such a sentence as "The Others are other than the One" ('iTEpOV pV T'XXO TOV ?v6s) is used predicatively and, to avoid embarrassing consequences, should not be taken to properly name a Form or character-

289



istic. A first line of defense for Plato is to claim that his involvement in such an infelicity is only apparent. He is merely maintaining that the word "other" does not change its meaning when it is uttered several times in different contexts. In order to say this it is not required to hold that "other" have the logical function of properly naming Otherness. Despite the use of "onoma" (name) it may be urged that the passage need not be taken so strictly. But, if we transcend the spirit of its message to the letter, it is hard to avoid doing so. Nonetheless, a second and stronger refuge remains, for it is not the case that Plato is committed in propria persona to what is said in the second part of the Parmenides where he is often deducing absurd consequences which follow if identity-predication distinctions are not made (see e.g., 137c). If, as Cornford says (Plato and Parmenides, [London, 19391, p. 165), with reference to Parmenides 147e, "This is perhaps the clearest statement in Plato that every word must have a definite meaning, which is a constant character or 'nature' (qw$Ls), for which the word is the 'name.' ", we are well on the way to clearing Plato of the charge of such a confusion as an essential part of his doctrine.

But, unfortunately, evidence of some genuine unclarity in this connection may be adduced from our second passage, Phaedo 103e- 104a, where one cannot suggest that Plato is presenting statements and inferences which he need not himself accept. He asserts that not only does the Form have the same name always, but also something else, which always has the morphe of the Form, merits this name as well. For example, the Odd (Oddness) always has its name (i.e., "Oddness" or "the Odd"). But there is, in addition, something else which not only has its own name, but also, in that it is inseparable from the Odd, has the same name as the Odd. Speaking of the Three (n Tptds) Plato asks, "Doesn't it seem to you that it is always to be called by its own name and by that of (the) odd? (a'pa OU bOXEL OOL TXr TE ai'ris 6vo6FaT &EL 'rpoGayopEVTEa ElVaL Xvi T4 TOD rEpLTTOV ...;) In speaking of two different things, Oddness and Three, having "the same name" Plato is conflating a proper and a common name. But is he here going beyond the type of relation described above under the sort of Form-particular homonymy'9 where the names in- volved are "F" used as a true predicate and "the F" used as a proper name? Plato is at pains to point out that three, albeit called odd, is not identical with the Odd.20 I think we may conclude that he does not regard "odd" in "Three is odd" as functioning as the proper name of the Odd and, hence, leading to an identity between the Three and the Odd. To attribute such a position to him is to make a much stronger claim than merely to hold that he did not in fact distinguish between the (proper) naming and the describing (or common naming) function of a word. The latter case would lead to unfortunate formulations, but would not be the basis for a faulty positive view about naming. At Sophist 256c Plato examines the statement "Motion is other than the other" and notes that Motion is not other, in a sense, and is other (0vx 'ETEpOV ̀p'?'rTL v xai ?TEpov) where the first use of "other" is equivalent to "Otherness" or "the Other", in that Motion is not identical with Otherness ('KTpov), but is other (than Otherness). Plato is clearly not led by the differing uses of the same term, "other," into an identity/predication confusion. The second use of "other" (italicized above) is. therefore, obviously not to be interpreted as a proper name of Otherness.

Plato, of course, does not start explicitly to address himself to the distinction between identity and predication, or proper naming and describing, until the Parmenides (second part) and the Sophist, but we need not suppose that he regarded his younger self as a late-leamer. The crux of the matter is that, in the middle dialogues at least, he did not have a clear focus on the relation between "just" and "the Just" and did not go beyond the bare surface facts of eponymy and homonymy. As we saw, homonymy was adequate

290



to show that "F" is a common name of Socrates in "Socrates is F", but it does not show that "F," in that context, is a proper name of F-ness.

It would appear that an adequate account of the role of the predicate adjective must allow for two functions. "Just" in "Socrates is just" must both describe (commonly name) Socrates and connote, signify, designate, express or introduce Justice. There is every reason to believe that Plato was not clear about the connotative relation of the predicate adjective to the Form but it does not follow that he clearly misperceived this as a denotative relation.21 In sum, it is possible that Plato could have taken the predicate adjective to properly name the Form. There is, I submit, no compelling reason to think he did so and we are therefore well advised to content ourselves with a safer account of the semantic function of the Forms such as that of Ryle, ("Plato's Parmenides," Mind 48 (1939), p. 134), who observes, "To put it roughly, a Form is taken to be something answering to any general predicate, noun, verb or adjective, in such a way that any significant abstract noun will be the proper name of such a something. And it is because there exist such somethings that many ordinary objects can be characterized by a common predicate."

University of California, Davis

NOTES

1 R. E. Allen, "Participation and Predication in Plato's Middle Dialogues," (hereafter PP), in Plato I, ed. Gregory Vlastos (New York, 1970), p. 183.

Thomas W. Bestor, "Common Properties and Eponymy in Plato," Phil. Quarterly, 28 (1978), (hereafter CPE), p. 190.

Thomas W. Bestor, "Plato's Semantics and Plato's 'Parmenides,' " Phronesis, 25 (1980), (hereafter PS), p. 38. 2 Cf. Allen, PP, p. 169, who claims that the given passages "imply that 'F' is a name, a name whose prime designate is a Form: 'F' names the F. But this name is also applied, through what we may call derivative designation, to particulars which are namedafter the Form." 3 It may seem incongruous to speak of a term such as "just" in "Socrates isjust" as a name of Socrates. But it seems preferable to treat Plato's use of onoma in eponymy and homonymy literally as name. In so doing one has only to follow, as does Allen, PP, p. 170, the traditional distinction between the proper and the common name which goes back at least as far as the Stoics (see Diogenes Laertius, Lives of the Philosophers, VII, 58). Adjectives were included under the classification of common name (noun) which came to be divided into (general) substantives and adjectives (see R. H. Robbins, A Short History oJ Linguistics [Indiana, 1967], pp. 33-34, 57). "Noun," of course, is now restricted to substantives. For further remarks on this topic see G. Vlastos, "The Unity of the Virtues in the Protagoras" (hereafter UVP) in Vlastos, Platonic Studies (Princeton, 1973), p. 238. 4 This possibility follows from Phaedo 103c ff. where we have the Hot, the Cold, the Odd and the Even, for one might well, on the basis of Phaedo 103b, extend the range of eponymy to include these. I will not enter into the question as to whether they are Forms proper or immanent characteristics. In either case we would get eponymy between something called "the F" and something(s) called "F." s "Things are called paronymous when they get their name from something whose name differs in grammatical ending; for example, the grammarian from grammar and the courageous from courage" (Categories 1 a 12-15). This similarity between eponymy and

291



paronymy has been noted by R. E. Allen in Plato's 'Euthyphro'and the Earlier Theory of Forms (London, 1970), (hereafter PE), p. 126. Allen cautions that for eponymy one need not have a difference in grammatical ending (see footnote 4 above). I have preferred to deal with such cases, where we have, for example, "the Just" and "just," as standard instances of homonymy (see below) and not as deviant cases of eponymy/paronymy. 6 Bestor, CPE, pp. 191ff., offers a very helpful discussion of different types of eponymy and, rightly I believe, calls in question the suggestion that self-predication builds on eponymy. I do not, however, see evidence that this suggestion was in any way a "majority opinion" among those who attribute self-predication to Plato. 7 That these ontological aspects of his philosophy did commit Plato to self-predication is affirmed by G. Vlastos, "The Third Man Argument in the 'Parmenides' "in R. E. Allen, Studies in Plato's Metaphysics (London, 1965), p. 248, and denied by Allen, PP, pp. 172- 176. 8 The second case of homonymy may perhaps be considered to be eponymy as well. See footnote 4 above. 9 This point has not been regularly observed. G. Vlastos, "Reasons and Causes in the Phaedo, " Philosophical Review, 78 (1969), p. 300, note 29, and G. E. L. Owen, "A Proof in the Peri Ideon," Journal of Hellenic Studies, 87 (1957), pt. 1, p. 106, note 19, take Parmenides 133d as giving us homonymy between particulars and Forms. 10 Allen, PP, p. 169, note 4, lists Sophist 240a and Republic 596a as well, but these do not explicitly say that the Forms and their particular instances have the same name though they, or their context, imply it. 11 It is, of course, possible to challenge the Third Man regress by treating the names of both Form and particular as common, but taking the common name "F" when applied to the Form to be of a sufficiently different sense than "F" when applied to the particulars to preclude the Form and the particulars from being brought together into a group of univocally F things.

Although this approach does fit the first type of Form/particular homonymy which I distinguished just above and then dismissed, I believe it raises issues well beyond the scope of this paper which merely questions the claim that a straight-forward analysis of Plato's concepts ot eponymy and homonymy is sufficient to embarrass the Third Man. If I am correct, both the eponymous relation and the second type of Form/particular homonymy (Phaedo 78e) require that the Form be properly named. One would have to urge on other grounds the existence, for Plato, of a case of (non-eponymous) homonymy consistent with self-predication and then try to show that "F" when a common name of F-ness is different in sense from "F" when applied to particulars (for example. through arguing by analogy from the claim that a man and a picture of a man are not men in the same sense).

Since this alternative does involve homonymy and semantic considerations, in order to stress its distance from the thesis examined in this paper one might, following a lead offered by the Editor of Phronesis, be tempted to re-title the paper Syntax and Self- Predication, on the grounds that the Bestor/Allen doctrine is that "F" has a different syntactical function (proper vs. common name) when applied to the Form. 12 See in this connection the remarks of F. C. White, "Plato's Middle Dialogues and the Independence of Particulars," Phil. Quart., 27, 1977, p. 208. 13 See, for example, Euthyphro lOd, Phaedo 76d, 77a, Republic 476a, 479a, Parmenides 130b, 135c, Sophist 250a, Philebus 60a. Let me illustrate with Euthyphro 1Od. "Therefore that which is beloved of the gods is not pietv ... nor piety that which is beloved of the gods

292



but this is other than that (ovtx apa to OEoWlXis OOOV ?OTLV ... OVSE TO O EoV OOtXis ...

&XX'iepov ToUTro TOUToV)." Since the point of the passage is to assert that piety is not the same thing as "beloved of the gods," it will not do to translate hosion as "pious." When the insight is recalled at 1 5c the articles are included: iiL"vrOaXL 'yap 'Tov O'TL 'Ev C 'rpoa0Cv T7

TE OuLOV Xotl rO OoELXis oUv r'rVTOV 'isv E'q&vTl. At Parmenides 1 29ab the Dissimilar is referred to as oE EOTLV &VO61OLOV, &voIloL6T1s and t&

&v6~ooLa. 14 As does Bestor, CPE, pp. 190, 203; PS, pp. 38, 54-59. He follows Allen, PP, p. 183. 15 We find this position clearly stated by K. W. Mills, "Some Aspects of Plato's Theory of Forms: Timaeus 49cff," Phronesis, 13 (1968), pp. 145-6, who writes, "Predicate-names pick out the sorts themselves - the common types or characteristics which things may have or of which they may be instances. The predicate just' is a name -a proper name, no less - of justice, the characteristic of being just, in very much the way in which 'Plato' is the proper name of Plato."

Evidence that this type of interpretation is still flourishing may be found in more recent work - Bestor, PS, p. 39 and Plato on Knowledge and Reality (Indianapolis, 1976), by Nicholas P. White. Mr. White writes, p. 9, "... it is no denigration of Plato to note, as many have, that his writings exhibit a tendency to confuse meaning and reference and in particular to assimilate the claim that a term is meaningful or significant with the claim that there is an entity in the universe that it in some sense stands for or picks out.... It is shown particularly by the ease with which some can assume, without argument, that in order to be meaningful and to function as an intelligible part of discourse, a general term such as 'wise' must, in contexts such as 'Socrates is wise,' pick out or refer to some piece of the contents of the world (such as, say, wisdom) in the way in which the name 'Socrates' refers to Socrates. " (My italics.)

White, (p. 74, note 54, and p. 141), even goes so far as to suppose a repeated application of the naming/describing confusion by Plato. First, Plato takes predicate adjectives to be proper names of Forms by mistaking describing as naming. Then he takes the proper name of the Form to be a description of this Form and thus gets the notorious self- predication by mistaking naming as describing. I have been urging, however, that the attrbution of self-predication to Plato stands or falls on considerations other than such semantic issues as Plato's treatment of the various ways of "naming." If 1 am correct, the best that can be said of White's elegant, but hyper-speculative, hypothesis is that Plato did not make the distinctions needed to preclude it. 16 On this point see the excellent remarks by Allen, PE, p. 126. 17 This is how I read Mills, pp. 145-6 and Bestor, PS, pp. 39-43. Mills, p. 148, sees that it is "the beautiful" which is a synonym for "beauty," but then concludes (pp. 148-9) that Plato "certainly did take the view that an adjective and its correlative abstract noun were to be treated as alternative proper names for the same one Form." It is vital, however, to hold that it is the adjective plus the definite article which is the more standard way to give the proper name of the Form. We may agree with Mills (p. 149) that when Plato asks with respect to the Form of the Beautiful, "ti esti kalon;" "kalon" functions as a proper name, but the question is equivalent to "ti esti to kalon;" (What is the Beautiful?). Such usage is no evidence at all that predicates are proper names of Forms or that "just," in "Socrates is just," names Justice. Mills, p. 149, attempts to support the thesis that "F," in such a context, is a proper name of "F-ness" by suggesting that the term "Plato" may be applied both to Plato himself and to statues of Plato. He is the Plato, they are Platos, i.e., each is a Plato. This analogy is quite in keeping with Plato's ontology, but I do not believe it is

293



decisive in the crucial respect. In the first place, it presupposes that "just" or "beautiful," as predicate adjectives, are to be taken as proper names of Forms and that is the very question at issue. In the second place, the attempt to assimilate the predicate adjective to the proper name may well end up assimilating the proper name to the predicate adjective. Mills remarks (p. 161), "Did I not suggest that, as the name 'Plato' is legitimately applicable not only to Plato himself but to statues of Plato, so too, according to Plato's theory, the name 'beautiful' is legitimately applicable not only to the Beautiful itself but to particular beautifuls?" (Note the "Plato"/Plato himself - "beautiful"/the Beautiful parallel where the "the" is ignored in the analysis.) But "beautiful" when said of particular beautifuls is not a proper name. Hence, if the parallel holds, "Plato," when applied to the many Platos, cannot be a proper name. So, if any overlap of function is to be seen between the two uses of the same word, it would appear that "Plato," when applicable to Plato, is not to be considered a proper name. But, in fact. no such overlap need be inferred from the example. When we say of a statue that it is a Plato we are not, contrary to appearances, properly naming Plato, but using his name as part of a "sortal expression" as in "That is a statue of Plato."

Bestor, PS, pp. 40-1, in the same context, uses "Ford," as in "Henry Ford," a proper name from which all Ford vehicles derive their label as Fords. This certainly does give homonymy, though not a precise parallel with the Platonic ontology, for, contrary to Mills' Plato/Plato-statue example, the many Fords, fortunately, are not copies or instances of Henry. Furthermore, when we say of such an entity "This is a Ford," i.e., "This is a vehicle made by the Ford Motor Company," we are not even giving the appearance of naming Henry, though we are, of course, using his name. 18 Note that the uses of"F" as equivalent to "the F" given in footnote 13 are not so used. 19 In looking for cases of possible common name/proper name confusion I have not limited myself to those involving particulars and Forms. Here at Phaedo 103e-104a, for example, Three is not to be read as a particular and there is some dispute as to whether the Odd is to be taken as a Form. (See D. Gallop's edition of the Phaedo [Oxford, 19751), p. 187). The logical point at issue, however, is that raised in connection with the relevant type of Form-particular homonymy. 20 See Vlastos, UlVP, p. 240. 21 The Cratvlus is sometimes cited in support of such a view (Mills, p. 145, A. Nehamas, "Predication and Forms of Opposites in the Phaedo," Rev. of Meta., 26, 1973, p. 481) but, ironically, it goes further than any Platonic dialogue to militate against it. Predicate adjectives in the Cratvlus are certainly presented as names (e.g., 41 2c, 41 6a-4 1 7c) but they name only the subject. The relation of the predicate adjective to the attributed characteristic is something other than naming. Names are applied to indicate (422d-&kxoiv, 428e-Fv&Ai.,rTat) the nature of the things named or to signify it (4l9a-aiLaivov). They do not have to be taken as naming these natures or, afortiori, as properly naming them. See also Charles Kahn, "Language and Ontology in the Cratvlus" in Exegesis and A rgumenz, ed. Lee, Mourelatos and Rorty (Assen, 1973), p. 173, who suggests that the Form is best understood as the sense (and not the referent) of a predicate adjective.

294



Documents

Semantics and Self-Predication in Plato