Socratic Eros and Platonic Dialectic Author(s): Jerry Stannard Source: Phronesis, Vol. 4, No. 2 (1959), pp.

120-134 Published by: BRILL Stable URL: http://www.jstor.org/stable/4181655 . Accessed: 19/04/2013 21:21
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 of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

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

http://www.jstor.org

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

Socratic Erosand PlatonicDialectic'

JERRY STANNARD

Introduction
PROBLEM for the student of Platonism is that of establishing the relation between Socrates and Plato. To this problem I can claim no comprehensive answer. But within the confines of the present topic some further details may be added to an already extensive literature. I propose to investigate two aspects of dialectic - certainly one of the links between Socrates and Plato - and to argue (i) that the Socratic eros serves a philosophic function later paralleled by Plato's dialectical method and (ii) that Plato's formulation of the dialectical method was the outcome of his reflection on the role played by the Socratic eros in philosophic inquiry.

HE MOST VEXING

Section i That Socrates employed a mode of argumentation, that of question and answer, and that this method is described frequently in terms of the verb 'to converse with' (aL0?1youocL), is so well known that there is no need to rehearse the evidence.z One of the most evident of characteristics used to describe Socrates, and one found in Xenophon as well as in Plato, is his love of conversation.3 As Robinson and others have pointed out, the 'What is X?' question is typical of the Socratic method.4 In order to get at the answer to questions such as 'What is friendship, piety; and the like?,' Socrates engaged a representative of nearly every class or type of Athenian. His especial interest was, however, with the youth. For it was when Socrates engaged them in such questions and
I Except for the addition of references and a few minor alterations, this is substantially the paper read before the FacultyResearchLuncheonGroup in May, 1958. 2 Apol. 33A, 37A; Phaedo 6iD, 63D; Charm. jg4E, IsSC; Symp. 213D, 217B; Xen. Mem. I i, x6; vi, i; II x, i; IV v, 12. Cf. Crito 49A, B ceLpCj &noxpEvea8oL r6 ... Ipaullo post] MvSpeq7tpo &CXouq a7rou8t gLoc2Xy6FLevot and Lysis 1pco,rEV.ov 21IC (cf. n. 3). 3 Charm. IS3A (Socrates' aut)OeL4 &Oarp43&4); Xen. Mem. I i, I0. Cf. Lysis 21 I C where to the request tWL&m'X&yoU au r Socratesreplies &LoOx?kovand Symp. 2 17B:

&WoE7tp C?WCL8tOXX?lC.

ThefactthatSocrates wascriticized several timesfor speaking

on such homely subjects as cobblers and cooks (Symp. 22I E; Gorg. 4go E; cf. Xen. Mem. I ii, 32-37) must refer to a typical feature of his conversations. 4 Cf. R. Robinson, Plato's Earlier Dialectic 2 pp. 49 sq. and K. W. Mills, Phronesis 2 (19g7) i4g-46. This characteristic was already recognized by Xenophon Mem. I i, 16.
120

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

forced them to re-examine the traditional mores, that we find the purposeback of his persistent questioning. This purpose, moral betterment, constitutes a second characteristicof the Socratic method.' When we turn from the early dialoguesto the later ones, the second characteristicis no less evident despite the fact that Socratesceases to who be the central speaker.It should be noted, however, that regardless is betterment moral always or actual be, might the speaker questioner assumedto be the outcome of the discussion.2This is to be explained by the fact that Socrates' dialectical method has been modified by Plato in such a way as to preserve what he believed to be the nucleus of the method yet, at the same time, placingthe entire method on a philosophic footing quite absent in the early dialogues. One reason why Socrateswas so successfulin the method of question and answer was the extra-philosophiccontext within which his conversations were placed: Socrates, especially when conversing with the youth, enjoyed a close, intimate friendshipbeyond what can properly be termed a teacher-pupilrelation. In the dialoguesof the later period, when Socrates is no longer the main speaker, what is the mechanism that assuresthe success, morally and intellectually, of the joint inquiry typifiedby dialectic? 3 If we consider the middle-period dialogueswe shall find the answer. It is only in these dialogues that the friendshipwhich existed between Socrates and the youth receives a philosophical justification. The with their theory of eros ground friendship and the Phaedrus Symposium on a philosophic basis and supply as well the purpose of eros. Eros is grounded philosophicallyin these two dialogues in that it enables the and to reach the world lover to transcendthe world of sense particulars of metaphysicalreality. This is, of course, an oversimplificationof two of Plato's richest dialogues. But for all its oversimplificationthe above serves to bring out two important points, (i) an exacharacterization minationof eros is inadequateunless it revealsthe hierarchicalstructure
I
2

Euthyphro2D; Laches i9oB; Charm. i66D; Symp. 218E; Xen. Mem. I i, iS. That this is one of the functions served by dialectic cf. Soph. 246D; Phaedr. 276E277A and Skemp, Plato's Statesmanp. 67. 3 Socratesis made to say on several occasions, "let us investigatetogether" (cf. Protag. 332 D; Meno 87-88; Hipp. I, 29gB. The belief that philosophyis a joint enterprise or
(Crito 48 B; Charm.
I g8

xoLVI cx4Lq

D; [Alcib. I], 124 B) is as much a Platonic dogma

as Socratic and persists even in the later dialectic, witness Pol. 2S8C. The Meno may noted (Die Stellungder Menonin der platonischenPhilosophie, be viewed, as Buchmann Philol. Suppl.-bd. 29 (1936)) as a turning point in the emergence of a distinctively Platonic dialectic from the earlier Socratic dialectic of n. 2 p. 120. Cf. especially Meno
75D, 78C, 8oD. 4
l

2I

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

of eros and (ii) this hierarchicalstructure is, in the end, to be explained in metaphysicalterms. To these two points I shall return in the following pages. I suggestedabove that there must be some strong yet intimatebond between the questioner and the person questioned. A bond such as this is a necessarycondition for the sort of discussionwhiclhSocratesbelieved always ended in moral betterment. As Plato continued with the composition of the dialoguesit became more and more apparentthat such a relation between questioner and person questioned must, if philosophy is to end in moral betterment, be such that the mutual inquiry will, however difficult, be sufficiently rewarding to maintain the necessary degree of concentrationand participationof both parties. For Plato this was literally an academic matter as well as a practical matter. In fact, the two cannot be separatedwhen emulation of Socratesis in question. The bond of which I have spoken that existed between Socratesand his respondents, particularlythe youth, was eros. When Plato came to 'idealize' Socratesin the middle and late dialoguesit was necessarythat eros be explained.' With Socratesdeparted, a substitute for eros must be found. Its explanationwas imperativeif the Socraticprogramwas to be carried on as a living traditionin the now-functioningAcademy. Thus, when we turn to the Platonismof the later dialogues, we must remember what his problem was. The first point I should like to establish is the hierarchicalstructureof eros. This will be followed by some further definingcharacteristics of the Socratic eros. Eachof the defining characteristics,I shall then argue, is paralleledin what Plato believed to be the philosophic method par excellence - dialectic. In conclusion I shall suggest that the parallels adduced between eros and dialectic were responsible for Plato's examinationof his own method thus permitting him to make the distinction between method and the theory of method, i.e. methodology. Section2 One of the obvious points of contact between the Symposium and is the explanation of eros in hierarchical terms. The the Phaedrus general structure of the hierarchical treatment of eros is that, proceeding from sensual or even carnal eros, a more worthy type can be
Socratessay of himself toac&rov ey6 Uvociuv ol1au gXevL ?t5 ra' a
Plato was frankly admitting this power of Socrates' in [Alcib. I1], ioSD when he has 7tpay.LFas xxl Etqak. Alcibiades also admires Socrates' ability, cf. Symp. 2iD, 2i6C (rV KuvoLv dg gXeL) and 2 I 8 E. OcaxVlxsaEv
I

1 22

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

engendered provided that certain conditions are fulfilled. The theory of eros in the Symposium,in its hierarchical structure, is that of transcending successive stages of a purified or re-directed desire (.iu[LOvd) while the Phaedrus, thought not ignoring the motive force of desire is more concerned with the establishment of a special sort of mechanism, madness (p.xvlx) by which sensual eros is transcended (24MB sq.; 249D sq.). In both dialogues the outcome is the same: starting with an undifferentiated psychic force or motive power it is first necessary to re-direct or re-educate it when it is misdirected and by continuing the re-educative process the lover is led on to more exacting stages of nloral and cognitive development until at such time the goal is reached.' This goal is of the kind that, once attained it is declared to be impossible to revert to a lower stage.2 The detailed exposition of the re-educative process, plus the statement of the goal at which the lover aims, occupies but a small portion of either of the two dialogues. Partly this is due to the other purposes served by the dialogues 3 and partly to the nature of the subject matter. In both dialogues there is, besides the portrait of Socrates the lover,4 an examination of the nature and function of eros. These two strata are not always distinguished but it seems clear that Plato intended more than a portrait, however idealized, of Socrates. What more he intended was, I believe, an examination of how eros actually functioned in, and in some cases was identical with, philosophic
I As Plato came to recognize the significanceof his dialecticalmethod the notion of the goal of philosophy was modified accordingly. In the early dialogues moral betterment was almost exclusively the goal (cf. n. I p. 1 2 X). In two of the middle dialoguesa single Form was made explicitly the goal - The Good in Rep. VI and Beautyin Symp. 2 Io0-I I. In the later period the goal, though not excluding moral betterment as is clear from the Phaedrus(cf. n. 2 p. 12 1), tended to become the knowledge of a pluralityof Formsand their interrelations, e.g. the yLywroc yk'nr of Soph. 254 sq. (cf. n. I p. 130). 2 See now, Bluck, The Phaedrus and Reincarnation, AJP 79 (1958) I56 sq. esp. p. i 6o, "...since the object of trying to attain to philosophic virtue is in fact eternal bliss, it is clear that only when such virtue has been attainedcan one hope for complete immunity from any further fall." 3 Thus some scholars, unconvincinglyto my mind, have stated or suggestedthat the aim of the Phaedrusis a "reformedrhetoric," e.g. Taylor Plato 5 p. 3oo and suggestedby Hackforth,Plato's Phaedrusp. 9. I believe a "reformedrhetoric" to be a contradictio in adiecto (cf. Phaedr. 26oE); failureto recognize this leadsto some peculiarstatements, e g. "Les deux discours [sc. in the Phaedrus]de Socrate illustrent magnifiquement le programmede la rhetorique dialectique" (V. Goldschmidt, Les Dialogues de Platon p. 324). As for the Symposium,I. Edman, Arts and the Manp. i i6, believes it to be a "defenseof the arts." 4 In addition to the portraits in the Symposiumand the Phaedruscf. Meno 76C; Xen. Symp. VIII2; Mem. [I vi, 28. 1 23

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

inquiry. There would be little point in discussingthe mechanismof eros were Plato intending only to reproduce Socrates' conversationon the two occasionswhich serve as the dramaticbackdropof the dialogues. Some further expansionof the hierarchical natureof eros is requiredin order to assess its philosophic significance. Startingwith the fact that eros is common to all, it follows that since not all persons act in the same manner, eros alone is insufficient to determine and regulate a course of action. Plato makes this point in other terms but the fact remains that in the Symposium and the Phaedrus the hierarchicalnature of eros subservesthe finalattainmentof the Forms. The majordifference between the 'Lesser Mysteries' and the 'Greater Mysteries' of the is that the latter make explicit what it is that guides or directs Symposium the successive stages of the erotic ascent. The explication is the final vision of the Forms in Symposium2ioE-2I iA and Phaedrus 247 CD. In
regard to the Mysteries of the Symposium, some scholars have asserted that the 'Lesser Mysteries' are substantially Socratic while Plato's own beliefs are embodied in the 'Greater Mysteries.'" Now the former are no less ennobling than the latter but there is no guarantee, philosophic or otherwise, that once started on such a path the end result will be, or will be reached in, precisely the manner Diotima claims. Given a Socrates it certainly follows that oLiyx6tovEq xoCrXT cvcWc[Mwill press onwards and become o'L?yxUp.ovs4 x&tra tv iu 'v (206 C). But what of others? There is nothing to ensure that such a program will be adopted nor that, in the end, a Ypoanv- re xaL &xOOaUoV- (209 A) will be engendered. It happened, to be sure, in Socrates' case and Plato is examining whether, with nothing further, the process can be duplicated by all and sundry. His answer is the 'Greater Mysteries' and takes the form of specifying the object of eros at each stage of the erotic ascent. The enumeration of the stages of eros corresponds to the objects peculiar to each stage of the hierarchy.2 The hierarchical structure of eros is
I Jaeger, PaideiaII, 192; Cornford, Unwritten Philosophyp. 7i; Stenzel-Allan,Plato's Method of Dialectic p. X. 2 This might be schematizedas follows. The left-hand columnn states the type of eros, while the object of that specific type is found in the right-handcolumn. (i) sensualeros a beautifulbody (Symp. 2io A cf. Charm. Is4B) (ii) humanitarian eros all beautifulbodies (Symp. 2 I O B) 4uxf or spiritus(Symp. 2 I OB; cf. [Alcib. I], 131C; Xen. Mem. (iii) spititual eros IVi, 2) (iv) socialeros institutions andlaws(Symp. 2 10 C; cf. Crito Si AB) (v) epistemic eros intellectual studies (Symp. 210 C) (vi) formal or philo- The Forms (Symp. 2 I OE sq.; cf. Phaedr. 247 CD). sophic eros

1 24

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

determined by a correlative hierarchyof TC& 6vt. We shall see below that the principlewhich establishesthe hierarchyof eros, viz. the Theory of Forms, is exactly analogous to the principle by which dialectical divisionsarejustified(cf. infra p. 13 I). The uppermostpoint of the eros hierarchyis the vision of the Forms; and this vision, clearly adumbrated in Rep.VIIis only dramatically,not historically, ascribableto Socrates. The Phaedrus, likewise, contains a hierarchicalaccount of eros. The fact that this account is embellished by and interwoven with the myth of the chariotand that of the lover's soul growing wings should not obscure the central fact that the lover must pass through various stages before he is permitted to attain his goal. The argument for the hierarchical ascent would seem to be that since erotic madnessis divine and that the divine is beautiful, intelligible and good (246 E), it follows that only the inspired lover can know the beautiful, intelligible and good (249 E). Plato even uses the imagery of the mysteries, reminiscent of the Symposium,to describe the final goal resulting from a transcendenceof the lower forms of eros. He who is not newly initiated, or has been corrupted, does not quickly rise from this world to that other world and to absolute beauty when he sees its namesakehere, and so he does not revere it when he looks upon it, but gives himself up to pleasure and like a beast proceeds to lust and begetting; he makes licence his companion and is not afraidor ashamedto pursue pleasure in violation of nature (2goE-25pA, trans. Loeb). Once the hierarchicalstructure of eros has been established, the remainingpoints canbe dealt with more briefly. A second characteristicof eros, one that is, so to speak, deducible from its hierarchicalcharacteris what I shall term the mediating function or intermediate role of eros. For eros, in spite of its hierarchicalstructure, is itself a process which renders continuousany development - be it psychic or epistemic. That is, the assertion that eros is hierarchicalstill leaves open the question whether the successive stages of eros can, in fact, be attained and each but the ultimate stage transcended. Further, even if this is so, is the passageupwardsone of discrete jumps with nothing more than an ad hoc justification, or is the ascent orderly, continuous and of such a nature that it follows a patternwhose conditions can be specified and for which rules are available?If eros is the sort of process which is orderly and continuous and one which can be plotted with reference to the object correlated with each stage of the hierarchy, then eros is, by its very 125

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

nature, the sort of self-motivated force which, once put into effect, needs no additional vis a tergo. It may, of course, need to be channeled even after the process is underway but such cases would seem to be rare.' And even if a case as this should arise there will not be required as stimuli anything beyond recognition that each stage, defined by its object, is superseded, save for the last, by a further stage whose object is the more real and hence more valuable. This second defining characteristic of eros can thus be stated in terms of the mechanism or motive power which, proceeding by its own intrinsic nature, carries the lover onwards until the process is complete. A process such as this is exactly and in the that depicted in the 'Greater Mysteries' of the Symposium Phaedrusby eros as the fourth type of mania. In both cases, should the premises by which the lower stages are established be accepted, the end result can be calculated. Further textual evidence for the second characteristic consists in Diotima's description of eros as analogous to the role assigned to Op0' 86ECt (202E). The latter two occupy an inter(Symp. 202A) or to a &oXEus[v mediate position, one in an epistemological, the other in a theological hierarchy, of such a sort that in virtue of their intermediate position a continuum is established. The importance of the mediating function of eros is that in its absence there would be no cognitive means of specifying how the ascent is to be made. That is, instead of restricting the role of intuition to the final vision of the Forms, a form of non-discursive reasoning would preempt the place where discursive reasoning is indispensable and alone suitable. As we shall see below, the mediating function of dialectic, itself very much a form of discursive reasoning, is the feature to which Plato attaches so much importance as regards dialectic as a method of inquiry.2 The intermediate position of eros is the mechanism by which the gap between sensual eros and 'heavenly' eros is bridged; eros thus is responsible for establishing a continuum between the world of sense
I The outstanding example is, of course, Alcibiades. Probably Socrates, as much as any of Alcibiades' contemporaries, was impressed by his capabilities and attempted to guide him along more socially-useful paths by introducing him to philosophy. The important point here is that Socrates took advantage of Alcibiades' advances and, on this basis, sought to make him a better man (cf. Symp. 2 i 8 E: xot rtL4 eat' &v ,xoLlKVO4LwL, VOLO O'ELevcv). It was this attempt which Plato probably had in mind in 8t r; CDv a{ Phaedr. 2s4E when he speaks of the vicious horse who, when humbled sufficiently, (cf. Symp. 21 8 A, 2 I 9 C) will eventuallyfollow the wisdom of the charioteer - in this case Socrates. 2 Non-discursive thought plays an important role in dialectic, however. The recognition 1 26

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

in the same fashion as OpOp 86ioc and, particulars and the world of Fornms later, dialectic serve to establish an epistemological continuum. The analogy with Op0' 8o6, is instructive because, it is Plato's frequent accusation, all too often individuals stop at the stage of doxa and aspire to nothing further.' The analogy should not be pressed, however. Although it is true that one may stop and rest content with sensual eros, none the less eros is itself, in a way in which doxa is not but reason and dialectic are, the means by which any temporary halting place may be transcended. This is so because eros and, as I shall show, dialectic, unlike doxa, are hierarchical in the sense that each stage is rigorously defined by its object. But this is precisely what is impossible with reference to doxa and its objects since the latter, sense particulars, are characterized by transiency, mutability, and indeterminateness and so never capable of specification.z Now it may be countered by saying that the objects of the lower stages of the eros-hierarclhy are sense particulars too. This cannot be denied but, on the other hand, since the ascent, at least in its lower stages, is one in which discursive reason plays a major role, the admission does not detract from the value and reality of the objects, known only in a non-discursive fashion, which define, at least, the ultimate stage. In virtue of their superior ontological and hence value status, the objects defining the ultimate stage, the Forms, serve as the goal for the completion of the ascent. In order to introduce the third characteristic of eros - what I shall call its discriminating function - let it be noted that Plato's purpose in the and the Phaedrusis an examination of the term 'eros' quite as Symposium much as an examination of those actions denoted by the term 'erosr' Here again the Socratic basis is evident for the 'What is X?' question is directed to an examination of a particular term as well as its referent.3 It is of slight consequence to praise or comdemn something until the nature of whatever is to be praised or condemned is clearly known and distinguished from what may superficially resemble it. All too often,
of what genus is to be chosen for the successive divisions is, as Cornford remarks, the "intuitive moment" in dialectic (PTK p. 83, i86). Campbell is referring to the same psychic mechanism when he states that 'divination seems always to be assumed as the first step in the dialectical inquiry" (The Sophistes and Politicus of Plato p. 1o8). Cf. the penetrating discussion of Kucharski, Les Chemins du Savoir dans les Derniers Dialogues de Platon pp. 169 sq. I Cf. Meno 8sC; Gorg. 4s4D sq., 4s8A; Pol. 278D. 2 The locus classicus is the Divided Line of Rep. VI. Cf. also Meno 96E-98A- Phireb.
52D.
3

Cf. Meno

7I

B and Gorg. 448E.

1 27

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

both semanticandbehavioristicanalysis are necessarypartsof philosophic inquiry because of the confusionproduced by misleadingsimilaritiesor hidden dissimilaritiesbetween the term or act in question and a related and the Phaedrus are or supposedlyrelated term or act. The Symposium lessons in analysis in the sense that it is necessary, prior to speaking about eros (either in praise or condemnation) to define its nature.I Socratesthus remarksthat he will be unable to follow the other speakers in their fulsome praise of eros since he is guided solely by the truth (Symp.I98DE).2 It turns out that this means that it is first necessaryto state the nature of eros before proceeding to enumerate its effects (ib. (237C) Socratesstates that lest the disi99C). Similarlyin the Phaedrus cussion about eros become a mere logomachy, the definition of eros must first be decided on. Following that, its nature and effects can be most profitablydiscussed. So far I have spoken only about the semantic discriminationof types of eros. But it is my contention that eros itself is discriminating.We have observed that each stage of the eros-hierarchyis defined by its object. to the purpose or object of the varioustypes of When this is transferred erotic activity it will be seen that each type of crotic activity can also be The locusclassicus defined by its particularobject or end-in-viewv. is the There each juxtaposition of the first six speeches in the Symposium. speakerpraisesa particularsort of erotic activity on the groundsthat the specific nature of the activity is evidence for a specific type of eros. What is noteworthy here is that each type of erotic activity discriminates between a variety of possible objects or ends and chooses one alternative to the exclusion of others. The lover discriminatesamong courses of action because eros itself is directed to one type of object to the exclusion of others. It is the nature of eros to be discriminating,for were it not the possibility of transcendingthe lower stages of the hierarchy would be closed to all savethose who possessedan educatedor philosophio eros. And the assumptionback of the Symposium is and the Phaedrus that man can be educated, that a perverse eros can be re-directed.3 In the Phaedrus Socrates contrasts 'left-handed' eros with 'right-handed'
I

Paralleled by Gorg. 463 C.

2 Socrates frequently claims that his sole motivation is to seek the truth (Charm. X6 i C;

Meno 7sC; cf. Euthyd. 273 E; Gorg. 473 B). Plato, likewise, stresses truth in his description of the results to be obtained by dialectic, cf. Soph. 246D; Phileb. I4B; Pol. 266D; Phaedr. 259E, 2o6D. 3 The educative function of eros comes out clearest in Phaedr. 2 5 3 B and in the Symposium (cf. n. I p. I26). But it is implied also in the early dialogues (e.g. Lysis 221) and in the late dialogues (e.g. Soph. 2 2 3 A). 128

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

eros on the grounds that the former aims at sensualgratificationwhile the latter discriminatesfrom amonga set of alternativesand chooses the
philosophic life (26sE-266B). 1

The fourth characteristicof eros is actuallytwo-fold - eros as synoptic and eros as systematic. I have suggested above that the purpose of Socratic inquiry is moral betterment. When this is coupled with the metaphysicaljustification of ethics supplied by the Platonic theory of Forms the promise is held out that the goal can be achieved by the adoption of the specified procedure. The dual characteristicof eros as synoptic and systematicarises from the fact that the completion of the erotic ascent is such that each but the last stage has been experienced and found to be wanting in ultimate, i.e. metaphysicalsanction and has thus prompted the ascent to the next higher stage. The completion of the ascent is, by definition, complete only if each stage but the last has been transcendedand that the means for the ascent have been the adoption of the procedures specified. Of these, the most important are the belief in the Forms, that knowledge of the Forms is virtue, and finally that such beliefs are capableof being communicatedand taught. In the Plato states that the marks of the good are perfection, suffiPhilebus ciency and choiceworthiness (2oD). These are the properties which most fittingly describe the final stage of the eros-hierarchy.The lover knows, upon realizationof the goal, that his ascent is rewardingand it is so because synoptic vision and systematic completeness, both of which accrue to the beholder of the Forms (Rep.VII, s37C), are guaranteed by the eros which places everythingin its correct perspective. The same characteristic,it will be shown below, applies to dialectic classification and the ordering of genera in their correct relations to one another.
Section 3

I should now like to turn to the second part of my argumentand to justify the contention that Plato's dialecticalmethod is characterizedby the identical features which I have asserted above characterizeeros. I shall suggest, finally, pending a fuller treatment elsewhere, that Plato's examinationof dialectic as method, that is, his methodology, is what
I The alternatives to a philosophic way of life are none other than the eight types of incarnations depicted in Phaedr. 248 C-E. This seems clear by comparing them with the Er myth whose moral consists in the fact that those who are without philosophic training will inevitably choose unwisely (cf. esp. Rep. X, 618 A-C).

129

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

could be expected of one who observed the significanceof the Socratic eros and its place in philosophic inquiry. Adopting the same sequence as above, the first characteristicof dialectic is its hierarchicalstructure. By 'dialectic' I shall mean that bi-polar process of collection and division which, from its first appearance in the Phaedrus, occupies a major portion in the dialogues immediately following - the Sophistes, Politicusand Philebus.Limiting my remarks to dialectic in this sense, I thus rule out that method, actuallycalled by the same name, found in the Phaedo and Republic on the grounds that the earlier method aims, not at the classificationof a pluralityof Forms, and a determination of their inter-relations, but at the discernment of a single, unique, supreme Form typified by the Form of the Good.' The hierarchical structure revealed by dialectic is nowhere more evident than in the process which Plato terms 'division'. Beginningwith a term, commonly accepted as one which denotes a class or genus and by repeatedly dividing in a dichotomous manner the class-term, operating only with the right-hand member of the previous division and ignoring the left-hand member, there is reached eventually the term which denotes the infimaspecies. This term is the name or definitionof the smallest class any further division of which would result in a statement about one or several of the particularscomprisingthe class.7 The term denoting the initial class plus the term denoting the final righthand dividend constitute a definition by genus and specific difference of the class-termin question. This may be illustratedby Plato's example of the definition of the class-term'angler.' An angler is definedas one who engagesin the art of acquiringsomething (the initial class, tIxvn XX1J, is the widest class) by striking upwardsfrom below (the last right-hand dividend and so the class narrowest in extension, Soph. 22i B). Plato emphasizesthat in the process of division the lines are drawn, not haphazardly,but only at the naturaljoints.3 That is, the division into genera and sub-g,enera is regulatedby a naturalcleavagebetween classes
Cf. Stenzel-Allan, op. cit. PP. 79-83: Ross, Plato's Theory of Ideas p. 8i; Cornford, PTK p. I 8 E. The aim of the later dialectic, on the other hand, is as Cornford PTK p. 18 3 puts it, "the mapping out of the realnmof Forms in all its articulations." Cf. Ross ibid. P. 113; Skemp, op. cit. p. 69. 2 Cf. Phileb. s6 D-x7 A and Stenzel-Allan op. cit. pp. xxxii-xxxiii. 3 This methodological requirement was introduced in Phaedr. 26sE and repeated often, e.g. Pol. 2 sg D, 261 A, 287 C; Soph. 265C-266B. Aristotle's criticism (partt. an. 624b s sq.) is premised not so much on this as on the fact that Plato sometimes says, what he may lave thought to be logically equivalent, that each class is divided in the middle (Pol. 262B, 26sA). These are not logically or even pragmatically equivalent for
1 30

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

or their class-namesthe ensemble of which, when representedschematically, depicts a descendingorder from that class havingthe widest extension to that with the narrowest extension., As in the preceding portion of this paper, the hierarchicalstructureis the most important defining characteristic. Once it is established the other characteristicscan be stated with a minimumof difficulty. Before I turn to these, however, a further remarkis called for. We have observed that the apex of the eros-hierarchyis the world of Forms. In a more complex fashion, dialectic too is based on the Theory of Forms. The Forms are operative in dialectic in a two-fold sense. First, dialectic is concerned, not with sensible particulars,but with the class-definingproperties in terms of which the Forms, present in the particulars,are the reason why the particularsare what they are. And because the Forms are present in rebuslanguage is made possible by which particularsare assignableto their respective classes. Plato frequently asserts that he is concerned with the referent of the class-term not its name.2 The reason for this is that dialecticaldivision may reveal classes and their inter-relations no trace of which exists in current linguistic usage.3 Secondly, the Forms are involved in dialectic in the sense that the divisionsby which one class is distinguishedfrom another are not random or arbitrary,as is the assignmentof names to a determinate class, but rather reflect the real divisionsor articulationsamong the Forms themselves (cf. n. X p. I 30). It is the latter, viz. the relations between the Forms,thatdeterminethe linguistically-sanctioned divisions and not vice-versa. Whether we turn to language,then, or to the real distinctions reflected byr language, the relation between the classes which is establishedby division is one that is essentially hierarchical. Terms vary in their extension because the classesof which they are the namesand to which they are isomorphicare themselveswider or narrower. At one point in the Sophistes it is establishedthat 'Being' is predicable of every Form. 'Sameness'and 'Difference' are also but only in the more
anyone who, like Aristotle, rejects the Theory of Forms. But as Cherniss has noted (ACPA 1, SS-S6)there are examplesof acapeaLq in Plato which are not dichotomous. Aristotle's criticism, then, is best directed at the employment of classes denoted by privativeterms the possibilityof which is never precludedby Plato's description of his method whether or not he actuallydid make use of such language. I Cf. CornfordPTK pp. I71, 172, I78, 324 for schemataof the 8XaykaetL in the Sophistes. 2 Soph. 2i8C; Phileb. 57B; Meno 87B.
3 Pol.

26oE; Soph. 229D.

' 3'

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

restricted sense that while the proposition 'each Form is the Same as itself and Different from every other Form' is true, it is true only on the assumption the proposition 'there are Forms' is true (cf. Soph. 2g4 D sq.). The relation between these two propositions is an instance of the hierarchicalstructure revealedby dialectic. The second characteristicof dialectic is its mediating role. This and the succeeding characteristic follow from the hierarchical structure embodied by the dialectical method. The importance of the present characteristicis that any sort of inquiry, if it is to be at all precise something for which Plato more than once praises dialectic I - is seriously handicappedif there is reason to believe that something has been overlooked. Thus the Strangerin the Politicuscriticizes the zeal of Young Socrates for proceeding in such a way that he makes a false division, thereby leaving out of account an important datum the inclusion of which alone guarantees the completeness of the dialectical definition of the statesman(Pol. 262 B). Stenzel, the eminent German Platoniker, has done good service by calling attention to the function of dialectic as mediatingthe worlds of doxa and logos. His argument is too complex to be stated here but, summarizedin the words of his British editor, it amounts-tothis: The method of ataCLpeCLg appears to indicate a rapprochement between the two worlds, not only in the sense that it shows the gradualapproachof the intelligible to the fringe of the perceptible, but also in the sense that both alike are analyzedand found to reveal
the same structure. The imposition of Form upon Matter, 7rspos

is a feature common to both worlds.2 upon the 0&CTCLpoV, After havingshown that there exists a close relation between the first two characteristicsof dialectic, the third follows from the first two taken together. The third characteristic, that of discrimination, has been alluded to above and is perhaps the most obvious feature of the dialectical method. It consists, essentially, in making a justifiable distinction between easily confused resemblances or similarities. This aspect of dialectic appearedalready in Socrates' attempt to distinguish between carnal or sensual eros and a philosophic eros.3 The contrast
I 2

Phaedr. 265 D; Phileb. 20C, 3iE, g8C. Stenzel-Allan, op. cit. p. xxxii. 3 Philosophic eros is the proper name for such activity because the Socratic method is predicated on the fact that successful emulation of a philosopher or a lover will amount to the same thing in practice. Compare Symp. 209 BC with Phaedr. 2 53 B. Robin writes,
1 32

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

between the art of dialectic and that of argumentation (&vrnXoyLx4) is stated by Plato to the effect that an art, to be worthy of its name, must be such that a man will be able to produce a resemblance between all things between which it can be produced, and to bring to light the resemblances produced and distinguished by anyone else (Phaedr. 26I E. trans. Loeb). There is a second sense in which the discriminating function of dialectic had its origin in the Socratic method. We have seen that Socrates' attempt to discriminate base eros from a noble eros was dictated by ethical considerations. In the same way, Plato's demand for precise discrimination is determined by his ethical convictions. He is concerned with distinguishing between what is real and what is apparent, between the true and the false, the model and the copy. I It was for this reason that the method of dialectic was employed in his attempt to settle, once and for all, the question of the relation between the sophist and the philosopher. The Sophistesis devoted to the thesis that so long as one is restricted to outward appearances alone there is no possibility of detecting the real and significant differences between the two competing In order to re-enforce the point, Plato offers not Lebensanschauungen. one but six definitions of the sophist, arrived at by collection and division and each of which emphasizes one side of the sophist's may-sided charlatanism and the totality of which leave no doubt that he differs toto genere from the philosopher. The fourth and final characteristic of dialectic - its synoptic and systematic role - parallels that of eros. The completion of a formally correct dialectical division enables the philosopher to speak of tle- essential nature of the class so defined and of its relation to all other classes contained in the same dialectical hierarchy. This he is entitled to do since one of the claims made for dialectic is its exhaustiveness. 2 On several occasions Plato makes the Stranger say that as a result of carefully applying the method, the entire class has been examined. 3 Dialectic is synoptic and systematic in this sense as well as in the sense that the location of any class, with respect to the other classes in the hierarchy,
"Lebut methodologiquede 1'educationscientifique ne differe pas en effet de celui que nous assignionstout i I'heure a l'interrogationou a I'amour"(Platon p. 84).
x Pol. 292D,
2 3

3ooDE; Phaedr. 262A; Soph. 253D.

Cf. ChernissACPA 1, 33 n. 26 to which may be addedSoph. 235 BC. E.g. PoI. 267C sq.; cf. Chernissloc. cit.
133

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions

is another way of saying that all the relations between the class in question and its neighbors is possible should dialectic be carried on indefinitely, i.e. each term appearingin the explication be itself subjected to a dialectical exploration. In the Phaedrus, Plato ruthlessly exposes the defects of rhetoric and eristic.1 Because such procedures - they cannot be called arts - are there is no assurancethat the results obtained by either will haphazard be correct.2 In order to compensate for this initial disadvantage, rhetoric and eristic rely on an irrationalmechanism,persuasion,by which they hope to accomplish their rather dubious goal, probability.3 Dialectic, on the other hand, since it aims at the truth, has no need for any such mechanismas eloquence or persuasion.Its motive force or that psychic mechanism by which conviction replaces doubt and truth replaces doxa is the ratioessendi of dialectic itself - reason.4 With the mention of the psychic mechanism by which dialectic is made so potent a tool of philosophic inquiry, I have come full circle in my argument, to wit that eros, as a psychic mechanismor motive force was the model for Plato's dialecticalmethod and that the role of eros in philosophic inquiry served as the impetus for Plato's examination of whether the art of philosophic inquiry can itself be subjected to philosophic inquiry.
I Rhetoric is criticized most severely in the Gorgias and the Phaedrus. In the former rhetoric is criticized more on moralistic grounds (cf. esp. 480 B sq.) whereas the latter undercuts on methodological lines the very possibility of a rhetoric (cf. esp. 2 59 E sq.). Actually the basis of Plato's censure is pretty much the same for the two dialogues in that rhetoric - whether it be Gorgian or Isocratean - is lacking in philosophic foundations. It is the latter which Plato is attempting to state in those dialogues where the central focus is on the method of dialectica,l division. 2 Dialectic as '&Xv* is favorably compared with rhetoric on the grounds that the latter possesses no rules and is consequently haphazard, a mere &trxvoq 'pLp0 (Phaedr. 26oE; cf. Gorg. 463 B; Legg. XI, 938 A). The type of speech resulting therefrom - eminently illustrated by Lysias' speech (Phaedr. 262 C) - iS lacking in structure and organization (Phaedr. 264C, cf. Hipp. I, 286A). 3 Persuasion is the essence of rhetoric according to Gorg. 4g4E. That it is historically attributed to Gorgias himself, cf. Gorg. A28 Vors. 5 (P1. Gorg. 4ssA) and Calogero in Ross Festschrift (J.H.S. LXXVII Part i) p. I 3. On probability as the aim of rhetoric cf. Gorg. 48 6 A (etx6 xal rL6mv6v) and Phaedr. 2 7 2 E sq. 4 Dialectic is essentially rational, primary emphasis being placed on conviction (cf. Symp. 2 X2 B, 2 I SD), certainty and truth of such a sort that &7Top(cL is removed (Phileb. 34D). The importance of dialectic is summed up thus: r& y&p xc&)XLarc 6vrno &Caw'[rTa, xOal jL ?6y9 06vov, &w 8i ou8evrlaacp6 8exvuroct Pol. 2 86 A, cf. Symp. 2 I9 A. a,

The Pennsylvania State University

' 34

This content downloaded from 200.26.133.57 on Fri, 19 Apr 2013 21:21:24 PM All use subject to JSTOR Terms and Conditions