Socrates Young and Old: From the Parmenides to the Phaedo. The Parmenides is the earliest of Plato s Socratic dialogues (I mean the dialogues in which

Transcription

1 Socrates Young and Old: 1 From the Parmenides to the Phaedo Introduction The Parmenides is the earliest of Plato s Socratic dialogues (I mean the dialogues in which Socrates appears). Only in the Parmenides is Socrates a young man. Everywhere else he seems to be at least middle-aged. In the Phaedo, by contrast, he is an old man of 70 and at the end of it he dies. The Phaedo is thus the last of Plato s Socratic dialogues. The other Socratic dialogues come somewhere in between. Such matters are beyond dispute. What is not beyond dispute is the significance to be attached to them. Should we, for our reading and understanding of the dialogues, follow the dating of the dialogues according to when Socrates engaged in them (the dramatic chronology)? Or should we instead follow their dating according to when Plato wrote them (the historical chronology)? The scholarly world has, for the most part, decided to adopt the latter course. I wish to explore adopting the former.1 My first reason for doing this is simply one of curiosity. I would like to try out the experiment of interpreting the dialogues, or rather some of them, according to dramatic chronology to see where that will take us. We have had plenty of interpreting of the dialogues according to historical chronology and we know pretty well where that takes us and what sort of Plato (and Socrates) it produces. It would seem worth trying the other chronology too to see where that takes us. Or at least it would seem worth doing this as regards those dialogues which have a 1 Several scholars have recently suggested readings of Plato s dialogues other than that according to historical chronology. I would instance in particular Julia Annas Platonic Ethics, Old and New, Ithaca and London, Cornell University Press (1999), who is in this respect expressly following some of the ancient commentators, and John M. Cooper in his introduction to Plato. The Complete Works, Indianapolis and Cambridge, Hackett (1997): viii-xviii. Such various ways of reading Plato have, in fact, been in favor for some time among certain groups of scholars. For a helpful summary, with extensive references, see Gerald A. Press s introduction to his edited volume Who Speaks for Plato? Studies in Platonic Anonymity, Lanham, Maryland, Rowman and Littlefield (2000): 1-6, as well as his earlier The State of the Question in the Study of Plato, The Southern Journal of Philosophy 34 (1996):

2 clearly marked dramatic chronology (for not all of them do), and a chronology that is made key to the setting of the dialogue (as is true, in particular, of the Parmenides and the Phaedo). A second reason is that the dramatic chronology would seem, at least initially, to be the better one to follow if our aim is to understand what Plato himself meant to convey. For he went to some lengths to tell us that chronology2 but none to tell us the historical chronology, so at least in his mind the former was important for understanding what he wrote but the latter was not. Of course it could be that Plato was mistaken, or devious, and the historical chronology is important in ways that he ignored or hid. But it can hardly be reasonable to start with that assumption. A sane hermeneutics should counsel us to read a work according to the intention of its author before we try to read it according to some other intention. A third reason is that the dramatic chronology is the only chronology that we know anything definite about. The historical chronology, for all its sophistication, cannot profess to be more than speculation. Although Plato could not and did not write the dialogues all at one sitting, he could and did go on revising them throughout life.3 Hence even if, say, he wrote the Republic and the Phaedo before the Parmenides it does not follow that the Parmenides does or must represent a later stage in his thinking. Not only can an author choose to write a logically earlier part of a series of works after he has written a logically later part but, in addition, if all Plato s dialogues were continually being revised in the light of each other, they would, in the form we now have them, all represent the final stage of his thinking and not any stages in between. So the effect would be as if Plato had written them all at one sitting even though he did not. Stylistic analysis will not help much here. For even if we can differentiate dialogues according to overall differences 2 2 At least in many cases, and certainly in the case of the dialogues I am particularly concerned with here. 3 See Jacob Howland, Re-Reading Plato: the Problem of Platonic Chronology, Phoenix 45 (1991): , and especially

3 of style, and can associate these styles with different periods in Plato s life (neither of which 3 points is easy to establish),4 it still does not follow that Plato could not have used an early style when revising an early dialogue or a late style when revising a late dialogue. Or does anyone claim to know so much about Plato the man and about how he aged as to be able to say that imitating the writing of the young Plato was beyond the skill of the old Plato? For such reasons, therefore, I want to try out interpreting Plato s dialogues, or several of them, according to the dramatic chronology. As a result I will not look for the sort of development in Plato that scholars propose who separate the dialogues into the historically early, middle, and late. Instead I will look for the development, or at least the differences, in the portrayal of the characters, or rather in the portrayal of one character, that of Socrates. Hence my interest in the Parmenides and Phaedo, since these portray Socrates at the limits of youth and age in his philosophical life. Accordingly, while most scholars interpret the Parmenides as marking a turning point in Plato s thinking from the classical theory of forms in the Republic and Phaedo to the later dialectic of the Sophist and Statesman, I propose rather to explore seeing it as marking a turning point in Socrates thinking from an immature approach to the classical theory of forms in the Parmenides itself to a mature approach in the Republic and Phaedo. I shall, therefore, first concentrate on the theory of forms as Socrates presents it in the Parmenides and on Parmenides criticisms of this presentation. My aim will be to show both how Socrates presentation can be seen as immature and also how Parmenides criticisms, while effective against Socrates immaturity, have no force against the theory itself as Socrates presents it later in the Republic and Phaedo.5 4 See Howland, Re-Reading Plato, passim. 5 My claims here are, interestingly enough, supported by an article that argued the opposite, How Parmenides Saved the Forms, by Samuel Rickless in Philosophical Review 107 (1998): Rickless contends that the several

4 4 The Young Socrates in the Parmenides Consider, then, Socrates speech at the beginning of the Parmenides where he criticizes Zeno s attack on plurality (Parmenides 128e5-130a2). A point of particular interest about this speech is that Socrates is not content to question the validity of Zeno s arguments. He insists on going further and proposes a particular theory of his own that, as he thinks, gives an account of plurality which escapes Zeno s attack. This approach of Socrates is not only unnecessary (one can question another s criticisms of a certain position without putting forward rival theories of one s own), it is also what gets him into trouble. The way Socrates behaves here is, indeed, not dissimilar from the way Thrasymachus behaves when he criticizes Socrates in the first book of the Republic, or, to keep to the present dialogue, from the way Zeno himself behaved, according to Zeno s own account, when he wrote his attack on plurality namely out of a young man s love of victory (Parmenides 128e1-4).6 The young Socrates seems to be caught by the same love. For, as just noted, in his criticism of Zeno he does two separable things. The first is simply to point out that Zeno s arguments against plurality, as so far stated, do not establish any problem for believers in plurality. For there is nothing paradoxical or contradictory about something s having contrary predicates if it has these predicates in different respects. Socrates himself, for example, is both one theses that make up the theory of forms as presented in the Parmenides are also found in the theory as presented in the Republic and Phaedo. He also contends that the theory can be saved if one of these theses is removed and that the main purpose of the Parmenides is to show us this solution. Thus, he says, the Parmenides is intermediate, in the way many scholars have maintained, between the Middle Period dialogues of the Republic and Phaedo and the Late Period dialogues of the Sophist and Statesman. But the thesis that Rickless identifies as needing to be removed from the theory is not found in the Phaedo or Republic but only in the Parmenides. His argument therefore suggests, contrary to his intention, that the Phaedo and the Republic were already avoiding the fault that the Parmenides was meant to expose, and therefore that the former dialogues are logically posterior and not prior to the latter. The thesis in question, that no form can have contrary properties, is presented by Socrates at the beginning of the Parmenides and is what, I contend, shows his youth and immaturity. 6 This likeness between the young Zeno and the young Socrates is expressly noted by Kenneth Sayre in Parmenides Lesson: Translation and Explication of Plato s Parmenides, Notre Dame, Indiana, University of Notre Dame Press

5 and many: many in having many bodily parts and one in being one man. To show there is a 5 problem here Zeno has to do more than point out the junction of manyness and oneness in the same thing; he must go further and show that the same thing is one and many in the same respect. This challenge to Zeno is fair enough and is really just a request for clarification. If Zeno responded well to it the conversation could continue to the next stage; if he did not the conversation would end there. But we do not know how Zeno would respond. For Socrates is not content with making his challenge and waiting for a reply. He has, as it seems, the haste of a young man in pursuit of victory and insists straightaway on doing the second thing of putting forward his own theory of forms. Had he been less impatient, or less anxious for victory, he could have avoided the counterattack that follows and that gets Zeno off the hook of explaining how Zeno s own attack on plurality really works. Moreover Socrates puts forward his theory, not hypothetically as in the Phaedo (99d4-100a8), but dogmatically, and not through an appeal to the medium of human speech but directly as his own. Indeed, as others have noted, 7 Socrates in the Parmenides seems not unlike a pre-socratic philosopher proposing his own theory of beings, or of the cosmos, to rival that of others, and specifically to rival that of Zeno and Parmenides. If so, then this would fit in rather nicely with Socrates own portrait of himself as a young man in the biographical excursus of the Phaedo (96a6-98b6). What I am suggesting here can perhaps be confirmed by the consideration that the theory of forms as Socrates presents it in the Parmenides shows some signs of youthful haste. For instance, Socrates proposes that the forms of One, Many, Like, Unlike, Motion, Rest do not (1996): 64, For example, Robert S. Brumbaugh, Plato on the One: The Hypotheses in the Parmenides, New Haven, Yale University Press (1961): 34.

6 possess contrary features.8 The One is not many, the Like is not unlike, and so on. Only the 6 particular sensibles possess such contrary features as they variously share in the respective forms. But such a thesis cannot be sustained. The Many is one because it is one form; the Like is unlike because it is unlike the Unlike; the Unlike is like because it is like itself and, indeed, like every form (including the Like) in being a form; Motion is at rest because it remains the same form and does not change into any other form.9 So much can be discerned already before Parmenides begins to criticize Socrates and, in fact, Parmenides criticisms seem less serious for Socrates theory. Admittedly problems of the sort I have just mentioned do come up in the second part of the Parmenides, but why do they not come up straightaway? Perhaps and to keep to the hypothesis I am exploring in this paper because Parmenides and Zeno want to spare Socrates youth. It would discourage Socrates, as it would any enthusiastic young man, to make an immediate and obvious fool of him. Better to start with less obvious problems that while driving such a young man to perplexity will not dismay him but provoke him to further and deeper thought. The other problems can then be allowed to emerge afterwards in the safer context of the dialectical exercise where the young man and his theory are not directly under attack. One can note here, though, by way of anticipation, that in later dialogues Socrates does not say, as he does in the Parmenides, that no form can possess contrary features. Or, to be more precise, he is more careful about what must be meant if one does say that forms do not admit of their opposites. The Phaedo is, in fact, the classic case in point. Socrates does say in that dialogue that no opposite ever becomes its opposite, as tallness short or shortness tall (102d5-103a2), but 8 This is the thesis that Rickless identifies as causing the problems in the theory of forms; see note 5 above. 9 Proclus drew attention to such points in his commentary on the passage in question and notes that they show Socrates to be at an incomplete stage of thinking about the forms though at a stage at which he can already begin to see the full theory of forms (as later Neoplatonists like Proclus himself expounded that theory). Proclus Commentary on Plato s Parmenides, translated by Glenn R. Morrow and John M. Dillon, Princeton University Press (1987): 93-99, ,

7 then, in response to an interjection, he immediately clarifies that this thesis must not be 7 understood of particular opposed things (about which it is indeed true that the opposite can become its opposite, as when a short thing becomes tall) but of the opposites themselves (103a4-c8). In other words, the thesis is not about something tall or short but about tallness and shortness, and the meaning is that an opposite (tallness) cannot become its opposite (shortness).10 This, of course, leaves open, as far as it goes, the possibility that, although an opposite cannot become its opposite, an opposite could participate in its opposite. Manyness, for instance, could never become oneness (the form of manyness is not the form of oneness), but it could participate in oneness (as the form of manyness participates in oneness because it is one form). In the Phaedo, therefore, Socrates says, as he does not say in the Parmenides, that the thesis about forms not becoming their opposites (or not having opposite predicates) must be taken strictly and only applied to the opposites as such and not to things (whether particulars or forms) that may come to share in the opposites. Hence it will apply to the case in the Phaedo of tallness being shortness but not to the case in the Parmenides of manyness being one. The fact that this point is put in as a sort of aside to the main discussion in the passage in question of the Phaedo only serves to bring it more to our attention. Plato has, as it were, especially highlighted for his readers a point that, if left out, might mislead them into taking Socrates remarks in a way that would leave these remarks vulnerable to objections (in particular those of the Parmenides). With the point expressly put in, by contrast, Plato has warned us against falling into this mistake. One should also note that the point being stressed here does not undermine the ensuing argument in the Phaedo about the immortality of the The point needs reading in this way and not, for instance, as meaning that tallness cannot become small (or manyness one). For then tallness (manyness) would not be taken as an opposite to smallness (oneness) but as a thing that possessed or shared its opposite, which is not the point at issue. Socrates is talking about the opposites themselves and not, at this point in the discussion, about things that have the opposites as a feature or property.

8 soul. For while that argument does turn on opposite things (snow and fire, soul and being dead) rather than on the opposites themselves (coldness and hotness, life and death), the opposite things turn out to be special cases where, as Socrates puts it, the thing always has the form of the opposite (as snow always has the form of coldness and soul always has the form of life, 103d5-e5, 105c9-d11). Hence, the opposite thing behaves like the opposite itself, whose form it has, by never admitting its opposite as snow never admits hotness but instead perishes, and as soul never admits death but instead goes away. But to return to the Parmenides. Before Parmenides comes to his direct criticisms of Socrates he asks for clarification about what things are forms (Parmenides 130b7-d2). That Socrates allows there to be forms of Likeness, Largeness, Beauty, Goodness, but probably not of Man and fairly definitely not of Mud or Hair or Dirt suggests again that he is trying, after the manner of the biographical excursus in the Phaedo, to develop some sort of cosmological theory of beings. For such a theory looks like an attempt to say what things a mind would need to make a world. Principles of Largeness, Likeness and so forth could thus, perhaps, be the primary principles for the construction of bodies and in particular of Man and Horse, say, and also of Water and Earth, and then these things could in their turn be used to explain Mud and Hair and Dirt (as combinations of Water and Earth and excrescences from Man and Horse).11 The Timaeus contains a cosmology along these sorts of lines and the Timaeus can, in fact, be viewed as a suggestion of what would be the case if one took seriously the claim that mind made the world. Anaxagoras, we know, made that claim but did not, in Socrates view, take it seriously enough. Socrates in the Phaedo says that he himself wanted to take the claim with full seriousness but was 8 11 See Ian M. Crombie, An Examination of Plato s Doctrines, Volume Two: Plato on Knowledge and Reality, New York and London, Routledge (1963): 2: Crombie helpfully refers to the Timaeus to fill out this sort of suggestion.

9 unable to find a satisfactory account and abandoned the project (Phaedo 98b7-99d2). The 9 Timaeus is not Socrates taking up the project again because Timaeus speaks in the Timaeus and not Socrates. Socrates is certainly an eager listener of Timaeus theory (because it seems to be what he had looked for in vain in Anaxagoras), but he never asserts it as his own, presumably because, as even Timaeus admits, it is no more than a likely account and so not something one can claim to know (Timaeus 29c4-d3). My suggestion is then and again to keep to the hypothesis I am exploring in this paper that in the Parmenides Socrates is presented as still at his early Phaedo-stage of trying to work out a cosmology and as struggling with doubts and difficulties about how his chosen theory of forms can be made to do the trick. The criticisms he is subjected to by Parmenides would, accordingly, be part of what subsequently induces him to give up cosmology and pursue the second sailing or second best course mentioned in the Phaedo (99c8-100a8). But I will come back to this point. The criticisms to which Parmenides subjects Socrates may be reduced to five: the problem of participation; the Largeness regress; forms as thoughts; the Likeness regress; the separation between forms and particulars. The first criticism gives some support to the suggestion just made that Socrates is trying to do cosmology (Parmenides 130e4-131c11). It takes participation in a materialistic way after the fashion of Anaxagoras (that fiery things, for instance, are fiery because they partake of more parts of fire than non-fiery things), and Socrates is sufficiently materialist, or, I would suggest, still sufficiently under the spell of Anaxagoras, to be caught by the criticism. Even his analogy with the day, that a form can be in many particulars and still be one form as a day is in many places and still is one day, is taken materialistically by Parmenides and Socrates does not protest. But if Anaxogoras could have a materialist view of participation and not be embarrassed by the fact (for he has infinite and divisible amounts of each kind of thing), Socrates

10 10 could not. His principles are each meant to be one and indivisible and so a materialist view will not work. The analogy with the day is, however, a step in the right direction and a further step in the right direction is taken in the Republic when Socrates proposes an analogy with the sun instead. This does not fully settle the problem of participation since the sun is an analogy for the Good rather than for the several forms, but it does escape Parmenides criticisms. The sun is a single source of light that, while remaining single, spreads its light everywhere and, if participation is like that, one can get a better idea of how forms could be single and yet present in many different particulars. Socrates does not profess, even at the end of his life in the Phaedo (100c9-e3), to have found a wholly satisfactory account of participation, but at least he avoids putting forward or accepting bad, materialistic ones. One might, however, say that there is an adequate explanation given in the Timaeus (28c2-29b2), where the forms are the craftsman s models or blueprints and the particulars participate the model because they are made, by the craftsman, to be like the model.12 But, again, this explanation is not that of Socrates even if Socrates would be attracted by it. It is a plausible account, to be sure, but it remains sufficiently doubtful, or sufficiently a matter of clever guesswork, not to count as something that the older Socrates (the Socrates after the Parmenides) can commit himself to. I do not mean to imply by this that we should not read the Timaeus as providing a possible answer to the puzzle of participation. I only mean that we should not ignore the fact that it is an answer that Socrates is never presented as putting forward or endorsing for himself. Parmenides second criticism raises the problem of an infinite regress (Parmenides 132a1-b2). If all large things are large because there is a form of the Large in which they share, and

11 if the form of the Large is itself large, then must there not be another form of the Large for the first form of the Large to share in so that it too can be large? And if this other form of the Large is also large then must there not be a further form of the Large for it to share in, and so on ad infinitum? Hence each form, instead of being single as Socrates wants, will turn out to be infinitely many. The precise structure of this argument has been the subject of much scholarly debate.13 I will not enter myself into this debate here but only use some of its more settled results to summarize what I take to be the main elements of the regress. So first we have the claim that where many things share the same feature, there is some form that is or at least corresponds to that feature. This claim is the so-called one over many principle (cf. Aristotle, Metaphysics 990b13). Parmenides states it in the following way: wherever there appear to be many larges, there appears to be some one form the same for all of them. Added to this principle we have, second, the claim that this form of the Large is also large, or is one of the larges. This claim has come to be known as the self-predication assumption, that a form receives its own feature as one of its predicates. Finally we have, added to these two claims, the third claim that the Large, in order to be large, must be large by sharing in some form of the Large other than itself. This is the so-called non-identity assumption. Socrates would seem to be implicitly making both assumptions in his response to Parmenides, since he takes for granted, or at least does not protest against, the thought that to say the Large is large is like saying this man is large. For this implies that the Large will receive this predicate in the same way as this man does, namely as an extrinsic addition. Thus, all that is necessary for Socrates to be caught in the regress is for him to be sufficiently unsure of See Sarah Broadie, The Third Man s Contribution to Plato s Paradigmatism, Mind 91 (1982): See in particular the essays by Geach, Owen, and Vlastos in Studies in Plato s Metaphysics, ed. Reginald Allen, Routledge and Kegan Paul, 1965; also Gregory Vlastos, Plato s Third Man Argument: Text and Logic, Philosophical Quarterly 19 (1969): ; also, among many others, Reginald Allen, Plato s Parmenides: Translation with Comment, Yale (1997):

12 himself, or sufficiently unclear about how he ought to speak, to be tempted by the one over 12 many principle and by the view that to say the Large is large is like saying this man is large. Now Socrates is in this condition in the Parmenides, but he seems not to be so later. In the Republic and other dialogues where the theory of forms arises Socrates is careful not to endorse anything like a full-blown one over many principle. In addition, he rejects the view that saying the Large is large is like saying this man is large. The later Socrates avoids saying, or allowing, that the is in this man is large is the same as the is in the Large is large. To say that this man is large is really to say that this man participates in the Large. But to say that the Large is large is really to say that the Large is identical with the Large or with largeness.14 To use the words of the Republic (476e4-480a13), the former is is the is of being and not being (einai kai me einai) and the latter is is the is of being (einai). With this distinction in place no regress can be generated, for the second part of the argument that to say the Large is large is like saying this man is large is false. But the effect of the distinction is to undermine any value that the one over many principle might have had to establish or prove the theory of forms. For now one has to read that principle as saying: wherever there are many larges that are not identical with largeness there is some form of the Large that is identical with largeness in which they share. But this principle assumes the theory of forms in its very statement and so cannot, without begging the question, be used to establish that theory. One should note, in fact, that the regress argument, properly understood, only creates problems for one of the reasons for positing the theory of forms. It does not create problems for the theory itself once posited. For since, as just mentioned, the theory separates out the necessary senses of is, it is as such not vulnerable to the regress. One must, therefore, distinguish between

13 criticisms of the theory of forms that attack reasons for positing the theory, on the one hand, 13 from criticisms of the theory that attack the theory once posited, on the other. This distinction I borrow from Aristotle and Aristotle is careful to note that the regress argument is of the former type and not of the latter (Metaphysics 990b8-17). It is an attack on the one over many principle, as already noted, and even if some Platonists (those whom Aristotle attacks) were inclined to put forward this principle as a reason for positing forms, it is far from clear that Socrates does this in any dialogue after the Parmenides.15 Perhaps it is also worth noting that the classical theory of forms in the Republic and Phaedo, while it does say things like the Large is large or the Beautiful is beautiful, does not go on to say, as Socrates does say in his response to Zeno in the Parmenides, that the Large and the Beautiful and forms generally cannot mix together and be divided, that they are just themselves and nothing else besides (129d6-130a2). The Large and the Beautiful, while being identical with nothing but largeness and beauty, could, as far as the classical theory is concerned, participate in lots of other forms, such as Being, Oneness, Otherness, Goodness, and so on. Of course such things are said in the Sophist by the Eleatic Stranger but they are exploited by Parmenides in the second part of the Parmenides where several of the paradoxes seem to trade on equivocation over the identity sense of is and the participation sense of is. Perhaps among the things Socrates is supposed to learn from these paradoxes is not only that this distinction between senses of is needs to be carefully drawn but also that the distinction holds at the level of forms themselves and 14 See Allen, Plato s Parmenides, 192; also Crombie, Examination of Plato s Doctrines, This conclusion is not brought into doubt but rather confirmed by the passage in Book X of the Republic (596a6-7) where Socrates is made to put forward a kind of one over many principle: we are accustomed to posit one form for each of the manys to which we apply the same name. For Socrates pointedly does not say here, as Parmenides does in the Parmenides, that wherever there is a many a form for that many appears, but rather that we posit (tithesthai) a form for the many. This looks more like an assertion of what we, who already accept the theory of forms, are accustomed to do in the light of that theory than like a proof or an argument to justify accepting the theory in the first place. Hence it is not the sort of one over many principle that gets Socrates into trouble in the Parmenides or that

14 not just between forms and particulars. However, we have to be cautious here because, in the 14 Sophist, it is not Socrates but the Eleatic stranger who introduces the participation of forms with forms. Socrates himself, even to the last day of his life, focuses rather on the participation of particulars with forms. Nevertheless he does this in a more chastened and tentative way. He does it, as he says in the Phaedo (100a3-7), by way of hypothesis, not by way of definitive assertion. In other words, we should be careful about how we read the suggestions in the Sophist just as I said earlier that we should be about those in the Timaeus. What Socrates is willing to say or to hypothesize is not the same as what he is willing, without complaint, to listen to being said or hypothesized by others. The Parmenides, we may suggest therefore, is not about making Socrates abandon the theory of forms; it is about stopping him, in the words of Parmenides, from trying to mark off the forms too soon before he has been trained in dialectic (135c8-d1). It is about making him less assertive and dogmatic and so about making him the one who knew that and what he did not know even if he was willing sometimes to hypothesize it. The analogies, for instance, of the Sun and the Divided Line in the Republic are tentative; they are attempts to explain what Socrates does not assert definitively. He gives the analogies only in response to the dialectical demands of the conversation. He would prefer not to speak (Republic 506b2-507a6, 509c1-4). But since he is forced to speak (Socrates, remember, is compelled to stay at the beginning of the Republic 327a1-328b3), he speaks according to the assumptions or hypotheses already made earlier in the conversation. The talk about the Sun and about the hypotheses of the Divided Line is itself hypothetical. It is hypothetical talk about hypotheticals. Socrates silence in the Sophist and Statesman (where theses about forms mixing with forms that are congenial to the analogies of the generates the regress.

15 15 Republic appear), may perhaps be best explained, therefore, by the fact that the Eleatic Stranger speaks more dogmatically, and wants to speak more dogmatically, than Socrates can or wants to. Eleactic theses, we may say, have a special fascination for Socrates, so he willingly listens; but he is unable to say that he knows them or that they are required for discourse, so he refrains from adopting them himself. The remaining criticisms of the young Socrates by Parmenides illustrate the same points. The criticism of Socrates suggestion that the forms are thoughts (Parmenides 132b3-c12) is to the effect that thoughts have a content in addition to being thoughts in a mind, so that one cannot avoid the previous criticisms about forms by retreating to the mental realm. Forms have some kind of objective being over and above their being thought about. This criticism does not so much raise a new difficulty as block a way to counter the two previous difficulties (namely that one cannot escape the problems of participation by reducing forms to mental concepts). It rests for its power, therefore, on those previous criticisms. If those criticisms can be answered, as they can, nothing further needs to be said to answer this criticism. The Likeness regress is directed at Socrates suggestion that particulars participate in forms by being made like them (Parmenides 132c11-133a7). For if particulars are thus like because they share a common form, and if the form is itself like the particulars, then it too must be like because it shares a form. Since that other form must again be like the form that shares it, there must be another form for it to share in, and so on ad infinitum. The way out of this regress is similar to the way out of the Largeness regress. The form is like the particulars, not because there is some one form they all share, but because it is like, or rather is identical with, what the particulars are like, namely itself. There is no need to bring in some other form of likeness here. The self-identity

16 of the form with itself (that the Large is identical with the Large) is sufficient for that Parmenides final criticism concerns the apparent unknowability of the forms that arises from the idea of correlatives (Parmenides 133b4-134e8). For each form is supposed to be what it is with respect to itself, and none of them exists among us but only the particulars do. A particular slave, for instance, is slave relative to a particular master, but the form of Slave as such is relative to the form of Master as such and not relative to some particular master. This seems plausible as far as it goes, but it leads to paradoxical results when applied to knowledge. Knowledge is similarly relative, since knowledge is knowledge of truth. Hence knowledges will be differentiated according to the truths they are relative to. The knowledge we have will accordingly be relative to what we know, and what we know are the particulars that share the forms, not the forms by themselves. The forms will, by this account, become unknowable to us and will play no role in our knowledge. The theory of forms will thus be rendered useless. Conversely the knowledge that is knowledge of forms will be possessed by the god if by anyone at all and the god will, with this knowledge, know nothing about us and the things that exist among us. He will be ignorant of all human things, which for those who, like Socrates and Parmenides, believe in the gods is a scandalous supposition. There might be a number of ways of responding to this argument but the older Socrates actual response would seem to be twofold. The first part we can find especially in the Republic (475e9-480a13), and it is the denial that there are two kinds of knowledge, knowledge of particulars and knowledge of forms. On the contrary there is only one kind of knowledge in the strict sense, namely knowledge of forms. Particulars as such are perceptible to the senses but not knowable objects for the mind; they only become knowable to the extent that they participate in 16 See Sarah Broadie, The Third Man s Contribution to Plato s Paradigmatism, 345.

17 forms. This is surely one of the messages of the analogy of the Divided Line. What is in the 17 visible part of the line, the particular sensible things and their images or likenesses, are not the object of knowledge but of imagination (eikasia) and belief (pistis). Only in the intelligible part of the line, and in the higher of its two divisions, is there knowledge or understanding (noesis). Only then do forms as such appear and only then can human beings be said to get to realities or beings, and no longer to images (Republic 509d1-511e5). The second part of Socrates response is the doctrine of recollection or anamnesis from the Meno, Phaedo, and Phaedrus. Particulars are not themselves objects of knowledge; they are only occasions for recollecting the forms, the true objects of knowledge. The soul pre-existed its own birth and was then like the god in seeing the forms directly by themselves; birth, however, has induced a certain forgetting of this previous knowledge and the soul needs to be reminded of what it once knew. Particulars do the reminding. They do this, presumably, because they are likenesses or participations of the forms and the likeness or participation is what does the triggering of the dormant memory. Thus there is no problem of separate knowledges. If we know anything it is forms that we know, not particulars; and to the extent that particulars are knowable, god knows them as much as we do, if not more so, since they are knowable to the extent that they participate the forms, and god knows the forms more perfectly than we do. The Old or Mature Socrates after the Parmenides My conclusion, therefore, is that none of Parmenides criticisms, properly considered, contains anything that is fatal to the classical theory of forms found in dramatically later dialogues. Indeed the classical theory can be seen deliberately to avoid these criticisms. It either provides answers where it can or refrains from any definite assertion where it cannot or need not. And even there

18 answers are available though ones that Socrates only listens to and does not himself assert. I 18 attribute this change, in part at least, to the dialectical exercise in the remaining sections of the Parmenides. The need to distinguish between the is of identity and the is of participation can be found there as also the need to allow, or at any rate not to deny, that forms themselves participate in other forms. The need to be much more cautious in stating theories is a message of the whole of the Parmenides. It is a message that Socrates seems to follow in his later life. The Parmenides also, of course, recommends a method for ensuring this cautiousness. Socrates learns from the method but he does not seem to have adopted it himself, or not in its totality. The method of provisional hypothesizing, as we may call it, that he attributes to himself in the Phaedo is not exactly the same as that of Parmenides. It does, to be sure, involve turning attention to things through the medium of speeches and not directly, and this is a marked feature of Parmenides discussion with Aristoteles in the Parmenides. We may conjecture that Socrates adopted this procedure, or what he calls his second sailing, in imitation of Parmenides. He carries it out rather differently, though. Socrates does not recommend taking any hypothesis whatever and giving it a systematic examination by seeing what follows for itself and other things if it is asserted and then again if it is denied (Parmenides 135d7-136c5). He recommends rather adopting some plausible hypothesis and seeing what follows from its assertion, as he does in the Phaedo when he examines what follows as regards the immortality of the soul from the hypothesis of the theory of forms. He also says, and again in accordance with the analogy of the Divided Line from the Republic, that if the hypothesis is challenged one should retreat to some higher hypothesis from which the first hypothesis can be seen to follow. But how is one supposed to arrive at a plausible hypothesis in the first place? Parmenides offers no advice on that score but speaks as if one might adopt any

19 hypothesis at will. Socrates, however, does have his own method. It is his maieutic or 19 midwife method from the Theaetetus (149a1-151d3). He examines the hypotheses or opinions of others on some topic of interest that arises from the context of discussion, such as, in the Republic, what justice is. He himself propounds nothing but poses questions to others in order to bring their ideas to birth and to examine them. From this questioning and examination, however, there does result a need to posit some theory of forms, if only to explain the possibility and nature of discussion.17 For the discussants are talking about something, justice, say, or courage, or even knowledge, and understand enough about it to be able to see, through questioning, that this or that proposed account of it will not work. But they are not able to see, or not easily, what account will work instead. What will explain this fact? Some theory of forms, and of forms separable from particulars, seems to be the answer. For we need a theory to the effect that justice or courage or knowledge are fixed natures (else there would be nothing, or nothing definite, to discuss) but not natures that are immediately manifest (else the correct account would not need discussion to be found but could be read directly off the particulars). Hypothesizing a theory of forms seems, therefore, to be the most reasonable procedure, and that is what Socrates does, at least when he is not questioning others opinions. Still even this hypothesis he does not so much assert on his own account as take from the agreement of others who, with him, have come to see its reasonableness (Republic 475e9-476a8; Phaedo 100b1-c8). Once the hypothesis is in place the discussion proceeds, in Parmenidean fashion, to examine what follows. Moreover, only as much of the theory of forms is put forward as is needed for the matter under discussion at the time, unlike Socrates 17 It might be argued that the Theaetetus is precisely the dialogue that shows Socrates has abandoned the view that Forms are necessary for discussion, since it is devoted to a discussion of knowledge, where Forms would seem to be absolutely indispensable, if they are indispensable anywhere, and yet Socrates does not discuss the topic. But this is not the only way of reading the Theaetetus and other ways of reading it will not yield the same conclusion. At least that is my contention.

20 20 procedure in the Parmenides with Zeno when he goes over the top, as it were, in talking of the theory. Socrates has learned to be cautious and not to propound, like his Pre-Socratic former self, a full-blown theory of his own regardless of dialectical context. One should note, however, about Socrates method of midwifery, that it can be practiced with or on anyone, young or old, and that it involves considerable input from the one being questioned. Such is the style of a typical Socratic dialogue (the sort that scholars tend to regard as early), where someone s opinion is examined and found at the end to be wanting. These sorts of dialogues could, of course, be early, but that is not my interest here nor is it, I think, the important thing to notice. What is important is rather that the maieutic method itself, regardless of when it is being used, requires this sort of dialogue. By contrast the Parmenidean method does not. All it requires is a respondent to agree that some proposed consequence does, or does not, follow from the hypothesis laid down. And the younger the respondent the better in this regard so that no unnecessary or irrelevant issues are raised but each point gets accepted or clarified at once. Parmenides himself, of course, provides a perfect example of this method in the Parmenides, but so does the Eleatic Stranger in the Sophist and Statesman. Socrates nowhere provides as clear examples but there are certainly some, as precisely in the books of the Republic where the analogies of the Sun, the Divided Line, and the Cave appear, though Socrates never insists on youth in his respondent as Parmenides and the Eleatic Stranger do. Socrates, because of his maieutic method, is more friendly to genuine discussion. The differences of style between the several dialogues, therefore, as also between the first and the subsequent books of the Republic, need not be construed as indicative of time of writing by Plato. They can, instead, be read as indicative of the different methods the main discussant is presented as using, whether maieutic or Parmenidean. Hence elements of both these styles can be found in the same dialogues and from all

21 periods of Socrates life. 21 Socrates presumably got the Parmenidean part of his method from his meeting with Parmenides. The method, when practiced by itself, is presented as typically Eleatic, as in the Sophist and Statesman. Part of the reason for this is, perhaps, that Eleatics have a definite theory of their own to propound. Socrates by contrast does not. He has a theory, to be sure, the theory of forms, but he takes it from the dialectical context and in the way the context requires. He does not try to impose it from outside, at least not after the Parmenides. The reason for this is, I suggest, because he adds to the Parmenidean method the maieutic method and this latter is logically prior. It is required to get a hypothesis going in the first place, either from some interlocutor or, in the case of the theory of forms, from the internal necessities of the nature of discussion. The full development of this method and its combination with the Parmenidean method seems to be typically Socratic, or perhaps if we follow the Delphic Oracle in the Apology (21a4-23c1) typically divine. At least such is the picture, or part of the picture, that one can put together from reading Plato s Socratic dialogues according to their dramatic chronology. How this picture, or any picture based on the dramatic elements of the dialogues, will ultimately fare against the currently more accepted one that is derived from reading the dialogues according to their historical chronology, whether it gives us a better understanding of the dialogues, or of Socrates or of Plato, is an interesting question, and a question that is rightly, I think, attracting more scholarly attention. A note of caution, however, needs to be added in my own case. One should not suppose that the alternative picture I have here proposed implies that the later the dramatic date of a dialogue the more philosophically mature Plato intends Socrates discussion to appear. For, to mention only the most obvious, the maturity of a discussion will also depend on the maturity of the interlocutor. All,

22 22 I think, that this alternative picture implies is that the method Socrates uses, and the character he displays, in the dramatically post-parmenides dialogues, however mature or immature the interlocutor, will at least display the lessons of the Parmenides. But that is not insignificant. On the contrary, if my suggestions in this paper are anywhere near right, it should often prove decisive.

23 23