The Epic Rhapsode and His Craft: Homeric Performance in a Diachronic Perspective

  González, José M. 2013. The Epic Rhapsode and His Craft: Homeric Performance in a Diachronic Perspective. Hellenic Studies Series 47. Washington, DC: Center for Hellenic Studies. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_GonzalezJ.The_Epic_Rhapsode_and_his_Craft.2013.


9. The Rhapsode in Classical Athens

9.1 Of Transcripts and Scripts

My point of departure is Gregory Nagy’s proposal of a textual fixation [4] of the Iliad and the Odyssey in five stages spanning a continuum in which, at one end, writing plays no role at all, while, at the other, a largely uniform written text of the poems is slavishly memorized and adhered to in performance. [5] My particular focus here is in the transition from what Nagy calls ‘transcripts’ to ‘scripts.’ These are his definitions: “A transcript can be a record of performance, even an aid for performance, but not the equivalent of performance”; [6] a script, in turn, is “a narrower category, where the written text is a prerequisite of performance.” [7] If we assume that rhapsodes down to the fifth century BC trained and performed without the aid of writing—transcripts may have been in limited circulation, but the professional himself did not rely on them—what intellectual trends and cultural developments might have encouraged their increasing reliance on written records (whether self-produced or acquired through the burgeoning book trade), so that in time (towards the late fourth century and beyond) their performance largely depended on scripts memorized and rehearsed for delivery? Strictly speaking, training and performance that do not use writing might still depend on the slavish memorization of a fixed, but orally transmitted, uniform text, possibly written down and archived where it may be consulted and appealed to by disagreeing schools of performers. [8] For reasons of historical plausibility, [9] however, and in agreement with Nagy’s evolutionary model of textual fixation, I assume instead that at the transcript stage Homeric bards composed their poems in performance using traditional language, themes, and sequence, and that their technique produced a measure of variation consistent with long-standing traditions of composition and delivery. In other words: there was no controlling written ur-text, and memory, always of capital importance, was primarily not an instrument of slavish reproduction, but an aid to a traditional sort of creativity that we may call, not without paradox, ‘traditional improvisation.’In this and the following chapters I explore the transition from transcripts to scripts as follows: first, by situating rhapsodes (ῥαψῳδική) against actors (τραγική) and orators (ῥητορική), highlighting the connections that existed between these three performance trades and their professions, connections recognized already in antiquity; second, by focusing on orators and rhetorical theory in the late fifth- and fourth-century BC Athens and on the cultural forces responsible for the, at the time, polemical introduction of writing into their practice: I consider the ancient terms of this debate and the causes that brought about this disputed development; third, by arguing that, after making the necessary generic adjustments, one can recognize how similar social trends and expectations impinged on rhapsodes and their art, and to this we owe, at least in part, the move from transcripts to scripts in the performance of the Homeric epics. In Part Ⅳ (pp. 435ff.) I also examine the late fourth-century age of Lykourgos, the Athenian statesman, and of Demetrios of Phaleron, the autocratic pro-Macedonian ἐπιμελητής, both historical figures who, I believe, reinforced the cultural dynamics already at work, hastening the transition from the transcript to the script stage of rhapsodic performance. I end with a survey of the classical, Hellenistic, and Roman imperial records, documentary and epigraphical, on epic performers, many of whom will have declaimed the Homeric poems in a largely fixed, canonical form, with very little, if any, exercise of compositional creativity; and, to that extent, they will have limited their personal artistry largely to the manner of delivery before the audience—voice, dress, gestures—at times, but for the absence of a mask, nearly indistinguishable from the stage presence of tragic and comic actors.

9.2 The Rhapsode as Ὑποκριτής

It was a commonplace of classical antiquity that tragedy had its roots in epic; [10] it is therefore only natural that the acting profession would have looked back to the rhapsode as a model. Even the term for ‘actor’, ὑποκριτής, reflects this relationship. Homeric usage connects the cognate verb with the interpretation of signs and dreams. [11] One can therefore affirm that originally ὑποκριτής and ὑποκρίνεσθαι must have had their home among μάντεις, προφῆται, and others responsible for interpreting divine oracles to inquiring seekers. As Else (1959:101–102) notes, the activity denoted by ὑποκρίνεσθαι often took place in the context of rendering an interpretive judgment in reply to a questioner’s concern. [12] Koller (1957:102) agrees in substance, glossing the verb as ‘to decide for, explain, or clarify to someone else’. The derivatives ὑπόκρισις and ὑποκριτής are only attested much later. The former, first in Pindar’s fr. 140b Sn-M, [13] sometimes paraphrased ‘in the manner of’, already shows role-playing as an established meaning. [14] The latter first appears in the inscriptional Fasti (IG Ⅱ/Ⅲ2 2318) in connection with the actors’ contest instituted in 449 BC at the Dionysia; [15] and a few years later in Aristophanes’ Wasps 1279, dated to 422 BC, where its meaning cannot be settled by an appeal to either its context or the scholia. [16] ὑποκρίνεσθαι is arguably not a natural choice for stage acting: neither etymology nor the earliest attested meaning commends its use in this setting. Therefore, as Koller (1957:103) pointed out, we must surely assume—the absence of direct evidence to this effect notwithstanding—that ὑποκριτής was already in use as the nomen agentis applied to the one who rendered the answer of the deity by oracular pronouncement or clarified its meaning or both. [17] Homeric usage makes clear that this can be the only natural setting for the term.

But how, then, does the actor come to be named ὑποκριτής? On the basis of Plato’s Iōn, Koller concluded that rhapsodes had long been called ὑποκριταί of Homer, not in the sense of ‘dramatic actors’ but of ‘expounders’ of his poetry, and that this ‘explanation’ is the alleged original burden of tragic actors. [18] Although I agree with the general outlines of Koller’s argument as it regards the rhapsode, my own detailed examination of this question below (see Appendix) yields a somewhat different rationale for the application of the label ὑποκριτής to the actor. For now, I will only adduce two well-known passages that support Koller: Iōn 532d7, where Sokrates refers to Ion and his ilk as ὑμεῖς οἱ ῥαψῳδοὶ καὶ ὑποκριταί; [19] and Iōn 535e9–536a1, where Ion, ‘the middle ring’ in the chain of inspiration, is once again described as ὁ ῥαψῳδὸς καὶ ὑποκριτής. [20] One may counter that there is something of a stage quality to Ion’s performance—this is true, and more on this below. [21] But he is surely not an actor in the traditional sense, however dramatic his delivery. By Plato’s time, ὑποκριτής in the sense of tragic or comic actor was well established. And after painting the portrait of an epic performer who disavowed interest in any poet other than Homer, we must surely expect that, to remain in character, Plato would have had Ion protest as demeaning the epexegetic καὶ ὑποκριτής that Sokrates repeatedly appends to ὁ ῥαψῳδός if it could only suggest a generic association with poetry other than Homer’s. [22] Not so: ὑποκριτής was a traditional description of the rhapsode, and this agrees with Ion’s unequivocal insistence that an essential part of his profession, on which he expended the greatest effort, was speaking well about Homer (περὶ Ὁμήρου λέγειν): [23] he must not only know the poet’s ἔπη but his διάνοια (530b10–c1); he must both declaim the former and be a ἑρμηνεύς of the latter (530c3–4, 535b2). [24]

We do not know what this rhapsodic exposition of Homer was like, and sadly Sokrates prevents Ion from making a demonstration, so we are left to conjecture. It is an error, however, to argue (as some have) that Ion being our only evidence for such an exegetical rhapsodic practitioner, he must be an idiosyncratic creation of Plato, peculiar to this dialog and without a real-life parallel. Surely the force (and point) of Plato’s dialog would be lost if Ion were so unrepresentative of the rhapsodes of his day. And one could only judge the focus of his argument misconceived, with its unrelenting insistence on his superb ability to ‘expound Homer’, [25] ‘speak about Homer’, [26] and ‘adorn Homer’ [27] (not, nota bene, ‘sing Homer’ or ‘declaim Homer’). For ‘to sing’ and ‘to declaim’ the verbs ᾄδειν, καταλέγειν, or even λέγειν—but λέγειν τὰ Ὁμήρου [ἔπη], not περὶ Ὁμήρου—would have been quite adequate and obvious. But only on two occasions is ᾄδειν used of rhapsodes (at 532d8 and 535b4) and we find λέγειν with ἔπη just once (at 535b2). [28] The emphasis on exposition and critical commentary is everywhere evident: rhapsodes learn not only Homer’s ἔπη but also his διάνοια (530b10–c1); a good rhapsode must understand (συνείη) what is said by the poet (530c2–3); Ion classes himself with Metrodoros and Stesimbrotos (530c9–d1); the verb ἐξηγέομαι is applied to his interpreting what Homer and Hesiod say ‘regarding divination’ (531b5–6); Ion’s being ‘terrific at Homer’ (περὶ Ὁμήρου δεινός) involves competent judgment (κριτὴς ἱκανὸς 532b5; cf. 538d4–5, 539d3); Sokrates compares Ion to one who gives his opinion (ἀποφήνασθαι γνώμην) about a painter (533a4); and he accuses Ion of not making a demonstration (ἐπιδείξειν) of all he knows about Homer (541e3–4; cf. 542a2–3). [29] Even the topos of the rhapsodes’ stupidity, [30] because they knew Homer’s ἔπη by heart but were ignorant of their meaning, [31] must be regarded as proof that they regularly did attempt an exposition: for how else would their ignorance be so obviously exposed? This, to me, does not seem a case of revealing their deficiency in private conversation only; the accusation has an official ring to it, as if their excellence at performing and poverty in expounding were of a piece with their trade. The contempt in which, according to Xenophon, the milieu of Sokrates held rhapsodes must surely be set in the context of the challenge that rhetors, sophists, and philosophers variously mounted to their cultural authority as custodians of the traditional educational curriculum. Far from standing alone as an exceptional distortion of a tendentious Plato, the Iōn corroborates this state of affairs. [32]

Now, how are we to think of this hermeneutic function? Koller (1957:105) suggested that it consisted of “prose speech, mixed with verses, portions of verses, kola, kommata, with all the characteristics of poetic speech,” in sum, “a mixture of poetic speech and everyday prose.” [33] This, he proposed, represented the origin of ‘literary’ prose—though a very poetic one at first [34] —and hence there was a direct line extending from rhapsodic to sophistic ἐπίδειξις. I believe that the outlines of this proposal are essentially correct, and I would only modify it by conjecturing that initially the explanatory function of the rhapsode would have largely consisted of hexameters composed in performance, joining well-known episodes or speeches, effecting transitions between them, elaborating the twists and turns of the poems’ plots, as well as the thoughts, feelings, and motivations of participating gods and heroes. [35] On one level, τὸν Ὅμηρον κοσμεῖν, ‘to adorn Homer’, [36] must have denoted such an elaboration: not the work of an autonomous creativity expressing a rhapsode’s individuality, but the traditional recomposition controlled since time immemorial by the notion of ‘order’, i.e. the κατὰ κόσμον explored in chapter 7. [37] But with the passing of time and the increasing theatricality of rhapsodic performance, the practice of ornamenting Homer would have been gradually reinterpreted, partly as elaborate ὑπόκρισις, [38] partly as the metaphorical adornment of ‘praise’. Ion’s gorgeous attire answers to the former; [39] rhapsodic ἔπαινος, which I shall consider presently, to the latter. [40]

The scenario I have outlined here would readily explain Plato’s apparent blurring together of the inspired poet (notionally the source of the tradition) and the rhapsodes (notionally the performers and expounders). Indeed, were we to reduce the bard’s performance merely to reproducing (as a mechanical feat of memory) what purported to be Homer’s ipsissima verba, how could this action even begin to suggest the sort of manic inspiration Plato foists on rhapsodes? There would hardly be anything mysterious and godlike in a feat of memory, for Xenophon makes clear that the layman, too, could boast in his ability to recall the entire poems. [41] To Ion’s puzzled question, ‘How can it be that when someone talks about another poet, I pay no attention, cannot contribute anything worthwhile, and just doze off, but when Homer is mentioned I am immediately awake, pay attention, and have much to say?’, Sokrates’ answer must be, ‘Is it not obvious? Because you have only memorized Homer!’ Memorizing thousands of lines of poetry, however impressive an accomplishment to the modern, is quite different from the notion of ‘social memory’ embodied by the poetic tradition. The latter alone is shrouded in mystery, for it reenacts the past as only autopsy or divine omniscience can (Apollo’s or the Muses’; cf. Β 484–486 θ 487–491). [42] On the lips of the poet a claim to that memory fully justifies some notion of divine influence. One can hardly say the same when access to the tradition is gained by the hard but ordinary labor of a capacious memory exercised in retaining a widely accessible text. It would be hard to tell in that case who looks more foolish: Ion, for asking the question, or Sokrates, for failing to give the obvious answer. On the other hand, an exposition of Homer strictly as a prose lecture would not suffice: without the creative contribution of the rhapsode’s own poetry [43] it is again hard to see how the matter of inspiration would be apposite and, if it is, why it should not be made extensive to the likes of Metrodoros of Lampsakos and Stesimbrotos of Thasos (Iōn 530c9–d1). An understanding of rhapsodic practice along the lines I have just described accords a unity to the Iōn that has been denied to it by many a scholar. Whether poetic inspiration or rhapsodic exposition was its theme has been debated by many, and various strategies devised to subordinate the one to the logic of the other. [44] Yet the troubling dichotomy largely dissipates if rhapsodic exposition shares to some extent the character of poetic composition. This explains why the composing poet and the performing (declaiming and expounding) rhapsode can be handled under a single scheme and assigned the same kind of manic inspiration. At the same time, it preserves the notional priority of the poet as source (author and authority) of the bard’s performance, for only through ‘Homer’ does Ion have access to the Muse. The afflatus is mediated, and passes from poet to rhapsode; but, by the same token, the rhapsode is not without his creative inspiration.

Over time, the verse component must have given way to an increasingly large proportion of explanatory prose, a development that would readily account for the implicit comparison of rhapsodes with sophistic orators, who made a living of epideictic displays and educational lessons for a fee and for whom Homeric material often furnished a point of departure. [45] These sophists may have proceeded at first in conscious imitation of, and competition with, rhapsodes. As O’Sullivan (1996:117) writes (displacing the focus from rhapsodes onto poets): “Although rejecting the mode of communication the poets had adopted, and although subjecting them to searching and hostile criticism, the Sophists also seem to have imagined themselves as their successors.” I believe that O’Sullivan’s emphasis on “mode of communication” (prose versus poetry) is misplaced and that the real fight was about cultural capital and authority, especially in education, broadly conceived. [46] Hence the contempt in which rhapsodes were held by the Greek intelligentsia of the time. As immediate exponents and (at least implicit) claimants of performative authority, rhapsodes made readier and more vulnerable targets than their poetry or—once their traditional repertoire had been ascribed to a distinct author—than the poets whose work they alleged to perform. O’Sullivan is doubtless right to infer from the following facts a live rivalry between sophists and rhapsodes: [47] Protagoras apud Plato identified Homer and Hesiod as sophists (Prōtagoras 316d6–7; cf. Kratinos fr. 2 K-A); like rhapsodes, sophists dressed in purple robes (Aelian Varia historia 12.32); both appeared at the same festivals (Philostratos Vitae sophistarum 1.9.493 on Gorgias; Plato Lesser Hippias 363c7–d4 on Hippias); and sophists composed speeches and dialogs featuring Homeric characters. [48]

9.3 The Rhapsode as Ἐπαινέτης

One aspect of the Iōn where the connection between the rhapsodic and sophistic performance traditions comes clearly to the fore is Plato’s use, characteristically pointed, of the terms ἐπαινέω and ἐπαινέτης. [49] On four different occasions do the interlocutors employ them. The first comes after the celebrated metaphor of the magnetic chain, which Sokrates closes with the following words: ‘Of this, the cause you are asking me about, why you are at a loss in regard to the rest but not in regard to Homer, is that you are a terrific epainetēs of Homer by divine dispensation, not tekhnē’; [50] to which Ion replies: ‘Well said, Sokrates; but I would be surprised if you spoke well enough to convince me that it is raving and possessed that I praise (ἐπαινῶ) Homer’. [51] The other two instances are in the closing exchange between the philosopher and the rhapsode: ‘But if you are telling the truth, Ion, that by art and science you are able to praise (ἐπαινεῖν) Homer, you are wronging me’. [52] Faced with this verbal challenge, Ion at last yields his point to Sokrates, who condescendingly ends the dialog with a declaration of victory: ‘This nobler title, then, you have in my eyes, Ion, to be an epainetēs of Homer not by dint of art, but divine’. [53] One should mark the technical nature claimed by Ion and contested by Sokrates for his ‘commendation’ of Homer. In 536d6, ‘I praise Homer’ (Ὅμηρον ἐπαινῶ) is paralleled by ‘my speaking about Homer’ (μου … λέγοντος περὶ Ὁμήρου 536d7). The mere fact that the agent noun ἐπαινέτης is used acknowledges that ‘praising Homer’ is an integral part of the rhapsode’s profession. Discharging this function well corresponds in 536d3 to being ‘well supplied about Homer’ (περὶ μὲν Ὁμήρου εὐπορεῖς 536d1–2). Sokrates makes clear at 536b8–c2 that this ample supply goes beyond recalling large amounts of epic verse; for, when Ion has plenty to say (εὐπορεῖς ὅτι λέγῃς), he speaks not Homeric verse but about Homer (περὶ Ὁμήρου λέγεις). It is plain, then, that the artistic and scientific quality that Sokrates denies to the rhapsodic profession cannot be reduced to mnemonics.

To these four passages from the Iōn we may add three others. In the Prōtagoras Sokrates gently chides a hetairos for his unwitting departure from Homer’s view of a youth’s charm when he first shows facial hair: ‘But are you not an epainetēs of Homer, who said that youth is at its most charming with its first beard, just the age of Alkibiades now?’ [54] And in the Republic Homer’s theology is censured for its questionable morality: ‘Though we praise much of Homer, yet this we shall not praise: Zeus’ sending of the dream to Agamemnon’. [55] But it is the next passage that makes clear the ultimate implications of ‘praising Homer’, the claim purveyors of culture—educators, rhapsodes, sophists—staked out in connection with the alleged social benefits of his poetry: ‘Surely then, Glaukos, said I, when you meet epainetai of Homer who say that this poet has educated Greece and that, with a view to the management of human affairs and our instruction therein, he is worth taking up, learning, and living one’s entire life arranged according to him; you must welcome and salute them for being as virtuous as they can, and agree that Homer is the Poet of poets and first among tragedians, but you must know that of poetry, only hymns to gods and encomia of virtuous men should be admitted into the city’. [56] The point made here by Homer’s ‘supporters’ (the epainetai) is clearly one of pedagogical preeminence, comprehensiveness, and absolute sufficiency. It is hard for us fully to grasp the foundational significance of such a totalizing cultural narrative, whose impact descends into the triviality of a bearded youth’s charm as easily as it ascends to statesmanship, military leadership, [57] and the most serious matters of state. [58] But it is precisely against this background that the word ἐπαινέτης acquires the connotation of the Homer enthusiast who makes his poetry the sum total of all necessary learning. This type is best represented by the rhapsode, who makes his living not only declaiming but also exploring and expounding to his audience the significance of his epic repertoire. Several scholars have sensed and commented on the marked character of the term in this context. So Stallbaum 1857a: “Nempe videntur Homeri ἐπαινέται non tam ii dicti esse, qui Homerum laudabant, quam potius illi, qui unice eius sapientiam probabant eamque ita commendabant, ut inde etiam vitae recte sapienterque regendae ac moderandae praecepta haurienda esse arbitrarentur. In quorum numero certe imprimis etiam Homeridae habendi sunt … . His igitur viris sapientiae Homericae consultis sese adiunxerunt sine dubio etiam rhapsodi …” (331–332). [59]

A good fourth-century BC illustration, not by a rhapsode but by a statesman who nevertheless adopts the epideictic pose of one, is Lykourgos’ Against Leōkratēs, dated to 330 BC, where Lykourgos the son of Lykophron (of the deme Boutadai) indicts a certain Leokrates for his flight to Megara right after the defeat of the Athenians and Thebans by Philip Ⅱ of Makedon at Khaironeia. [60] In §102 he commends to the jury the example of Hektor as follows: βούλομαι δ’ ὑμῖν καὶ τὸν Ὅμηρον παρασχέσθαι ἐπαινῶν. [61] As an ἐπαινέτης of Homer, Lykourgos extols patriotism and courage in fighting for one’s country. Velardi (1989:34–35) sees in the outline of §§102–104 the pattern that the more professional rhapsodic epainetai might have followed in praising Homer: “1) un breve discorso di carattere generale finalizzato ad introdurre i versi, una sorta di prologo contenente anche informazioni storiche sull’istituzione dell’agone rapsodico delle Panatenee e probabilmente, nel caso di una recitazione rapsodica vera e propria, notizie biografiche su Omero (par. 102); 2) la declamazione dei versi (par. 103); 3) un commento del brano recitato (par. 104)” (35). Though perhaps a little too schematic, if one allows for ad hoc adjustments to the particulars of each performance his proposal seems, in the main, sound for the late-classical rhapsode. [62] Indeed, the section that precedes Lykourgos’ commendation of Homer, with its focus on Euripides’ Erekhtheus, is explicit about the politician’s intent to make poetry serve pedagogical ends (ταῦτα ὦ ἄνδρες τοὺς πατέρας ὑμῶν ἐπαίδευε §101); further emphasis flows from the ensuing contrast between laws that, owing to their brevity, cannot educate (διὰ τὴν συντομίαν οὐ διδάσκουσιν §102) and poets who, by mimesis of choice deeds, persuade men with logos and apodeixis (equivalent, respectively, to drama’s speech and argument on the one hand, and stage production or opsis on the other). [63] The goal is the public demonstration, both ἐπίδειξις and the related ἀπόδειξις, which, as Koller (1957:105) argued (see above, §9.2), made the sophists the intellectual heirs (and competitors) of the earlier rhapsodes. [64]

9.4 The Rhapsode and Ὑπόκρισις

So far we have considered how the rhapsodic trade shaped not only the emerging actor, but also the sophist (who was, in effect, the first professional orator and teacher of oratory). But this is not the only direction in which influence was exerted. In Athens tragedy soon rivaled and even surpassed epic poetry in popular appeal as the queen of all cultural productions, and in time it also exerted a profound reciprocal influence upon the performance of Homer. Evidence for this can be found in many places, including the Iōn itself, for, as I already observed, Ion’s performance was powerfully dramatic, with an exaggerated evocation of πάθη, [65] especially ἔλεος and φόβος (535c5–8), the very emotions Aristotle singles out in connection with the κάθαρσις of tragedy (Poetics 1449b27–28). [66] Plato, who is never straightforward, must have been playing with the double referent of ὑποκριτής: not only the traditional one, ‘expounder [of Homer]’—which is surely the way we are to think Ion heard it—but the far less flattering ‘stage actor’, with all the negative mimetic nuances the philosopher assigned to it. This, Plato’s demeaning ὑπόνοια, would tend to reduce poetry to an instrument for arousing disorderly emotions, primarily through a markedly mimetic delivery. The same unfavorable implication is found in Aristotle’s Poetics (chapter 26), where, while criticizing exaggerated acting, the philosopher observes that even an epic rhapsode may overdo his gestures.

This pairing of rhapsode and stage actor resurfaces in Aristotle’s Rhetoric Ⅲ.1, where we learn that the matter of ὑπόκρισις, ‘oral delivery’, had only lately come to τραγική and ῥαψῳδική. [67] Happily, the philosopher does not leave us in doubt as to his meaning, but explains that ὑπόκρισις lies in the voice, how its loudness, melodic line, and rhythm should be used to convey πάθος in every circumstance. [68] It is clear, however, that for him delivery was inextricably linked with writing. [69] For in defining the scope of his investigation of λέξις he specifies that it comprehends such principles as can be assembled into a scientific study of ‘delivery’; [70] and to illustrate this distinction he remarks that ‘written speeches are more effective on account of their lexis than their thought’. [71] This comment is most naturally construed to involve writing in the successful deployment of all the expressive resources of ‘style’, resources that Aristotle views strictly as instruments in the service of effective delivery. Chapter 12 of Rhetoric Ⅲ shows that writing is not a passing concern, even if it seems superficially paradoxical in the context of ὑπόκρισις; for here a distinction is drawn between the graphic and agonistic styles, the former lending itself best to precision, the latter to delivery. [72] It soon becomes clear, however, that he assumes that both will be written, and that the distinction between them is only how successful a reader will judge the written text, now deprived of its performative setting and voice, as compared to the hearer who sits in the audience when it is delivered. [73] Despite Aristotle’s strong emphasis on oral delivery, it is telling that he should state that ‘one thing is to know how to speak correct Greek; another, not to be forced to keep silent, should one wish to share with the others—which is precisely what happens to those who do not know how to write’. [74] The connection the philosopher establishes between rhapsodic, dramatic, and oratorical ὑπόκρισις in Rhetoric Ⅲ.1 is all the more significant in that it belongs to a technical discussion with careful definitions and analysis, one that explicitly considers the service that writing can offer to what is a preeminently performative task: the oral delivery of rhetorical speeches. This parallel, I believe, suggested itself to Aristotle because in his own time these professionals of public performance shared the broad outlines of a common methodology, and this as much in their preparation and rehearsal as in their delivery before an audience. We readily understand that, however different in the particulars, the delivery of actor, rhapsode, and orator were but three species of one genus: they all had recourse to the same resources—voice, gestures, and outward appearance—through which they were to express ēthos and pathos. But the recurrent late-classical association of rhapsodes with ὑποκριταί and ὑπόκρισις, and the contexts in which this happens, also confirm that, by this time, they shared with actors and orators an important aspect of their training: the use of written texts (first in the character of transcripts, then of scripts) to prepare for and secure a successful performance. This was an innovation gradual in coming, for since time immemorial the Homeric rhapsode had relied on recomposing epic poetry in performance; but, as we shall see, it was also a departure from the original practice of orators, which some of the more seasoned among them hotly decried as enervating and harmful to the speaker and his art.

9.5 Alkidamas’ On the Sophists

Given the relation between sophists and rhapsodes I have argued for and the role of the latter as performers of poetry, we might anticipate that Alkidamas may mention ῥαψῳδική as a comparandum to the orators’ use of writing. And indeed we are not disappointed, as we shall see below. But consider first how different the talents are that improvisation and scripted delivery call for. [85] The former requires speaking fittingly about the happenstance (περὶ τοῦ παρατυχόντος), a swift articulation of thoughts and arguments, a felicitous abundance of vocabulary (the resources of language at the level of expression), aiming accurately at the ‘opportunity’ of the matter at hand (καιρός), and a sensitive speaker-audience interaction that takes adequate account of the desires of the hearers and their shifting moods: in view is τὸ προσῆκον, a goal open indiscriminately neither to every nature nor to whatever training the orator may chance upon. Writing, on the contrary, enjoys leisure of composition (κατὰ σχολήν) and length of time (ἐν πολλῷ χρόνῳ), and benefits from comparing and collating other sophists’ συγγράμματα, culled for compelling thoughts (ἐνθυμήματα) and felicitous expressions (τὰς τῶν εὖ λεγομένων ἐπιτυχίας). Add to these correction, revision, and rewriting (this last, ἀνακαθῆραι, literally, ‘cleaning up’ the text), and we have the full apparatus of a developed editorial practice.

We can easily transfer this to the rhapsode. He would have been schooled from youth up in traditional composition in performance: his παιδεία was demanding and called for rare natural talent. [86] He would declaim a sort of improvisation that remained within the parameters of traditional themes, diction, and sequence. The unexpected would arise from interaction with his audience—whose desires he would seek to respond to and gratify—and from competing with other rhapsodes under time constraints and the rules that regulated transitions between one bard and the next (e.g. as to thematic continuity). Depending on the format, interruptions from the audience or amoibaic rhapsodic exchanges may not be out of the question. Speed of thought and expressive resourcefulness would be crucial to his success. Conversely, the leisurely drafting and memorizing of a script for oral delivery, away from the competitive pressures of actual performance and assisted by previous transcripts, in search of apposite ἐνθυμήματα and the felicitous verses of others—all the while correcting, revising, and rewriting, both on the advice of others and his own—would have been to the advantage of any who aspired to be a rhapsode but had not enjoyed the life-long apprenticeship that was the sine qua non of recomposition-in-performance. The same elitist criticism may be levied in this case: we would have, on the one hand, what called for rigorous, life-long παιδεία and was only open to singularly gifted talent; and, on the other, what was readily acquired, derivative, and easy even for the uninstructed (ἀπαίδευτοι).

Strictly speaking, these two strategies were not mutually exclusive. Surely, the ability to improvise must have spanned a range. Some performers may have employed texts only as transcripts, to be used in rehearsal not for strict memorization but to hone improvisatory delivery. Memory, after all, remained always an essential tool even of the most creative extempore traditional performance, for which Homeric diction functioned as a language, albeit a special one—just as memory is essential for any natural-language ability. [87] Those orators who did not depend primarily (or even regularly) on writing could, by a change in what Alkidamas calls a ‘frame of mind’ (τὴν τῆς ψυχῆς ἕξιν μεταρρυθμίσαντες §6), draft fitting scripts for their speeches. The same, I think, might be said about the production of poetic transcripts by classical rhapsodes who exhibited a limited and occasional dependence on writing. Ultimately, the opposition between αὐτοσχηδιασμός and scripted delivery was one of opportunity versus time and leisure (i.e. καιρός versus χρόνος and σχολή), and it marks a shift away from the creative primacy of the performative setting. Improvisation possessed flexibility, but also uncertainty; scripts, in turn, could only thrive where competitive rules minimized the unexpected and put the emphasis on the stylistic finish of the delivery, with the corresponding depreciation of the skills that must have been the guarantee of victory at the ἀγῶνες among the more traditionally schooled singers: responding to audience feedback and interruptions, thematic contraction and expansion to adjust to the available time and the interests of the hearers, mastery of relay poetics, [88] and so on. It must be the case, therefore, that the transition from transcripts to scripts was facilitated by competitive rules at the Panathenaia that enforced an increasingly fixed thematic sequence. This sequence must have been predictable enough before the actual performance to give a competitive advantage to rhapsodes who chose to draft their recital in advance and commit it to memory.

Unfortunately, we are too poorly informed to know even the outlines of what must be a crucial piece of this cultural puzzle: the change in the extraction—and, hence, the training—of the rhapsodes that competed at the Athenian Panathenaia. [89] The fame of Ionian bards was well established, and we must assume that among them the traditional skill of extempore recomposition still flourished during the classical age. According to a dominant tradition, Homer himself hailed from the region, specifically from Khios, and thus it is only to be expected that, for his dialog with Sokrates, Plato would choose a bard from a city like Ephesos. But the love of Homer, the distinguished place his poetry held in Athens (emphasized by Lykourgos in Against Leōkratēs §102), and the social and material benefits to successful prize winners must have tempted the more ambitious Athenians to try their hand at the competition. [90] Since the Iliad and the Odyssey were the pedagogical mainstay (Isokrates Panēgyrikos 159), attaining a level of proficiency adequate for a festival appearance might have seemed to aspiring performers not to go much further than committing to memory large portions of the poems. A book-trade copy might serve as a handy script; to be sure, concentration and recall would have to be honed, but the creative center of gravity would be in the delivery (i.e. in the use of voice, gesture, and dress). Extraordinary facility and a quick mind might enable one thus trained to save the day by supplying a suitable line of his own making here and there, should his memory suffer a minor lapse in performance. But this would no longer qualify as true traditional oral composition. There may be a hint of this state of affairs in Xenophon Αpomnēmoneumata 4.2.10, where Euthydemos’ imposing collection of Homeric writings (τὰ Ὁμήρου ἔπη πάντα) gives rise to Sokrates’ gibe that perhaps he wishes to become a rhapsode. And from Symposion 3.5–6 we learn that Nikeratos’ father, wishing to raise his son to be an ἀνὴρ ἀγαθός, had forced him to learn all of Homer’s epē (πάντα τὰ Ὁμήρου ἔπη μαθεῖν); apparently, this called for daily exposure to rhapsodic recitation (πῶς ἄν … λελήθοι ἀκροώμενόν γε αὐτῶν ὀλίγου ἀν’ ἑκάστην ἡμέραν;), [91] an apprenticeship of such a sort as an ambitious Athenian of means might have been able to procure for himself. In fact, some have even suggested that this may be the same Nikeratos who according to Aristotle (Rhetoric 1413a6–9) was bested by Pratys at a rhapsodic competition. [92] If Xenophon’s report in the Symposion is accurate and his and Aristotle’s Nikeratos are one and the same, Nikeratos must have competed as an amateur or else, despite his earlier criticism of rhapsodes, he must have decided to become one himself. [93]

It is curious that scripted material is likened here to ὑπόκρισις for its precision, whereas in Aristotle’s Rhetoric Ⅲ.12 it is the agonistic style—the one contrasted with the graphic as comparatively less precise—that is considered the most appropriate for delivery. This is primarily a matter of emphasis, for Rhetoric Ⅲ.1, as we shall see below, [98] also associates ὑπόκρισις with the precision and technical competence that follow from reducing delivery to an art (in particular, to an art that makes use of writing). [99] Nevertheless, the varying emphasis shows that, conceptually, Alkidamas stops short of the more developed notion of a style enabled by writing to which Aristotle takes us, namely, the ‘graphic style’, best represented by epideictic speeches. For Alkidamas, just as the orator as writer of speeches resembles a poiētēs, a ‘speech maker’ or ‘poet’, and his speech recalls a poiēma sooner than a logos, so also in delivery he smacks more of a rhapsode than a rhetor. Though my argument here does not turn on the precise meaning of ὑποκρίσει καὶ ῥαψῳδίᾳ, I rather incline to ‘rhapsodic hypokrisis’ in the sense of ‘performance’. [100] Indeed, one need not read into Alkidamas the terminological distinction between tragic and rhapsodic delivery Aristotle observes in Poetics 26 and Rhetoric Ⅲ.1; and though ‘poet’ may well lurk behind the word ποιητής (§34) and Isokrates shows us that ποιητὴς τῶν λόγων need not mean more than ‘maker’ or ‘composer of speeches’, [101] the emphasis clearly falls on careful crafting that almost certainly involves writing, yet all the while is oriented towards delivery. [102]

9.6 Ῥαψῳδέω in Isokrates and Plato

Isokrates and Plato both use ῥαψῳδέω in a sense broader than the declamation of verses learned by rote. [103] Indeed, at Panathēnaikos 18 [104] Isokrates lampoons some in the Lyceum—whom he calls sophists—who claimed to know everything and were quick to show themselves everywhere; in discussing (διαλέγοιντο) the poets, especially the poetry of Hesiod and Homer, they contributed nothing of their own but merely rhapsodized the poets’ material (τὰ δ’ ἐκείνων ῥαψῳδοῦντες) and called to mind (μνημονεύοντες) [105] the most sophisticated things that others before them had said. [106] Though satirical in tone, we recognize the basic description of rhapsodes presented by Koller 1957 and defended above: itinerant, they not only declaim what are notionally the verses of Homer (or another poet) but also expound them as modern scholars would expect a sophist to do. This suggests, once again, that the boundary between rhapsode and sophist was more permeable then than it seems to us now. Note also that they were performing before an audience (perhaps small, but a real audience all the same), [107] and that the sophists’ ensuing criticism of Isokrates (§19) pertained to his view of poetry and its precise role in education. [108] Some believe that Isokrates reserves ῥαψῳδέω strictly for the verse (the ἔπη) and uses another verb for the exposition (here μνημονεύω, for what is alleged to be rote learning). But without an immediate specific referent for the τά of τὰ ἐκείνων, [109] its indefinite character encourages us to think of the expository part as of a piece with their rhapsodizing.

I am not, of course, denying that the sophists also declaimed poetry. My point is only that ῥαψῳδέω by itself (without the aid of μνημονεύω) would have sufficed to denote the interspersing of declamation and commentary that was characteristic of the rhapsode. The clause with μνημονεύω tells us about the caliber and authorship of the comments, but, as the structure of the passage makes clear, the presence of interpretation does not hinge on it. Indeed, the main verb, διαλέγοιντο, is qualified by a participial μὲν … δὲ opposition: on the μέν hangs ‘saying nothing of their own’; the δέ, in turn, subdivides into ‘rhapsodizing’ and ‘mentioning’. If ‘rhapsodize’ referred strictly to the declamation of ἔπη, it would be an unnecessary intrusion into the logic of the passage; for it would have been sufficient to say that ‘they were talking about the poetry of Homer and Hesiod, saying nothing they could claim as their own, but merely mentioning the finest thoughts others before had said about them’. On this assumption, ‘rhapsodizing their poetry’ would clearly go beyond ‘they were talking about their poetry’ and would hang limp as a curious addition to the polarity ‘not their own thoughts but what others had said’. Moreover, how could ‘reciting their verses’ be fairly held against the sophists as indicative of their lack of originality? If, however, ‘to rhapsodize’ means ‘to treat in the manner of a rhapsode’, and rhapsodes, as a rule, both declaimed poetry and commented on it; and if, among sophists, rhapsodes were infamous for their intellectual mediocrity and lack of original thinking (and they arguably were)—then it would have made sense to note, in a show of contempt, that they were ‘saying nothing they could claim to be their own original thinking, but merely treating the poets’ material with the same mediocrity rhapsodes are wont to, compensating for this deficiency with the most sophisticated commentary of earlier thinkers.’ Then ῥαψῳδέω does not hang limp as a superfluous detail, but becomes central to the criticism Isokrates levels against the sophists.

At any rate, whatever else we may say, the entire episode is summarized by διαλέγοιντο, which unites under one conceptual label of verbal exchange both the poetry itself and its exposition. This episode is anomalous only to this extent: the performance dynamics were such that the outcome was a conversation of sorts (rather than a declaimed monologue or a series of them); [110] and yet, though the interlocutors are clearly in basic agreement, the dialog may well have had a competitive dimension (each trying to outdo the others). [111] In a more formal agonistic setting, a sequence of rhapsodes taking in turn to the βῆμα would answer to such emulous repartee. In Panathēnaikos 33, [112] looking back on the same event, Isokrates reiterates his characterization of the sophists; only, he now substitutes μνημονεύοντες by ληροῦντας περὶ αὐτῶν, disparaging as sheer prate the exposition of these sophists (just as was done so often to rhapsodes). [113] Once again, when Isokrates in this passage writes of silencing them, he does not merely say ‘I think I could silence those who prattle about them’, but ‘those who rhapsodize their things (τἀκείνων) and prattle about them’. Thus, in effect, he witnesses once more to the unity of declamation and commentary that underlies Sokrates’ conversation with Ion the rhapsode. Isokrates’ insistent use of ῥαψῳδέω in connection with the sophists supports the view that he is drawing attention to their manner of commentary, not just their (otherwise unobjectionable) declamation of poetry. Since sophists arguably despised rhapsodic commentary, this description amounted to an effective put-down. It is not clear whether the αὐτῶν is neuter or masculine plural, i.e. whether it refers to the τά of τἀκείνων or the ἐκεῖνοι. If to the former, ληροῦντας περὶ αὐτῶν would overlap in sense with ῥαψῳδοῦντας τἀκείνων for added emphasis; if to the latter (Norlin’s choice in the Loeb Classical Library), it would stand for ‘these poets’ and refer more specifically to the biographic accounts that ancient scholars, including rhapsodes, were so fond of. [114]

Here Sokrates denies serious consideration to written speeches (ἐν μὲν τῷ γεγραμμένῳ λόγῳ): hence γραφῆναι οὐδὲ λεχθῆναι cannot represent two independent alternatives. At issue is the fixity of writing, its inflexibility, which makes it unsuitable to dialog: hence we need both γραφῆναι and λεχθῆναι. To illustrate the point, Sokrates mentions ‘rhapsodized speeches’ (οἱ ῥαψῳδούμενοι [sc. λόγοι]), which are spoken (ἐλέχθησαν) [117] without questioning (ἀνάκρισις) or instruction (διδαχή). [118] The words οὐδὲ λεχθῆναι ὡς οἱ ῥαψῳδούμενοι ἄνευ ἀνακρίσεως καὶ διδαχῆς πειθοῦς ἕνεκα ἐλέχθησαν, which Burnet’s OCT isolates with commas as a separate clause, are read by some as an afterthought. Yunis (2011:238) illustrates this reading, which I believe neither the context nor the overarching argument of the dialog commends: “no speech in verse or prose that is worth much serious attention has ever yet been written or even spoken as the [speeches] performed by rhapsodes are spoken for the sake of persuasion without [oral] examination and teaching.” [119] According to Yunis, Sokrates here “extends the category of non-serious discourse beyond written discourse to include oral discourse that eschews individual engagement and the dialectical pursuit of knowledge” (2011:238, my emphasis). I submit that Sokrates does not have an ‘extension’ of his argument in view, but rather illustrates the deprecated mode of communication—oral delivery that is seen as either subservient or else inferior to the written word—with its foremost exponent in the culture of his time: rhapsodic delivery of epic poetry and commentary that, by this late time in the chronology of textual fixation, was largely (perhaps often decisively) dependent on scripted performance. Yunis has no warrant for introducing “even” in “written or even spoken,” an insertion that serves to sever ‘written’ from ‘spoken’. οὐδέ cannot be both conjunctive and adverbial. [120] His translation has made it into both by glossing it as the conjunctive “or” and adjoining the adverbial “even.” [121] Since standing between two infinitives it is surely a conjunction, the only question here is whether, following as it does ‘not one’ (οὐδένα), its meaning is that of a disjunctive ‘or’ or a conjunctive ‘and’. If the former, we have the negation of both terms of an alternative; if the latter, we have the negation of both members of a combination. [122] The clause is negated by the οὐ of οὐδένα, while the οὐ of οὐδέ merely confirms the first negative (Smyth §2761). That οὔ … οὐδέ can have a conjunctive sense, i.e. ‘not (A and [then] B)’, is clear e.g. from Demosthenes 1.8: ‘You should not (οὐ δεῖ), men of Athens, dismiss such an opportunity as has befallen you and [in consequence] have that happen (οὐδὲ παθεῖν) to you which you have already experienced many a time before’. [123] From the examples Demosthenes goes on to adduce, it is clear that what he thinks the Athenians will ‘suffer’ if they waste the opportunity is the loss of power and prosperity consequent on inaction. [124]

Several textual and contextual indices strongly suggest that, throughout the conversation in the Phaidros, Sokrates views writing and speaking as complementary aspects that are integral to the process of public delivery. On several occasions we find τε καὶ joining forms of λέγειν and γράφειν. Thus, at 258d Sokrates remarks that writing speeches is not in itself shameful; what is shameful is ‘not to speak and write them well’ (τὸ μὴ καλῶς λέγειν τε καὶ γράφειν 258d4–5). In its context, the use of a single article τό with two infinitives joined by τε καί suggests (although admittedly it does not require the view) that these verbs articulate two stages of a single process in which ‘writing’ is held to be of greater relative significance. [125] This is what one would expect if in fact the focus is on the shortcomings of writing as a mode of communication. Sokrates could have pointed in the direction of two independent alternatives by saying λέγειν ἢ γράφειν (cf. 277b6). It would have been, in fact, more natural to counter the charge that writing speeches is inherently shameful with the reply that what is shameful is writing speeches poorly. Why say instead that ‘what is shameful is both speaking and writing speeches poorly’? Why introduce the distracting and (ex hypothesi) not intrinsically germane oral delivery, which is otiose if it has no connection to the matter at issue, namely, the propriety of writing of speeches? The situation is otherwise and hardly puzzling if Sokrates is addressing oral delivery built upon, or seen as second best to, written scripts. This very subject already appears at the beginning of the dialog, where the scroll of Lysias’ speech takes center stage. Even the order of the infinitives, first λέγειν and then γράφειν, is commended by Sokrates’ argument. In 257d Phaidros observes that men of influence in their polities are ashamed to write speeches and bequeath their writings to posterity because they fear being branded as sophists. The focus here is arguably on deliberative speeches, a view confirmed by Sokrates’ witty and tendentious use of the subsequent inscribing of successful motions (the presumed core of a deliberative address) as a counter argument to his interlocutor’s observation (257e–258a). Although some deliberative speeches may have been partially written in advance, [126] ordinarily the public would not have thought of them as having preexisted the actual assembly address so much as, if at all, postdating it, their written text offering a possibly edited and polished version of what had transpired. [127] Not so with forensic speeches by logographoi, which, whatever the editorial modifications predating their ‘publication,’ were clearly understood as scripts to serve the litigant’s delivery at the trial. In other words, deliberative speeches would commend the order ‘both speak and write’; forensic, the order ‘both write and speak’. [128] Concerning the former, in 277d7–9, Sokrates also criticizes those who privilege writing as being superior to the spoken word in its capacity for ‘permanence’ (or ‘certainty’, βεβαιότητα) and ‘clarity’ (σαφήνειαν). The text (as punctuated, I think correctly, by the OCT contra Yunis 2011 ad loc.) suggests that, spurred by the supposed advantages of writing, the individual in question authors a document in the context of passing laws (νόμους τιθείς). It is possible that Sokrates is resorting again tongue-in-cheek to the inscribing of the assembly’s resolutions as the target of his criticism; but since he pairs this public dimension of writing with writing in a private capacity (cf. 258d9–10), it is more natural to take his statement to apply more broadly to publishing an edited version of a previously delivered demegoric speech. This motivates the order, first ‘speaking’ and then ‘writing’, in the formulation with which Sokrates retakes at 277d1–4 the investigation of whether speaking and writing speeches is honorable or base (περὶ τοῦ καλὸν ἢ αἰσχρὸν εἶναι τὸ λόγους λέγειν τε καὶ γράφειν 277d1–2).

It is true that in the latter part of the Phaidros, starting at 257d, Sokrates (for good Platonic reasons) is also preoccupied with the question whether the subject matter of rhetoric is true and not merely persuasive, irrespective of the intervention of writing at any stage in the delivery and of any supposed advantage of writing as a mode of composition and communication. This is not to say that he necessarily denies the possible involvement of writing so much as that his focus is elsewhere and he sidesteps the polemic altogether. Therefore, Phaidros 259e1–2, which summarizes the scope of the investigation thus, τὸν λόγον ὅπῃ καλῶς ἔχει λέγειν τε καὶ γράφειν καὶ ὅπῃ μή, σκεπτέον, precedes Sokrates’ insistence that one who would speak well and nobly must know ‘the truth about what he intends to speak’ (τὸ ἀληθὲς ὧν ἂν ἐρεῖν πέρι μέλλῃ 259e5–6). In good consequence, when the tekhnē of rhetoric personified upbraids the orator who is ignorant of his subject, she uses λέγειν and neglects to mention γράφειν. [129] But Phaidros soon brings the conversation back to the realities of Athenian life when he draws attention to the courts as the preeminent context for rhetoric; and he does so by reintroducing writing in tandem with speaking. When Sokrates in effect asks, ‘Is that what you have been told about the art of rhetoric?’, Phaidros replies: ‘No by Zeus, not at all so, but that it is chiefly in connection with lawsuits that there is speaking and writing by [oratorical] art, and speaking [by oratorical art] in connection with assembly speaking’. [130] Note the explicit involvement of ‘writing’ in the forensic setting and its corresponding neglect in the deliberative, which agrees with my suggestion above that the average Athenian would have considered ordinary the involvement of logographoi in lawsuits, while writing would not have been immediately associated with preparing and training for demegoric addresses, only with the exceptional publication in writing of assembly speeches after their delivery. [131] Sokrates, however, does not restrict his analysis to courtroom speeches, and when consideration of the role of writing in rhetoric is retaken as the focus of the dialog, its scope embraces private and public speeches (277d7), whether prose or poetry (277e7), with the explicit inclusion of passing laws and writing ‘political compositions’ (νόμους τιθείς, σύγγραμμα πολιτικόν 277d7). [132] In one of the few places where a possible advantage of writing is even envisaged (only to suggest that it does not exist), Sokrates gently mocks Lysias by hinting that perhaps there is a technical ‘logographic’ reason, inaccessible to the layman, for the arrangement of his written speech (264b7–8). [133] Once he finally addresses the limitations and impact of writing, the focus is on teaching through listening to written treatises or manuals. Attention to teaching is only to be expected, for it answers to Sokrates’ characteristic interest in deprecating sophistic rhetorical education in favor of dialectic.

Whatever insight we may derive from all this about the role of writing in rhapsodic performance in the fourth century BC follows inferentially or tangentially from the main thrust of the argument. At the same time, one must not fail to notice that the spoken word remains the vehicle for instruction—only, it is no longer the word of dialectic exchange but the reading aloud of written manuals. Faced with Theuth’s invention of γράμματα, Thamos observes: ‘You provide your students with apparent, not true, wisdom; for, once they become [mere] assiduous hearers of you (πολυήκοοι … σοι) without instruction, they will think that they are broadly learned, while for the most part ignorant and hard to be with because they have grown wise only in appearance’. [134] Thamos’ objection pertains to the manner of ‘instruction’: Theuth has discovered various branches of knowledge—number, calculation, geometry, and astronomy—but instead of teaching his disciples face-to-face, exchanging questions and answers, he commits these sciences to writing and expects the students to learn them by reading them aloud or hearing others read them. That this is the sense in which we are to take πολυήκοοι … σοι is confirmed by Plato’s Laws 7.810e11. Here the Athenian probes the claim that youths are to be educated by making them assiduous listeners of the poets as they are read aloud and by making them acquire through such exposure a broad knowledge of them (πολυηκόους τ’ ἐν ταῖς ἀναγνώσεσιν ποιοῦντας καὶ πολυμαθεῖς), even to the point of learning entire poets by heart. Not so; true ‘instruction’ calls for dynamic oral exchange between teacher and pupil. Therefore, Thamos asserts that the students’ listening to Theuth’s writings is ‘without instruction’ (ἄνευ διδαχῆς). [135] Once again, the focus is squarely on denatured discourse that, wedded to writing, has lost its dialectical dynamism.

In the final analysis, Alkidamas’ criticism of writing strikes us as the swan song of a fast vanishing practice, a swimming against the current whose flow, though the volume be small at first, has shifted its direction irreversibly. Noting the paradox of his writing against writing (§29), he is forced to offer a justification of the written speech. And it is particularly telling that, faced with proving his excellence to those who are not acquainted with his work, he prefers offering a sample of his writing over making a demonstration of his improvisatory prowess, lest his audience, accustomed to the finish of the written piece, should judge him inferior to others who have made scripted delivery the centerpiece of their art (§31). The same competitive pressure, one can only assume, must have been felt by the Athenian rhapsodes: an uneven field that pitted the finish and flourish of a rehearsed script against the older skill of improvisatory recomposition in the hands of lesser practitioners. Truly gifted traditional bards must have dazzled audiences with their singing and would have had little to fear from those who tried to make up with memory and script what they lacked in recompositional mastery. This must have been a matter of gradation, with performers falling at various points along a continuum, not the often cited, but simplistic opposition between “creative singers” and “reproducing rhapsodes.” But as soon as the festival rules, whether intentionally or not, accommodated scripted performances, writing conferred a clear competitive advantage to the majority of rhapsodes whose talent, as is always true in any profession, did not approach either extreme. This shift in practice is of a sort that tends to build on itself and is hard to reverse. Once it began, it must have taken over and displaced most of the rhapsodes unwilling to commit themselves fully to it.

Footnotes

[ back ] 1. The English translation is taken from Wolf 1985:109–110, chap. xxⅳ. Wolf is quoting from (and slightly modifying) Caesar’s Bellum Gallicum 6.14. His Latin (from the second edition) runs as follows: “De his quidem postremis [sc. Druidibus] Caesar et Mela referunt, propriam eorum fuisse disciplinam, in qua nonnulli ad vicenos annos permanserint, ut magnum numerum versuum ediscerent, litteris non mandatorum” (Wolf 1876:62, his emphasis).

[ back ] 2. “Quam vellem tantillum nobis Graeci tradidissent de vatibus et rhapsodis suis!” (Wolf 1876:62).

[ back ] 3. Important works on the rhapsode and rhapsody include Aly 1914; Patzer 1952; Sealey 1957; Ritoók 1962; Ford 1988; Boyd 1994; Collins 2001a and 2001b. See also the bibliography cited in chapter 10 below, §3.1.

[ back ] 4. By ‘textual fixation’ I mean the gradual process of fixation, not necessarily in writing, of the text of the poems as to their themes, sequence, and form. This objective process is not to be confused with the emic ‘notional fixity’ considered in §7.1. Notional fixity contributes nonetheless to textual fixation in various ways, for example, by facilitating the rise of ‘Homer,’ the individual author. By ‘objective process’ I mean a measurable, actual fixation from the perspective of a cultural outsider.

[ back ] 5. First articulated in Nagy 1981, it is developed further in Nagy 1996b:41–42 and 1996c:107–113.

[ back ] 6. Nagy 1996c:112.

[ back ] 7. Nagy 1996c:112

[ back ] 8. I am neither conceding the intrinsic likelihood nor the historical plausibility of such a scenario. Nor is it material to it, logically speaking, whether one can conceive that the declamation of so long a poem as, for example, the Iliad could have accurately reflected, down to its phraseology, a written text that for many generations was only—or primarily—transmitted by word of mouth. A scheme for the textual fixation of the Iliad and the Odyssey on such terms is very different from the one espoused in these pages.

[ back ] 9. Part I of this book offers a critique of several theories variously adduced to support a conjectural early-archaic recording of the Homeric poems in writing. Proponents of an early fixation of the text resort to speculations that depend on cultural discontinuities and exceptional circumstances. For example, the existence of a ruler of great power and means who realizes the outstanding quality of ‘Homer’s’ performance and insists on owning a written transcript of it. As we have seen, sometimes even the invention of alphabetic writing is tied to the recording of Homeric poetry. The ruler’s literacy is tacitly assumed, as is his interest in the written artifact as a natural and culturally obvious aid to memory. Furthermore, advocates of such views must grant the availability of a sufficiently abundant substrate (papyrus? wood? stone?) and the existence of a scribe capable of writing down thousands upon thousands of lines. The poet, never before engaged in such an ambitious recording session, must have been able to adapt to a much slower performance rate than the one he was accustomed to; and yet, somehow, he excelled, without the encouragement of an ordinary audience, and without losing his train of thought while the scribe made haste to inscribe every word he heard. And if this chain of hypotheses is not sufficiently tenuous, one must explain how the diffusion of this extraordinary cultural artifact over so large a geographical extension could have been so effective as to arrest, soon thereafter, a centuries-old habit of composition in performance and impose the uniform, now fixed, text on dozens of itinerant rhapsodes who roamed the Mediterranean basin and almost certainly must have been illiterate. I find reconstructions of this sort difficult to believe.

[ back ] 10. Plato makes frequent reference to this: τῶν ποιητῶν οἱ ἄκροι τῆς ποιήσεως ἑκατέρας, κωμῳδίας μὲν Ἐπίχαρμος, τραγῳδίας δὲ Ὅμηρος (Theaitētos 152e4–5), where ἄκροι may refer to Homer’s chronological priority (note his pairing with Epikharmos, who some ancient traditions of scholarship, e.g. Aristotle Poetics 1448a33–34, placed at the source of comedy; cf. Pickard-Cambridge 1927:353–363, esp. 355n3); or else it may denote his superior skill (so LSJ s.v. Ⅲ.1), as other passages suggest (pace Gudeman 1934:109). Passages from the Republic further underline the point: ἔοικε μὲν γὰρ [sc. Ὅμηρος] τῶν καλῶν ἁπάντων τούτων τῶν τραγικῶν πρῶτος διδάσκαλός τε καὶ ἡγεμὼν γενέσθαι (595b10–c2); μετὰ τοῦτο ἐπισκεπτέον τήν τε τραγῳδίαν καὶ τὸν ἡγεμόνα αὐτῆς Ὅμηρον (598d7–8); οἱ γάρ που βέλτιστοι ἡμῶν ἀκροώμενοι Ὁμήρου ἢ ἄλλου τινὸς τῶν τραγῳδοποιῶν (605c10–11); συγχωρεῖν [χρὴ] Ὅμηρον ποιητικώτατον εἶναι καὶ πρῶτον τῶν τραγῳδοποιῶν (607a2–3). It may well be, as Naddaff (2002:40–41) remarks, that Sokrates emphasizes (even exaggerates) the similarities between epic and tragedy to open the way for his own ‘atragic’ reading of the Iliad (cf. esp. Naddaff 2002:144n10). But this rhetorical strategy can only succeed if Homer as the father of tragedy is already a cultural commonplace. We remember, of course, Aiskhylos’ celebrated comment (apud Athenaios 7.347e) that his tragedies were ‘slices from Homer’s great dinners’ ([Αἴσχυλος] τὰς αὑτοῦ τραγῳδίας τεμάχη εἶναι ἔλεγεν τῶν Ὁμήρου μεγάλων δείπνων); this statement probably amounts to more than an acknowledgment that his μῦθοι were cognate with Homer’s. A broader reading indeed is supported by Aristotle’s parallel between Homer and tragedy (Poetics 1448b31–1449a2), which makes clear that their point of contact is a proportionality (ἀνάλογον) of form (σχῆμα): ὥσπερ δὲ καὶ τὰ σπουδαῖα μάλιστα ποιητὴς Ὅμηρος ἦν (μόνος γὰρ οὐχ ὅτι εὖ ἀλλὰ καὶ μιμήσεις δραματικὰς ἐποίησεν), οὕτως καὶ τὸ τῆς κωμῳδίας σχῆμα πρῶτος ὑπέδειξεν, οὐ ψόγον ἀλλὰ τὸ γελοῖον δραματοποιήσας· ὁ γὰρ Μαργίτης ἀνάλογον ἔχει, ὥσπερ Ἰλιὰς καὶ ἡ Ὀδύσσεια πρὸς τὰς τραγῳδίας, οὕτω καὶ οὗτος πρὸς τὰς κωμῳδίας. (Cf. Lucas 1968:77 ad 1448b35–36.) This cultural topos recurs so frequently among later writers that it can hardly be assumed to go back to the distortion and conflation by Plato of two genres generally perceived as independent. Thus, for example, the Homeric scholia to Α 332: πρῶτος δὲ Ὅμηρος πρόσωπα κωφὰ παρήγαγεν εἰς τὴν τραγῳδίαν; and to Ζ 466: πρῶτος παῖδας εἰσάγει τῇ τραγῳδίᾳ; so also [Plutarch] De Homero 213: ἡ τραγῳδία τὴν ἀρχὴν ἔλαβεν ἐξ Ὁμήρου. Cf. Gudeman 1934:109 for further witnesses (but note that Philostratos Vita Apollonii 6.11 does not refer to Homer, as he claims, but to Aiskhylos). See also Schmid 1908 on the scholia to the Iliad and dramatic μίμησις.

[ back ] 11. See below, Appendix. The later acceptation ‘to answer’—or something that approaches it—does not lack representation in the diachronic layering of the poems (e.g. at Η 407, though cf. Nagy 2003:21–22). Koller (1957:101) believes that this was a semantic development largely restricted to the Ionic dialect and only rarely present in Athenian authors as a literary affectation (e.g. Thoukydides 7.44.5). I take a somewhat different view of this matter.

[ back ] 12. See below, Appendix, for a critique of Else’s formulation.

[ back ] 13. The relevant lines are as follows: ἐγὼ μ[ ‖ παῦρα μελ[ι]ζομεν[ ‖ [γλώ]σσαργον ἀμφέπω[ν ⌞ἐρε ‖ θίζ⌟ομαι πρὸς ἀϋτά̣[ν ‖ ⌞ἁλίο⌟υ δελφῖνος ὑπ⌞όκρισιν⌟, ‖ ⌞τὸν μὲν ἀκύμονος ἐν πόντου πελάγει ‖ αὐλῶν ἐκίνησ’ ἐρατὸν μέλος⌟ (fr. 140b.11–17). I quote the translation in Henderson 1992: “I (however indeed, while hearing him) playing his brief (songs), and fostering (…) with a loud-sounding tongue, am provoked in response to it, acting like a dolphin of the sea, whom the lovely tune of flutes has excited in the expanse of a waveless sea” (148).

[ back ] 14. The construction is best understood as an accusative in apposition to a sentence; cf. Schwyzer GG Ⅱ.86, under ‘Akkusativ der Satzapposition.’ For the interpretation of this fragment, see Wilamowitz-Moellendorff 1922:500–502, Fileni 1987, and Henderson 1992. As to Lesky’s contention that this text once and for all banishes the possibility that originally ὑποκριτής might have meant ‘answerer’ (Lesky 1956:475), I must confess with Page 1956 that I do not understand the logic of his argument. Perhaps his point is that in Pindar’s time the use of ὑπόκρισις must have been intelligible to the audience in terms of its original meaning; and that one should therefore be able to make good sense of the passage if we use the alleged meaning to translate it. But if so, is ‘I am provoked … replying like a dolphin’ to be discarded in favor of ‘I am provoked … interpreting like a dolphin’? Clearly neither is admissible except as a metaphor, and this does not gain us any advantage in the argument.

[ back ] 15. It can hardly be doubted that the word is correctly supplied at IG Ⅱ/Ⅲ2 2318.70 (=Mette 1977:15 I col. 6.3; cf. Capps 1943:1–3) and its beginning ὑπο̣ is extant at 82 (=Mette 1977:16 I col. 6.15). Cf. Pickard-Cambridge 1988:71–72 and 101–107. I assume that ὑποκριτής is original and not a scribal anachronism. The year when the contest was instituted is variously given as 450/49, 449/48, or 448/47 BC, depending on the details of the reconstruction (see Capps 1903:16–17, 22n62).

[ back ] 16. Cf. Zucchelli 1962:52n98. ὑποκρινόμενον in Wasps 53, however, does mean ‘to interpret’.

[ back ] 17. Plato’s Timaios 72a6–b5, though much later, would lend support to this view: ὅθεν δὴ καὶ τὸ τῶν προφητῶν γένος ἐπὶ ταῖς ἐνθέοις μαντείαις κριτὰς ἐπικαθιστάναι νόμος· οὓς μάντεις αὐτοὺς ὀνομάζουσίν τινες, τὸ πᾶν ἠγνοηκότες ὅτι τῆς δι’ αἰνιγμῶν οὗτοι φήμης καὶ φαντάσεως ὑποκριταί, καὶ οὔτι μάντεις, προφῆται δὲ μαντευομένων δικαιότατα ὀνομάζοιντ’ ἄν (‘whence also it is customary to appoint the class of prophets as judges over inspired oracles; these same ones some call seers in utter ignorance that they are interpreters of utterances and visions [that come] through riddles, and they would most accurately be called not seers but prophets of things divined’). For more on this passage see above, §8.1.1.

[ back ] 18. See Koller 1957:106. The debate whether ὑποκριτής in drama originally meant ‘answerer’ or ‘expounder’ can be most readily joined by reading Lesky 1956, Else 1959, and Zucchelli 1962. See also, more recently, Ley 1983. Nagy (2003:21–38) offers an important modification that complements Koller’s insight. He argues that the traditional nature of Homeric poetry—that audiences over the years thought of the rhapsode in performance as ‘quoting’ the notionally unchanging speeches by the characters in the poems—gives to the instances of ὑποκρίνεσθαι in epic the connotation of “responding by way of performing” (21). This would readily lead to its use in the context of drama.

[ back ] 19. Iōn 532d6–e1: βουλοίμην ἄν σε ἀληθῆ λέγειν, ὦ Ἴων· ἀλλὰ σοφοὶ μέν πού ἐστε ὑμεῖς οἱ ῥαψῳδοὶ καὶ ὑποκριταὶ καὶ ὧν ὑμεῖς ᾄδετε τὰ ποιήματα, ἐγὼ δὲ οὐδὲν ἄλλο ἢ τἀληθῆ λέγω, οἷον εἰκὸς ἰδιώτην ἄνθρωπον (‘I wish you were right in saying that, Ion. But surely you yourselves, the rhapsodes-hypokritai, are the wise ones, and those whose compositions you sing; while I myself speak the simple truth, as one would expect from an ordinary man [without expertise]’).

[ back ] 20. Iōn 535e7–536a1: οἶσθα οὖν ὅτι οὗτός ἐστιν ὁ θεατὴς τῶν δακτυλίων ὁ ἔσχατος, ὧν ἐγὼ ἔλεγον ὑπὸ τῆς Ἡρακλειώτιδος λίθου ἀπ’ ἀλλήλων τὴν δύναμιν λαμβάνειν; ὁ δὲ μέσος σὺ ὁ ῥαψῳδὸς καὶ ὑποκριτής, ὁ δὲ πρῶτος αὐτὸς ὁ ποιητής (‘Do you realize then that this one, the spectator, is the last of the links which I said receive their force one from another under the influence of the magnet? And you, the rhapsode-hypokritēs, are the middle one, while the first is the poet himself’).

[ back ] 21. Bölte 1907 offers a helpful exploration of the dramatic potential of Homeric epic and the ways in which a rhapsode might take advantage of it. Cf. also Throop 1917 and Hornung 1869:9–13.

[ back ] 22. The use at 532d7 of one article οἱ with two coordinate nouns, οἱ ῥαψῳδοὶ καὶ ὑποκριταί, should suffice to prove that ὑποκριταί is epexegetic and that Sokrates does not have in view two different types of performers (rhapsodes and actors). But, should there be any doubt, when he calls Ion at 535e9–536a1 ὁ ῥαψῳδὸς καὶ ὑποκριτής, he renders the point incontestable. Dramatic actors are never under discussion in the Iōn, however much their trade has arguably influenced the rhapsode’s practice. From the beginning Sokrates addresses himself solely to performers of epic (τοὺς ῥαψῳδούς 530b5), and ὑποκριτής (or its plural ὑποκριταί) merely draws out to ironic effect an important diachronic facet of their craft. Incidentally, Sokrates takes for granted in Republic 395a7 that one and the same individual cannot be a good rhapsode and an accomplished actor, even if in real life this was not always necessarily the case.

[ back ] 23. E.g. 530c9; cf. 533c5–7 and d2.

[ back ] 24. For more on the rhapsode and the poet as ἑρμηνῆς see above, §8.3.2 and §8.4. It is clear that ὑποκριτής passed from the rhapsode to the actor (and not vice versa), since only for the former can one compellingly motivate a term derived from the ambit of μάντεις, προφῆται, and ὑποφῆται. Indeed, those who refuse to embrace the dichotomy between ἀοιδός and ῥαψῳδός disputed in this book have no reason to withhold from the rhapsode the description ‘ἑρμηνεύς of the inspiring deity’ that Plato applies to the poet in the Iōn (534e4–5 and 535a5). When, in the evolution of his profession, an aspect of rhapsodic performance was conceptualized as explanatory of the poetic tradition—when it was thought of and referred to as ‘expounding the poet’—it was natural that, to designate the rhapsode in this capacity, the same term would be used that had previously designated his hermeneutic function vis-à-vis the deity (hence Iōn 535a6–7 and a9).

[ back ] 25. Iōn 530d2–3, 535a6–10.

[ back ] 26. Iōn 530c8–9, 532c6, 533c5–6, 533d2, 534c1, 536c1, 536d7, 542a5.

[ back ] 27. Iōn 530d6–7, 536d3, 536d5–6, 541e2.

[ back ] 28. Cf. 533e7–8 with epic poets, and 537c1–2 with Homer, as the subject.

[ back ] 29. See Capuccino 2005:275 for a helpful “diagram of the exegetic activity of the rhapsode.” I cannot, however, accept her understanding of this exegesis, which reduces it to ἑρμηνεύειν and excludes ‘explanation’ in favor of ‘[vocal] expression’ (194–195). Lacking diachronic depth, her treatment espouses a reductive view of rhapsodic mediation: the rhapsode as mere mouthpiece of the poet (131). There is, to be sure, some truth to her argument: in advocating Homer’s διάνοια, rhapsodic exegesis will have offered a positive appraisal of it. But it may none the more thereupon be reduced to mere ἔπαινος: how are we to imagine that the performer gave voice to his approbation? Surely not just by declaiming Homer’s verse with verve (as Capuccino herself acknowledges at 166n217). Whatever the subject in which the poet was purported to instruct his public, at the very least the rhapsode would have had to describe in sufficient detail the content of Homer’s teaching. Even if this did not call for glossing archaic formulas or illuminating difficult epic syntax, only a cramped definition of ‘exegesis’ could fail to embrace such activity. This is Pepin’s weakness, whom Capuccino follows, when he reduces “exégèse” to “un mouvement d’entrée dans l’intention d’un texte ou d’un message” (Pépin 1975:291), forgetting the reciprocal outward motion entailed by the preverb ἐξ (cf. Latin ‘explico’). Capuccino’s analysis is pulled in opposite directions at once. On the one hand, she wants to preserve the rhapsode’s active critical engagement; on the other, to the extent that it detracts from his function as ‘mere spokesman,’ she wishes to marginalize this engagement as ‘mental’ and ‘unofficial.’ Regarding such ‘mental exegesis’ she writes: “Insomma, quello che Ione dovrebbe poter fare per essere un buon mediatore di Omero, è conoscerne a fondo la mente … ; vale a dire ripercorrere quel dialogo silenzioso dell’anima con se stessa che è il pensiero (scil. l’attività del pensare) di qualcuno. … Ma tutto questo non riguarda alcuna attività esegetica o interpretativa; riguarda piuttosto l’importanza che Socrate attribuisce al dialogo …” (192–193, her emphasis; cf. 196). And she elaborates as follows her claim that rhapsodic laudatory exegesis was ‘unofficial’: “Questo potrebbe far credere che la locuzione Ὁμήρου ἐπαινέτης designi una precisa figura professionale … . Se il rapsodo (soprattutto il rapsodo omerico) era famoso per le sue declamazioni pubbliche, non è necessario che lo fosse per frequenti discussioni … con altri ‘esperti’ di Omero di fronte a un uditorio meno numeroso” (163–164n211; the limitation to ‘discussions with other experts’ is a red herring). In her exposition, Capuccino credits Pépin 1975, who refers to Philo’s distinction between two λόγοι: the ‘mental’ one of discursive thought (ὁ ἐν διανοίᾳ) and the audible one of discursive speech (διὰ στόματος καὶ γλώττης); this latter he names the ἑρμηνεύς of the former (De migratione Abrahami §§71–73). Rightly understood, however, Philo’s distinction hardly justifies Capuccino’s collapse of ‘interpretation’ onto mere ‘[vocal] expression’. The Alexandrian regards speech as a complex faculty which God has given man to articulate and communicate his thought (Quis rerum divinarum heres sit §§105–111). Should one choose to call this faculty ‘expression’, then it is certainly not just ‘vocal expression’, as though some mechanical conveyance of thought into words. If ‘expression’ in all of its manifold aspects is reconceptualized as mediating communication that issues from a separate source—i.e. as ‘putting it into words’ (τῇ λέξει σημαίνειν Aristotle Sophistical Refutations 166b15–16)—then it is rightly called ‘interpretation’, for it expounds its content and makes it explicit (cf. Pépin 1975:293). In this sense, ἑρμηνεύειν complements ἐξηγεῖσθαι. This ‘interpretation’, moreover, is perfectly consonant with the function I have ascribed to the rhapsode as interpreter of the will of Zeus. Divine thought-λόγος and speech-λόγος concur in the figure of the mediating rhapsode whom the Muses inspire.

[ back ] 30. So Xenophon Αpomnēmoneumata 4.2.10: τί δὲ δὴ βουλόμενος ἀγαθὸς γενέσθαι, ἔφη, ὦ Εὐθύδημε, συλλέγεις τὰ γράμματα; … ἀλλὰ μὴ ῥαψῳδός; ἔφη· καὶ γὰρ τὰ Ὁμήρου σέ φασιν ἔπη πάντα κεκτῆσθαι. μὰ Δί’ οὐκ ἔγωγ’, ἔφη· τοὺς γάρ τοι ῥαψῳδοὺς οἶδα τὰ μὲν ἔπη ἀκριβοῦντας, αὐτοὺς δὲ πάνυ ἠλιθίους ὄντας (‘“Euthydemos,” said Sokrates, “you collect bookrolls in order to become good at what?” … “Perhaps [you want to be] a rhapsode then?” he said. “For they say that you also own all of Homer’s poetry.” “By Zeus, not I!” he said. “For I know that rhapsodes hone their verses, but they themselves are very silly”’). And Xenophon Symposion 3.5–6: ὁ πατὴρ ὁ ἐπιμελούμενος ὅπως ἀνὴρ ἀγαθὸς γενοίμην ἠνάγκασέ με πάντα τὰ Ὁμήρου ἔπη μαθεῖν· καὶ νῦν δυναίμην ἂν Ἰλιάδα ὅλην καὶ Ὀδύσσειαν ἀπὸ στόματος εἰπεῖν. ἐκεῖνο δ’, ἔφη ὁ Ἀντισθένης, λέληθέ σε, ὅτι καὶ οἱ ῥαψῳδοὶ πάντες ἐπίστανται ταῦτα τὰ ἔπη; καὶ πῶς ἄν, ἔφη, λελήθοι ἀκροώμενόν γε αὐτῶν ὀλίγου ἀν’ ἑκάστην ἡμέραν; οἶσθά τι οὖν ἔθνος, ἔφη, ἠλιθιώτερον ῥαψῳδῶν; οὐ μὰ τὸν Δί’, ἔφη ὁ Νικήρατος, οὔκουν ἔμοιγε δοκῶ. δῆλον γάρ, ἔφη ὁ Σωκράτης, ὅτι τὰς ὑπονοίας οὐκ ἐπίστανται (‘“My father, taking care that I would become a good man, forced me to learn all the poetry of Homer; even now I could recite by heart the entire Iliad and Odyssey.” “But has this escaped your notice,” said Antisthenes, “that all rhapsodes know these verses too?” “How could it,” said he, “when I used to listen to them almost daily?” “Do you know then any tribe,” he said, “sillier than rhapsodes?” “No, by Zeus!” said Nikeratos, “I think certainly not.” “For it is clear,” said Sokrates, “that they do not know the underlying meaning”’). Cf. Graziosi 2002:22.

[ back ] 31. Xenophon’s Sokrates uses the word ὑπόνοιαι (Symposion 3.6.7). Cf. the section on Theagenes above, §6.2.

[ back ] 32. Cf. Graziosi 2002:45 and Capuccino 2005:271–272 with 272n22.

[ back ] 33. “Sie konnte nur Prosarede sein, untermischt mit Versen, Versteilen, Kolen, Kommata, mit allen Eigenheiten poetischer Sprache, d. h. die Form der Hypokrisis ist eine Mischung von Dichtersprache und Alltagsprosa.”

[ back ] 34. Cf. Aristotle Rhetoric 1404a24–28.

[ back ] 35. The work of Kelly 1990 on the differing rates of correption in Homeric speeches and narrative would seem to support this conjecture. I quote here from his conclusion: “[T]he formulas for the speeches were composed at an earlier date than the formulas for the narrative. The supposition of a proto-epic composed of speeches and a connecting prose narrative is strengthened by the existence of such poems in … the Indo-European community, and by the dominance of speeches in the Homeric text as it now stands. The transition from speeches to [versified] narrative was effected by means of quoted narrative. … Narrative, then, not the speeches, remains the locus for innovation” (80–81). By “quoted narrative” Kelly means narrative in the mouth of a Homeric character. Garner 2011 takes issue with Kelly’s theory and views correption in Homer not “as an accidental or residual archaism,” but “as a dynamic enabler of poetic flexibility” that remained an active, “essential and integral tool” to composition (50).

[ back ] 36. Iōn 530d6–8: καὶ μὴν ἄξιόν γε ἀκοῦσαι, ὦ Σώκρατες, ὡς εὖ κεκόσμηκα τὸν Ὅμηρον· ὥστε οἶμαι ὑπὸ Ὁμηριδῶν ἄξιος εἶναι χρυσῷ στεφάνῳ στεφανωθῆναι (‘Really, Sokrates, it is worth hearing how well I have adorned Homer, [so well] that I think I deserve to be crowned with a golden wreath by the Homeridai’).

[ back ] 37. See above, §7.2.

[ back ] 38. That is, a stage presence marked by a characteristically histrionic use of voice, gestures, and dress.

[ back ] 39. Cf. Iōn 530b6 and 535d2.

[ back ] 40. See below, §9.3. For more on rhapsodic κόσμησις, see Boyd 1994:118–120.

[ back ] 41. See above, §9.2 n. 30. This does not hold for the archaic appeal to ‘memory’, which at that time was not conceived as rote learning and did connote mystery and divine influence. But in the environment of late fifth- and fourth-century BC Athens, as just remarked, laymen might commit to memory large portions of Homer in an ordinary, mechanical way.

[ back ] 42. For more on poetry and ‘social memory’ see above, §7.2.3. For the poet and divine omniscience see pp. 183ff.

[ back ] 43. Or without a prose exposition that was fused with his creative poetic elaboration and viewed in immediate continuity with it.

[ back ] 44. For various approaches to the interpretation of the Iōn, a sample of the relevant bibliography might include Verdenius 1943, Ladrière 1951, Diller 1955, Wyller 1958, Flashar 1958, Tigerstedt 1969:13–29, Partee 1971, Dorter 1973, Schousen 1986, Velardi 1989, Ott 1992, Janaway 1995:14–35, Harris 1997, and Capuccino 2005.

[ back ] 45. As the beginning of Plato’s Lesser Hippias shows. Cf. Culverhouse 2010.

[ back ] 46. To this fight, prose and the new technology of writing became instrumental. See below, §9.5.

[ back ] 47. O’Sullivan 1992:67.

[ back ] 48. O’Sullivan 1992:67n32.

[ back ] 49. See Capuccino 2005:159–167 and Elmer 2013.

[ back ] 50. Iōn 536d1–3: τούτου δ’ ἐστὶ τὸ αἴτιον, ὅ μ’ ἐρωτᾷς, δι’ ὅτι σὺ περὶ μὲν Ὁμήρου εὐπορεῖς, περὶ δὲ τῶν ἄλλων οὔ, ὅτι οὐ τέχνῃ ἀλλὰ θείᾳ μοίρᾳ Ὁμήρου δεινὸς εἶ ἐπαινέτης.

[ back ] 51. Iōn 536d4–6: σὺ μὲν εὖ λέγεις, ὦ Σώκρατες· θαυμάζοιμι μεντἂν εἰ οὕτως εὖ εἴποις, ὥστε με ἀναπεῖσαι ὡς ἐγὼ κατεχόμενος καὶ μαινόμενος Ὅμηρον ἐπαινῶ.

[ back ] 52. Iōn 541e1–3: ἀλλὰ γὰρ σύ, ὦ Ἴων, εἰ μὲν ἀληθῆ λέγεις ὡς τέχνῃ καὶ ἐπιστήμῃ οἷός τε εἶ Ὅμηρον ἐπαινεῖν, ἀδικεῖς.

[ back ] 53. Iōn 542b2–3: τοῦτο τοίνυν τὸ κάλλιον ὑπάρχει σοι παρ’ ἡμῖν, ὦ Ἴων, θεῖον εἶναι καὶ μὴ τεχνικὸν περὶ Ὁμήρου ἐπαινέτην.

[ back ] 54. Prōtagoras 309a6–b1: οὐ σὺ μέντοι Ὁμήρου ἐπαινέτης εἶ, ὃς ἔφη χαριεστάτην ἥβην εἶναι τοῦ ⟨πρῶτον⟩ ὑπηνήτου, ἣν νῦν Ἀλκιβιάδης ἔχει;

[ back ] 55. Republic 383a7–8: πολλὰ ἄρα Ὁμήρου ἐπαινοῦντες, ἀλλὰ τοῦτο οὐκ ἐπαινεσόμεθα, τὴν τοῦ ἐνυπνίου πομπὴν ὑπὸ Διὸς τῷ Ἀγαμέμνονι.

[ back ] 56. Republic 606e1–607a5: οὐκοῦν, εἶπον, ὦ Γλαύκων, ὅταν Ὁμήρου ἐπαινέταις ἐντύχῃς λέγουσιν ὡς τὴν Ἑλλάδα πεπαίδευκεν οὗτος ὁ ποιητὴς καὶ πρὸς διοίκησίν τε καὶ παιδείαν τῶν ἀνθρωπίνων πραγμάτων ἄξιος ἀναλαβόντι μανθάνειν τε καὶ κατὰ τοῦτον τὸν ποιητὴν πάντα τὸν αὑτοῦ βίον κατασκευασάμενον ζῆν, φιλεῖν μὲν χρὴ καὶ ἀσπάζεσθαι ὡς ὄντας βελτίστους εἰς ὅσον δύνανται, καὶ συγχωρεῖν Ὅμηρον ποιητικώτατον εἶναι καὶ πρῶτον τῶν τραγῳδοποιῶν, εἰδέναι δὲ ὅτι ὅσον μόνον ὕμνους θεοῖς καὶ ἐγκώμια τοῖς ἀγαθοῖς ποιήσεως παραδεκτέον εἰς πόλιν.

[ back ] 57. E.g. Iōn 541b3–5.

[ back ] 58. Perhaps the closest parallel in recent history might be the place the Bible held among the early Puritan settlers of Massachusetts.

[ back ] 59. Cf. Wilamowitz-Moellendorff 1920: “Ὁμήρου ἐπαινέτης, das hier 536d und als letztes Wort des Dialoges steht, von Ion an der ersten Stelle mit Ὅμηρον ἐπαινῶ aufgenommen wird, bezeichnet den, der zu Ehren Homers redet, seine Sache führt, auf ihn schwört. … Noch der Kreter der Gesetze ist Διὸς ἐπαινέτης 633a” (41n2); and Albini 1954:35 ad 536d: “ἐπαινέτης significa ‘panegirista’, ‘encomiatore’. La parola, con cui il dialogo si chiude, ha un valore pregnante.” Thus, Velardi (1989:32) concludes: “In realtà ci sono elementi sufficienti per affermare che i due termini [epaineîn ed epainétes], almeno in riferimento ad Omero, hanno un preciso valore tecnico.”

[ back ] 60. For a cultural analysis of the speech and, more generally, the times of Lykourgos, see Mikalson 1998:11–45.

[ back ] 61. I have restored τὸν Ὅμηρον … ἐπαινῶν, the reading of the manuscripts that Reiske and Koraēs had unnecessarily emended to τῶν Ὁμήρου … ἐπῶν (which Conomis prints). Not only is ἐπαινέω here paralleled at Against Leōkratēs §100, where it is used in connection with Euripides, but it also reflects the conventional diction of epideixis, one that, I am arguing, goes back to long-standing rhapsodic practice. Cf. Velardi 1989:33–35.

[ back ] 62. See above (§9.2) for my own suggestions about the way in which the practice of rhapsodes in performance may have changed with time, and the corresponding evolution in the character of the ‘stitching’ and exposition (ἐπίδειξις or ἑρμηνεία, depending on the point of view) they would have engaged in.

[ back ] 63. Cf. Humphreys 1985a:216–217. On this passage see further below, §11.1.

[ back ] 64. ἐπίδειξιν ποιούμενοι πρὸς τοὺς Ἕλληνας (§102); οἱ δὲ ποιηταὶ … μετὰ λόγου καὶ ἀποδείξεως τοὺς ἀνθρώπους συμπείθουσιν (§102). While the traditional term for the display of oratory and of performance generally is ἐπίδειξις, ἀπόδειξις is closely related to it. And though the latter can be employed in the specialized sense of rhetorical or logical proof, it still retains the flexibility to evoke the demonstrative ‘show’ of ἐπίδειξις. So, for example, of three instances of ἀποδείκνυμι in Aristotle’s Poetics, only one (1456a38) more narrowly relates to proving one’s point (complementing the refutation of contrary arguments); the other two (1450a7 and 1450b11) pertain more generally to communicating the thoughts of the characters on stage and associate the word with ἀποφαίνεσθαι, the term used by Sokrates in Iōn 532e8 and 533a4 for the exegetical function of the specialist (specifically, in painting, but, by implication, also in the rhapsodic art). In other words, ἀπόδειξις is part and parcel of the ἐπιδείξεις of the σοφοί: ἐπίδειξις especially regards the public display before the audience, ἀπόδειξις the public display of the subject matter. Hence the use of ἀποδείξεις to designate the rhetorical speeches by Themistokles of Ilion (Θεμιστοκλῆς … ἀποδείξεις πεπόηται τῶν ῥητορικῶν λόγων) given in Xanthos in 196 BC (see Robert and Robert 1983:154–156; cf. Pernot 1993:1.50). Robert and Robert (1983:162n27) write: “Il a donné des ‘démonstrations’, ἀποδείξεις, de son art, à savoir dans des ἐπιδείξεις.” And in his Progymnasmata 106 (Spengel), Theon of Alexandria explained that topoi differ from encomia and invective and are “concerned simply with their subjects and involve no demonstration (χωρὶς ἀποδείξεως),” whereas the other two “are concerned with specific persons and include demonstration (μετὰ ἀποδείξεως)” (the translation is taken from Kennedy 2003 ad loc.; cf. Pernot 1993:2.679). This connection is very old indeed: it is attested in [Plato] Hipparkhos 228b6–7 ([Ἵππαρχος] ἄλλα τε πολλὰ καὶ καλὰ ἔργα σοφίας ἀπεδείξατο), in Herodotos’ proem (Ἡροδότου Θουρίου ἱστορίης ἀπόδεξις ἥδε), and, significantly, in Alkidamas’ On the Sophists §§29–32, where a failure to understand it has led scholars to emend the ἀποδείξεις of §29 to ἐπιδείξεις, so as to harmonize it with the τῶν ἐπιδείξεων of §31 (see, for example, Mariß 2002:281–282). Cf. Nagy (1990c:217–222), who insists on ἀπόδειξις as ‘performance’ (rather than ‘public presentation’) because of its implicit reference to the artistic medium which effects the display. I am in essential agreement with him on this point, but would accept ‘public display’, either as a direct reference to performance (in the case of rhapsodes, orators, etc.) or else as a metaphor from the world of the performer and his audience. Cf. Bakker 2002a.

[ back ] 65. The metaphor of the ring embodies the lines of transmission along which the emotions flow: though we know nothing of the poet, his allegedly manic possession (which portrays him as a bacchant) is a safe index of his own emotional ecstasy; as to the other links, there is no doubt that the magnetic influence takes the form of a rhapsode overcome by his feelings, who in turn overwhelms his audience emotionally. (If we attend to the details, however, the picture is not so tidy; for, contemplating the prize as all but won, Ion combines inner laughter with his and his audience’s outward wailing, 535e4–6.) The inclusion of the audience as the terminal link should caution us not to push the metaphor too far, since the audience neither composes nor performs under the influence of such ‘possession’—it merely feels a powerful emotional influence. The presence of strong πάθη confirms the divine activity, mediated by poet and rhapsode, whose source is the Muse, but its particular effect on each link is peculiar to the link itself: whether it be a poet, a rhapsode, χορευταί, διδάσκαλοι, etc.

[ back ] 66. Note also the mention of ἐκπλήττειν at Iōn 535b2–3, to be compared with ἔκπληξις at Poetics 1455a17–18. Cf. Flashar 1958:67–69.

[ back ] 67. See below, §13.1.

[ back ] 68. ἔστιν δὲ αὕτη [sc. ὑπόκρισις] μὲν ἐν τῇ φωνῇ, πῶς αὐτῇ δεῖ χρῆσθαι πρὸς ἕκαστον πάθος, οἷον πότε μεγάλῃ καὶ πότε μικρᾷ καὶ μέσῃ, καὶ πῶς τοῖς τόνοις, οἷον ὀξείᾳ καὶ βαρείᾳ καὶ μέσῃ, καὶ ῥυθμοῖς τίσι πρὸς ἕκαστα. τρία γάρ ἐστιν περὶ ἃ σκοποῦσιν· ταῦτα δ’ ἐστὶ μέγεθος ἁρμονία ῥυθμός (1403b26–31). See below, §13.2.

[ back ] 69. See below, §14.2.

[ back ] 70. See below, §14.1.

[ back ] 71. καὶ ἔστιν φύσεως τὸ ὑποκριτικὸν εἶναι, καὶ ἀτεχνότερον, περὶ δὲ τὴν λέξιν ἔντεχνον. διὸ καὶ τοῖς τοῦτο δυναμένοις γίνεται πάλιν ἆθλα, καθάπερ καὶ τοῖς κατὰ τὴν ὑπόκρισιν ῥήτορσιν· οἱ γὰρ γραφόμενοι λόγοι μεῖζον ἰσχύουσι διὰ τὴν λέξιν ἢ διὰ τὴν διάνοιαν (1404a15–19).

[ back ] 72. ἔστι δὲ λέξις γραφικὴ μὲν ἡ ἀκριβεστάτη, ἀγωνιστικὴ δὲ ἡ ὑποκριτικωτάτη (1413b8–9).

[ back ] 73. καὶ παραβαλλόμενοι οἱ μὲν τῶν γραφικῶν ἐν τοῖς ἀγῶσι στενοὶ φαίνονται, οἱ δὲ τῶν ῥητόρων, εὖ λεχθέντες, ἰδιωτικοὶ ἐν ταῖς χερσίν. αἴτιον δ’ ὅτι ἐν τῷ ἀγῶνι ἁρμόττει τὰ ὑποκριτικά· διὸ καὶ ἀφῃρημένης τῆς ὑποκρίσεως οὐ ποιοῦντα τὸ αὑτῶν ἔργον φαίνεται εὐήθη, οἷον τά τε ἀσύνδετα καὶ τὸ πολλάκις τὸ αὐτὸ εἰπεῖν ἐν τῇ γραφικῇ ὀρθῶς ἀποδοκιμάζεται, ἐν δὲ ἀγωνιστικῇ οὔ, καὶ οἱ ῥήτορες χρῶνται (1413b14–21 from Ross’s OCT, quoted for convenience; cf. below, §14.2).

[ back ] 74. τὸ μὲν γάρ ἐστιν ἑλληνίζειν ἐπίστασθαι, τὸ δὲ μὴ ἀναγκάζεσθαι κατασιωπᾶν ἄν τι βούληται μεταδοῦναι τοῖς ἄλλοις, ὅπερ πάσχουσιν οἱ μὴ ἐπιστάμενοι γράφειν (1413b5–8). This statement has confounded many (e.g. Rapp 2002:1.932), and a ‘suitable’ meaning is extracted by interpolating much that is allegedly tacit (so, for example, Kennedy 1991 ad loc.)—material that, in my view, is extraneous to the context. See below, §14.2, for my discussion of this passage and an alternative translation for τὸ μὲν … τὸ δέ that does not affect the material point here.

[ back ] 75. He does not consider epideictic as a genus dicendi (though ἀποδείξεις makes an appearance in §29; see above, §9.4 n. 64). Alkidamas’ life and oeuvre have recently been the focus of much work. In the past, the interest of scholars has been the alleged rivalry between him and Isokrates. More recently, he has been studied on his own terms for his contribution to the development of rhetoric and the origin of literary criticism. See, for example, Brown 1914:27–42, Milne 1924, Walberer 1938, Gastaldi 1981, Eucken 1983:121–132, Friemann 1990, Ritoók 1991, O’Sullivan 1992, Bons 1998, Liebersohn 1999, Graff 2000 and 2001, Muir 2001, Schloemann 2002, Mariß 2002, and McCoy 2009.

[ back ] 76. The distinction between the λογογράφος and his client is only once acknowledged: λογογραφήσουσι in §6 does not demand (though it may hint at) the technical meaning familiar to the modern scholar (cf. Mariß 2002:129); in §13, however, the professional speech writer seems in view. On the client/logographos relationship see Worthington 1993; on the logographos generally see Lavency 1964, Wolff 2007 [1967], Dover 1968b, Usher 1971, Todd 1990b:163–167, Todd 1993:94–96, Iannucci 2001, and Grethlein 2004.

[ back ] 77. See the still essential study by Brown 1914. See also Hudson-Williams 1951, Hammerstaedt and Terbuyken 1994–1996, Klawitter 1998, and Schloemann 2000.

[ back ] 78. ἐπειδή τινες τῶν καλουμένων σοφιστῶν ἱστορίας μὲν καὶ παιδείας ἠμελήκασι καὶ τοῦ δύνασθαι λέγειν ὁμοίως τοῖς ἰδιώταις ἀπείρως ἔχουσι, γράφειν δὲ μεμελετηκότες λόγους … (‘since some of the so-called sophists neglect learning and training and, like the layman, do not know how to speak, but being practiced in writing speeches …’ §1); πρῶτον μὲν οὖν ἐντεῦθεν ἄν τις καταφρονήσειε τοῦ γράφειν, ἐξ ὧν ἐστιν εὐεπίθετον καὶ ῥᾴδιον καὶ τῇ τυχούσῃ φύσει πρόχειρον. εἰπεῖν μὲν γὰρ ἐκ τοῦ παραυτίκα περὶ τοῦ παρατυχόντος ἐπιεικῶς … οὔτε φύσεως ἁπάσης οὔτε παιδείας τῆς τυχούσης ἐστίν (‘first, then, one might deprecate writing for this reason, that it is easily acquired, facile, and readily available to just any natural disposition. For to speak fittingly on the spot on whatever matter presents itself … is neither within the reach of every natural disposition nor attainable by just any training’ §3).

[ back ] 79. οὐχ ὡς ἀλλοτρίαν ἐμαυτοῦ τὴν δύναμιν αὐτῶν ἡγούμενος, ἀλλ’ ὡς ἐφ’ ἑτέροις μεῖζον φρονῶν (‘not considering their ability alien to me, but priding myself more on other grounds’ §2).

[ back ] 80. So Radermacher 1951, who emends the text to τὸ γράφειν ἐν παρέργῳ τοῦ ⟨λέγειν⟩ μελετᾶν οἰόμενος χρῆναι (§2).

[ back ] 81. Even where writing plays a role, it is ultimately for the hearing (not reading) of the people: ἐπὶ τὴν ἀκρόασιν (§11).

[ back ] 82. τοὺς ἐπ’ αὐτὸ τοῦτο τὸν βίον καταναλίσκοντας ἀπολελεῖφθαι πολὺ καὶ ῥητορικῆς καὶ φιλοσοφίας ὑπειληφώς, καὶ πολὺ δικαιότερον ἂν ποιητὰς ἢ σοφιστὰς προσαγορεύεσθαι νομίζων (‘being of the opinion that those who spend their life on this very thing are greatly deficient in their oratorical skill and in philosophy, and considering that one would much more justly call them makers than sophists’ §2).

[ back ] 83. From various sources we know that they kept and used books when the Athenian book trade was still in its infancy and owning scrolls was still considered extraordinary. Cf. Dover 1993:34–35, Pöhlmann 1994:19n19, and O’Sullivan 1996:115–116.

[ back ] 84. Later on he mentions ποιήματα: οἱ τοῖς ὀνόμασιν ἐξειργασμένοι [sc. λόγοι] καὶ μᾶλλον ποιήμασιν ἢ λόγοις ἐοικότες (‘speeches of contrived diction and more like artifacts than speeches’ §12). Mariß 2002:180 writes: “Im Hinblick auf die Metapher in dem vorhergehenden ἐξειργασμένοι schwingt bei ποιήματα auch hier die Bedeutung, ‘künstl(er)i(s)ch gefertigte Gebilde’, mit; den ursprünglichen Sinn von ποιητάς hatte Alkidamas, in einer sprachlich ähnlichen Wendung in §2, polemisch genutzt … . Der abwertende Unterton klingt auch in ποιήματα an.” For the early history of the word ‘poet’ in its literary sense see Ford 2002:131–157.

[ back ] 85. All the references in this paragraph are to On the Sophists §§3–4.

[ back ] 86. Here, too, there would be an opposition between the ἰδιῶται and the δημιουργός (§§1, 4), and true traditional skill would hardly be εὐεπίθετον (§3).

[ back ] 87. So, for example, in Plato’s Iōn memory is mentioned at 537a2–4, and again humorously at 539e7–540a3. Cf. Lesser Hippias 368d6–7 and 369a4–8.

[ back ] 88. That is, a follow-up of one performer by the next that is not only smooth and polished at the thematic level, but also respectful of the generic constraints observed in transitions between poetic sections.

[ back ] 89. Athens played the central role because of its dominance over the recitation and diffusion of Homeric epic during its defining stage. Cf. Nagy 2001.

[ back ] 90. There was a comparable shift in the pool of competitors participating in athletic events all throughout Greece. To be sure, this shift did not consist, as must have been the case with Panathenaic rhapsodes, in a gradual displacement of a largely Ionian (i.e. non-Athenian) professionalism by a growing Athenian ‘semi-professional amateurism’ (to use an oxymoron). Athletic events witnessed instead an increasing involvement of upper middle- and lower-class athletes. While the elite (in Athens, those who belonged to the liturgical class) never withdrew from festival competitions, citizens from the lower rungs of the leisure class, from the hoplite ‘middle class,’ and even a small number of lower-class individuals gained access, a development that led a few disgruntled καλοὶ κἀγαθοί to withdraw into equestrian events, which were beyond the means of all but the wealthiest. Pleket (1975:74) argues that in Athens the ephēbeia “functioned as a bridge for members of the urban elite between gymnasium sport and the world of the public contests.” I am not, of course, claiming for rhapsodic performance in Athens that a particular social class of citizens had thitherto monopolized the competition. My suggestion is that just as a world of expanding social opportunities for the Athenian upper classes through the ephēbeia and the athletic training of the gymnasium enabled their participation in public athletic contests, so also the expanding educational resources and opportunities for the lower rungs of the leisure class (i.e. the upper middle class) and even the growing popular orientation of the Athenian upper-class engagement with the democratic process (an orientation to which their interest in rhetorical training witnesses) must have helped and encouraged some of them to train for, and participate in, the rhapsodic events at the Panathenaia. On wealth, social status, education, and the Athenian democratic process see Ober 1989. For a recent survey of Greek education see Griffith 2001. On social status and Greek sports, see Pleket 2001 [1974, lightly updated]; Pleket 1975:71–74, Young 1984, Pleket 1992, and Golden 2008:23–26. On the ephēbeia (whether it already existed in the fifth century and included the thētes) see Pleket 2001:183; Rhodes 1993:503; and Raaflaub 1996:157, with 172–173nn148–149. For the Athenian ‘middle class’ see Hansen 1991:115–116 and Hanson 1996 (with qualifications from van Wees 2001 and 2002; and Gabrielsen 2002).

[ back ] 91. Cf. Pelliccia 2003:111.

[ back ] 92. Cf. Ford 2002:196n31. See further below, §12.3.1, item 2.

[ back ] 93. Alternatively, he might have found it preferable to acquiesce in the criticism of his dinner companions, feigning to share it rather than mount a defense of rhapsodes in the face of peer pressure. That he reportedly sought the ὑπόνοιαι of the likes of Stesimbrotos and Anaximander shows that he intended to overcome the commonly perceived limitations of the average rhapsodic training.

[ back ] 94. περὶ πάντων μὲν γὰρ τῶν πραγμάτων γεγραμμένους ἐπίστασθαι λόγους ἕν τι τῶν ἀδυνάτων πέφυκεν· ἀνάγκη δ’ ἐστίν, ὅταν τις τὰ μὲν αὐτοσχεδιάζῃ, τὰ δὲ τυποῖ, τὸν λόγον ἀνόμοιον ὄντα ψόγον τῷ λέγοντι παρασκευάζειν, καὶ τὰ μὲν ὑποκρίσει καὶ ῥαψῳδίᾳ παραπλήσια δοκεῖν εἶναι, τὰ δὲ ταπεινὰ καὶ φαῦλα φαίνεσθαι παρὰ τὴν ἐκείνων ἀκρίβειαν (‘For knowing written speeches about every subject matter is naturally one of life’s impossibilities. And, when one improvises some things and [carefully] molds others, it is inevitable that the speech, being uneven, furnish grounds for censuring the speaker; and that the molded portions appear closely to resemble rhapsodic performance while the improvised, next to the precision of the former, seem mean and careless’ §14).

[ back ] 95. Schloemann (2000:215n53) suggests “dichterische Deklamation.”

[ back ] 96. Whose style would be that of the “gebundene Rede der Kunstprosa” (Koller 1957:104). Koller connects the ὑπόκρισις of §14 to sections 16–17, which regard those who are accustomed to work out speeches in detail and compose them (doubtless, in writing) with precise diction and rhythms. Alkidamas implies that, unable for lack of talent or training to take advantage of the freedom offered by improvisation, sophists are brought back to the figures and rhythms (εἰς ἐκεῖνα τὰ σχήματα καὶ τοὺς ῥυθμοὺς ἀποφέρονται §17) constrained by writing.

[ back ] 97. So Mariß 2002: “Vielmehr steht ὑπόκρισις als Begriff für den Vortrag von dramatischer Dichtung eigenständig neben dem Rezitieren von—in erster Linie—epischen Texten” (195).

[ back ] 98. See section 14.1, pp. 579ff.

[ back ] 99. Although I do not agree with Schloemann (2000:214) that Aristotle contradicts himself in his use of ὑπόκρισις in Rhetoric Ⅲ.1 vis-à-vis Rhetoric Ⅲ.12, I join him in his conviction, convincingly substantiated by his excellent article, that ‘delivery’, as a tekhnē and not a natural talent, is throughout connected with writing and hence susceptible of displaying a range of stylistic devices and a varying degree of precision.

[ back ] 100. Pace Mariß 2002:195.

[ back ] 101. Isokrates Against the Sophists 15: ἡ δὲ παίδευσις τοὺς μὲν τοιούτους τεχνικωτέρους καὶ πρὸς τὸ ζητεῖν εὐπορωτέρους ἐποίησεν· οἷς γὰρ νῦν ἐντυγχάνουσι πλανώμενοι, ταῦτ’ ἐξ ἑτοιμοτέρου λαμβάνειν αὐτοὺς ἐδίδαξεν, τοὺς δὲ καταδεεστέραν τὴν φύσιν ἔχοντας ἀγωνιστὰς μὲν ἀγαθοὺς ἢ λόγων ποιητὰς οὐκ ἂν ἀποτελέσειεν, αὐτοὺς δ’ ἂν αὑτῶν προαγάγοι καὶ πρὸς πολλὰ φρονιμωτέρως διακεῖσθαι ποιήσειεν (‘Instruction makes such men more technically skillful and more resourceful in their discovery; for it teaches them to take from a readier source the things which they now randomly hit upon. And while it could not make good competitors or speech makers out of those who have an inferior nature, it could lead them on beyond themselves and render them more discerning in many respects’); Isokrates Antidosis 192: περὶ μὲν οὖν τῆς φύσεως καὶ τῆς ἐμπειρίας ταῦτα γιγνώσκω· περὶ δὲ τῆς παιδείας οὐκ ἔχω τοιοῦτον λόγον εἰπεῖν· οὔτε γὰρ ὁμοίαν οὔτε παραπλησίαν ἔχει τούτοις τὴν δύναμιν. εἰ γάρ τις διακούσειεν ἅπαντα τὰ περὶ τοὺς λόγους καὶ διακριβωθείη μᾶλλον τῶν ἄλλων, λόγων μὲν ποιητὴς τυχὸν ἂν χαριέστερος γένοιτο τῶν πολλῶν, εἰς ὄχλον δὲ καταστὰς, τούτου μόνον ἀποστερηθεὶς, τοῦ τολμᾶν, οὐδ’ ἂν φθέγξασθαι δυνηθείη (‘This, then, is what I know about natural talent and experience. But I cannot make a similar point about instruction, for it is neither similar nor about equal to these in power. For if one should exhaustively hear all that concerns oratory and should grow more accomplished than others, he might perhaps become a more pleasing speech maker than the rest. But were he to stand before the crowd lacking this alone, courage, he would not even be able to utter a word’).

[ back ] 102. Mariß (2002:99–100) essentially agrees with this view, noting that in On the Sophists, as in Plato’s Phaidros, the boundary between ‘poet’ and ‘speech maker’ is fluid. With ποιητής Alkidamas may well intend, as Mariß remarks, a disparagement of the orator who places too great a value on precise drafts, suggesting that he rather resembles a manual artisan. In §34, however, it is clear that ποιητὴς λόγων is the rhetorical wordsmith (not the poet) who, excessively dependent on writing, never rises to the level of the true orator. I agree with Mariß (2002:307) that δεινὸς ῥήτωρ and ποιητὴς λόγων are mutually exclusive at the rhetorical level of Alkidamas’ diatribe; but I do not think she is correct in denying Isokrates Against the Sophists 15 and Antidosis 192 a connection with delivery. In Against the Sophists 15, ἀγωνιστὰς μὲν ἀγαθοὺς ἢ λόγων ποιητὰς does not establish an opposition (“Gegenbegriff”) between ἀγωνισταί and ποιηταί: Isokrates does not specify the relationship between them. But since both are possible outcomes of the sophistic παίδευσις in the case of gifted men, both must a fortiori be included in the promise that πολιτικοὶ λόγοι hold for the formation of successful ῥήτορες (Against the Sophists 9). Thus, one should also see both as connected with delivery, for this is the end of such logoi. A man like Isokrates, who mostly writes but does not deliver them, is still at that time an anomaly. Isokrates Antidosis 192 confirms this view (pace Mariß 2002:306: “[Der Redenschreiber] hat bei ihnen [sc. Isokrates und Platon] jedoch, anders als bei Alkidamas, mit dem Vortrag nichts zu tun”); for here, Isokrates notes, however refined a ποιητὴς λόγων one might be in attaining mastery of all that pertains to speech-making, only let him lack courage and he will not be able to utter a word before a crowd. The obvious implication of this statement is that the ποιητὴς λόγων is not merely a writer, but also a speaker (as Norlin’s Loeb Classical Library translation renders it; Mathieu’s Budé, in turn, offers “un inventeur de discours”).

[ back ] 103. The sense offered by LSJ s.v. for Phaidros 277e, ‘to repeat by heart or rote’, is much too narrow; and restricting it to ‘reciting poems’ in Isokrates Panathēnaikos 18 and 33, though true, is misleading if we do not observe its contextual tie to ‘exposition’.

[ back ] 104. Isokrates Panathēnaikos 18: ἀπαντήσαντες γάρ τινές μοι τῶν ἐπιτηδείων ἔλεγον ὡς ἐν τῷ Λυκείῳ συγκαθεζόμενοι τρεῖς ἢ τέτταρες τῶν ἀγελαίων σοφιστῶν καὶ πάντα φασκόντων εἰδέναι καὶ ταχέως πανταχοῦ γιγνομένων διαλέγοιντο περί τε τῶν ἄλλων ποιητῶν καὶ τῆς Ἡσιόδου καὶ τῆς Ὁμήρου ποιήσεως, οὐδὲν μὲν παρ’ αὑτῶν λέγοντες, τὰ δ’ ἐκείνων ῥαψῳδοῦντες καὶ τῶν πρότερον ἄλλοις τισὶν εἰρημένων τὰ χαριέστατα μνημονεύοντες (‘some of my friends met me and told me that three or four of those herdlike, know-it-all sophists, prompt to be everywhere, were sitting together in the Lyceum and discussing the poets, especially the poetry of Hesiod and Homer, saying nothing of their own vintage but rhapsodizing the poets’ material and mentioning the finest things uttered before by certain others’).

[ back ] 105. ‘Repeating by rote’ or ‘from memory’ (so Norlin in the Loeb Classical Library) is too tendentious a translation for a verb that simply means ‘to call to mind, mention, say’ (cf. Lykourgos’ Against Leōkratēs §110.1). Such a marked gloss (for which there is no support in the LSJ) reads into the text a preconceived view of the activity that these sophists are engaging in.

[ back ] 106. Cf. Nagy 1996c:122–124.

[ back ] 107. ἀποδεξαμένων δὲ τῶν περιεστώτων τὴν διατριβὴν αὐτῶν … (§19).

[ back ] 108. Cf. §§26 and 33. This may have been en nuce the bone of contention between rhapsodes and those whom we today call sophists.

[ back ] 109. ἔπη is not found in the context and, although ποίησις is, it appears as the object of διαλέγεσθαι περί; i.e. as a subject of discussion, not of declamation.

[ back ] 110. From Plato’s Phaidros 277e8–9 and Alkidamas’ On the Sophists §14, one may infer that in the fourth century BC the average rhapsodic showpiece was largely, if not exclusively, monologic, a discourse that made little allowance for substantial input from the audience. With this in view, Hippias’ boast in the Lesser Hippias 363d that he is ready to answer any question posed to him by his audience can be seen, by comparison, as a pretentious claim for the superiority of his skill as a sophist.

[ back ] 111. Dialectic, of course, may prove competitive in actual practice.

[ back ] 112. Isokrates Panathēnaikos 33: περὶ μὲν οὖν τῶν πεπαιδευμένων τυγχάνω ταῦτα γιγνώσκων. περὶ δὲ τῆς Ὁμήρου καὶ τῆς Ἡσιόδου καὶ τῆς τῶν ἄλλων ποιήσεως ἐπιθυμῶ μὲν εἰπεῖν, οἶμαι γὰρ ἂν παῦσαι τοὺς ἐν τῷ Λυκείῳ ῥαψῳδοῦντας τἀκείνων καὶ ληροῦντας περὶ αὐτῶν (‘These then happen to be my opinions about the educated. Now, I am eager to speak about the poetry of Homer, Hesiod, and the rest, for I think that I could stop those in the Lyceum from rhapsodizing their things and prattling about them’).

[ back ] 113. Cf. the Suidas s.v. ῥαψῳδοί: ῥαψῳδῆσαι δέ ἐστι τὸ φλυαρῆσαι (4.287 Ρ no. 71.2–3 Adler); s.v. καταρραψῳδήσει: φλυαρήσει (3.52 Κ no. 748 Adler); and s.vv. εἰκῆ ῥαψῳδεῖ: ἀντὶ τοῦ φλυαρεῖ (2.524 ΕΙ no. 78 Adler).

[ back ] 114. Cf. DK 8 1 on Theagenes of Rhegion, on whom see §6.2 above.

[ back ] 115. Hence, it supports my preferred understanding of Alkidamas’ ὑποκρίσει καὶ ῥαψῳδίᾳ as ‘rhapsodic hypokrisis’ and, by implication, the reading of the Iōn offered above.

[ back ] 116. I am following the punctuation of Moreschini’s Budé text, placing with Stallbaum 1857b after λεχθῆναι the comma that Burnet (in the OCT) had put after γραφῆναι. Read on for my rationale.

[ back ] 117. An empiric aorist (Smyth §1930) or an aorist of description (Smyth §1932). Yunis (2011:238) calls it “gnomic.” The three types are closely related.

[ back ] 118. Heitsch (1993:64–65) translates, correctly, “ohne die Möglichkeit von Einrede und Erläuterung.” Some take ἀνάκρισις as ‘preliminary investigation’ vel sim. But this misses Plato’s point that the give-and-take of true conversation cannot happen when the text is already fixed. (Fixity, of course, is relative, allowing for multiple gradations. But, for Plato, shades of gray would only spoil his black and white analysis, and for this reason γραφῆναι should be read as an accurate description of the dominant tendency.) In this case the semantic component ‘preliminary’ is beside the point (LSJ correctly, contra the DGE s.v. ἀνάκρισις), for what matters is the probing of divergent points of view by careful questioning (the basic meaning of ἀνακρίνω). In other words, ἀνάκρισις amounts to the Sokratic ἔλεγχος by another name, and διδαχή is the learning netted by the investigation. Cf. Heitsch 1993:210n481. There is nothing to object, of course, to taking ἀνάκρισις for the question-and-answer format of the preliminary judicial procedure, so long as in the context of Phaidros 277e8–9 one limits its import strictly to this format and strips it of the notion ‘preliminary’. For more on the Athenian ἀνάκρισις see Harrison 1968–1971:2.94–105 and Todd 1993:126–127.

[ back ] 119. A similar criticism applies to Hackforth’s translation in Hamilton and Cairns 1961.

[ back ] 120. A quick survey of the adverbial instances adduced by various grammars shows that, in such cases, οὐδέ is never simultaneously conjunctive. As one would expect, δέ can only perform one or the other function. Thus, for example, Smyth §2931 illustrates the adverbial use with ἀλλ’ οὐδέ, ὅτ’ οὐδέ, οὐδ’ εἰ—all three of which exhibit accompanying conjunctions—and οὐδέ with a participle (ὅπου γὰρ ἐγὼ μὲν οὐδὲ πεπονθώς from Demosthenes 21.205). One must not confuse the structure of our Phaidros passage with the intensifying progression τὲ … δέ (Smyth §2981 and K-G Ⅱ.2.244 §520 Anm. 3). In Phaidros 277e we have an outer coordination τὲ … καὶ depending on ἡγούμενος (τε ἡγούμενος … , καὶ …), whose second clause (the καὶ-clause) includes the subordinate coordination γραφῆναι οὐδὲ λεχθῆναι. In actual τὲ … δέ constructions, what begins as if the speaker expects to add a second member is rounded instead by a contrast. This has the effect of placing emphasis on the second (contrasting) member. Of course, Yunis and those of his persuasion allege no contrast (or even intensification) between γραφῆναι and λεχθῆναι—only that the latter further illustrates the criticism levied against the former. The related οὔτε … οὐδέ construction, also by way of a contrastive δέ, implies an emphatic second member: ‘(both) not this but not that’. This can be freely rendered ‘neither this nor yet/even that’. Quite apart from the fact that this is not the construction of Phaidros 277e, for my claim that δέ cannot be simultaneously conjunctive and adverbial it is important to realize that the ‘yet/even’ is added to bring out the contrast in translation, and not because δέ performs both functions at once (again, the literal translation is ‘neither A but not B’). The intensifying effect—strictly speaking, the outcome of what Smyth considers a “harsh” “combination of copulative and adversative particles” (§2981)—is often brought out further by the addition of emphasizing particles like αὖ, γέ, or μήν (Smyth §2949 and K-G Ⅱ.2.290 §535g). It bears repeating that Sokrates intends no contrast between ‘written’ and ‘spoken’; the only interpretive point at issue is whether these are two successive facets of a singly conceived delivery or else the second is added as a further independent illustration of the deficiency imputed to writing.

[ back ] 121. Note that he does not treat thus the immediately preceding ἐν μέτρῳ οὐδ’ ἄνευ μέτρου.

[ back ] 122. To speak about the logical structure of Sokrates’ statement, stricto sensu ‘(not A) and (not B)’ (i.e. ¬A ʌ ¬B) does not in itself necessarily imply anything about the relationship of A to B; whether these two are conceived as independent alternatives or as somehow connected to each other is a judgment that must be made contextually. Naturally construed, however, Yunis’s ‘not A or even B’ precludes the notion that A and B may be two stages of one larger action. To illustrate with an example: the statement ‘I did not swindle the investors or spend other people’s money on my pleasures’ does not necessarily presuppose (though it may suggest) the charge that the ‘money’ spent came from the ‘investors’; but, naturally construed, ‘I did not swindle the investors and spend other people’s money on my pleasures’ would precisely deny a charge that involves a two-stage process: first, swindling the investors; then, spending the swindled money. The latter corresponds to what I have called the conjunctive, the former to the disjunctive, οὐδέ. I believe that conjunction, not disjunction, is the point of Sokrates’ words: ‘no speech has ever yet been written and (then) delivered that is worth much serious attention, as the ones rhapsodized are spoken for the sake of persuasion, without examination or instruction’. On my view, ‘rhapsodized’, as applied to the manner of delivery, by Sokrates’ (and, a fortiori, Plato’s) time would have implied the use of written scripts. To be clear about the nature of my objection to Yunis’s translation: perhaps his ‘or even’ is merely a free rendering intended to convey his judgment that Sokrates’ inclusion of λεχθῆναι is an afterthought to γραφῆναι. If so, I believe that he does not do full justice to the local context of Phaidros 277e and the overarching argument of the dialog. But by writing ‘(n)or even’, a common (and proper) gloss for οὐδέ in ‘οὔτε … οὐδέ’ statements, Yunis gives the impression that this choice, and hence the interpretation built on it, is recommended by an objective appeal to semantics and syntax. Footnote 120 immediately above is intended to foreclose this textual (as opposed to contextual) justification. Read on, in turn, for my contextual objection.

[ back ] 123. οὐ δεῖ δὴ τοιοῦτον, ὦ ἄνδρες Ἀθηναῖοι, παραπεπτωκότα καιρὸν ἀφεῖναι οὐδὲ παθεῖν ταὐτὸν ὅπερ ἤδη πολλάκις πρότερον πεπόνθατε.

[ back ] 124. The Loeb Classical Library translation by Vince reads: “Men of Athens, you must not let slip the opportunity that offers, nor make the blunder you have so often made before.” Contextually, it is clear that ‘making the blunder’ is one and the same with ‘letting the opportunity slip’. Hence, ‘nor’ is not equivalent to ‘or not’ (‘you must not let slip the opportunity or [you must] not make the blunder’) but to ‘and not’: ‘you must not let slip the opportunity and [you must] not make the blunder’. The same logic applies to Trevett’s translation: “You must not pass up such an opportunity, men of Athens, when it has fallen into your lap, nor suffer the same fate as you have suffered many times already” (2011:33–34).

[ back ] 125. Note how the statement that makes mention of both speaking and writing (258d4–5) is bracketed on either side by a sole reference to ‘writing’: τοῦτο μὲν ἄρα παντὶ δῆλον, ὅτι οὐκ αἰσχρὸν αὐτό γε τὸ γράφειν λόγους (258d1–2); τίς οὖν ὁ τρόπος τοῦ καλῶς τε καὶ μὴ γράφειν; (258d7). The latter (258d9–11) is of special significance because, by having 277d6–e8 echo its detailed formulation (including the oppositions ‘private vs. public’ and ‘poetry vs. prose’), Sokrates signals the unity of focus of the intervening discussion which he is then bringing to a close.

[ back ] 126. Cf. Plutarch’s interesting remarks apropos Demosthenes in Dēmosthenēs 8.4 and Yunis 1996:175n4, 241–247.

[ back ] 127. Cf. Hudson-Williams 1951 and Trevett 1996, with 425n2.

[ back ] 128. This forensic order is not invariably observed where expected. At 261b4, λέγεται precedes γράφεται (περὶ τὰς δίκας λέγεταί τε καὶ γράφεται τέχνῃ). This may well be because at that point Sokrates’ focus for a while has been off the use of writing and on the truth of the orator’s subject matter. In redirecting attention, at least in part, to the technology of writing, it is natural for Phaidros to mention Sokrates’ focus first (λέγειν) and then add to it what has meanwhile been left out of sight in their conversation. In fact, Sokrates has just defined rhetoric without reference to writing as ψυχαγωγία τις διὰ λόγων, οὐ μόνον ἐν δικαστηρίοις καὶ ὅσοι ἄλλοι δημόσιοι σύλλογοι, ἀλλὰ καὶ ἐν ἰδίοις (261a8–9).

[ back ] 129. The focus on truth also motivates the disjunction λέγει ἢ γράφει at 277b6, where true knowledge of one’s subject is again at issue. A comprehensive formulation while making a point that is indifferent to technology—the need to suit with a view to persuasiveness the particular variety of speech to each type of soul—accounts for the disjunctions ἐνδεικνύμενον ἢ λεγόμενον τέχνῃ ποτὲ λεχθήσεται ἢ γραφήσεται (271b7–8). A slightly different rationale explains λέγων ἢ διδάσκων ἢ γράφων (272b1): although their ostensible subject is ‘the intending orator’ (τὸν μέλλοντα ῥητορικὸν ἔσεσθαι 271d1), in actual fact, as the inclusion of διδάσκων intimates, the teacher of rhetoric is ultimately in view (i.e. a συγγραφεύς like Lysias 272b2–3; cf. 258c8). Hence, the three verbs may well refer respectively to the practice of rhetoric, its instruction, and its codification in a manual. To the extent that such manuals offered illustrative examples and models to imitate, as (partial) instantiations of written oratory they were not dissimilar from written speeches. For this reason, it is not inappropriate to apply γράφων more broadly to them also.

[ back ] 130. οὐ μὰ τὸν Δί’ οὐ παντάπασιν οὕτως, ἀλλὰ μάλιστα μέν πως περὶ τὰς δίκας λέγεταί τε καὶ γράφεται τέχνῃ, λέγεται δὲ καὶ περὶ δημηγορίας (261b3–5).

[ back ] 131. Yunis 2011: “Athenian politicians did not normally engage in forensic speech-writing or publish texts … until Demosthenes changed the practice in the mid fourth century” (171).

[ back ] 132. In fact, as Thompson 1868:142 ad 277d observes, “[t]he original question which has been so long delayed, is now shown by Socr[ates] to have been virtually disposed of in the course of the foregone discussions.” It is somewhat incongruous that criticism levied against Lysias by τὶς … τῶν πολιτικῶν (257c4–5) might prompt broad consideration of the role of writing in rhetoric. Lysias, after all, was not an Athenian citizen and could not participate in the city’s politics (cf. Yunis 1996:175–176). The speech by Lysias that opens the dialog is best classified as epideictic, a type that according to Aristotle Rhetoric 1414a17 did especially invite the written style. The status of Lysias as a resident alien explains Sokrates’ expressly widening the scope to … ἤ τις ἄλλος (277d6). At any rate, the political domains of assembly and courtroom were notoriously porous, and even the profile of the alleged critic is unclear: described as ‘one of the politikoi’, was he “one of our politicians” (with Hackforth in Hamilton and Cairns 1961:503), which suggests a public figure, a professional of politics, or simply “a man [involved] in political life” (with Thompson 1868:83), which suggests an engaged but private citizen?

[ back ] 133. See Phaidros’ reply at 264b9-c1. Seemingly missing Sokrates’ ironic use of λογογράφος and its derivatives at 257e3, 258b4, 258c2, and 264b7, Yunis holds that he is referring broadly to “composing speeches”: “With regard to λογογραφία S[ocrates] shifts from the conventional sense ‘forensic speechwriting,’ as used by Ph[aedrus] (257c5), to a literal sense ‘composing speeches,’ i.e. of any kind” (Yunis 2011:172; cf. 193). Though doubtless right to observe that the term is not restricted to forensic writing, this fact does not imply that it designates anything other than professional wordsmiths who depend on writing for their trade. Only ironically, by a tendentious abuse of its denotation, is it made extensive generally to those who move legislation in the assembly and specifically to Lykourgos, Solon, and Dareios.

[ back ] 134. σοφίας δὲ τοῖς μαθηταῖς δόξαν, οὐκ ἀλήθειαν πορίζεις· πολυήκοοι γάρ σοι γενόμενοι ἄνευ διδαχῆς πολυγνώμονες εἶναι δόξουσιν, ἀγνώμονες ὡς ἐπὶ τὸ πλῆθος ὄντες, καὶ χαλεποὶ συνεῖναι, δοξόσοφοι γεγονότες ἀντὶ σοφῶν (Phaidros 275a7–b2).

[ back ] 135. Cf. Phaidros 275c5–d1.

[ back ] 136. A similar criticism can be found in Plato’s Gorgias 448d1–10, now directed against sophistic rhetoric, which falls short of the give-and-take that the much more interactive and responsive Sokratic method is capable of. This is perhaps why Gorgias’ willingness to take questions as part of his display is noted as remarkable (447c5–8).

[ back ] 137. Stallbaum 1857b:216 ad loc. translates the passage thus: “Qui vero in scripto de quacunque re sermone necessario multum lusum inesse arbitratur, nec unquam ullum sermonem versibus vel sine versibus multo dignum studio scriptum putat aut dictum esse, sicuti ῥαψῳδούμενοι illi sermones nulla adhibita disquisitione et explicatione persuadendi causa recitari consueverunt … .” And he adds the following note: “In his verba: ὡς οἱ ῥαψῳδούμενοι — ἐλέχθησαν, nescio cur interpretibus tam multum negotii creaverint. Continent enim explicationem praecedentis οὐδὲ λεχθῆναι. Nam quum Socrates negavisset ullam unquam orationem magno dignam studio scriptam aut dictam esse, alterum hoc accuratius erat explicandum. Monet igitur hoc de iis tantum valere orationibus, quae sine accurata rerum pervestigatione recitarentur eo tantum consilio, ut efficeretur aliqua persuasio. Quod nemo est quin videat de orationibus sophisticis et forensibus esse accipiendum” (216). I would only take issue with the aut between scriptam and dictam: Sokrates does not worry so much about written speeches as about orally delivered speeches that, by adhering to a script, speciously supplant the true orality of dialog.