1997.4.2, Murray, Plato on Poetry

This review will deal with the edition of the Ion contained in this volume, with particular reference to its pedagogical value. The commentary on parts of Republic Bks. 2 and 3 is welcome as the only recent one available in English, while that on the first part of Rep. Bk. 10 is more up-to-date than J. Ferguson’s edition (1957), and more authentically scholarly than S. Halliwell’s (1988), which is partially oriented to a translation. But since this material is likely to be read only by more advanced students, I shall not comment on it much for the purposes of the present review. Murray (M.) does not direct her treatment of these central texts of Platonic literary criticism to any specific group. She offers an undefined “reader” a commentary “which takes account of modern scholarship on the subject, and which explores the ambivalence of Plato’s pronouncements on poetry through the analysis of his own skill as a writer” (Preface, p. vii). Now many, if not most, of the earlier volumes in this series (CGLC) have tried to accommodate students. A recent edition of another Platonic dialogue, C.J. Rowe’s Phaedo (1993), was said to offer help to “comparative beginners in the language”, while also being “intended for any student, classical scholar, or philosopher”. My experience of reading the Ion recently with a class of senior undergraduates and M.A. students was that Murray’s edition needed to be supplemented by a more elementary commentary, such as A.M. Miller, Plato’s Ion, Bryn Mawr Commentaries, Bryn Mawr, 1st. ed. 1981, 2nd. rev. ed. 1984. ¹ (M. cites its first edition [p. 242], but never refers to it.) CGLC‘s general editors (with one of whom M. herself [see p. viii] read the Ion in her first term as an undergraduate) might have realised that the largest body of potential readers of this dialogue would be comparative beginners. As it is, it is mostly advanced students who will be helped by M.’s abundance of loci paralleli, and her references to secondary literature in English, French, German, and Italian. Beginners get some help from M.’s translations of difficult phrases and sentences, but she rarely cites standard works on grammar and syntax, as do the Bryn Mawr commentaries, or, to cite a CGLC edition, Dover on the Symposium. She is also inconsistent. She manages to explain the attraction of the relative pronoun into the case of its antecedent (a staple of first-year Greek) six times (on Ion 532e1, 533a3-4, 535c2, 535e7-8, 536e4, 538b1); notes an obvious double accusative (535d8-9); and identifies prolepsis (531c6-7, 531e5). Yet she ignores the use of the deliberative subjunctive in indirect questions (532c4, 533a3, 533a5, 533b4, 536b7, 536c1), as well as several other points of language (see below) that will surely challenge even advanced readers. Given the availability of Miller’s commentary, what university instructors really need, and will not find here, is an edition of the Ion along the lines of Ian Walker’s useful presentation of the Euthyphro (Scholars Press: 1984), in the A.P.A. textbook Series. Hereafter I comment in sequence on M.’s treatment of the Ion. My remarks primarily concern areas where I believe that this edition could have been more serviceable to both undergraduates and graduate students.

Pp. 3-12: M.’s preliminary discussions of mimesis and of inspiration ² are derived from her article “Inspiration and Mimesis in Plato,” 27-46 in A. Barker and M. Warner eds., The Language of the the Cave (Edmonton [not Alberta, pace M., p. 239], 1992). But in neither place does she explain for the benefit of students of the Ion why Plato abandoned inspiration for mimesis in his literary and artistic criticism in the Republic (see art. cit. p. 39). Ion’s performative capacities are of course mimetic (see 535b1-535e6), and, as M. notes (on Ion 535c7-8 and Rep. 393c5-6, and 395d1-3), the strong emotional effects that he produces in his audience exemplify the kind of mimesis that is the object of criticism and censorship in the Republic. But why is poetic inspiration absent from the later work? Could it be that it explains only how poetry that can evoke a strong emotional response is adventitiously composed, and transmitted to an audience? To evaluate such effects Plato needed the concept of mimesis, which, when deployed within the psychosocial theory of the Republic, could identify bad poetry (or in Rep. 10 perhaps identify all poetry as bad; see M. pp. 6 and 229). Inspiration, like magnetism ( Ion 533d3-e2, 535e8-536a7), was, by contrast, a fact of nature, and identified poetry as valuable just because of its communicability, or “magnetizing” effect, regardless of its moral significance. ³

P. 10: Socrates’ long speech on inspiration (533c9-535a2; 535e7-536d3) is the highlight of this dialogue and the main reason for its popularity with beginners. M. challenges us to reflect on its purpose by claiming that it is “ambiguously” constructed. That is, it is eulogistic in linking poetic composition with divine sources; yet at the same time it “undermines the authority traditionally accorded to poets by depriving them of tekhne”. ⁴ M. concludes that “like Ion himself we are left in a state of aporia, unable to decide how to read S[ocrates]’ apparent eulogy of the poets.” But if the eulogy is insincere (as it must be if “apparent”), then it is not a genuine eulogy at all, and so there should be no aporia. There is also nothing inherently ambiguous about Socrates’ thesis that poetry and rhapsody do not meet the standards of craft-knowledge. M. therefore presumably means that an audience might be ambivalent about the implications of the speech: that is, “we” may know what Socrates means, but can let our empathy with emotionally stirring epic and lyric poetry override his thesis. If indeed it was Plato’s intention to show that in its reception Socrates’ encomium could conflict in this way with his arguments, that is a further reason why in the Republic there is no reference to the theory of poetic inspiration. The very theory would be seductive, as well as its subject matter. M. never asks why Socrates delivered this speech at all, instead of confining himself to elenchic arguments. It is, after all, “unique among Plato’s earlier compositions in its exuberance of poetic imagery”, where “in explaining the poet to us, Plato lets Socrates speak like a poet for the nonce” (G. Vlastos, Socrates: Ironist and Moral Philosopher [Cambridge and Ithaca, 1991], p. 287). (M. notes only word play, at p. 117.) When Alcibiades ( Symp. 215d5-6) later describes Socrates’ own logoi he uses terms reminiscent of the analysis of poetic composition and performance in the Ion when he says that Socrates makes people “amazed” (ἐκπεπλεγμένοι; cf. Ion 535b2) and “possessed” (κατεχέσθαι; cf. Ion 533e7, 534a4, etc.); cf. also the leaping heart and weeping at Ion 535c6-8 and Symp. 215e1-2. But those logoi were clearly elenchic arguments; cf. Symp. 221d7-222a1. Why in the Ion is Socrates shown resorting, without explanation or apology, to seductive poetic language in a prolonged epideictic oration that “touches” Ion’s soul (cf. 535a3-4)?

M., as we have seen, finds this whole exercise an ambiguous construction, designed to generate ambivalence in the reader. But perhaps Socrates was also parodying the style of Ion’s lectures on Homer, which were probably similarly replete in imagery. Ion is certainly a “eulogist” (ἐπαινέτης) of Homer (536d2; 542b4), who “embellishes” (κοσμεῖν at 530d6-7; M.’s gloss) his author. Such a presentation would also be more intelligible to Ion than Socratic dialectic (as his emotional reaction at 535a3-4 shows), even though he might not have appreciated the irony of being criticised in his own rhetorical style (cf. on 530b6 below). ⁵ On this view Plato would be letting Socrates appropriate Ion’s epideictic (and exegetical) genre to analyze his limitations, after refusing (at 530d9-531a1 and 536d8-e1), in the greatest irony of all, to let Ion himself show his talents in this area. M. at “Inspiration and Mimesis”, p. 35, though not in the Introduction to this edition, asked of the speech in the Ion: “It is ironic, certainly, but how ironic?” If it is a deliberate parody, we have an answer to that question.

Pp. 30-31: M. compares Shelley’s Defence of Poetry with Plato’s views. But Shelley also translated the Ion with the title Ion, or Of the Iliad; see The Complete Works of Percy Bysshe Shelley, ed. R. Ingpen and W.E. Peck (New York, 1965), Vol. 7, 233-254 (earlier reprinted in Five Dialogues of Plato bearing on Poetic Inspiration [London, 1910].) M. notes (p. 31) that Shelley stated in a letter of February 1821 that he was reading the Ion; but in a letter of 22 October 1821 ( Complete Works, Vol. 7, p. 311) he also reported that he had translated it.

P. 96: The Ion“probably belongs to [Plato’s] early period”, while the Republic“belongs to his so-called middle period” (132). This means that the Ion offers evidence for Socratic method, yet this is a topic that M. almost entirely neglects. Ion by the end, we are told, seems “to have learnt nothing from Socrates’ cross-questioning” (M. on 541b1-2). But such “cross-questioning” (an inappropriate forensic metaphor, since Ion has not been previously questioned) has involved the Socratic elenkhos, a subject recently of some intense debate. M. uses the word “elenchus” once (on 531a1), but for an analysis of its role in this dialogue, and its relation to the speech on rhapsodic inspiration (533c9-536d3), we have to turn (as M. should have) to G. Vlastos, Socrates: Ironist and Moral Philosopher, 267-8 and 287-8. M. does refer twice (on 531e9 and 540a2-5) to the use of a “syllogism”, a term taken from Richard Robinson’s Plato’s Earlier Dialectic (2nd. ed., Oxford, 1953), where, in his introductory survey of the elenchus (Robinson, pp. 20-22), “syllogism” identifies only the general process of reaching a conclusion from preceding premises. But to cite Robinson on Socratic method in general while ignoring Vlastos on the specifics of Socratic epagoge at 540b-d (Socrates, 267-8) is not perhaps the way to take account of “modern scholarship” (M., p. vii). ⁶

530b6: This is the first of M.’s sixteen references to irony in the dialogue. But nowhere is this term defined, despite a benchmark modern treatment by Vlastos ( Socrates ch. 1; originally at CQ 1987). This is disconcerting, since it prevents us from learning what general purpose Socratic irony serves in the dramatic context of this particular work. Are we, for example, to assume that Ion invariably fails to realise that Socrates is saying the opposite of what he means? M. implies this (“Ion’s vanity remains unassailable” on 542b1-2, at the end of the dialogue), but she might have addressed the question directly in the Introduction. In fact most of M.’s examples of irony fall into the category of irony as extravagant praise, on which see Nightingale, Genres (see n. 4), pp. 114-19.

530d3: On καλὰς διανοίας μ. remarks that “Ion’s opinions about Homer may be ‘fine’, but they may not necessarily be true. The ambiguity in the meaning of the word καλός is exploited throughout the dialogue.” M. herself returns to this major point only once and briefly on 533e7. But Plato nowhere claims that the beauty of poetry is incompatible with its truth; he only at most implies that any true statements made by a poet, and repeated or explicated by a rhapsode, are not known in a comprehensive or reliable form, as they would be if poetry were itself a comprehensive tekhne that embraced all other tekhnai. That is why any claim to knowledge by poet or rhapsode is better assessed by someone who possesses that knowledge in an authentic form.

Later at Meno 99b11-d5 (quoted by M. at Appendix pp. 235-236), Plato notes that inspired (or “divine”, theios) poets have true belief rather than knowledge; cf. εὐδοξία (99b11) anticipated by δόξα ἀληθής (99a5). At the end of the Ion, the description of Ion as θεῖος and μὴ τεχνικός (542b4), has the same general implication. It is not only “the theme of poetic inspiration” (M., p. 235) that links the Ion with the later passage from the Meno, but a deeper epistemological distinction between adventitiously formulated true belief and systematically acquired knowledge. The Meno spells out what is implicit in the Ion (as well at Apology 22b8-c6 [= M., p. 235]). ⁷

530d10-11: dianoia here and at 530c4 is (as M. rightly notes) used in the sense of “‘meaning’ or ‘thought’, as opposed to form of expression”; Plato also notes this sense in marking the same contrast at Lysis 205a9-b3.

531a5: Socrates asks if there is that “about which (περὶ ὅτου) Homer and Hesiod say the same things.” Instead of just noting that the antecedent of the relative is omitted, M. might have stressed that this question introduces a crucial principle for the dialogue: namely, that poetry has a subject matter “about which” it speaks. It is as though the Socrates of the Ion respects only one of the three criteria by which mimeseis are differentiated at Aristotle, Poetics chs. 1-3: the ἕτερα (“different objects”, Janko), as opposed to the means (ἐν ἑτέροις) and “manner” (ἑτέρως) of poetic expression; cf. Poet. 1447a16-18. This crude emphasis on content overrides Ion’s fumbling attempts to refer to style (see M. on 531d6), or expression in performance (see M. on 540b3, and my note on this passage below). M. should have highlighted this dominant principle in her Introduction. It drives the analogy between poetry and the standard crafts, the analysis of crafts themselves (see below on 537c5-6), and also the classification of poetic genres (cf. 534c6-7).

531d12: Socrates addresses Ion as ὦ πηίλη κεφαλή, and, as M. notes, this can be paralleled from other Platonic texts. But if it is “clearly ironic in tone” (as M. also claims), this may be because it is an echo of epic and tragic usage (cf. Iliad 8.281, and Soph. O.C. 1631), appropriately used to address a professional performer. Its irony is not well conveyed by M.’s British “My dear chap”; perhaps “Noble wight” might capture its archaizing theatricality more effectively.

531e10-532a2: In commenting on the comprehensiveness required of someone who has mastered a tekhne, M. is right to note Symp. 223d3-6, where Socrates seems to open up the possibility of a comprehensive poetic tekhne embracing tragedy and comedy, in contrast with the position maintained in the Ion and elsewhere. M. also suggests that the argument in the Symposium is “purely hypothetical”. But it has a wider role in that dialogue; see D. Clay, “The Tragic and Comic Poet of the Symposium,” 186-202 in J. Anton and A. Preus (eds.), Essays in Ancient Greek Philosophy, vol. 2 (Albany, 1993).

532b5: Someone able to assess all poetry is identified as a krites hikanos. Cf. Plato, Euthyphro 7d3-4 where ἱκανὴ κρίσις is used to identify a device for solving disputes. The parallel is apt since at Ion 531e4-532a4 the argument that an expert must have comprehensive knowledge is developed in terms of an ability to assess different statements on the same subject.

532b8: Ion here requests an aition for his enthusiasm only for Homer, and at 536d1 Socrates claims to have provided it. This highlights the fact that while Socrates’ long speech may be an ambivalent euology (see above), it is also an aetiology that exploits myth (here in the form of the Muses). In this respect it resembles Protagoras’ “Great Speech” at Protagoras 320c8-328d2, where the Sophist offers an aition for Athenian democratic practice in partly mythological form. Such epideictic aetiology (which all the speeches in the Symposium exemplify) is part of a wider body of “aetiological stories”. On these see K.J. Dover’s study of the background to Aristophanes’ speech in the Symposium; see JHS 86 (1966) 41-50, repr. at Dover, The Greeks and their Legacy (= Collected Papers Vol. II) (Oxford, 1988), ch. 11 (pp. 102-114), at 42-44 or 104-107 of the reprint.

532c8-9: M., like MacGregor, takes τὸ ὅλον here as the subject of a question about the unity of the poetic art (“for the whole thing is poetry, isn’t it?”), when in an identical question thirteen lines later (532e4-5) she takes it as an adverbial accusative: “is there an art of painting as a whole?” Smith and Miller are similarly inconsistent. But to holon must be adverbial in both places, as it is in similar statements elsewhere (e.g. Meno 79c1, Phaed. 261a7, and Rep. 455d4). Thus translate 532c8-9 as “there is an art of poetry as a whole” (Cooper, Rouse, Woodruff).

532e1: Socrates rejects any claim to wisdom, and expresses an interest in the truth οἷον εἰκὸς ἰδιώτην ἄνθρωπον. ἄνθρωπος is self-deprecatory, since it standardly denotes disrespect (Smyth GG 986b; Gildersleeve, Syntax, p. 487). By the same token, when later at 539e8-9 Socrates refers to Ion as a rhapsodos aner this is surely irony, since aner with the name of a profession normally denotes respect.

533a7: Socrates refers to Daedalus, and then later to Epeius (533b1), as well as to legendary musicians, Olympus, Thamyras, Orpheus, and Phemius (533b8-c1). M. says (on 533c1) that there is “more than a little irony in comparing Ion with these great masters of the past.” But does Ion realise this? This is the question that M. consistently ignores in identifying irony; see on 530b6 above.

533c7-9: Ion urges Socrates ὅρα τοῦτο τί ἔστιν, referring to his exgetical fluency exclusively with regard to Homer. Socrates replies: ὁρῶ καὶ ἔρχομαί γέ σοι ἀποφανούμενος. M. points in the right direction: “the nuance of these words is practically ‘you explain (if you can)’, to which Socrates answers, ‘Yes, I can and will.'” But perhaps it is best to say that ὅρα here means, as it often does in Plato, “consider [sc. something as the next stage in the discussion]”, and so Socrates’ ὁρῶ is a quasi-future, reinforced by the next clause (whether or not it is emended; see next note). Cf. Plato, Crito 48d6-7 where Crito says ὅρα DE TI/ DRW=MEN (“consider what we are to do”), and Socrates’ reply (σκοπῶμεν) refers to the future with a hortatory subjunctive; and also Sophist 268a10-11 where the sequence between interlocutors is ὅρα SU/ and σκοπῶ: “—That’s for you to consider.—I will” (tr. F.M. Cornford). Those who render the sequence at 533c7-9 along the lines of “See”/”I do see” (Cooper, Miller, Rouse, Smith, Stock, Woodruff) just produce confusion. If the imperative “See” means “consider” or “investigate”, then Socrates cannot reply that he does already see (in the sense of recognize) that which he has yet to consider. ὁρᾶν would hardly change its sense that rapidly.

533c9: M. rejects C.G. Cobet’s emendation ἀποφανούμενος after ἔρχομαι (for the vulgate ἀποφαινόμενος), which was accepted by Burnet in the O.C.T. (and derivative texts in Stock and Miller), and by Nihard, Albini, and Verdenius. (This emendation is one of only two entries in M.’s apparatus for the Ion; the other, a trivial gloss at 533c2 is not discussed. I would like to have seen Sydenham’s instructive emendation ἂν at 540d5 included.) There are, however, Platonic parallels for the use of the future participle in this locution ( Protag. 313a2; Rep. 449a7, and 562c4), although, as M. (following MacGregor and Miridier) notes, in one case ( Phaed. 100b3) a present participle is used, and the idea of futurity derived from the main verb. Rowe ad loc., however, argues otherwise, unconvincingly, I believe, although his note might have been cited, if only because it appears in a CGLC edition.

533e8-9 (with 536c2-3): The references here to οἱ κορυβαντιῶντες were carefully discussed by I.M. Linforth at Univ. of Calif. Publics. in Class. Philol. 13 (1946) 137-40, and more briefly by E.R. Dodds, The Greeks and the Irrational (Berkeley, 1951), p. 79 with n. 102. M.’s notes (on 533e8-534a1 and 536c2-6) fail to distill the issue raised in this secondary literature. The first passage is an analogy between the ecstasy of lyric poets and the ritualistic transport of Corybantes, the second one between Ion’s obsession with Homer and the way that different Corybantes focus obsessively on different gods. M. should have recorded (on 536c2-6) that Linforth thought that the multiplicity of deities involved in the second analogy was incompatible with a precise reference to Corybantic ritual, and argued for a more general sense of the verb κορυβαντιᾶν (“telestic enthusiasm in general”) in that passage. Dodds disagreed, and tried to keep this second passage as a further reference to the ritual of the cult, and M. reports only his views on 536c2-6. Linforth, however, was probably right, since the analogies in the Ion are only with the outcome of Corybantic ritual, not its specific procedures, and particularly not its therapeutic role (which M. notes without elaboration on 533e8). Poets and rhapsodes are surely not being compared to troubled souls who need solace in ritual.

534c1: On θείᾳ μοίρᾳ μ. refers to an obscure article by J. Souilhi; the standard treatment in English is E.G. Berry, The History and Development of the Concept of theia moira and theia tukhe down to and including Plato (Chicago, 1940).

534e1: ἐν τούτῳ goes unexplained as the instrumental use of ἐν with a verb of demonstrating (cf. ἐνδείψασθαι, 534e2) (see LSJ at ἐν A.III). Translate “by this [example]” (Cooper), not “in him” (T. Taylor), “in this way” (Jowett), “in this” (Woodruff), “here” (Rouse). Albini compared ἐν τούτῳ with an almost identical verb at Meno 82b1-2, where the “instrument” is the slave-boy being questioned. McKirahan (Bryn Mawr comm., 1986) ad loc. (“by means of him”) has the description of that demonstration right; Jowett and Sharples (demonstrate “on him”), and Rouse (“in him”) do not.

535c5: In οὐ γάρ σε ἀποκρυψάμενος ἐρῶ, the participle is essentially adverbial: “I will tell you frankly” (L. Cooper). (Cf. a direct parallel at Apology 24a5-6.) This is the coincident use of the aorist participle, which identifies an accompanying rather than a preceding action. For the classic modern treatment (improving on Goodwin MT para. 150) see W.S. Barrett on Eurip., Hippolyt. 289-292, a note known to two earlier CGLC editors (C. MacLeod on Iliad 24.170, and M. Griffith on PV 722). M. does not even insinuate this usage through a translation, as, for example, does Rowe (on Phaed. 62a8, and 86d5; though not on 60a6). Dover (in his CGLC edition) overlooks it at Symp. 189b3, and 202b10.

535d4: On 30,000 as “the traditional number of Athenian citizens” cf. a related passage at Herodotus 5.97.2.

536d8-9: At this transition to the final part of the dialogue M. claims that Socrates’ preceding speech on inspiration “concentrates exclusively on performance” (p. 126) (my italics) and ignores Ion’s role as a commentator on Homer. But in that speech Ion is said to speak “about Homer” (534c1 and 536b5-c1). No distinction is, or need be, drawn between Ion as performer and as commentator: he recites Homer just because he is only interested in commenting on this poet, and vice-versa. M. herself acknowledges (on 536d3) “Ion’s combined activity of reciting and commenting on Homer”. Hippias also adopted that combination in his lectures; see Plato, Hipp. Min. 369e2-4.

537c5-6: οὐκοῦν ἑκάστῃ τῶν τεχνῶν ἀποδέδοταί τι ὑπὸ τοῦ θεοῦ ἔργον οἵᾳ τε εἶναι γιγνώσκειν; “So has the god given to each skill the ability to understand a particular activity?” is M.’s translation; Cooper and Woodruff translate in a similar vein. Miller drew attention to an alternative construe, whereby τι ἔργον is the subject of ἀποδέδοται, and the concluding infinitive taken as an infinitive of purpose. Smith accepted such an analysis and translated, a little loosely: “Each of the arts has some subject assigned to it by the god, so that the art has the power to master the subject.” The association of ὑπὸ τοῦ θεοῦ with ἀποδέδοται arguably makes it awkward to take TI … ἔργον as the object of γινώσκειν.

As for exegesis, M. cites passages from Homer and Cicero to illustrate the inclusion of prophecy with more practical arts to anticipate the examples that will follow. For a more illuminating and near-contemporary parallel see the Hippocratic De Arte ch. 2, where the segregation of the tekhnai is grounded in a crude metaphysic of separate natural kinds (εἰδή) that each have different names. Cf. Socrates’ contrast (537e1-4) between distinct pragmata (subject matters) and the differentiating labels that we apply to tekhnai.

538c1: M. says that the Homeric quotation is “a conflation of three lines” ( Iliad 11.639-40 and 630). But the conflation involves Iliad 11.640 being cut off at the first syllable of the third foot (at the hemiepes), and the remainder of the line being supplied from 11.630. If this is an error of memory (as it surely is, given the similarity in subject-matter and the standard introduction [Pως οὕτως] for a quotation about which a speaker is uncertain; cf. Plato, Apol. 28c5) rather than the quotation of a different text, then its metrical character was worth noting. // 538d7: Socrates initiates a question with: σκέψαι δὴ, σοῦ ἐρομένου, εἰ ἔροιό με. We might say (as do M., Miller, and others) that the conditional clause indicates beyond doubt that the genitive absolute is conditional, and thus somewhat redundantly reinforces it. But dramatically the sentence highlights the absurdity of Ion asking questions at all: “Look, if you (!) were asking the questions, then were you to ask etc.” In other words, εἰ ἔροιο does not really clarify the semantic force of the preceding genitive absolute, since that clause is already an obvious hypothesis with a sarcastic spin. Such syntactically irregular (“pseudo-absolute”) clauses do often emphasize the idea that they express; see Smyth GG 2073. For other Platonic examples see Phaed. 77e3, or Symp. 183b5 and 190c1.

539e6: Here Ion claims that as a rhapsode he can investigate and assess everything (ἅπαντα) in Homer. On what “everything” might be over and above specialised knowledge see the next note. But cf. § 8 of the sophistic Dissoi Logoi (DK II, p. 415), where knowledge of logoi is said to entail knowledge of everything, a parallel noted by R.K. Sprague ( Plato’s Philosopher-King: a study of the theoretical background [Coumbia S.C., 1976], at p. 14 n. 5) (a work cited in M.’s bibliography).

540b3: Ion says that he knows what it is “appropriate” (πρέπει) for different types of individuals to say. M. sees him as “tentatively trying to formulate” what she calls “the distinction between content and expression” in the Homeric epics. The point might be more effectively put by noting Ion’s own performative, indeed histrionic, abilities, as he himself describes them at 535c5-8. There he notes in general terms (in ὅταν clauses) his capacity to simulate the emotions associated with pitiable and frightening situations, while here at 540b3-5 he identifies a set of typical characters (women, slaves, free men, rulers and subjects) whose “appropriate speech” he knows. The second claim is thus a consistent extension of the first, and resembles Aristotle’s claim ( Poetics 1451b5-11) that the dramatic poet deals with what is universal: “the sort of thing that a certain kind of person may well say or do in accordance with probability or necessity” (tr. Janko). The actor must thus develop a generic capacity to impersonate convincingly. If, as M. claims, Ion is trying to pinpoint “expression” here, then it must be in this performative sense.

540c1-3: Socrates asks whether a doctor or a rhapsode will know better the appropriate speech for someone in charge of a sick person, and Ion admits that it is the doctor. Note Plato, Gorg. 456b1-5 where Gorgias offers evidence of the persuasive power of rhetoric in getting patients to take drugs or undergo surgery. Gorgias was more sophisticated than Ion, and could represent his verbal art as comprehensive, or second-order, in nature; see Gorg. 456a7-8, and Sprague, Plato’s Philosopher-King, pp. 24-25.

540d4: Socrates confronts Ion with the question ἡ ῤαψῳδικὴ τέχνη στρατηγική ἐστιν; for a similar identification of one craft with another for the purpose of refutation by reductio ad absurdum cf. Euth. 14e7 where piety is identified as a form of barter (ἐμπορικὴ τις τέχνη). M. might have exploited this text and the parallel to reinforce her point (on 538b4) that the mere use of the phrase rhapsoidike tekhne does not imply that there is genuinely such an art, any more than there is one of piety.

540d5: Ion claims that a good rhapsode knows what a good general would say because, as M. (like Ion at 541b5) says, this expertise is prominent in Homer. But, as Grote ( Plato and the Other Companions of Socrates, II [London 1888], p. 135) noted, there is contemporary evidence of the office of general at Athens not being considered the province of trained experts. He refers to Socrates’ conversation with Nicomachides at Xenophon, Mem. 3.4, where Socrates seriously argues that the generic qualities of an oikonomos are transferrable to generalship. At 541b8-c2 it is perhaps the absurdity of such a transference from the rhapsodic art that Socrates is ridiculing, rather than the general principle that someone inexpert in generalship but possessing another expertise could become a general.

540d6-7: M. does not comment on the aorist in the apodosis of the present contrary to fact conditional at 540d6-7: εἰ ἐτύγχανες …, ἔγνως ἄν (“if you happened [to be skilled] …, you would know”), where it excludes any idea of duration in present time: i.e., “[the upshot is that] you would know”. On such aorists as Platonic idiom see Smith ad loc., and Goodwin MT 414. (For “upshot” as a felicitous generic term for the aorist see Gildersleeve, Greek Syntax, Part I, p. 104.) M. acknowledges this aspect elsewhere, but for a far more obvious use of the aorist (ἠγωνίσω at 530a8; cf. Symp. 194a1) to express “the result of the action”. M. also does not explain that the aorists (ἠρόμην … ἂν ἀπεκρίνω) in another conditional referring to present time immediately following at 540e1-3 similarly exclude duration; Goodwin, MT. 414., notes Plato, Protag. 311b7-c1 (a close parallel to the Ion); cf. also Meno 72b2-3.

541a1-b5: Socrates’ lesson in logic here (cf. Euth. 11e7-12d4 for a more elaborate version) needs comment. He is essentially trying to get Ion to admit that in a universal attribution expressed in the affirmative (all good rhapsodes are good generals [541a3-4]) “the terms of the affirmative must be convertible, not universally, but in part” (Aristot. An. Pr. 25a7-10, tr. Jenkinson); i.e., that it follows that only some good generals are good rhapsodes. But when at 541a5-7 Ion agrees that it is not the case that all good generals are good rhapsodes, he may, for all we know from this text, think that no good generals could be members of his profession.

541b8: M. notes that the reference to Ion “going around as a rhapsode” (ῤαψῳδεῖς περιιών) is paralleled in a description of Homer at Rep. 600d5-6. But περιιών has a derogatory sense, caught well by Saunders’ “traipse round” (Penguin Classics translation). It is used with ζητεῖν to describe incessant and irritating inquiry at Ap. 23b5, Protag. 348d4, and Symp. 209b3. At Ap. 31c5 Socrates even self-deprecatingly glosses SUMBOULEU/W περιιών with POLUPRAGMONW, which spells out this implication, and also fits well with the tone of the passage in the Ion.

1. The only other edition in English to which M. refers is J. M. MacGregor, Plato: Ion (Cambridge, 1912; repr. 1956, 1965). The others (all essentially “schools’ editions”) in English that I have seen (and will refer to on occasion) are by G. Smith (London, 1895) and St. George Stock (Oxford, 1909); I have not seen that by J. Thompson and T.R. Mills (London, 1899). In other languages I have seen those by: W.J. Verdenius (2nd. ed., Zwolle, 1959) (in Dutch); R. Nihard (Lihge, 1923; 2nd. ed., 1949); A. Annaratone (Naples, 1930); and U. Albini (Florence, 1954). I have not seen those by: L. Mertz (1st. ed., Paris, 1900); G. Abt and A. Chaumeix (Paris, 1900); C. Diano (Citta di Castello, 1930), and W.J.W. Koster (Amsterdam, 1937).

2. M. (p. 11 with n. 26) notes that the description of inspiration at Plato, Phaedr. 245a1-8 was influential in the Renaissance. Add a modern use by A.E. Housman, The Name and Nature of Poetry (1933) (at A.E. Housman: Collected Poems and Selected Prose, ed. C. Ricks [London etc., 1988], pp. 364-5), in a context in which Housman defended the notion that poetry could be understood without reference to its meaning, a view that interestingly parallels Socrates’ recurring claim that Ion speaks without knowledge or skill.

3. M. might have given more attention to C. Janaway, “Craft and Fineness in Plato’s Ion,”Oxford Studies in Ancient Philosophy 10 (1992) 1-23, since Janaway argues that the theory of inspiration is evaluative: it is “a theory of what makes good poetry good” (p. 23), and “an embryonic theory of one form of artistic production” (p. 19). But this “theory”, I would myself argue, is really an illegitimate inference from evidence of the effect of certain types of poetry on a performer and an audience with whom emotional contact is established. Such evidence does not show that this linkage can be explained only if poet and performer are “possessed from outside” (my italics) by some incomprehensible force (Janaway, p. 19). All that Socrates can legitimately infer is that they fail to meet the criteria of rationality based on the tekhnai; nothing follows as to how that failure occurs. It is tradition alone that generates the theory that it is due to inspiration from the Muses. M. is good on defining that tradition (pp. 6-8), but less effective in analysing the nature of Plato’s adaptation; see my comments on 530d3 below.

4. A.W. Nightingale’s recent study Genres in Dialogue: Plato and the construct of philosophy (Cambridge, 1995), ch. 3, discusses Plato’s critical treatment of “encomiastic discourse” with particular reference to the speeches in the Symposium. If, as M. proposes, Socrates’ speech in the Ion is a eulogy that undermines the object of its praise, then it is an early example of Plato’s analysis of this genre.

5. M. Miller, “‘The Arguments I Seem to Hear’: argument and irony in the Crito,”Phronesis 41 [1996] 121-137, has recently suggested that Socrates’ speech in another dialogue, the Crito, is also deliberately adapted from a contemporary genre to the needs of an interlocutor of limited intelligence.

6. When central philosophical issues arise in the commentary on the material from the Republic, M. invariably refers to handbooks on Plato or on this dialogue, such as those by I.M. Crombie, J. Annas, N. White, J. Gosling and, rather surprisingly, at p. 185 to a relic from the Victorian era, R. Nettleship’s Greats lectures. Where literary issues arise M. is considerably more au courant with the literature, and the work of non-Anglophone scholars.

7. P. Woodruff, “What Could Go Wrong with Inspiration?: why Plato’s poets fail,” ch. 6 in J. Moravcsik and P. Temko (eds.), Plato on Beauty, Wisdom, and the Arts (Totowa, 1982) addresses the epistemological aspects of inspiration in a programmatic study that ranges over a variety of texts from different dialogues. In juxtaposing the Ion and Meno, here, I am only proposing that at this pre- Republic and Phaedrus stage Plato thought that poets could hold true beliefs but that they were constitutionally unable to convert them into knowledge. I see no evidence to show that he yet thought (as Woodruff, p. 142-143 seems to argue) that they could not hold beliefs at all.