Homeric Responses

  Use the following persistent identifier: http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Homeric_Responses.2003.

Introduction. Homeric Responses

Four Questions

Question 1: About synchronic and diachronic perspectives

Question 2: About the evolutionary model

In terms of an evolutionary model, there were at least five distinct consecutive periods of Homeric oral / written transmission – “Five Ages of Homer,” as it were – with each period showing progressively less fluidity and more rigidity: [7]

  1. A relatively most fluid period, with no written texts, extending from the early second millennium BCE into the middle of the eighth century;
  2. A more formative or “pan-Hellenic” period, still with no written texts, from the middle of the eighth century BCE to the middle of the sixth; [8]
  3. A definitive period, centralized in Athens, with potential texts in the sense of transcripts, at any or several points from the middle of the sixth century BCE to the later part of the fourth; this period starts with the reform of Homeric performance traditions in Athens during the regime of the Peisistratidai; [9]
  4. A standardizing period, with texts in the sense of transcripts or even scripts, from the later part of the fourth century to the middle of the second; this period starts with the reform of Homeric performance traditions in Athens during the regime of Demetrius of Phalerum, which lasted from 317 to 307 BCE; [10] {2|3}
  5. The relatively most rigid period, with texts as scripture, from the middle of the second century onward; this period starts with the completion of Aristarchus’s editorial work on the Homeric texts, not long after 150 BCE or so, which is a date that also marks the general disappearance of the so-called eccentric papyri. [11]

A further distinction must be made between transcript and script. A transcript merely records a given performance and has no direct bearing on the traditions of performance. A script, on the other hand, controls the performance, making it uniform and keeping it from becoming multiform.

Question 3: About dictation models

The reconstruction of these distinctions is a far cry, however, from a blanket assumption of an “essential difference” between aoidos and rhapsōidos. There are points of convergence as well as divergence between the two words, and my evolutionary model posits a historical continuum linking one to the other.

Question 4: About cross-references in Homer

My twofold thesis is linked to my evolutionary model for the making of Homeric poetry. Such a model can be applied to the major question at hand, that is, how is cross-referencing possible within an oral medium?

I disagree with many aspects of this formulation, though I find it useful for pursuing the topic at hand. I disagree especially with the presuppositions at the end of the statement, as italicized. For similar reasons, I disagree even more with this statement by J. S. Clay:

Clay follows up with this description of my method:

What is missing in this description of my method is the most essential aspect: I insist on the need to find a poetic rationale, a teleology of meaning, in the process of cross-referring over time. In other words, there is a poetic system involved in the very act of Homeric cross-reference.

Further, I call into question the assumption that a “closed tradition” is to be equated automatically with “one that has become fixed into a text.” I resist the absolutism inherent in such a description. It is as if we could find some kind of absolute distinction between “closure” and “open-endedness,” corresponding to a surface distinction between written and oral traditions.

It seems to me that we could easily think of any given tradition, written or oral, as relatively “closed” – once we see it becoming internalized by the person who controls it. In other words, the concept of “closure” is relative to start with, not absolute. Any tradition can be considered “closed” to the extent that it becomes internalized by its practitioner. But the point is that the same tradition can be considered “open-ended” to the extent that it remains externalized for other potential practitioners. Even a relatively fixed tradition, notionally closed, cannot necessarily prevent successive reopenings of that same tradition by new practitioners.

We see at work here the communicative power of tradition. Even if a given tradition is closed to outsiders, it will remain open to insiders. Further, even if such a tradition is kept internalized by one person, it can become externalized by some other person who has also already internalized it. That is what I mean by the “Nostradamus effect.”

Such an illustration of independent internalizations by way of written tradition can help us appreciate all the more the relative open-endedness of oral traditions, which are generally not constrained by ideas of fixity suggested by the very existence of a written text.

Having outlined my overall approach to questions of “text,” “tradition,” and “cross-reference,” I now offer a direct application, focusing on a specific passage taken from Homeric poetry. As a way of introducing the passage I have in mind, I quote what Austin has said about my approach, with specific reference to a theme signaled by this same passage. The reference concerns a quarrel between Odysseus and Achilles, as narrated briefly in Odyssey 8.72-83: {12|13}

Let us take a close look at this Homeric passage:

72αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος ἐξ ἔρον ἕντο,
73Μοῦσ’ ἄρ’ ἀοιδὸν ἀνῆκεν ἀειδέμεναι κλέα ἀνδρῶν,
74οἴμης, τῆς τότ’ ἄρα κλέος οὐρανὸν εὐρὺν ἵκανε,
75 νεῖκος Ὀδυσσῆος καὶ Πηλεΐδεω Ἀχιλῆος,
76ὥς ποτε δηρισαντο θεῶν ἐν δαιτὶ θαλειῃ
77ἐκπάγλοις ἐπέεσσιν, ἄναξ δ’ ἀνδρῶν Ἀγαμέμνων
78χαῖρε νόῳ, ὅ τ’ ἄριστοι Ἀχαιῶν δηριόωντο.
79ὣς γάρ οἱ χρειων μυθήσατο Φοῖβος Ἀπόλλων
80 Πυθοῖ ἐν ἠγαθέῃ, ὅθ’ ὑπέρβη λάϊνον οὐδὸν
81χρησόμενος. τότε γάρ ῥα κυλινδετο πήματος ἀρχὴ
82Τρωσι τε καὶ Δαναοῖσι Διὸς μεγάλου διὰ βουλάς.
83ταῦτ’ ἄρ’ ἀοιδὸς ἄειδε περικλυτός· αὐτὰρ Ὀδυσσεὺς …
72But when they had their fill of drinking and eating,
73The Muse impelled the singer to sing the glories [ kleos plural] of men,
74from a story-thread [
70] which had at that time a glory [ kleos ] reaching the vast heavens:
75the quarrel [ neikos ] of Odysseus and Achilles son of Peleus,
76how they once upon a time [ pote ] fought at a sumptuous feast of the gods,
77with terrible words, and the king of men, Agamemnon,
78rejoiced in his mind that the best of the Achaeans were fighting.
79For [ gar ] thus had oracular Phoebus Apollo prophesied to him, {13|14}
80at holy Delphi, when he [Agamemnon] had crossed the stone threshold
81to ask the oracle. For [ gar ] then [ tote ] it was that the beginning of pain [ pēma ] started rolling
82upon both Trojans and Danaans, on account of the plans of great Zeus.
83These things, then, the singer sang, whose fame is far and wide. But Odysseus…

Odyssey 8.72-83

I see in these verses an Odyssean cross-reference to the Iliadic tradition. [
71] My focus is on the wording of verse 81, tote gar ‘for then it was…’, where the gar ‘for’ refers back to the time of the neikos ‘quarrel’ at verse 75, not to the time of Agamemnon’s consultation of Apollo’s oracle at verse 79. The pēma ‘pain’, prophesied by Apollo, kulindeto ‘started rolling’ at the precise moment when the neikos ‘quarrel’ got under way. The wording of verse 81, I must stress, cross-refers not only to the mention of the quarrel at verse 75 but also to the precise moment of that quarrel: “By virtue of cross-referring to a specific point in epic time, the wording tote gar ‘for then it was…’ at verse 81 cross-refers also to a specific point in a notionally total and continuous narration extending into the current narrative.” [72] What we see at work here is “the essential notion, inherent in oral poetic traditions, of a total and continuous narration, of {14|15} which any given performance is but a part.” [73] The tote ‘then’ of verse 81 is a precise cross-reference to the pote ‘once upon a time’ of verse 76. [74] The tote ‘then’ marks a “return to the time-frame introduced by the earlier temporal adverb [pote].” [75]

In terms of my argument, such a return to an earlier time-frame is a matter of performance, not just composition. That is, the cross-reference represented in this story-within-a-story is performative as well as compositional. To put it yet another way, the blind singer is here being represented as cross-referring by way of performance.

Precisely because the “return” to the time-frame is performative, it can work not only from the present to the past but also from the present to the future. As we will see in Chapter 1, the prophecy of an event in the plot of Homeric narration can “repeat” in advance the wording of that given event. In the case of Odyssey 8.79-81, which signals the prophecy of Apollo about the future quarrel between Achilles and Odysseus, we can imagine a hypothetical performance where the prophecy becomes a retelling of the story already at the moment of the prophecy. In other words, the “return” to the time-frame of the quarrel can be previewed by “retelling” it in advance.

In Hainsworth’s commentary, although he speaks of an “allusion” that is “invented” in Odyssey 8.72-83, we find that he accepts the traditionality of at least one of the themes in this Homeric passage:

The problem is, this summary of a traditional theme makes that theme appear to be more of side-effect than a driving force of traditional Homeric narrative. I will not repeat here my arguments for the presence of this particular driving theme as revealed by the diction of the Iliad itself. [
83] Instead, I turn to a related theme that happens to reveal more overt matches, on the level of diction, between the microcosm or micro-narrative of Odyssey 8.72-83 and the macrocosm or macro-narrative of the Iliad. This driving theme is made evident by the word pēma ‘pain’ in Odyssey 8.82, described as some colossal boulder that has just started rolling downward from the towering heights above, heading straight at the doomed Achaeans down below. This ‘pain’ signals an Iliadic theme, which can be summarized as follows: “Achilles is a pēma for the Trojans when he is at war and a pēma for the Achaeans both when he withdraws from war and when he dies.” [84]

In our Iliad, this pēma ‘pain’ is realized in the death of Patroklos, which prefigures the “offstage” death of Achilles:

Ἀντίλοχ’ εἰ δÕ ἄγε δεῦρο διοτρεφὲς ὄφρα πύθηαι
λυγρῆς ἀγγελίης, ἣ μὴ ὤφελλε γενέσθαι.
ἤδη μὲν σὲ καὶ αὐτὸν ὀΐομαι εἰσορόωντα
γιγνώσκειν ὅτι πῆμα θεὸς Δαναοῖσι κυλίνδει,
νίκη δὲ Τρώων· πέφαται δÕ Êριστος Ἀχαιῶν
Πάτροκλος, μεγάλη δὲ ποθὴ Δαναοῖσι τέτυκται. {17|18}
Antilokhos! Come, so that you may learn
of the ghastly news, which should never have happened.
I think that you already see, and that you realize,
that a god is letting roll a pain [ pēma ] upon the Danaans,
and that victory belongs to the Trojans; the best of the Achaeans has been killed,
Patroklos, that is; and a great loss has been inflicted on the Danaans.

Iliad 17.685-690

I find it essential to compare the words spoken by Menelaos [
85] in referring to any mortal who dares to fight Hektor and thus undertake a confrontation with Apollo himself:

ὁππότ’ ἀνὴρ ἐθέλη πρὸς δαίμονα φωτὶ μάχεσθαι
ὅν κε θεὸς τιμᾷ, τάχα οἱ μέγα πῆμα κυλ ί σθη.
Whenever a man willingly, in defiance of a daimōn, fights a mortal
whom a god honors, surely a great pain [ pēma ] rolls down upon him.

Iliad 17.98-99

Patroklos had dared to confront Apollo, thus prefiguring Achilles, but Menelaos dares not (Iliad 17.100-101).

In the first song of Demodokos, Odyssey 8.72-83, we are looking at a micro-narrative framed by the macro-narrative of the overall Odyssey. In this {18|19} micro-narrative, we see the oracular god Apollo engaged in the act of prophesying the macro-narrative of the Iliad – or, better, an Iliad. This micro-Iliad, framed by the macro-Odyssey, is ominously encapsulated in a single word, pēma ‘pain’, a superhuman force that threatens to crush the heroic ancestors of Hellenism (Odyssey 8.81-82).


[ back ] 1. Saussure 1916: 117.

[ back ] 2. Saussure, ibid.: “Est synchronique tout ce qui se rapporte à l’aspect statique de notre science, diachronique tout ce qui a trait aux èvolutions. De même synchronie et diachronie désigneront respectivement un état de langue et une phase d’évolution.”

[ back ] 3. The paragraph that just ended and the one that follows are an abridgment of my longer formulation in the new introduction to BA 2 (= Nagy 1999a), paragraphs 23-25. In the first edition of BA (= Nagy 1979), I had decided to avoid using the terms “synchronic” and “diachronic” altogether (though I had used them earlier in Nagy 1974, esp. pp. 20-21). Part of the reason for my return to Saussure’s usage has to do with the remarks of Lord 1995:196-197 on the term “diachronic.”

[ back ] 4. PH 4.

[ back ] 5. PH 21 n. 18, following Jacopin (1988: 35-36), who adds: “Both synchrony and diachrony are abstractions extrapolated from a model of reality.”

[ back ] 7. The scheme that follows is based on the outline in HQ 41-42, with details in HQ ch. 3; also in PP 110, with details in PP ch. 5- ch. 7. I originally used the term “Ages of Homer” as a tribute to Emily Vermeule (Nagy 1995a). The HQ and PP discussions, as I have just cited them, emphasize respectively the earlier and later phases of my evolutionary model.

[ back ] 8. For this period, I do not insist on the absence of texts; my point is simply that there is no compelling evidence for the existence of Homeric texts at such an early date. Cf. HQ 31-32. In a new project, I plan to rework my definition of “period a” in the light of a forthcoming work by Douglas Frame on the Ionian cultural contexts of epic performance traditions.

[ back ] 9. HQ ch. 3; cf. PP 69-71.

[ back ] 10. PP 153-186.

[ back ] 11. PP 107-152, 187-206.

[ back ] 12. PP 112.

[ back ] 13. HQ 34-36, 65-69.

[ back ] 14. HQ 34-36, with bibliography.

[ back ] 15. PP 153-186, HQ 32-34.

[ back ] 16. PP 187-206.

[ back ] 17. Nagy 1999c, 2000b, 2001b; book version: Nagy 2002a.

[ back ] 22. The original theory appeared in an article, Lord 1953, which was later reprinted in Lord 1991 (pp. 3-48, with an appendix), a book published in the “Mythology and Poetics” series that I have edited.

[ back ] 23. For elaboration on this point, see below, Chapter 3.

[ back ] 24. The first formal version of my evolutionary model was published in a Festschrift for Albert Lord (Nagy 1981). For an earlier informal version, see Nagy 1974:11. See also in general above, “Question 2.”

[ back ] 25. HQ 100.

[ back ] 26. See again Janko 1982:191; Jensen 1980:92; West 1990:34; Powell 1991:221-237.

[ back ] 27. See also HQ 31-34, 36-37, 100.

[ back ] 28. Powell 1991:221-237 (his chapter title is “Conclusions from Probability: How the Iliad and the Odyssey were written down”) and 1997.

[ back ] 29. Powell 1997.

[ back ] 30. Powell 1991:232-233.

[ back ] 31. Lord 1953 = 1991:44; see further in HQ 32-34.

[ back ] 32. Lord 1991:44.

[ back ] 33. Ibid.

[ back ] 34. PP 110; ch. HQ 42. See “Question 2” above.

[ back ] 35. Powell 1997.

[ back ] 36. Cf. [M.] Parry MHV [1932]: 347-361.

[ back ] 37. Lord 1995:22-68.

[ back ] 38. Powell 1997.

[ back ] 39. PP 26-27.

[ back ] 40. Ibid.

[ back ] 41. Lord 1995, esp. chs. 1, 8, and 10.

[ back ] 42. Ibid.

[ back ] 43. See HQ 34-43. For more on the semiotics of “writing” as “composition,” see “Question 4” below.

[ back ] 44. Powell 1997.

[ back ] 45. PP 59-78.

[ back ] 46. Cf. HQ 75-76, esp. n. 37, with bibliography. Clarke (1999:179) observes that my argumentation in HQ, as a “sustained hypothesis,” can “offer a clear and manageable alternative to the now weary guess that transcripts of Homer’s oral performances were handed down and used as secret cribs by generations of rhapsodes.”

[ back ] 47. Ibid. See also HQ chs. 2-3, esp. pp. 111-112.

[ back ] 48. Powell 1997.

[ back ] 49. For a stance similar to Powell’s, see Janko 1998c, whose rhetoric I confront in Chapter 3 below.

[ back ] 50. In the original version of this section, Nagy 1997b, I offered further comments that I omit here.

[ back ] 52. For the hermeneutics of the term Verschriftlichung ‘textualization’ as opposed to Verschriftung ‘transcription’, I turn to the engaging work of Oesterreicher 1993.

[ back ] 53. On the semiotics of writing and composing in general, I find Habinek 1998 particularly helpful.

[ back ] 54. See, for example, Aristotle Poetics 1451a25 (Ὀδύσσειαν γὰρ ποιῶν οὐκ ἐ ποίησεν ἅ παντα ὅσα αὐτÓ συνέβη). The usage of graphein in referring to poetic composition is a relatively late phenomenon: see for example Pausanias 3.24.11, where he speaks of Homer as “writing” (Ὅμηρoς δὲ ἔγραψε μὲν τῆς ποιήσεως ἀρχόμενος ὡς Ἀχιλλεὺς;…) cf. also 8.29.2.

[ back ] 55. BA 40.

[ back ] 56. I disagree with Rutherford (1991-1993:42), who adds different italics to my formulation, thereby distorting it: “When we are dealing with the traditional poetry of the Homeric (and Hesiodic) compositions, it is not justifiable to claim that a passage in any text can refer to another passage in another text.”

[ back ] 57. PH 53.

[ back ] 58. PH 53-54 n. 8, with examples from both Homeric and Hesiodic poetry. More on this point in what follows.

[ back ] 59. Pucci 1987 (see esp. p. 29 n. 30); see now also Pucci 1995.

[ back ] 60. Reference here to BA 15.

[ back ] 61. Austin 1991:229; the italics are mine.

[ back ] 62. Clay (see next note) refers here to BA 20-21. I disagree with the idea that my pages there reflect a discomfort with the concept of artistic self-consciousness in Homeric poetry. The point was, and always has been, that such self-consciousness is different from what we find in written literature. Also, I call into question the vague reference to “questionable Romantic notions.” On such negative uses of the term “romantic” in polemics about the making of Homeric poetry, see PH 1-2, where I point out that the negativity cuts both ways: those who assume an individual genius called “Homer” can be just as naïve as those who envision some kind of collective self-expression by the ancient Greek “Volk.”

[ back ] 63. Clay 1983:243. At this point, she refers to Pucci 1979:125.

[ back ] 64. Clay 1983:243.

[ back ] 65. See examples in PH 54-55, 80.

[ back ] 66. Dumézil 1984/1999. My first attempt at a formulation of the “Nostradamus effect” appeared in Nagy 1999d.

[ back ] 67. The focus of interest is on Nostradamus’s vision of a blackfriar in gray, as quoted in the title of Dumézil’s 1984 book (“…Le moyne noir en gris dedans Varennes”). In Nagy 1999d:vii, I drew attention to Dumézil’s success in isolating a passage from the Roman historian Livy (first century BCE and CE) as a point of convergence in the readings of Nostradamus and, indirectly, of literati in the era of Louis XVI.

[ back ] 68. See especially Dumézil 1984:93-94.

[ back ] 69. Austin 1991:233-234. At the end of this statement, he refers to my formulation in BA 21-25. On the semiotics of “crystallization,” see HQ 109 and PP 108-109. With reference to Austin’s wording, I distance myself from the notion of “history” in the expression “Mycenaean history,” though I welcome the notion of a poetic tradition that can be traced back historically to the Mycenaean era.

[ back ] 70. On oimē in the sense of the ‘story-thread’, see PP 63 n. 20; see also PP 63 n. 19 on the use of the genitive of this word here to mark the point of departure for the performance of the blind singer.

[ back ] 71. Cf. BA 22.

[ back ] 72. For a valuable analysis of the syntax that drives the wording of verse 81, see Pelliccia 1985:185-186 n. 18: [ back ] Nagy … understood line 81 τότε γάρ [‘for then it was…’] to refer back to the time of the νεῖκος [‘quarrel’], with the γάρ [‘for…’] clause [of verses 79-80] taken as an isolated parenthetical aside reporting events prior to the quarrel. R. Fowler [1983:125] called this “a shaky interpretation of Greek,” and corrected it thus: “The second γάρ depends on χρησόμενος (for successive γάρ’s in Homer cf. J.D. Denniston [1954] p. 58); lines 81-82 thus imply that troubles are already besetting Agamemnon, probably at Aulis.” – But the matter is not so simple. Fowler is referring to Denniston’s entry on “successions of γάρ clauses or sentences, each clause dependent on the previous one” (loc.cit.); he does not mention Denniston’s entry six pages later (64f.) on “successive γάρ’s [that] have the same reference,” nor the comment within that entry (65): “We must distinguish from the above passages others in which the first γάρ clause is parenthetical, and the references in the two γάρ clauses are therefore not parallel” (citing Od. 20.30506 [sic]). The arguments from γάρ are therefore inconclusive.

[ back ] 73. BA 2 xvii paragraph 30 n. 2. See also HQ 77-82. By italicizing notion, I am stressing that the continuity and the totality are merely notional, not necessarily “real” for the empirical observer on the outside looking in, as it were. For comparative evidence on notional totality, see Flueckiger 1996:133-134.

[ back ] 74. See again Pelliccia 1985:185-186n18: [ back ] It might further be argued against Nagy that τότε in 81 most naturally refers to what immediately precedes it; but in fact it can with equal ease refer back to ποτε̠ [‘once upon a time’] in 76; cf. [Hesiod] Th. 58-68, where between correlated ὅτε (58) and τότε (68) there intervenes a passage (63-67) whose time reference is completely different from that of the correlated clauses: τότε here serves (as per Nagy it does in Od. 8.75-82) to dismiss the parenthesis (corresponding to 79-81 in Od. 8) and to return to the time-frame introduced by the earlier temporal adverb with which it is correlated. This point has been repeatedly misconstrued by W. J. Verdenius [1972] 225-60; 247 esp. [ back ] Italics mine. Pelliccia (ibid.) adduces further examples, quoting from West 1985:129: “In Th. 68 τότε is resumptive, looking back to the account of the Muses’ birth in 60 after the digression on their present-day activities.” The scholia to Odyssey 8.75, we need to add, claim that the quarrel between Achilles and Odysseus took place after the killing of Hektor.

[ back ] 75. I am quoting again the italicized part of the formulation by Pelliccia in the preceding footnote.

[ back ] 76. HQ 77-82.

[ back ] 77. Most of this paragraph is excerpted from HQ 82.

[ back ] 78. HQ 82 n. 53; cf. Foley 1991 and my review in Nagy 1995b. For an important study of Homeric Fernbeziehungen, with conclusions different from mine, see Reichel 1994.

[ back ] 79. See Marg 1956:21. For a similar position, see Clay 1983:241-246.

[ back ] 80. Hainsworth 1988:351.

[ back ] 81. I disagree also with Clay 1983:243.

[ back ] 82. Hainsworth 1988:351. At the end of this statement, he refers to Rüter 1969:247-254 (also to BA 42-58).

[ back ] 83. BA 45-49.

[ back ] 84. BA 64. This theme is linked to the name of Achilles, which can be explained morphologically as *Akhi-lāwos ‘he who has akhos [“pain”] for the laos [“host of fighting men”]’. For more on the morphology and on the semantics of that morphology, see Nagy 1994a, where I develop arguments that anticipate the objections of Létoublon 1995:289-290. In a separate work, I criticize her general assumptions about the semantics of naming heroes.

[ back ] 85. “Agamemnon” has been corrected to “Menelaos” in BA 2 63n.

[ back ] 86. BA 2 xviii.

[ back ] 87. BA 65. I have added italics in order to highlight “evolving” and “evolved.” This formulation is quoted by Clay (1983:242), but she omits the final sentence, through which I had meant to convey the idea that the Iliadic cross-reference, as we have it, is a highly sophisticated artistic device. Clay’s omission is symptomatic, I think, of her tendency to flatten out the diachronic dimension of my argumentation: see also Clay pp. 244-245, for her objections to my formulation of “Monro’s Law,” quoted from BA 21. From this quotation, I repeat my own wording: “Or rather, it may be a matter of evolution [emphasis added]. Perhaps it was part of the Odyssean tradition to veer away from the Iliadic.” One last time, I stress the diachronic perspective of cross-reference.