Homeric Echoes in Posidippus

Gregory Nagy
[Originally published in 2004 as chapter 5, pp. 57–64, of Labored in Papyrus Leaves: Perspectives on an Epigram Collection Attributed to Posidippus (P.Mil.Vogl. VIII 309), edited by B. Acosta-Hughes, E. Kosmetatou, and M. Baumbach. Hellenic Studies 2. Center for Hellenic Studies, 2004. In this online version, the original page-numbers of the printed version are indicated within braces (“{” and “}”). For example, “{57|58}” indicates where p. 57 of the printed version ends and p. 58 begins.]
The word ‘echoes’ in the title is meant as a substitute for ‘allusions’, which is inadequate for conveying the awareness of Homeric poetry in the poetics of Posidippus. The subtle ways in which Posidippus demonstrates this awareness reveal an understanding of Homeric poetry that transcends conventional views in later eras, as typified by Aristarchus of Alexandria in the middle of the second century BCE. In the earlier era of Posidippus, in the third century BCE, Homeric poetry was viewed more broadly. In this earlier era, as most prominently typified by Callimachus, the Homeric text was understood differently from the later era of Aristarchus, whose more narrow understanding of Homer has set the standards of Homeric textual criticism to this day.
In the era of Posidippus, Homeric poetry was thought to include a periphery of meanings and forms that later generations of Homeric scholars excluded as non-Homeric. The poetry of Posidippus, following a poetic vogue best exemplified by Callimachus, cultivated this Homeric periphery. Even if Posidippus may be considered a rival of Callimachus, we can still think of these two poets as parallel forces in the cultivation of this vogue. A strong interest in what I am calling here the Homeric periphery was typical of Hellenistic poetry in general.
It was not a question of simply alluding to Homer. It was more, much more, than that. Poets like Posidippus—especially Callimachus—alluded to a special kind of Homer, distancing themselves from the core of Homeric poetry while privileging its periphery. Their self-distancing from the Homeric core helps explain my description of their allusiveness in the title of my presentation. This allusiveness was a kind of distant echoing.
What was being echoed by these Hellenistic poets was indeed Homer, but this Homer was destined to become something other than Homer in the post-Callimachean era of Aristarchus and beyond. {57|58}
Aristarchus’ rigorous system of excluding various meanings and forms as non-Homeric elements in the Homeric tradition had actually been anticipated by a contemporary of Callimachus and Posidippus. This was Zenodotus, generally acknowledged as the first editor of Homer at the Library of Alexandria. Although the approach of Zenodotus in the third century may have been less systematic than that of Aristarchus in the second, what stands out is the overall similarity between the methods of these two editors of Homer. The overall criteria used by Zenodotus for excluding various meanings and forms as non-Homeric turn out to be remarkably convergent with those of Aristarchus. Such convergence raises a question that is essential for my presentation: why did Callimachus and other Hellenistic poets, including Posidippus, cultivate precisely those Homeric meanings and forms that were deemed to be non-Homeric by the leading contemporary editor of Homer?
The answer has to do with the poetic sensibilities that characterized the age of Callimachus. As scholar poets of the Hellenistic era, the likes of Callimachus and Posidippus preferred the peripheral to the central, the exotic to the conventional. They were interested in variation for the sake of variation, treating variants not as distractions from the core of Homeric poetry but instead as sources of fascination about exotica deemed peripheral to this poetry. By contrast, Aristarchus and such predecessors as Zenodotus stayed focused on what they considered to be the essential core of Homeric poetry—whatever they thought belonged to the real Homer. Everything non-Homeric was for them post-Homeric and ‘newer’.
The rigorous methodology of Aristarchus in excluding ‘newer’ elements in the Homeric text is reflected in his rigorous usage of the word neôteroi as a designation of ‘newer’ poets and neôterikos as an adjective describing features that distinguish the ‘neoterics’ from the genuine Homer. As the exhaustive study of Severyns has made clear, this rigorous usage of Aristarchus is reflected even indirectly in the usage of these same words by the excerpters and epitomators of Aristarchus, as reflected in the Homeric scholia. [1]
What Aristarchus was doing, and what his predecessor Zenodotus had been doing beforehand, though perhaps not as systematically, represented a radical departure from earlier ways of thinking about Homer. The Classical idea of Homer, as reflected in the conventional usage of most authors in what we know as the Classical period, pictured the poet of the Iliad and Odyssey as a figure more recent {58|59} than such primeval and superannuated figures as Orpheus and his ‘scribe’ Musaeus. [2] The idea of Homer as the oldest of all standard poets became a standard idea in its own right only in the era of Alexandrian scholarship. A moving force that led to this shift was Aristotle, who thought of Orpheus as a mythical figment postdating Homer. [3] It needs to be emphasized that fifth-century authors like Herodotus were far ahead of their times in thinking of Homer as the oldest poet—older even than Orpheus and Musaeus. [4]
This rethinking of Homer in the Hellenistic era affected radically the editing of Homer. For example, Aristarchus made it his editorial practice to athetize or even omit Homeric verses that could be characterized as ‘Orphic’, treating them as interpolations; also, within individual verses, he preferred variant wordings that seemed to him Homeric as opposed to other variants that seemed by comparison ‘Orphic’. [5]
Zenodotus anticipated the editorial practice of Aristarchus, in treating Homer as the oldest poet and in treating all other poets as neôteroi ‘newer’. This predecessor’s approach to the Homeric text was in many ways even more radical than that of Aristarchus. For example, Zenodotus athetized the Shield of Achilles passage in Iliad XVIII (483–608), condemning it as a non-Homeric interpolation. Nevertheless, even within the athetized text of the Shield, Zenodotus went on make judgments about individual words within given verses, evidently on the grounds that some variant words were more Homeric than others (as at XVIII 485).
Such exercising of editorial judgment in making distinctions between what was Homeric and what was non-Homeric—or even between what was relatively more likely and less likely to be Homeric—amounted to a scholarly skill in the procedures of diorthôsis—that is, in the ongoing process of ‘correcting’ the Homeric text in order to recover, ultimately, the true or ‘correct’ Homer. [6] The Hellenistic scholar-poets of Alexandria, such as Callimachus as well as rivals like Posidippus, transformed this scholarly skill into a poetic skill. In the ‘Seal of Posidippus’ (AB 118 = SH 705), the poet alludes to his mastery of diorthôsis by referring to himself as orthoepês (verse 24).
The Hellenistic poets of Alexandria were exponents of the kind of poetry that concentrates on displaying—artistically—the learning it took to distinguish the newer from the older kind of poetry. What was being echoed in their poetry was {59|60} not so much the idea that something happens to be not Homeric but rather that it must not be Homeric. What they cultivated was the poetics of the anti-Homeric, not just the non-Homeric. To do this successfully, they had to know, and know well, what was considered to be genuinely Homeric and what was post-Homeric.
In actively applying to their own poetry their notion of whatever was ‘newer’ than Homer, the Hellenistic poets of Alexandria cultivated usages of post-Homeric poetry that seemed overtly distinct from Homeric usage, as they understood it. This way, their own poetic usage became an implicit demonstration of their ability to discriminate post-Homeric from Homeric usage. Because their discrimination favored, at least on the surface, the ‘newer’ usage, the poetics of these Hellenistic poets may be characterized as ‘neoteric’.
For my part, I prefer to use this term ‘neoteric’ instead of ‘non-Homeric’ or ‘post-Homeric’ or ‘anti-Homeric’ in describing the stance of these Hellenistic poets, since their own criteria for determining what is or is not truly Homeric need to be treated objectively—and since I do not agree with these criteria.
Despite whatever disagreement we may have with whatever criteria developed by these Hellenistic poets, we need to treat objectively the actual application of these criteria to their own poetry, resulting in conscious choices of what they judged to be non-Homeric forms in preference to what they judged to be Homeric forms. We may disagree with their judgments, but these judgments still add up to a genuine poetics of discrimination, which can serve as a body of evidence for studying the history of variations in the Homeric tradition.
As we are about to see, Aristarchus already understood the evidentiary value of this poetics of discrimination. He and his followers treated the poetic usage of Hellenistic poets as a litmus test for isolating post-Homeric variants in the Homeric textual tradition. Before I turn to a specific example, however, I need to repeat that I prefer to call these variants ‘neoteric’—merely describing an ancient value-judgment and without accepting the implicit idea that any ‘neoteric’ variant found in the Homeric textual tradition must be ergo post-Homeric and ergo non-Homeric or even anti-Homeric.
In a pioneering study, Franco Montanari has shown that Aristarchus in his Homer commentaries evidently took great interest in Callimachean usage. [7] Building on Montanari’s work, Rengakos went on to show that Aristarchus studied the usage of Hellenistic poets in general—including Posidippus—in an ongoing search for forms and themes that he deemed post-Homeric. [8] What evidently made the study of these poets most valuable for Aristarchus and the Aristarcheans is that the Hellenistic poets cultivated poetry that was deemed in their own time to be post-{60|61}Homeric. Invoking again the concept used by the Aristarcheans, I stress that these Hellenistic poets gravitated toward the ‘neoteric’ aspect of Homer, turning this aspect into a ‘neoteric’ poetics of its own. [9] By contrast, the usage of earlier poets studied by the Aristarcheans was ‘neoteric’ only in the sense that it was post-Homeric. To put it another way, the Hellenistic poets were actively ‘neoteric’ in their outlook, whereas earlier ‘post-Homeric’ poets were only passively ‘neoteric’—only insofar as they were supposedly preceded by Homer.
My specific example of the ‘neoterism’ that characterized the Hellenistic poets is an attestation, in the poetry of Posidippus, of the name ‘Berisos’ (SH 701). The witness to this attestation is Aristarchus himself, as reported by his Aristarchean successor, Didymus (whose life overlapped the first centuries BCE and CE). The reportage of Didymus survives in this compressed account, as transmitted by the A scholia to Iliad XI 101:
αὐτὰρ ὁ βῆ ῥ᾿ Ἶσον τε. Ζηνόδοτος ἔξω τοῦ ρ “βῆ Ἶσον”. μὴ ἐμφέρεσθαι δέ φησιν ὁ Ἀρίσταρχος νῦν ἐν τοῖς Ποσειδίππου ἐπιγράμμασι τὸν “Βήρισον”, ἀλλ᾿ ἐν τῷ λεγομένῳ Σωρῷ εὑρεῖν. εὔλογον δέ φησιν ἐξελεγχόμενον αὐτὸν ἀπαλεῖψαι.
(The lemma is) “αὐτὰρ ὁ βῆ ῥ᾿ Ἶσον.” Zenodotus has it without the ρ, “βῆ Ἶσον”. Aristarchus says that “Βήρισον” is not attested now in the epigrams of Posidippus, but that he [= Aristarchus] had found it in the so-called Soros. He [= Aristarchus] says that it was reasonable that he [Posidippus] deleted it [= the word Βήρισον] upon being challenged [concerning it].
We may infer that Aristarchus thought of the Soros as a publication that predated the publication of the epigrams of Posidippus. [10] In the version of the epigrams of Posidippus that were in circulation ‘now’ (νῦν)—that is, in the time of Aristarchus—the name of the hero Berisos was no longer attested (μὴ ἐμφέρεσθαι δέ φησιν ὁ Ἀρίσταρχος), whereas it was indeed attested in the Soros. To put it another way, this particular attestation involved an epigram by Posidippus that Aristarchus had ‘found’ (εὑρεῖν) in the Soros but not in the epigrams that were ‘now’ (νῦν) in circulation.
Aristarchus may have thought that the Soros was an earlier publication on the grounds that the epigram he found there about Berisos was in his opinion an earlier epigram. This supposedly earlier poem seems to have been part of a series of epigrams for Trojan heroes who were mentioned in the Iliad, arranged in acrostic {61|62} order from alpha to omega. [11] The Trojan hero Berisos filled a gap by supplying a name starting with beta, since otherwise there is no attested Trojan hero whose name starts with beta. [12]
According to Didymus, as we infer from the reportage of the A scholia, Aristarchus thought that Posidippus must have changed his mind about using a form like Βήρισον. That is why, Aristarchus reasoned, a Berisos epigram was no longer attested in the publication of Posidippean epigrams circulating ‘now’ in the time of Aristarchus, about a century after the era when Posidippus flourished.
But why would Posidippus change his mind? According to Aristarchus, it was because Posidippus was being ‘challenged’ (ἐξελεγχόμενον) concerning his use of the form Βήρισον. For Aristarchus, the most obvious source of such a challenge would have been the Homer edition of that eminent precursor of his, Zenodotus himself, who flourished in the era of Posidippus and Callimachus. Aristarchus must have thought of Zenodotus as his forerunner in the task of guarding against ‘neoteric’ forms that threatened to infiltrate the Homeric text as he saw it. As we learn from the same reportage of Didymus, Aristarchus here invokes a variant reading adduced by Zenodotus, βῆ Ἶσόν instead of βῆ ῥ᾿ Ἶσόν. Now ΒΗΡΙϹΟΝ could be read as either Βήρισον or βῆ ῥ Ἶσόν, but the reading adduced by Zenodotus, ΒΗΙϹΟΝ, necessarily ruled out Βήρισον and left no choice: now the reading must be βῆ Ἶσόν, and the unique name Berisos becomes a phantom, to be replaced by another name unique to the Iliad, Isos.
Following the lead of Zenodotus, who was in turn followed in this case by Aristarchus, we would have to read the relevant passage of the Iliad (XI 101–102) in the form that we find in the editio maior of T. W. Allen:
αὐτὰρ ὃ βῆ Ἶσόν τε [13] καὶ Ἄντιφον ἐξεναρίξων
υἷε δύω Πριάμοιο νόθον καὶ γνήσιον ἄμφω
What Posidippus must have read, on the other hand, was something like this:
αὐτὰρ ὃ Βήρισόν τε καὶ Ἄντιφον ἐξενάριξεν
υἷε δύω Πριάμοιο νόθον καὶ γνήσιον ἄμφω {62|63}
Both versions, in fact, are justifiable on the basis of the internal evidence provided by Homeric diction.
On the one hand, to start with the ‘Posidippean’ version, it is essential to note that the form ἐξεναρίξεν is actually attested in one Iliad manuscript, Vat gr. 26 (= V1 Allen, V West). Moreover, as George Huxley has shown, ‘[i]n taking the letters ΒΗΡΙϹΟΝ in [Iliad 11] 101 as one word [Posidippus] understood them to be the name of the eponymous hero of the polis of the Βηρύσιοι in the Troad’. [14] The syntax of XI 101 is made straightforward by ἐξεναρίξεν, whereas the ἐξεναρίξων necessitated by the αὐτὰρ ὃ βῆ of the ‘Zenodotean / Aristarchean’ version produces a syntax that seems strained by comparison. The closest parallels I can find to the αὐτὰρ ὃ βῆ … ἐξεναρίξων at Iliad XI 101are βῆ δ᾿ ἴμεν ἀνστήσων at Iliad X 32 and βῆ δ᾿ ἴμεν αἰτήσων at Odyssey xvii 365.
On the other hand, the reading of verse-initial αὐτὰρ ὃ βῆ for XI 101 in the ‘Zenodotean / Aristarchean’ version is closely paralleled by verse-initial αὐτὰρ ὃ βῆ ῥ᾿at Iliad V849, X 73, XX 484, and XXI 205. Moreover, the omission of the ῥ᾿ in some medieval manuscripts (at XX 484 and at XXI 205) provides evidence for the authenticity of the wording αὐτὰρ ὃ βῆ Ἶσόν, without ῥ᾿, as adduced by Zenodotus. Finally, the name Isos, unique to the Iliad just like Berisos, is attested in a fragment of a hexameter poem ‘The Kathodos of the Atreidai’, quoted by Athenaeus 9.399a (Nostoi F 9 Davies = Nostoi F 11 Bernabé).
On the whole, then, both the ‘Posidippean’ and the ‘Zenodotean / Aristarchean’ versions of Iliad XI 101 can be justified on the basis of the internal evidence provided by the textual attestations and by formulaic analysis (ἐξεναρίξεν argues for the ‘Posidippean’ version, while the attestations of αὐτὰρ ὃ βῆ without ῥ᾿ at XX 484 and at XXI 205 argue for the ‘Zenodotean / Aristarchean’). Both versions can also be justified on the basis of existing external evidence for names like Berisos and Isos.
Still, the fact remains that the ‘Posidippean’ Berisos variant is exotic, whereas the ‘Zenodotean / Aristarchean’ variant Isos seems by comparison generic. Aristarchus and his followers made it their practice to purge as ‘neoteric’ the exotic forms and meanings cultivated by the Hellenistic poets in their echoings of Homer. [15]
In this light, even the primary genre of Posidippus as poet, the epigram, can be viewed as an echo of a ‘neoteric’ Homer. The non-Aristarchean and non-Zenodotean ‘Homer’ was himself an epigrammatist: according to the Lives of Homer traditions, Homer himself was commissioned to compose the epigram for {63|64} the tomb of Midas. [16] Such a ‘neoteric’ Homer became the non-Homer of later Alexandrian scholarship, but echoes of this alternative Homer can be heard in the epigrams of earlier Alexandrian scholar-poets like Callimachus—and like Posidippus as epigrammatist par excellence.


[ back ] 1. Severyns 1928. A particularly noteworthy follower of Aristarchus in applying the criterion of neôteroi was Apollodorus of Athens. (On Apollodorus as a student of Aristarchus, see Apollodorus FGH 244 T 1; cf. Pfeiffer 1968:261; Rusten 1982:32n10) A striking example is P. Col. inv. 5604, where Apollodorus comments on a manuscript of the Meropis (SH 903A) that he found (on which see especially Henrichs 1993:188–189; cf. Rusten1982:32). Apollodorus describes the anonymous author as neôteros tis ‘someone newer’; he explains his interest in the manuscript in terms of the idiôma of its content (ἐδόκει δέ μοι τὰ ποήμα [τα] νεωτέρου τινὸς εἶναι … τὸ ἰδίωμα τῆς ἱστορίας).
[ back ] 2. In the traditions of fifth- and fourth-century Athens, the figure of Orpheus—along with Musaeus—are conventionally pictured as if they were earlier than Hesiod and Homer; the locus classicus is Aristophanes Frogs 1032–1035. Other important testimonia include Hippias 86 B 6 DK; Plato Apology 41a; Republic 363a, 377d, 612b; Ephorus FGH 70 F 101.
[ back ] 3. Cicero De natura deorum I 38.
[ back ] 4. Herodotus II 53.3.
[ back ] 5. Nagy 2001.
[ back ] 6. More on diorthôsis in Nagy 1996:115–116, 118–122, 125–127, 186, 191, 198–200, 203–204.
[ back ] 7. Montanari 1995.
[ back ] 8. Rengakos 2000.
[ back ] 9. This formulation is meant as an alternative to the views of Cameron 1980, especially pp. 135–137.
[ back ] 10. See Rengakos 2000:329, with bibliography on the problem of identifying the Soros.
[ back ] 11. See the comments of Lloyd-Jones and Parsons p. 339 on AB 148 (SH 700), which is an epigram by Posidippus for Pandaros, a hero who fights on the Trojan side in the Iliad (II 827, IV 88, V 168, 171, 246, 795); cf. also Lloyd-Jones 1963:95-96 = 1990:190–191.
[ back ] 12. Merkelbach 1956:123–124 = 1996:152–153.
[ back ] 13. Aristarchus read Ἶσόν τε, not Ἶσον, according to a scholion in the margin of Vindobonensis 39 (= Vi2 Allen): Ἀρίσταρχος δὲ Ἶσόν τε. From what we read in the T scholia to XI 101, it seems as if Zenodotus too may have read Ἶσόν τε· Ζηνόδοτος δὲ χωρὶς τοῦ ῥ᾿, “βῆ Ἶσον τε” (pace Erbse III p. 144, who deletes the τε).
[ back ] 14. Huxley 1992:153.
[ back ] 15. See Rengakos 2000:331–334.
[ back ] 16. Vita Herodotea 131–140 pp. 198–199 Allen; Certamen 260–274 pp. 235–236 Allen. The epigram, VII 143 in the Palatine Anthology, is alternatively attributed to Kleoboulos of Lindos. The sources for the alternative attributions are summarized in the apparatus of Allen 1912:198–199.