On the paraphrase of Iliad 1.012–042 in Plato’s Republic 3.393d–394a

2016.09.23 | By Gregory Nagy

For Plato, mimesis is not a re-enactment as it is for rhapsodes: it is mere imitation. And it is easier to discredit such imitation when you hear a rhapsode paraphrase Homer in prose.

The tears of the old father Priam, pictured here at the moment of Hector’s final farewell to his parents, will in the end be respected by Achilles in Iliad 24. To be contrasted is what happens in Iliad 1, where the tears of the old father Chryses are harshly disrespected by Agamemnon. Attic red-figure neck-amphora: Departure of Hector, detail, Priam in tears. Attributed to the Peleus Painter, ca. 475-425 BCE. Vatican Museums, Museo Gregoriano Etrusco Vaticano, 16570.
The tears of the old father Priam, pictured here at the moment of Hector’s final farewell to his parents, will in the end be respected by Achilles in Iliad 24. To be contrasted is what happens in Iliad 1, where the tears of the old father Chryses are harshly disrespected by Agamemnon.
Attic red-figure neck-amphora: Departure of Hector, detail, Priam in tears. Attributed to the Peleus Painter, ca. 475–425 BCE. Vatican Museums, Museo Gregoriano Etrusco Vaticano, 16570. Illustration by Valerie Woelfel, with refinements by Jill Curry Robbins.


This comment is inspired by a point that David Elmer makes in his comment on verses 366–392 of Iliad 1, where Achilles retells to his mother Thetis what was told in the Homeric narrative at the beginning of the Iliad, at verses 12–42 of Iliad 1. Just as the words of Socrates in Plato’s Republic 3.393d–394a retell that same narrative, verses 12–42 of Iliad 1, without “quoting” any of the speakers who were featured in the action, so also the words of Achilles himself at verses 366–392 of Iliad 1 retell the same action without any “quotation” of direct speech. This point made by Elmer is relevant, I argue, to the concept of paraphrase as it is conventionally applied to what Plato has done here at Republic 3.393d–394a: supposedly, he is simply turning into prose the poetry that we read at the beginning of the Homeric Iliad, verses 12–42. In terms of my argumentation, however, things are not all that simple.

Paraphrasing in poetry as well as in prose

To begin, I highlight a formulation added by Elmer to supplement the point that he makes about the wording of Plato where Socrates paraphrases Homer: this wording, Elmer says, is “mimetic” just as the wording of Achilles is “mimetic,” insofar as the diēgēsis (διήγησις) or ‘narrative’ that is performed here by Plato’s Socrates “is embedded within Plato’s mīmēsis [μίμησις] of Socrates.”

In terms of Elmer’s interpretation, Plato’s mīmēsis of Socrates—where I translate mīmēsis as ‘re-enactment’—is parallel to the mīmēsis ‘re-enactment’ of Homer, as it were, by Achilles. The master narrator of the Iliad—let us continue to call him ‘Homer’—is telling the diēgēsis or ‘narrative’ at the beginning of the Iliad and is thus making a mīmēsis or ‘re-enactment’ of what was done and what was said in the Iliad, but then Achilles, in retelling the same narrative, performs a mīmēsis ‘re-enactment’ of Homer in the act of narrating that narrative.

Plato’s simplification of mimesis

This interpretation, I must emphasize, differs from what is claimed by Plato’s Socrates after he finishes his retelling of the Homeric narrative as it was originally told at verses 12–42 of Iliad 1: the retold narrative of Socrates, says Plato’s Socrates at Republic 3.394b, ‘becomes a simple diēgēsis without mīmēsis’, ἄνευ μιμήσεως ἁπλῆ διήγησις γίγνεται.

In terms of this claim made by Plato’s Socrates, there is no mīmēsis ‘re-enactment’ in your diēgēsis ‘narrative’ if you do not actually quote whatever is being spoken by the characters inside the diēgēsis. That seems at first to be the whole point of the mental exercise performed by Plato’s Socrates within the framework of Republic 3.393c–394b.

Paraphrasing the quotations that inaugurate the Homeric Iliad

The two characters who are actually quoted by ‘Homer’ at verses 12–42 of Iliad 1 are:

• a grieving father who is trying to free his captive daughter: he is a priest named Chryses
• a cruel despot who possesses the captive woman: he is Agamemnon, over-king of the Achaeans.

At Republic 3.393d, before he begins his paraphrase of what these two Homeric characters are quoted as saying in the Iliad, Plato’s Socrates claims that he will not paraphrase the entire beginning of the Iliad but will start where ‘Homer’ starts quoting the characters in the narrative. While saying that he will not paraphrase what happens at the beginning of the narrative before the quotations start, however, he is actually doing exactly what he says he will not do: at Republic 3.393d, Plato’s Socrates starts by paraphrasing verses 12–15 of Iliad 1, which precede the verses containing the quotations: he tells how the grieving father Chryses came with offers of supplication to Agamemnon in a quest to free his daughter from captivity. All this, as told at verses 12–15 of Iliad 1, is paraphrased by Socrates at Republic 3.393d in the course of his saying that he will not paraphrase this introductory narrative preceding what happens at verses 16–21 of Iliad 1, where the words of supplication spoken by the priest are then quoted directly.

After the quotation of the priest’s words at verses 16–21 of Iliad 1, the narrative resumes at verse 22, where we hear that the Achaeans ‘gave their approval’, as expressed by the verb ep-eu-phēmeîn, of what the priest was imploring Agamemnon and his Achaean warriors to do by way of supplicating them. More specifically, as we hear at verses 22–23, the Achaeans approved of the request that Agamemnon should show respect for the divine order by heeding the supplication of the old father. But Agamemnon refuses the supplication, as we hear from the narrative at verses 24–25, and now at verses 26–32 we hear a quotation of the harsh words of refusal as uttered by the king, who threatens the old priest with harm if he persists with his supplication and does not go away. Then at verses 33–36 we hear how the old father, frightened by the king’s threat, goes away without saying another word and finds a remote place where he may pray to the god Apollo in private. And then at verses 37–42 we hear the priest Chryses being quoted as he prays to Apollo and implores the god to punish Agamemnon and his Achaeans for disrespecting the tears of an old father.

In the mental exercise of retelling in Republic 3.393d–394a this whole narrative as found in verses 12–42 of Iliad 1, all the Homeric quotations are missing, and such a retelling is for Plato’s Socrates a thing that is now ἄνευ μιμήσεως ‘without mīmēsis’, 3.394b. Both for Plato’s Socrates and for Plato himself, as we see here, a diēgēsis ‘narrative’ that contains no quotations of the speakers involved in the action is no mīmēsis ‘re-enactment’ at all: rather, such a narrative becomes a ἁπλῆ διήγησις (haplê diēgēsis), ‘a simple narrative’, 3.394b.

A rhapsodic view of mimesis

I challenged such a view of mīmēsis ‘re-enactment’ in Chapter 3 of the book Poetry as Performance (1996). In the era of Plato and Aristotle,  I argued there, the rhapsōidos ‘rhapsode’ as a professional performer of Homer was engaged in a kind of mīmēsis ‘re-enactment’ that re-created not only the characters of heroic song but also the composer and prototypical performer of that song, Homer himself. Here is how I formulated the argument (PP 80):

From a modern point of view that sees Homer as the author of a text, the re-creating of a real Homer in the performance of a rhapsode may not even seem to be a matter of mimesis. Even for Plato and Aristotle, a straightforward third-person narration in heroic song is technically a matter of diegesis or ‘narration’ as opposed to mimesis. In contemplating the ‘I’ of ‘tell me, Muses’ or ‘tell me, Muse’, we find ourselves at a loss in finding the element of the mimetic—or, to say it in a more modern way, the dramatic. And yet, my claim is that this ‘I’ is perhaps the most dramatic of all the characters in heroic song—once we see this song on the level of performance as well as composition. This ‘I’ is Homer speaking. For us, however, his role is no longer overt as it had been for audiences of Homeric song, and Homer has lost his power as a dramatic persona.

In line with this formulation, there is a chain of mimesis here from the standpoint of a professional rhapsode. First, the narrative at verses 12–42 of Iliad 1 is mimetic in the sense that the rhapsode is making a mimesis of ‘Homer’ in the act of narrating the Iliad. Second, the narrative is double-mimetic in the sense that ‘Homer’ makes a mimesis of himself making a mimesis of the old priest and the over-king as they interact and speak with one other. Third, the retelling by Achilles at verses 366–392 of Iliad 1 is triple-mimetic in the sense that ‘Homer’ makes a mimesis of himself making a mimesis of Achilles making a mimesis of the old priest and the over-king.

To push this chain even further, the retelling of Homer as performed by Plato’s Socrates at Republic 3.393d–394a is quadruple-mimetic in the sense that Socrates is shown in the act of re-enacting —or let us say simply imitating—a professional rhapsode in the act of interpreting Homer. I analyze this aspect of rhapsodic activity in the book Plato’s Rhapsody and Homer’s Music (2002), where I argue that the professional rhapsode in the era dramatized by Plato not only re-enacted Homer by way of performing the poetry of Homer: besides performing the poetry, the rhapsode could also interpret Homer by way of performing in prose his own paraphrases of what Homer said—and then by performing interpretations based on such paraphrases.

Evidence for rhapsodic technical language in Plato’s Ion

Here is a compressed summary of my relevant argumentation in Plato’s Rhapsody and Homer’s Music, organized by way of listing ten examples of rhapsodic technical language as brought to life in the Platonic dialogue Ion, where Socrates compares notes, as it were, with a professional rhapsode named Ion (PR 25–32):

1. hormân + accusative, in the sense of ‘move, get [the performer] started, inspire’. The key passage is at Ion 534c3. In this context, where Plato’s ear catches the technical aspects of a rhapsodic expression, we read that the Muse ‘moves’ various different kinds of poets to produce their various different kinds of poetry: the Muse ‘moves’ Homer to make epic just as she ‘moves’ other poets to make dithyrambs, encomia, and so on. The Homeric context of the word hormân makes it clear that inspiration by the Muse happens in the context of performance, and it has to happen from the very start: the Muse has to ‘start’ the performer. At Odyssey 8.499, we see the blind singer Demodokos about to start his performance: hormētheis theou arkheto (ὁρμηθεὶς θεοῦ ἄρχετο) ‘getting started, he began with the god’. That is, the performer got started or was ‘moved’ by the Muse and then he began his performance, starting by hymning a god. What follows this start, as we hear it paraphrased by the Odyssey, is an epic account of the Iliou Persis, the destruction of Troy (Odyssey 8.500–520).

2. āidein ‘narrate’ (= ‘sing’) + accusative of a given topic of song, which must be named at the very beginning of the performance. The topic, signaling a given epic event or a given epic character defining the event, must be in the accusative case. When Homer or the rhapsode ‘sings’ in the accusative that given event or character, he notionally conjures them, bringing them back to life in the process of performance. It is a common feature of oral poetics that the events mentioned in performance become part of the event that is the performance and that the characters featured in the events become members of the audience attending the performance in the here-and-now (on this point, see Martin 1989:xiv). The key passage is at Ion 535b3–7, where we find a veritable catalogue featuring “accusatives of the rhapsodic topic” following āidein (ᾄδῃς): (1) Odysseus at the epic moment when he leaps upon the threshold, ready to shoot arrows at the suitors; (2) Achilles as he lunges at Hector; (3) some other highlighted thing (ti, accusative) from epic moments, as when (3a) Andromache bids farewell to Hector, or from other similar epic moments involving (3b) Hekabe or (3c) Andromache. We may compare the Homeric usage of aeidein = āidein ‘narrate’ (‘sing’) + accusative of the topic, such as the wrath of Achilles in Iliad 1.1. Thus the rhapsode’s topics are put into the same dimension of heroic-age “reality” as Homer’s topics. The rhapsode performs as someone who is parallel to and in continuity with Homer (further analysis in PP 60–64). In the case of ‘Homer’, he of course starts his topic at the beginning—as at Iliad 1.1. As for the rhapsode, his topics can start anywhere in Homeric poetry, as we have just seen from the catalogue of heroic topics at Ion 535b3–7.

3. ep-aineîn + Homēros (in accusative) ‘quote Homer’. This expression refers to the ‘quoting’ of Homer in medias res, in a specific context and for a specific purpose. The key passages are at Ion 536d6, 541e2 (agent noun ep-ainetēs, 536d3, 542b4; see also the comment at BA 98n). The specific purposes, as in the Ion, have to do with arguing specific points. We may compare the usage of ep-aineîn in Lycurgus Against Leokrates 102, where the orator ‘quotes’ Homer in order to make his specific case. Aside from the various specific purposes involved in this activity of ‘quoting’ Homer, there is of course one overriding general purpose, from an Athenian point of view: that is, the State officially ‘quotes’ Homer to its assembled citizens on the occasion of its highest holiday, the Panathenaia, in the format of rhapsodic competitions. On this occasion, each competing rhapsode gets the chance to ‘quote’ Homer before a general audience of 20,000 persons (535d3)—a round figure that seems notionally equivalent to the body politic of Athens (details in PR 28). In this case, each competing rhapsode would be required to take up the Homeric narrative continuum where the previous rhapsode had left off (PR 28). This rhapsodic imperative of maintaining continuum is relevant to the etymology of ep-aineîn: ‘to continue [epi-] making praise [ainos] for’ (+ accusative of the laudandus as the receiver of praise or of the laudator as the ultimate giver of praise). By implication, rhapsodic art is a continuation of praise poetry (details in PR 28). The idea of continuum is explicit in the epi- of ep-aineîn.

4. dianoia ‘train of thought’, applying primarily to Homer’s train of thought, not to the rhapsode’s. The key passages are at Ion 530b10, c3, d3. The rhapsode can enter into Homer’s train of thought at any point of the continuum that is the narrative. He can enter into it midstream, in medias res. To be able to join the Homeric narrative in progress is to know the dianoia of Homer. As such, the rhapsode is the hermēneus ‘interpreter’ of the dianoia of Homer (530c3; see no. 5 below). Since the rhapsode can become part of Homer’s train of thought, of Homer’s dianoia, he can also tell the thoughts of Homer as a verbal commentary (so, not necessarily a written commentary) about Homer (530c9; see no. 6 below). Such ‘commenting’ thoughts become, by extension, dianoiai as well: 530d3. On Socrates’ different ‘understanding’ of dianoia, see no. 6 below. The idea of continuum is explicit in the dia- of dianoia.

5. hermēneus ‘interpreter’, applied to the rhapsode as one who must know the dianoia of Homer on behalf of his audiences. The key passage is at Ion 530c3: τὸν γὰρ ῥαψῳδὸν ἑρμηνέα δεῖ τοῦ ποιητοῦ τῆς διανοίας γίγνεσθαι τοῖς ἀκούουσι ‘you see, it is imperative that the rhapsode [rhapsōidos] should become an interpreter [hermēneus] of the poet’s train-of-thought [dianoia] for his listeners’. Here we see the essence of the rhapsode’s “hermeneutics”: everything depends on his knowing Homer’s dianoia ‘train of thought’ (see no. 4; for more details, see PR 29). There are further applications of the word hermēneus at 535a6, a9 (see also 534e4). This concept of an ‘interpreter’ or ‘go-between’ acknowledges the reality of a mental gap between Homer on one side and his audience in the here-and-now on the other side. That gap can be bridged by the rhapsode, whose mind can implicitly neutralize the distance that separates the two sides.

6. legein peri + Homēros (in genitive) ‘make a verbal commentary on Homer’. The key passages are at Ion 530c9, d2–3. Here Ion is reacting to the claim of Socrates that a rhapsode is expected to be a hermēneus ‘interpreter’ of a poet like Homer, and that therefore Ion must surely know the poet’s ‘intention’, that is, his dianoia (531c). By using the literary word “intention” here, I am seeking to find a common ground between the specialized Socratic/Platonic understanding of dianoia as ‘intellect’ (as at Republic 6.511d) and a more general understanding of the word as reflected by the primary definition in the dictionary of Liddell, Scott, and Jones: “thought, i.e. intention, purpose” (details in PP 124). When Socrates uses the word dianoia at Ion 531c, he understands it to mean Homer’s intellectual capacity as revealed by his words (see also Aristotle Poetics 1450a6, etc.). Affirming his own rhapsodic understanding of dianoia as ‘train of thought’, Ion replies that he can indeed ‘speak most beautifully about Homer’, more so than any of his predecessors could speak about Homer (καὶ οἶμαι κάλλιστα ἀνθρώπων λέγειν περὶ Ὁμήρου 530c; see also 533c–d), and that the dianoiai that he ‘speaks about Homer’ are more beautiful than those spoken by any of his predecessors, including Metrodorus of Lampsacus, Stesimbrotus of Thasos, Glaucon, etc. (ὡς οὔτε Μητρόδωρος ὁ Λαμψακηνὸς οὔτε Στησίμβροτος ὁ Θάσιος οὔτε Γλαύκων οὔτε ἄλλος οὐδεὶς τῶν πώποτε γενομένων ἔσχεν εἰπεῖν οὕτω πολλὰς καὶ καλὰς διανοίας περὶ Ὁμήρου ὅσας ἐγώ, 530c–d; more at PP 124–125).

7. exēgeîsthai ‘speak authoritatively, make an exegesis’ about Homer. The key passages are at Ion 531a7, b8, b9; 533b8; cf. 533b1; at 531a7, the word picks up the idea of legein peri + Homēros (in genitive) at 530c9 (PP 125n81). See no. 6 above.

8. diatribein ‘perform’ (that is, perform rhapsodically). The key passage is at Ion 530b8. We may compare Isocrates Panathenaicus 19, where diatribē refers to the ad hoc performances of ‘sophists’ at the Lyceum who are described at 18 as ‘performing rhapsodically’ (rhapsōidountes) the poetry of Homer, Hesiod, and other poets; at 33, Isocrates refers again to the same ‘sophists’ at the Lyceum who are ‘performing rhapsodically’ (rhapsōidountas) and who also ‘speak about’—stupidly—Homer, Hesiod, and other poets (lēreîn peri + genitive; to be compared with no. 6 above; see also PP 123–124). Their activity of speaking about Homer, Hesiod, and other poets is described as dialegesthai (διαλέγοιντο, 18), on which see no. 10 below.

9. mnēsthēnai (and related forms) ‘make mention’ concerning a sequence from Homer within an exegetical frame, that is, to ‘quote’ it within such a frame and also to make comments or make a commentary. The key passages are at Ion 532c2, 536c7. As in the case of no. 6 above, what is meant is to ‘make a verbal commentary’ (further details at PP 122–125). Where mnēsthēnai takes the accusative case, it means ‘recall’, as when Socrates is trying to recall some verses from Homer (ἐὰν μνησθῶ τὰ ἔπη, 537a2; see also HQ 152). The rhapsode notes that his attention is always awakened when someone mnēsthēi ‘makes commentaries’ about Homer (ἐπειδὰν δέ τις περὶ Ὁμήρου μνησθῇ, 532c2). Later on in the Ion, the same theme of the rhapsode’s awakened attention is transferred from the act of making commentaries about the poet (περὶ μὲν Ὁμήρου ὅταν τις μνησθῇ, 536c7) to the act of actually performing or ‘singing’ something from a poet (ἐπειδάν μέν τις ἄλλου του ποιητοῦ ᾄδῃ 536b6; see also PP 125n88). On ‘singing’, see no. 2 above.

10. dialegesthai in the sense of ‘engage in dialogue’ about a given poet. The key passage is at Ion 532b9 (ὅταν μέν τις περὶ ἄλλου του ποιητοῦ διαλέγηται). This term appears in a context that is parallel to the context of mnēsthēnai at 532c (see PP 125n80). See no. 9 above.

Thoughts about Plato’s discrediting of mimesis in the Republic

Viewing these ten examples of technical rhapsodic terminology as used by Plato, we can see that he fully understands the mentality of rhapsodes: these professional performers and interpreters of Homer claim to possess the authority of Homer because they can supposedly re-enact Homer’s thinking as well as his performance. For the rhapsodes, even a paraphrasing of Homer is such a re-enactment. For Plato, however, mimesis is not a re-enactment as it is for rhapsodes: it is mere imitation. And it is easier to discredit such imitation when you hear a rhapsode paraphrase Homer in prose: now the poetic power is lost, since the voice of the supposedly original speaker, Homer, is no longer there. To imitate Homer is not to be Homer, and the imitation of Homer is most evident when a rhapsode tries to speak Homer’s mind without even speaking by way of poetry. That is why Plato’s Socrates demonstrates the most vulnerable aspect of rhapsodic authenticity by performing a Homeric paraphrase as if he were a rhapsode himself.

But Plato here and there shows off his own rhapsodic expertise when he lets Socrates paraphrase Homer, since the prose of the paraphrase performed by Socrates is not always ‘simple diegesis’. At times Plato’s Socrates slips into Homeric poetry when he is ostensibly performing non-Homeric prose. I find the most obvious example at Republic 3.394a, at the moment when the old priest Chryses is praying to Apollo to punish Agamemnon and his Achaeans. The god, says the grieving old man, must make the transgressors pay for his fatherly tears, and the paraphrase refers to these tears by way of wording that is perfectly Homeric: τὰ ἃ δάκρυα ‘his tears’. This is not prose. It is poetry, Homeric poetry, with a transformation of the Homeric phrase ἐμὰ δάκρυα ‘my tears’, spoken by the old man at verse 42 of Iliad 1, into the equally Homeric τὰ ἃ δάκρυα ‘his tears’ as spoken by that prodigious pseudo-rhapsode, Socrates himself.



BA. See Nagy 1979.

HQ. See Nagy 1996b.

Martin, R. P. 1989. The Language of Heroes: Speech and Performance in the Iliad. Ithaca. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Martin.The_Language_of_Heroes.1989

Nagy, G. 1979/1999. The Best of the Achaeans: Concepts of the Hero in chaic Greek Poetry. Baltimore. Revised ed. with new introduction 1999. https://chs.harvard.edu/book/nagy-gregory-the-best-of-the-achaeans-concepts-of-the-hero-in-archaic-greek-poetry/.

Nagy, G. 1996a. Poetry as Performance: Homer and Beyond. Cambridge. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Poetry_as_Performance.1996.

Nagy, G. 1996b. Homeric Questions. Austin. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Homeric_Questions.1996.

Nagy, G. 2002. Plato’s Rhapsody and Homer’s Music: The Poetics of the Panathenaic Festival in Classical Athens. Cambridge, MA and Athens. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Platos_Rhapsody_and_Homers_Music.2002.

PP. See Nagy 1996a.

PR. See Nagy 2002.