Greek, Roman, and Byzantine Studies 45 (2005) 331–357 © 2005 GRBS Intra-Socratic Polemics: The Symposia of Plato and Xenophon Gabriel Danzig N THIS PAPER I review the relationship between the two Symposia of Plato and Xenophon, supporting H. Thesleff’s hypothesis that Xenophon wrote his Symposium first, and adducing new evidence for it. The more widely accepted position, that Plato wrote first, implies that Xenophon modeled Critobulus’ speech on the one speech in the Platonic corpus (Phaedrus’ speech) which he disliked so much that he attacked it openly, in chapter eight of his own Symposium. While one cannot rule out this possibility, no one has explained why Xenophon would have had such divergent responses to a single speech. The other possibility is that Xenophon wrote first and that Plato modeled Phaedrus’ speech on that of Critobulus. On this account, Phaedrus’ speech is attacked by Socrates precisely because it is modeled on the speech of Critobulus and contains an offensive and distorted version of Critobulus’ suggestion concerning the role of homosexual affection in the military. While Critobulus suggested that infatuation with the military leader might lead to greater achievements in battle, Phaedrus suggests that the soldiers engage in homosexual relations with each other. The desire to set the record straight would provide a plausible motive for Xenophon’s modification of his existing Symposium, by adding the eighth chapter with its attack on Plato and its emphasis on heterosexual relations. It would explain why Xenophon took the unusual step of breaking the dramatic unity of his composition by having Socrates attack a speech that occurs in the composition of another author. Evidence is presented for further rounds in the dispute. I 332 INTRA-SOCRATIC POLEMICS Ancient critics believed that Xenophon and Plato were en- gaged in some form of literary rivalry.1 Although they had no special information about it, and based their conclusions on the same texts that we have before us,2 being more familiar with the role publishing played in the polis than we are today, their judgements may have been in some ways sounder than ours. Personal invective was not uncommon in ancient letters, and in fact a whole literature of gossip and slander concerning the philosophers existed.3 Ancient critics believed that the philos- ophers themselves participated in this kind of activity. Ath- enaeus’ character Masurius says that “time would fail me if I were to be inclined to quote the attacks which philosophers have made on people.”4 He lists the open attacks of Aeschines Socraticus, Lysias, and Antisthenes on other persons. Isocrates, too, sometimes names his targets openly (e.g. Bus. 1–9). But other writers may have avoided naming names, possibly out of prudential considerations. In the case of Plato, it is particularly difficult to uncover the roles his writings might have played in contemporary controversies, since his dialogues were generally set in the fifth century, and the characters he names are for the most part dead. He rarely even mentions other students of Socrates who became his rivals in Socratic composition.5 This may have been part of an oblique strategy of attack, but it makes it especially difficult to pin down his targets. Not all ancient critics believed that Plato and Xenophon 1 See Ath. 504F–505A, Gell. NA 14.3.2–4, Diog. Laert. 3.34. 2 This is obvious from the evidence they cite. Gellius refers to the argu- ments as argumenta quaedam coniectaria ex eorum scriptorum (14.3.1) 3 See E. N. Tigerstedt, Interpreting Plato (Stockholm 1977) 36–40, for a brief review of the subject. 4 220F, transl. C. D. Yonge (London 1854). 5 Xenophon on the other hand does name some of the other Socratic writers, usually offering them uncomplimentary roles. His reticence on Plato may be a reaction to Plato’s reticence about him. It is remarkable that Xenophon figures in one of the few fragments of Aeschines that we possess (Cic. Inv.rhet. 1.51 = fr.70 Giannantoni), a fact which not only demonstrates Xenophon’s prominence in the Socratic circle, but also suggests that his practice of including rivals in his compositions may have been closer to the norm than Plato’s reticence. GABRIEL DANZIG 333 were engaged in personal rivalry. Gellius rejects this as some- thing beneath their dignity, and argues instead that they were involved in a rivalry in virtue (14.3.3–4). It is certainly true that neither Plato nor Xenophon attacks the other by name: Plato never mentions Xenophon, and Xenophon mentions Plato only once, in a non-hostile context (Mem. 3.6.1). But even Gel- lius acknowledges some form of rivalry between them, and he sees it partly in the fact that they virtually never mention each other’s names (14.3.2).6 But if the non-mention of a rival is a sign of rivalry, and not of genuine disinterest, it seems more a sign of personal rivalry than of rivalry in virtue. Modern scholars too have found signs of intertextuality in the writings of Plato and Xenophon. For the most part they believe that Xenophon is reacting to Plato. Indeed, in his Apology, Xenophon speaks of “others” who have written about Socrates’ behavior in court (Ap. 1). Clearly he is referring to someone, and it may be Plato. At Memorabilia 1.4.1 Xenophon refers to other portrtaits of Socrates which show him as capable of ex- horting others to virtue but incapable of bringing them to it. Scholars have suspected that this too is a reference to Plato.7 And there are numerous other places in Memorabilia and in his other writings where Xenophon seems to have a Platonic text in mind.8 But does Plato ever refer to Xenophon? The most likely allusion to Xenophon occurs in Laws (694C– 695B). The brief reference to Cyrus’ faulty education appears to be a rather rude dismissal of Xenophon’s masterpiece, Cyropaedia. In Laws as a whole Plato does not take seriously or respond to any of the ideas that Xenophon presents in Cyropaedia, and the relation between the works has been described as that of two ships passing in the night.9 If Laws is in part a response to Cyropaedia—and I emphasize the if—it 6 Diogenes Laertius (3.34) makes the same point. 7 See L.-A. Dorion, Xénophon Mémorabiles I (Paris 2000) CXXVI–CXLIV. 8 See A.-H. Chroust, Socrates, Man and Myth (South Bend 1957) 230 n.39; C. Kahn, Plato and the Socratic Dialogue (Cambridge 1996) 393–401; and M. Stokes’ comments in the introduction to his Plato: Apology (Warminster 1997) 7. 9 J. Tatum, Xenophon’s Imperial Fiction (Princeton 1989) 40. 334 INTRA-SOCRATIC POLEMICS responds only by ignoring all of the serious issues that Xeno- phon raises, and by proceeding in another direction entirely.10 The two Symposia provide another possible case of intra- Socratic disputation. Although no consensus has been reached concerning the order in which the two Symposia were com- posed, scholars do agree that they show signs of influence one way or the other. Numerous formal and linguistic parallels be- tween the two works have been discerned, not all of which can be attributed to the fact that they are written in a common literary genre.11 These parallels are not limited to personal attacks, but include literary and philosophic matters as well.12 My prime intention here is to investigate the possibility of a personal dispute between the two. I In the eighth chapter of Xenophon’s Symposium (8.32–34), Socrates appears to criticize comments made by Phaedrus (whom he mistakenly identifies as Pausanias) in Plato’s Sym- posium: 10 See G. Danzig, “Did Plato Read Xenophon’s Cyropaedia?” in S. Scolnicov and L. Brisson (eds.), Plato’s Laws: From Theory into Practice (Sankt Augustin 2003) 286–297. 11 See H. Thesleff, “The Interrelation and Date of the Symposia of Plato and Xenophon,” BICS 25 (1978) 157–170, and B. Huss’ excellent commen- tary, Xenophons Symposion: Ein Kommentar (Stuttgart/Leipzig 1999) 449–453, for detailed lists of the similarities. I would mention the following, some of which are not included in those lists. In both compositions Socrates has a propensity to make big scenes (Xen. 2.3–4, 2.16–19, 4.19, 6.9–10, 8.6; Pl. 198A–199B, 214D–E). In both, Socrates claims to have a special expertise in the area of love: in Xenophon it is pimping (3.10), in Plato he is simply an exert in eros (177D–E). In both, Socrates is accused (by Charmides or Alcibiades) of engaging in excessively warm relations with other young men, and in both cases the charge seems to stem from jealousy (213B–C ≈ 4.27, see also 8.4–5). In both, Socrates engages in dialectics where the others made speeches. Both compositions mention a hair (190E ≈ 6.2), and an award of ribbons (212E, 213E ≈ 4.19–20, 5.1–10). Note also the very inter- esting parallels Thesleff finds between the characters that appear in each composition. 12 As examples, Xenophon thinks virtue is teachable (Xen. Symp. 2.12), believes that akrasia is possible, and sees beauty as functionalism (Symp. 5.4– 7, cf. Mem. 3.8). GABRIEL DANZIG 335 But Pausanias, the lover of Agathon the poet, defended those who wallow together in licentiousness and said that an army composed of lovers and beloveds would be strongest. For he said that they would be ashamed to abandon each other in battle. But it would be quite extraordinary if those who are used to pay- ing no attention to censure and to having no sense of shame before each other should nevertheless be ashamed to perform a shameful action. As proof he brought the example of the Thebans and the Eleans who are experienced with such things, and he claimed that even though they sleep with their beloveds, they still set them together in their ranks for battle. But there is no proof from this, for the situation is not similar: for them this practice is acceptable, but for us it is exceedingly shameful. Some scholars have argued that Xenophon refers here not to Plato but to another work on the subject that served also as the basis of Plato’s Symposium. K. Joel argued that this was An- tisthenes’ Per‹ dikaiosÊnhw ka‹ éndre¤aw.13 But this is highly speculative, and Thesleff argues reasonably that any common source must have been a “dialogic symposium,” which is not what Antisthenes’ work sounds like.14 Since Xenophon refers to the words of Pausanias, it has been thought that both Plato and Xenophon base themselves on some lost work by Pausanias.15 But as Dover has shown, chronological considerations make it extremely unlikely that the hypothetical source was a work by Pausanias.16 Athenaeus (216F) knew of no work by Pausanias, and assumed that the reference was to Plato. The discrepancies between Xenophon’s reference and the actual contents of Pausanias’ speech in Plato do not show that he is referring to another speech, but that precision was not expected in such matters.17 Dover and Cooper have shown convincingly that 13 K. Joël, Der echte und der xenophontische Sokrates II (Berlin 1901) 912. 14 See Thesleff, BICS 25 (1978) 157–158. He also notes that no other So- cratic symposium was known in antiquity. 15 See A. Hug, “Über das gegenseitige Verhältnis der Symposien des Xenophon und Plato,” Philologus 7 (1852) 638–695. 16 K. Dover, “The Date of Plato’s Symposium,” Phronesis 10 (1965) 1–20. 17 Michael Stokes has suggested to me that this might be an example of Xenophon’s deliberately dismissive way of referring to Plato, who himself was capable of such tactics. It may also serve as a hint to readers that 336 INTRA-SOCRATIC POLEMICS Plato’s Symposium is the target of Xenophon’s Socrates’ words here.18 But if this is a reference to Plato it is a singularly anomolous one. Nowhere else in his corpus does Xenophon single out another author for direct criticism in this way. What was so offensive to him about Phaedrus’ speech? No one criticizes everything he disagrees with, and the idea which Xenophon criticizes here is not so far from an idea found in a speech by Critobulus in Xenophon’s own Symposium. Critobulus suggests that the good-looking be chosen as military commanders, since this will inspire love in the troops. And yet rather than have Socrates criticize this, Xenophon takes the unusual step of having him criticize a speech that was not made anywhere in his own Symposium, but in Plato’s, and in doing so destroys the dramatic illusion of his own work. This suggests that there is something more at work than divergent opinions. I will argue that Xenophon is responding to Plato in this way because he viewed Phaedrus’ speech as a response to and distortion of the speech of his own Critobulus. On the surface this seems implausible. How can Xenophon and Plato each be responding to the other? Someone must have written first. Since Xenophon clearly attacks Plato, under- standably most scholars believe it to have been Plato.19 But the case is not quite closed. Thesleff has argued that the eighth chapter, in which these references occur, was added to the Symposium by Xenophon at a later date, and I will offer some additional arguments for this conclusion. As long as this is a reasonable possibility we cannot use these references to deter- mine the relative dates of the two Symposia as a whole. Other arguments that have been presented are not more ___ Xenophon is in fact referring to Plato’s Symposium where both of these speakers appear. But it may simply be the kind of careless mistake that arises when one quotes from memory. On the imprecision of Socratic writers in referring to each other, see Danzig, in Scolnicov/Brisson 286– 297. 18 Dover, Phronesis 10 (1965) 11–13; J. Cooper, “Notes on Xenophon’s Socrates,” in Reason and Emotion (Princeton 1999) 16–21. 19 See Huss, Xenophons Symposion 13–18, for references to the history of this debate. GABRIEL DANZIG 337 decisive. The one point on which there is agreement is that there is some direct literary relationship between the two com- positions—one author has read and reacted to the other. But the arguments for the priority of one or the other remain speculative. Dover argues for the priority of Plato by observing the different ways in which both compositions refer (or do not refer) to the Thebans’ Sacred Band.20 As he points out, the re- mark of Plato’s Phaedrus seems to imply that such an organiza- tion does not yet exist (“If a means could be found,” 178E), whereas Xenophon’s Socrates clearly refers to the Band as an existing institution (8.34). For this reason, Dover would date Plato’s Symposium prior to its formation in 378, and Xenophon’s Symposium later. But the argument is not tenable. As Dover acknowledges, “Plato could of course have put into the mouth of Phaidros a hypothesis the testing of which had become a practical possibil- ity between the dramatic occasion and the time of writing.”21 He rejects this possibility on the grounds that Plato does not usually aim for historical accuracy in his writings. But that does not mean that Plato could never take chronology into con- sideration. Despite his tolerance of anachronism, Plato rarely allows Socrates or his interlocutors to discuss matters that occurred long after Socrates’ death, as would be the case if they discussed the Sacred Band. The chronological difficulty prob- ably does explain Plato’s reluctance to allow Phaedrus to refer to it as already existing at the dramatic date of the Symposium.22 Rather than proof that Plato was unaware of its existence at the time of writing, this reference is evidence that he was quite aware of it, and that his Symposium was written after 378.23 20 Dover, Phronesis 10 (1965) 1–20. See also H. B. Mattingly, “The Date of Plato’s Symposium,” Phronesis 3 (1958) 31–39; J. S. Morrison, “Four Notes on the Symposium,” CQ 14 (1964) 42–55, at 42–46. 21 Phronesis 10 (1965) 15. 22 Xenophon on the other hand is not bothered by having his Socrates do this. He also has his Socrates refer to the battle of Cunaxa in Oec. 4.18– 19. In Mem. 3.4.4, by contrast, Socrates refers to the ascendancy of the Thebans over Athens by referring only to fifth-century victories. 23 In fact it would be dated after 371 if Symp. 178E–179A refers to the defeat of Sparta at Leuctra, as Gilbert Ryle has argued (Plato’s Progress 338 INTRA-SOCRATIC POLEMICS Huss offers a different kind of argument: Xenophon’s Sym- posium must post-date Plato’s, and hence be dependent on it, because it contains references to Plato’s Symposium, Phaedrus, and Theaetetus. As we have said, the Symposium references are not conclusive. But what of the references to the Phaedrus and the Theaetetus? One objection is that we do not know the relative dates of Phaedrus, Theaetetus, and Symposium. Huss has to assume that Phaedrus and Theaetetus post-date Symposium, which may not be true. But even if we grant this assumption, the argument still will not hold. Remarkably, all the references to the Phaedrus, like the two clear references to Plato’s Symposium, are found in chapter eight of Xenophon’s work. Not only does this mean that these references provide no evidence for the posteriority of Xeno- phon’s Symposium as a whole, it actually provides further sup- porting evidence for Thesleff’s theory of the later addition of the eighth chapter. If Xenophon had read these works before writing his own, and wished to respond to them, it is hard to understand why he would have confined his references to a single chapter. It is of course true that the eighth chapter as we have it is the only chapter that deals with the theme of choos- ing a lover, and hence the only one in which a reference to Phaedrus is appropriate. But that is begging the question. If Xenophon had read Symposium and Phaedrus and was inspired by it to compose something in response, why would he limit his response to this one section of his work? But what of the references to Theaetetus? These occur in the third and fourth chapters of Xenophon’s Symposium, and no one would argue that these sections were added later. The problem here is that these references are far less convincing than the others.24 The similiarity is that both texts discuss Socrates’ abilities at matchmaking, procuring, or pimping, contain the term procurer (proagoges) and mention Prodicus. But these ___ [Cambridge 1966] 38). Even if Dover’s argument were right, it would only imply that the eighth chapter of Xenophon’s Symposium, where Xenophon’s Socrates refers to the Sacred Band, is later than Plato’s Symposium. 24 See Ole Thomsen, “Socrates and Love,” ClMed 52 (2001) 117–178, at 135, for further doubts about the relationship between the passages. GABRIEL DANZIG 339 broad similarities may be attributable to the (hypothetical) fact that the historical Socrates did actually describe himself as a match-maker or midwife (possibly hinted at in Clouds 135–137), or that the theme was a common one in Socratic circles. And if there is influence here, it is more likely in the other direction. While the Platonic passage is relatively unusual in the Platonic corpus, the Xenophontic passage resembles other passages, such as Memorabilia 2.6 where Socrates refers to him- self as erotikos and explains that he is an expert in arranging matches. In Xenophon’s Symposium, Socrates refers to himself as a pimp, and explains that he teaches others how to win grace in the eyes of others, and in the sight of the city (Symp. 3.10, 4.56–60). This is what Xenophon’s Socrates does in Books 2 and 3 of Memorabilia. So the Xenophontic passage seems to be characteristic of Xenophon, while the Platonic pas- sage is somewhat anomalous. Moreover, the character of the two passages offers some ad- ditional support for Xenophontic priority. While Xenophon’s Socrates claims to be a pimp, Plato’s Socrates attributes to him- self the virtues of midwives, including their ability to arrange matches that produce good offspring. While it is hard to imagine that Xenophon’s far more original, surprising, offen- sive, and humorous comment on Socratic pimping was a reaction to Plato’s serious statement concerning midwifery, it is easy to explain the influence in the other direction. Plato does not use a derogatory term, but compares Socrates only to a midwife or match-maker. The term procurer is applied by him not to the Socratic art, but to a disreputable pseudo-art-form akin to match-making. It is easier to imagine Plato “cleaning up” Xenophon’s coarse language, and implicitly disparaging Xenophon, than Xenophon corrupting Plato’s. Why would Xenophon want to attribute Plato’s disreputable pseudo-art of procuring to his beloved Socrates? While the use of the terms “procurer” and “pimp” seem like a humorous provocation when considered as an original witticism, they lose some of their charm if Plato had already referred to Socrates as a mid- wife or match-maker and distinguished him explicitly from a procurer. If Xenophon did want to adopt and rehabilitate this title and apply it to Socrates, after Plato had used it deroga- torily, one would think that he would make clear why he does 340 INTRA-SOCRATIC POLEMICS so. In short, it is far from certain that Xenophon has read Theaetetus and is reacting to it, and it would be perilous to de- termine the relative dates of the two compositions on this assumption. Other efforts to deny Xenophon’s precedence on the basis of general considerations have not been more persuasive. The suggestion that Xenophon could not have written first, since he was too mediocre a writer to have invented the genre of the Socratic Symposium,25 is highly speculative: whatever his talents as a writer, Xenophon surely was an innovator when it comes to literary genre. His writings are the most diverse of any Greek writer that we know. In addition to symposiastic literature, he wrote both Socratic and non-Socratic philosophical dialogue, history, historical novel, short essay, encomiastic-biographical essay, and of course he included oratorical compositions in many of his works. Plato on the other hand stuck almost exclusively to philosophical dialogue and oratory. His most original piece of writing from a formal point of view is the semi-autobiographical seventh letter, if he wrote it.26 There is, however, some sense in one of the arguments for Xenophontic priority. Some scholars have argued that Xeno- phon’s Symposium must be prior because it is the simpler of the two.27 Of course, the simpler is not necessarily the earlier. But Xenophon’s Symposium is not merely simpler, it is also more life- like and lacks some of the literary polish we find in Plato. For example, all the preceding speeches in Plato, while interesting and serious in their own right, are also designed to provide material for Socrates’ subsequent speech. It hardly seems life- like to have Socrates make a unified speech which incorporates 25 See Huss’ brief discussion, Xenophons Symposion 14–15. 26 See Momigliano’s comment in The Development of Greek Biography (Cam- bridge [Mass.] 1971) 61: “I am reluctant to admit that forgery preceded reality in the matter of autobiographical letters. The letter seems to me an exceptional creation by an exceptional man, namely Plato.” L. Edelstein’s arguments against Platonic authorship are based on the questionable as- sumption that if the letter is authentic it must be consistent with Plato’s teachings in the dialogues (Plato’s Seventh Letter [Leiden 1966]). 27 See references in Huss, Xenophons Symposion 14. GABRIEL DANZIG 341 responses to each and every one of the previous speeches.28 This feature gives Plato’s Symposium a powerful literary unity that is, frankly, lacking in Xenophon’s.29 Xenophon’s Symposium has the virtue and vice of creating the impression of a spon- taneous conversation. This kind of effect seems to be the result of reflecting on living examples of symposia, not on literary compositions such as Plato’s. But this argument too is highly speculative: it is surely possible that after reading Plato’s highly- wrought composition, Xenophon decided to present a more realistic picture. In order to resolve the question, we will need to find more compelling arguments. II One further reason why scholars are reluctant to grant Xenophontic priority is that, as we have seen, there is good reason to believe that in other cases, particularly in Apology and Memorabilia, Xenophon is reacting to Plato. Why should the Symposia be any different? But there is a difference. While in both Apology and Memorabilia Xenophon makes it clear near the opening of the work that he is engaged in a dialogue with other authors (Ap. 1, Mem. 1.4.1), there is nothing of the sort in the opening of Xenophon’s Symposium. Moreover, the situation is actually reversed, for in the opening of Plato’s Symposium there is a reference to another version of the story told by one Phoenix. I will discuss the identity of this Phoenix below. For now, the presence of this reference suggests that if anyone is responding, it is Plato. A closer look at the two Symposia re- inforces this conclusion. Apart from the eighth chapter of Xenophon’s Symposium, the most demonstrable example of influence is found in the parallel speeches of Critobulus and Phaedrus. Huss notes five parallels between them, the most of any section of comparable length outside of chapter eight.30 If anything in the two Symposia shows 28 See R. G. Bury, The Symposium of Plato (Cambridge 1932) lvii–lx. 29 See G. Danzig, “Apologetic Elements in Xenophon’s Symposium,” ClMed 55 (2005) 17–48. Huss argues for a higher degree of literary unity (Xenophons Symposion 30–37). 30 See his appendix (a), Xenophons Symposion 450. 342 INTRA-SOCRATIC POLEMICS signs of influence, one or the other of these two speeches does. If we can determine which influenced the other we will have solved in principle the question of the relation between the two works. The similarities are both broad and detailed. The characters Phaedrus and Critobulus are both good-looking young men who enjoy the attentions of older men. Both are enthusiastic about love-relationships: Phaedrus suggests the topic of eros for the evening’s speeches in Plato’s, and Critobulus is the only one in Xenophon’s Symposium who makes love (or beauty) the topic of his speech—and his professed enthusiasm for his beloved Cleinias knows no bounds. This is the only pair of speeches from the two Symposia that have the same theme as their sub- ject. In fact, Critobulus’ speech is the only speech in the first seven chapters of Xenophon’s Symposium that concerns the subject of love or eros, the subject of all the speeches in Plato’s Symposium. Unlike the speeches in Plato, those in Xenophon aim at boasting about a different personal attribute. The asym- metry of this parallelism suggests that if Critobulus served as an inspiration for Plato’s Phaedrus, he may also have served as the inspiration for his choice of eros as the theme for his Symposium as a whole. The central aim of both speeches is to show that love (or beauty) inspires lovers (in Phaedrus’ case also beloveds) to act virtuously.31 Critobulus claims that love inspires lovers to perform acts of virtue, including displaying bravery on the field of battle. Among other things, he says that the beautiful inspire their lovers to be filoponvt°rouw d¢ ka‹ filokalvt°rouw §n to›w kindÊnoiw (4.15), a clear reference to heroic behavior on the field of battle. He adds that it is insane not to choose a beautiful man as general, since such a leader would inspire the troops to follow him through danger, as Critobulus claims he would follow his beloved Cleinias through fire (4.16). The idea that love can inspire successful war-making is the main theme of Phaedrus’ entire speech. Phaedrus’ claim at 31 This was a theme of some of Aeschines’ writings. See C. Kahn, “Aes- chines on Socratic Eros,” in P. Vander Waerdt (ed.), The Socratic Movement (Ithaca 1994) 87–106. GABRIEL DANZIG 343 179A that it is inconceivable that a lover would fail to come to his beloved’s rescue resembles Critobulus’ claim that he would go through fire with Cleinias. But Phaedrus’ speech expands the theme drastically. In terms of sheer length, the discussion of the effects of love in encouraging a willingness to die in battle or elsewhere runs from 178E to 180B. Most of Phaedrus’ speech is divided between the military theme (178E–179B) and the closely related theme of the willingness to die for love (179B–180B). For this reason, it is easy to see this speech as an expansion of one element of Critobulus’ speech. And there is little else in it. The only element in Phaedrus’ speech which does not seem directly related to Critobulus’ speech is the ac- count of the genesis of eros (178A–C). This seems designed primarily for the sake of Socrates’ later contrast with his own account of the genesis of eros (203A–C). In several ways, Phaedrus’ speech goes further than Crito- bulus’, and this too suggests posteriority. Where Critobulus insists that a handsome general is essential for inspiring the troops, Phaedrus makes the extreme suggestion that the entire army be composed of lovers and beloveds (178E–179B). While Phaedrus’ argument is provocative even after we have read Critobulus’, the reverse is not the case. It is hard to see how Xenophon could have thought that Critobulus’ speech would make much of an impression on an audience which has read its Plato. In some places Phaedrus’ speech seems like a summary of Critobulus’ speech (178C–D): ˘ går xrØ ényr≈poiw ≤ge›syai pantÚw toË b¤ou to›w m°llousi kal«w bi≈sesyai, toËto oÎte sugg°neia o·a te §mpoie›n oÏtv kal«w oÎte tima‹ oÎte ploËtow oÎt' êllo oÈd¢n …w ¶rvw. This contrast between love and a list of other qualities recalls Critobulus’ claim that love is more effective than wealth in in- spiring virtuous behavior in others. The remark has a much stronger motive in Xenophon, where it is a response to Callias, than it has here, where it is gratuitous. Phaedrus goes one bet- ter than Critobulus by insisting on the superiority of love to so many other qualities. Phaedrus also claims that love inspires men to feel tØn §p‹ m¢n to›w afisxro›w afisxÊnhn, §p‹ d¢ to›w 344 INTRA-SOCRATIC POLEMICS kalo›w filotim¤an (178D), which may be an attempt to sum- marize what Critobulus says in 4.15–16. While it is possible to explain almost all of Phaedrus’ speech as a reaction to Critobulus’ speech, it is much more difficult to see Critobulus’ speech as simply an expansion or summary of Phaedrus’. Critobulus’ speech contains many elements that are not alluded to in any way in Phaedrus’ speech. According to Critobulus, the willingness to face death is only one of the effects of love: it also inspires financial generosity, for example (4.15). Most of Critobulus’ speech is actually a hymn of praise to his beloved Cleinias. He claims that he would rather be blind to the rest of the world than to Cleinias alone; that he would rather be a slave to Cleinias than a free man without him; that he would rather give his money to Cleinias than gain it from another source. Critobulus mentions these personal feelings in order to show how strongly the others must feel for him. He argues that beauty is long-lasting, and is more effective than wisdom in winning the kisses of beautiful young persons. None of this has any close relation to anything in Plato’s Sym- posium and certainly not in Phaedrus’ speech. It is surely not necessary that a speech inspired by another speech appear to be an expansion or summary of the other, or that it go one better than the other in several ways. But if it does, this suggests influence. III These considerations are still somewhat subjective in char- acter. There is however one other consideration that points decisively in the direction of Xenophontic priority. As we have noted, Phaedrus’ speech is the only item in Plato’s Symposium (or anywhere else in Plato’s work) that Xenophon singles out for direct, harsh criticism, and Xenophon breaks the dramatic illusion to do this. At the same time, as we have seen, it is the only item in Plato’s Symposium which has a clear parallel in Xenophon’s own Symposium. But if Xenophon finds the idea that love can improve the soldierly virtues so objectionable that he attacks it so directly, then why would he himself, after read- ing this offensive speech, write a magnificent speech on this GABRIEL DANZIG 345 same theme and place it in the mouth of his own Critobulus?32 It hardly seems conceivable that Xenophon would admire and imitate the same speech that he hates and criticizes. We may be able to reinforce this conclusion by focusing on the difference between the two speeches. Critobulus’ speech differs in a subtle but important way from Phaedrus’ and is not open to the objection that Xenophon’s Socrates levels at Phaedrus. While Phaedrus suggested an army composed of lovers, Critobulus’ aim is to arouse love for the commander, so that they would follow him through fire, not to encourage coupling among the troops. Love of the commander will surely not include sexual relations with him. All Critobulus is saying is that it would be advantageous for the soldiers to be infatuated with him.33 There is a large difference between infatuation and sexual relations. Plato makes this distinction clear in Phaedrus (252e– 256E)34 and in Republic: “Then this pleasure [viz. sexual pleasure] must not come near love, and lover and boy who love and are loved in the right way must not be partner to it?” “No by Zeus, Socrates,” he said, “this pleasure certainly must not come near love.” “So then, as it seems, you’ll set down a law in the city that’s being founded, that a lover may kiss, be with, and touch his boy as though he were a son, for fair purposes, if he persuades him; but, as for the 32 It is of course possible that Xenophon disagrees with his Critobulus. In other places in Xenophon, Critobulus is presented as a very imperfect human being who needs correcting (Oec. 1–6, Mem. 2.6). But that does not seem to be the point here as Socrates does not criticize Critobulus’ words. In some ways Xenophon seems to identify with Critobulus (see Mem. 1.3). Some have suggested in fact that Critobulus represents Xenophon (Huss, Xenophons Symposion 68). That Socrates criticizes Critobulus so frequently is no objection, since he also criticizes Xenophon on every occasion where the two appear. This identification of Xenophon and Critobulus gains some support from Aristippus’ statement (at Diog. Laert. 2.48–49) that Xenophon was in love with Cleinias, a conclusion evidently derived from Critobulus’ speech in Symposium. 33 One may compare Pericles’ famous call on the citizens of Athens to regard themselves as lovers (§rasta¤) of the city (Thuc. 2.43). 34 Cooper points this out: Reason 19. 346 INTRA-SOCRATIC POLEMICS rest, his intercourse with the one for whom he cares will be such that their relationship will never be reputed to go further than this.”35 And this distinction is present implicitly in much of Xenophon. In Cyropaedia, for example, Cyrus capitalizes on his own good looks to inspire one soldier to perform valuable services, with- out of course engaging in anything suggesting sexual rela- tions.36 Critobulus is making the same suggestion; Phaedrus’ suggestion, on the other hand, transgresses these crucial boundaries.37 If conceived as a response to Phaedrus, Critobulus’ speech is a “cleaned-up” version. But unlike Socrates, whose whole speech aims to persuade Callias not to allow love to lead to sexual relations, Critobulus does not make this point clear. On the contrary, his speech about the value of good looks for com- manding officers could easily lead to the mistaken impression that he is in agreement with Phaedrus. This is difficult to ex- plain if Xenophon was composing after Plato with the aim of criticizing the speech of Phaedrus. If he has already read and come to dislike Phaedrus’ speech, why does he omit any hints of his criticism of Phaedrus here, leaving the reader to imagine that he agrees with him? In short, not only are there no signs that Xenophon has Phaedrus in mind when composing the speech of Critobulus, but, given Xenophon’s dislike of Phae- drus’ speech, the absence of such signs is a sign that he has not read it at all. 35 403B–C, modified from A. Bloom’s translation. Note Thomsen’s re- marks on the heavy eroticism present in this passage despite its insistence on no physical contact: ClMed 52 (2001) 161. 36 Cyr. 1.4.27–28, 4.1.22–23, 7.1.38. Cf. An. 7.4.7–11, Cyn, 12.20. At Cyr. 7.1.30 Xenophon speaks clearly of an army of friends (§k f¤lvn). 37 Elsewhere Xenophon makes statements which, though not as explicit as Phaedrus’ statement, nevertheless present a more indulgent attitude towards those who engaged in homosexual relationships. Socrates’ more severe attitude may aim to dispel accusations of misbehavior directed against him. But the references to Plato show that Xenophon is not merely concerned to preserve Socrates’ reputation from popular charges, but also to counter Plato’s account of Socrates. Contrast Cooper, Reason 17–20. GABRIEL DANZIG 347 Re-writing But if Xenophon wrote first, then it necessarily follows that chapter eight, and possibly nine as well, are later additions, as Thesleff argued. Scholars have not accepted this position partly because of a general skepticism about the prevalence of re- writing in the ancient world, and partly because of worries about the subjective application of such hypotheses to the in- terpretation of texts. These are long questions, and I will not be able to address them fully in this context. But some comments may be useful. We do know that re-writing occurred: Aristophanes, for example, re-wrote his Clouds after its initial performance.38 And rumours existed that Plato revised one or another of his dialogues.39 Even if these particular rumours are false, they provide evidence that re-writing was a well-known practice. Although we do not have textual traditions reflecting an earlier edition of Xenophon’s Symposium, that is also the case for Ari- stophanes’ Clouds. The absense of such evidence is not decisive, but it does mean that our argument remains uncorroborated. Another reason scholars have been reluctant to accept Thes- leff’s hypothesis is that some of his arguments seem subjective in character,40 and that he did not offer a compelling explana- tion for why Xenophon would have been motivated to re-write the ending of his composition.41 It seems to me that we can now offer an explanation. Socrates’ great speech in chapter eight makes clear that he did not condone sexual relations with young boys. Chapter nine contains a heterosexual skit, which 38 See K. J. Dover, Aristophanes Clouds (Oxford 1968) lxxx–xcviii. 39 Thesleff offers a discussion of the question and bibliography: Studies in Platonic Chronology (Helsinki 1982) 83–85. See also his “The Early Version of Plato’s Republic,” Arctos 31 (1997) 149–174; H. Tarrant, “The Composition of Plato’s Gorgias,” Prudentia 14 (1982) 3–22; D. Nails, Agora, Academy, and the Conduct of Philosophy (Dortrecht/Boston 1995) 116–122. 40 But many of them are strong. And if the hypothesis is right, then some of his more speculative suggestions, such as those on the ways in which Plato may have transformed episodes and characters in Xenophon, take on a special interest, not as evidence of dependence, but as a study of Plato’s methods of writing. 41 See his comments at BICS 25 (1978) 168. 348 INTRA-SOCRATIC POLEMICS culminates in the married guests riding home to be with their wives, while the unmarried ones vow to marry. These features suggest that if these sections were added, they were added in order to deflect criticism that Socrates encouraged homosexual activity. This could be explained as a reaction to criticisms of Xenophon’s own previously published Symposium. But then why the attack on Plato? The references to Plato show that Xeno- phon’s re-write was motivated by Plato’s publication. But did Xenophon feel compelled to react to every misrepresentation of his master that came to his attention?42 It is worth recalling that Plato’s Symposium is not especially offensive. In it Socrates offers reservations about homosexual relations, putting the love of bodies on a lower level than the love of souls, and Alcibiades testifies that this was no pose. So it is hard to see why the pub- lication of Plato’s Symposium would force Xenophon to react so critically. Nor would it explain why he put those criticisms in the penultimate chapter of an already existing work. But if Plato’s Symposium were itself a reaction to Xenophon’s all of this would make far more sense. After seeing an im- portant theme which he raised in his own Symposium distorted and caricatured by Plato, Xenophon would have wanted to set matters straight. He may have believed that Plato’s speech of Phaedrus was a caricature of his own speech of Critobulus. This would help explain the fact that this is the only place in Xenophon’s entire corpus where he responds with direct and harsh criticism to something in Plato. Xenophon may have felt compelled to respond directly to Plato precisely because Plato was responding, less directly, to him. I suggest then that Xenophon wrote a Symposium which contained the first seven chapters of the present Symposium, with another ending, possibly including an erotic dancing act (see Socrates’ suggestion at 7.5). After reading this, Plato decided to 42 Dorion’s argument (Xénophon CXXVI–CXLIV) that Memorabilia 1.4.1 contains an indirect attack on Plato’s portrait of Socrates as exclusively a practitioner of the elenchus implies that Xenophon was concerned to cor- rect mistaken portraits of Socrates. (It also suggests that Memorabilia precedes the non-aporetic dialogues, and may have had a hand in inspiring Plato’s adoption of a non-aporetic approach.) But the unusually direct manner in which he does this in Symposium suggests an additional motivation. GABRIEL DANZIG 349 improve on it by writing a Symposium devoted completely to the subject of eros. He may have done so because of what he perceived as Xenophon’s coarse and inaccurate treatment of Socratic eros in connection with the pimping remark, a remark which may have inspired some passages in Theaetetus as well. He was especially inspired by the very impressive speech of Critobulus on love, and modeled one of his speeches on it, also taking its theme as the theme of his own Symposium. But he turned it into a caricature of Critobulus’ speech by advocating homosexual relations among the soldiers. This distortion was offensive to Xenophon, and in reaction he re-wrote the ending of his Symposium, turning it into a critique of homosexual relations, including a scene that encourages heterosexuality, and singling out Plato’s work for criticism.43 IV How deep a re-writing should we imagine? It is of course possible that the original ending contained almost everything in the present version, except for the references to Plato and their immediate context. At the other extreme, it is possible that almost everything in the chapter as we have it is new. Thesleff 43 There are signs that Xenophon read at least the first book of the Republic before writing his Symposium. Note these similarities between the dramatic settings of the two works: In both, Socrates and his companion(s) are politely forced to join a celebration that they do not really want to participate in. In both, Socrates takes complete control of the party which has assembled at the home of a rich man. In both cases, the wealthy host speaks about his wealth, and explains the true nature of the benefit he derives from it. And in both, he claims that the primary benefit of wealth is moral, rather than material. Cephalus claims that wealth is useful because it enables one to be just: to pay one’s debts to men and gods, and to die with a good conscience. Callias claims it enables him to make others just. Callias makes explicit what is only implicit in Cephalus’ words: that while the others may speak beautifully about justice, he himself actually performs acts of justice. Cephalus lives up to Callias’ boast better than Callias does: he excuses himself from the discussion and goes out to offer sacrifice. Callias remains to enjoy Socrates’ speeches. In addition, Niceratus son of Nicias is mentioned in both (327C ≈ 1.3), as is a horse-race (328A ≈ 1.2). If Plato’s Symposium was written after Xenophon’s, and if Xenophon’s Symposium was written after Republic Book 1, then Republic one must have been written before Plato’s Symposium. 350 INTRA-SOCRATIC POLEMICS posits a maximalist view, arguing that chapter eight as a whole interposes itself unnecessarily between Socrates’ suggestion for a dance-performance and the Syracusan’s actual presentation of a skit. Moreover, he notes that the fact that we have a skit rather than a dance also suggests that an alteration has been made in the final chapter. There are other signs that these chapters are late. The whole character of Socrates’ speech makes it seem inappropriate in length, tone, content, and vocabulary to the rest of the compo- sition, and its unusual features point in the direction of Plato. The speech is far longer than anything else in Xenophon’s Symposium, more like the lengthy speeches in Plato’s Symposium. It is the only speech in Xenophon’s Symposium (with the partial exception of the speech of Hermogenes) that is not intended to be funny, and Socrates apologizes for that (8.41). In Plato’s Symposium, on the other hand, even Aristophanes’ speech has its serious moments. As Higgins notes, Socrates’ speech is the only one which contains a lengthy discussion of the god of love.44 Eros was the central theme of Plato’s Symposium, but was not discussed earnestly in the earlier part of Xenophon’s Symposium. After seeing Plato’s Symposium, and deciding to re-write, un- derstandably Xenophon would have wanted to say something serious on the subject. The point his Socrates makes, that love of the soul is superior to love of the body, has of course par- allels in Plato’s Symposium, especially in the speech of Pausanias, to which he refers, and in that of Plato’s Socrates. The unusual vocabulary of this section has been noted by Bowen,45 and in some cases it seems borrowed from Plato. All of this suggests that the eighth chapter as a whole is an addition, composed in reaction to Plato’s Symposium. As Huss’ citations show,46 chapter eight contains far more parallels to Plato than any comparable section of the work. There are ways in which this chapter is connected to the re- mainder of the work; but some of these only serve to reinforce 44 W. Higgins, Xenophon the Athenian (Albany 1977) 17. 45 A. J. Bowen, Xenophon: Symposium (Warminster 1998) 118. 46 Xenophons Symposion Appendix (a) and (b) 449–455; see also Thesleff, BICS 25 (1978) 164–167. GABRIEL DANZIG 351 the impression that it is an addition influenced by Plato. In the opening of the chapter Socrates comments on several of the previous speeches (8.2–8). This is a highly artificial way of tying the speech to the composition as a whole, and it is exactly the sort of thing that Xenophon might have done if he did add the chapter later.47 In fact, the technique seems borrowed from Plato’s Symposium: Plato’s Socrates also refers—implicitly at least—to the speeches of everyone else.48 But Xenophon has done a much clumsier job of it. Whereas Plato’s Socrates develops a thesis which naturally responds to the earlier speeches, and thus serves as a climax to the entire composition, Xenophon’s Socrates simply refers to some of the previous speakers, as though his sole purpose were to tie this chapter into the preceding text. He refers to Charmides, Critobulus, Niceratus, Hermogenes, Antisthenes, Callias, and Autolycus within the first seven sections of the chapter. The clumsiness may arise from the fact that this was not Xenophon’s original plan. And Xenophon goes further then this, as we have seen, even aiming to tie his speech in with the speeches in Plato’s Symposium!49 This reconstruction implies that Xenophon’s Symposium as it stands is not a perfect unity. Although this deficiency has been acknowledged in the past, it has been challenged recently by Huss with some interesting and powerful arguments.50 He argues that the eighth chapter develops several themes that were announced earlier but not fully treated. It deals with the relationship between Autolycus and Callias; it treats the themes of eros and gentlemanliness; and it offers a spiritual contrast to the praise of physical beauty which we find in the rest of the work. Some of the arguments that Huss offers on the unity of 47 See Higgins’ comments, Xenophon 18. 48 See Bury’s references, Symposium lvii–lx. 49 Note in particular Socrates’ reference to eros as a daimon who is as old as the eternal gods and still the youngest of them in appearance (8.1). This seems to combine a reference to the speech of Phaedrus (who claimed that eros was the oldest god, 178A–C) with a reference to the speech of Agathon (who claimed that he was the youngest, 195B–C). 50 Xenophons Symposion 12–25, 30–37. 352 INTRA-SOCRATIC POLEMICS the work imply not only that chapter eight fits in, but also that the earlier parts of the work were composed with an eye to the eighth chapter as we have it, and hence that this chapter must have been part of the original Xenophontic Symposium. He argues that Critobulus’ speech as a whole, with its emphasis on love of physical beauty, looks forward to Socrates’ later speech on love of the soul, and that by criticizing Critobulus—Are you going to say that because of your beauty you are able to make us better? (3.7)—Socrates is pointing towards his own cor- rective speech in chapter eight. Huss points out that Socrates’ speech answers specific claims made by Critobulus. He also argues that by handing off his art of pimping to Antisthenes Socrates implies the need for an additional account of his own erotic skill, which is what he offers in chapter eight. And he argues that Socrates’ defeat in the beauty contest in chapter five shows the insufficiency of an account of eros that is based on beauty, and therefore also looks forward to Socrates’ speech. These are intriguing arguments, but they may reflect a too hasty dismissal of Thesleff’s hypothesis. All the examples that Huss adduces as showing the organic unity of the work can also be explained without reference to the eighth chapter. Socrates’ criticism of Critobulus, for example, is wholely characteristic of Socratic humor, and is fully justified on its own. Socrates makes similar ironic remarks to Hermogenes concerning his claim to receiving special guidance from the gods (4.49), but these words do not look forward to any later account of Socratic divination. Although Socrates’ speech does answer many of Critobulus’ points, this does not seem to be its aim. If Xenophon wanted Socrates to be answering Critobulus, he could have had him refer to him in making the criticisms. Socrates does refer openly to earlier speakers when he wants to, and he refers positively to Critobulus’ erotic activity at one point in chapter eight (8.2). But when he comes to his great speech he addresses it not to Critobulus but to Callias and Autolycus, and he offers criticism not of Critobulus but of Plato’s Phaedrus (whom he calls Pausanias). That some elements of the speech run counter to claims made by Critobulus is simply a result of the fact that Plato gave Socrates a speech praising spiritual love, and one that criticizes a speech (Phaedrus’) that is based on that of GABRIEL DANZIG 353 Critobulus. Huss’ other arguments also fail to show that chapter eight is a necessary part of the whole. Socrates’ willingness to hand off his art of pimping to Antisthenes does not imply that he has another erotic art. As we have seen, Socrates has already claimed to be master of the art of pimping and he practices it throughout Xenonophon’s Memorabilia. The fact that he hands his profession over to Antisthenes is easily understood as a tit- for-tat response to Antisthenes’ claim that he received his wealth from Socrates (4.43). If Antisthenes wishes to take Socrates’ virtues, he must take his vices as well. With regard to the beauty contest, this should be seen as a satire of Socrates’ trial.51 Socrates’ expression of interest in the kisses of the judges surely does not suggest that this is a turning point at which the insufficiency of physical beauty becomes apparent. As it now stands, the eighth chapter provides a contrast to the rest of the work, allowing us to view the earlier antics from a higher perspective. It offers a more straightforward kind of apologetics for Socrates, presenting him as a reliable source of respectable advice on matters of love.52 This also suggests that it is written in a spirit which is different from that of the com- position as a whole. In any case, it is difficult to agree that the composition is missing something essential without this ending. That kind of argument, subjective in itself, runs up against special difficulties when dealing with a work like Xenophon’s Symposium which is not as fine an artistic whole as Plato’s Sym- posium. Further, since we do not know what the hypothetical original ending may have been, there is no way to say that it could not have filled whatever role we wish to attribute to the ending we have. V There is evidence for a further round in the dispute, in which Plato responded to Xenophon’s criticism of his caricature of Critobulus’ speech. If so, this would provide some additional 51 See Danzig, ClMed 55 (2005) 17–48. 52 On Xenophon’s apologetic strategies here, see Danzig, ClMed 55 (2005) 17–48. 354 INTRA-SOCRATIC POLEMICS confirmation of our account of the order of the two Symposia, and would show that Plato too was capable of making changes in a composition in order to voice a response to his rival’s criticism. In Plato’s Symposium, Aristophanes criticizes the very opinion that Xenophon’s Socrates expresses in Xenophon’s Symposium (8.33). Aristophanes remarks, referring to homosexuals (192A), fas‹ d¢ dÆ tinew aÈtoÁw énaisxÊntouw e‰nai, ceudÒmenoi: oÈ går Íp' énaisxunt¤aw toËto dr«sin éll' ÍpÚ yãrrouw ka‹ éndre¤aw ka‹ érrenvp¤aw, tÚ ˜moion aÈto›w éspazÒmenoi. m°ga d¢ tekmÆrion: ka‹ går televy°ntew mÒnoi époba¤nousin efiw tå politikå êndrew ofl toioËtoi. Who exactly are the people who say this? Dover takes the reference to be to comedy, and therefore concludes that Aristophanes is speaking “tongue-in-cheek”; Bury takes the remark as referring back to Pausanias’ words at 182A.53 But Pausanias was incensed at those who give pederasty a bad name, and his claim was that pederasty is a fine thing when performed by people such as himself. This is a far cry from the blanket condemnation of pederasty to which Aristophanes re- fers. A closer parallel is found in Xenophon’s Symposium. As we have seen, Xenophon’s Socrates makes exactly this point in speaking of Phaedrus’ speech (8.33), and he even uses the same word, énaisxunte›n, that is used by Aristophanes. This is the only place in Xenophon that I have found where one of his characters expresses this sentiment, and it happens also to be the only passage in Xenophon’s works which refers directly and critically to Plato’s Symposium. Is it merely a coincidence that in Plato’s Symposium Aristophanes utters words that are so ac- curate a reference to this very passage? My suggestion, that Plato has deliberately inserted these words into Aristophanes’ speech in order to answer Xenophon, is also unsettling, but perhaps ultimately less unlikely. It is in- evitable that Plato would have looked carefully at any direct attacks on him published by rival Socratic writers, or anyone 53 K. Dover, Plato Symposium (Cambridge 1980) 118; Bury, Symposium 64. GABRIEL DANZIG 355 else, and would have felt a need to respond. The insertion of a remark like this in a revised version of the piece that was attacked would be a very appropriate way of responding with- out acknowledging his opponent. The section in which Aristophanes makes this remark is short and could have been inscribed in the margins of the text without requiring a complete re-writing. Equally importantly, the passage can be removed without leaving a scar. Without the section we have a simple contrast between the behavior of homosexuals as youths (t°vw m¢n ín pa›dew Œsin, 191E) and their behavior as adults (§peidån d¢ éndrvy«si, 192A–B). The section in which Aristophanes criticizes those who say that homosexuals are shameless interrupts this contrast. His con- cern here is to show that homosexuals are admirable, while his concern in the surrounding discussion is simply to describe their way of life. This leads to a slight awkwardness: in order to show that homosexuals are admirable, Aristophanes includes a sentence which claims that when they grow up they enter pol- itics (ka‹ går televy°ntew mÒnoi époba¤nousin efiw tå politikå êndrew ofl toioËtoi, 192A). This means that together with the sentence mentioned above (§peidån d¢ éndrvy«si, paidera- stoËsi ka‹ prÚw gãmouw ka‹ paidopoi¤aw oÈ pros°xousi tÚn noËn fÊsei, éll' ÍpÚ toË nÒmou énagkãzontai, 192A–B) he has two sentences that describe how they behave when they grow up. Plato softens this doubling by describing their political activity as occurring televy°ntew, and their other activity as occurring when éndrvy«si, but there would have been nothing wrong if the first section were not there at all.54 It might seem odd for Plato to put criticism of Xenophon in the mouth of Aristophanes, with whom Plato’s Socrates dis- agrees, but what was his alternative? If he had put the comment into Socrates’ mouth that would have created the 54 At Resp. 498B–C Socrates lists three ages: youth, maturity (§n √ ≤ cuxØ teleoËsyai êrxetai), and an advanced age that is beyond the time of political activity. At Cyr. 1.2.4 Xenophon lists four ages: childhood, youth (ephebeia), maturity (tele›oi êndrew), and those who are past military age. In neither of these lists is there a distinction between the two ages that Plato offers here. 356 INTRA-SOCRATIC POLEMICS impression that Socrates was defending a not entirely reputable practice. Aristophanes is the most impressive non-Socratic per- sonality in the Symposium, and he made a speech which de- fended homosexuality in the clearest terms. Here then was the most likely place to put the comment. Placing the comment in the mouth of Aristophanes suggests that despite the dramatic nature of the compositions we are considering, authors could express themselves through the mouths of their characters. Similarly, Xenophon places his own criticism of Plato in the mouth of Socrates. And most sig- nificantly, he criticizes Plato for a speech which is given by one of the non-Socratic characters in his Symposium. It seems ob- vious that Xenophon is not intending merely to criticize the character Phaedrus, especially since he gets the name wrong. This suggests that authors could be held accountable for the things their characters say. This should complicate current ideas about the anonymity of Platonic writing.55 Final speculation In the opening scene of Plato’s Symposium, Apollodorus men- tions a rival account of the event told by one “Phoenix son of Philippus,” and comments éllå går oÈd¢n e‰xe saf¢w l°gein (172B). It is not clear why Plato mentions this. It has been suggested that it is a reference to a rival version of a Socratic symposium. Bury raised the possibility that it might be a reference to Xenophon’s Symposium, but he rejected it on the grounds that Xenophon’s Symposium was written later than Plato’s.56 If we are right, however, and Xenophon’s Symposium was written earlier, then this may well be a reference to it. As we noted above, ancient critics were not aware of any other Socratic symposium to which this might refer. If so, it raises an interesting possibility concerning the way in which Plato could refer to a rival. Readers have often won- dered what the significance of the name Phoenix is here.57 D. 55 On this whole subject, see the articles in G. Press (ed.), Who Speaks for Plato? (Lanham 2000). 56 Bury, Symposium xvii–xix. 57 See Thesleff’s hypothesis, BICS 25 (1978) 170 n.29. GABRIEL DANZIG 357 Nails says little about this mysterious figure.58 But the answer seems simple: Fo›nij contains some of the same consonants as Jenof«n in a different order.59 Although not an exact reflection of the letters in Xenophon’s name, it is about as close as Plato could get without inventing a new name. Philippus in turn is the idiotic and insipid clown in Xenophon’s Symposium.60 Apollodorus’ unnamed friend, whom he addresses through- out Symposium, and who seems to represent Plato’s audience, had heard something about the speeches on love made by Socrates and his friends at a banquet long ago, but he had not gotten a clear story. Apollodorus is called upon to correct the account of Phoenix son of Philippus. If this refers to Xeno- phon, it is a derogatory reference to his Symposium, and a prom- ise by Plato that he will do better. The derogatory reference together with the clownish paternity provides an additional reason why Xenophon would have wanted to respond to Plato in chapter eight. In that chapter (8.23) he does not fail to point out that Phoenix was the honored teacher of Achilles.61 May, 2005 Dept. of Classics Bar Ilan Univ. Ramat Gan, 52900, Israel danzigg@mail.biu.ac.il 58 The People of Plato (Indianapolis 2002) 241. 59 Along similar lines, is it possible that the reference in the Apology to a young man who does what Socrates does not, plãttonti lÒgouw (17C), is a sly reference to Plato himself? 60 See Thesleff, BICS 25 (1978) 168. 61 Earlier versions of this paper were read at the Bar Ilan Classics De- partment Colloquium, and at the 2005 APA meeting in Boston. I would like to thank the participants for their comments. I would also like to thank Michael Stokes and Jeffery Purinton, as well as two anonymous readers for GRBS, for their valuable comments on earlier drafts.