¹ Obbink, D., ‘4708: Archilochus, elegies (more of VI 854 and XXX 2507)’, The Oxyrhynchus Papyri 69 (2005), 19–42Google Scholar.

² It is possible, of course, that the poem was performed for Parian settlers on Thasos rather than on Paros, but since these new Thasian colonists were themselves Parian emigrants, and would consider their new colony as umbilically connected to Paros, the mythological resonances of the poem would be equally significant. Thus, when I refer to ‘Paros’ or a ‘Parian’ audience throughout the article, I do not mean to preclude the possibility of a Thasian performance context, but this would make little difference to the overall argument.

³ Most scholars agree that the mythological section is included as a paradeigma, on the basis of the first-person plural verb in line 4, which implies a separation between the ‘we’ outside the myth, and the third persons inside it, and the phrase καί ποτε (5), which suggests that the poet is looking back to the mythological past to find a suitable example. For a counter-view, however, see Bowie, E., ‘Historical narrative in archaic and early classical Greek elegy’, in Konstan, D. and Raaflaub, K.A. (edd.), Epic and History (Malden, MA, 2010), 145–166, at 151Google Scholar.

⁴ For the sake of simplicity, I print here the text of Obbink, D., ‘A new Archilochus poem’, ZPE 156 (2006), 1–9Google Scholar, which supersedes his original edition (n. 1). Various scholars have suggested alternative readings and supplements, some of which are arguably preferable to Obbink's text. However, my argument in this article relies on the poem's broad themes, and would still stand if alternative readings were printed, though I discuss some of the more essential textual points where they become relevant to the article. The only supplements on which my argument relies are ᾽ρκ[σίδης in line 5 and Ἡρακλ] in line 22 (discussed below); the first is universally accepted by scholars who have studied the papyrus, and the second almost universally so. For textual discussion and alternative readings, see Luppe, W., ‘Zum neuen Archilochos (P. Oxy. 4708)’, ZPE 155 (2006), 1–4Google Scholar; West, M.L., ‘Archilochus and Telephos’, ZPE 156 (2006), 11–17Google Scholar; D'Alessio, G.B., ‘Note al nuovo Archiloco (POxy LXIX 4708)’, ZPE 156 (2006), 19–22Google Scholar; Henry, W.B., ‘Archilochus, P. Oxy. 4708 fr. 1.18–21’, ZPE 156 (2006), 14Google Scholar; Magnelli, E., ‘On the new fragments of Greek poetry from Oxyrhynchus’, ZPE 158 (2006), 9–12Google Scholar; Bernsdorff, H., ‘Halbgötter auf der Flucht: zu P. Oxy. 4708 (Archilochos?)’, ZPE 158 (2006), 1–7Google Scholar; Tammaro, V., ‘Noterelle al nuovo Archiloco (P. Oxy. 4708)’, Eikasmos 17 (2006), 33–5Google Scholar; Nicolosi, A., ‘Sul nuovo Archiloco elegiaco (P. Oxy. 4708 fr. 1)’, Eikasmos 17 (2006), 25–31Google Scholar; Nicolosi, A., Ipponate, epodi di Strasburgo: Archiloco, epodi di Colonia (con un'appendice su P. Oxy. LXIX 4708) (Bologna, 2007)Google Scholar.

⁵ See Obbink (n. 4), 7; Mayer, P., ‘Krieg aus Versehen? Zur Funktion und Aussage der Telephos-Geschichte im neuen Archilochos (P. Oxy. 4708, fr. 1)’, ZPE 157 (2006), 15–18Google Scholar, at 17. For techniques and patterns of ring composition in epic and early poetry, see Notopoulos, J.A., ‘Continuity and interconnexion in Homeric oral composition’, TAPhA 82 (1951), 81–101Google Scholar; Lohmann, D., Die Komposition der Reden in der Ilias (Berlin, 1970), 5–7CrossRefGoogle Scholar; van Groningen, B.A., La Composition littéraire archaïque grecque: procédés et réalisations (Amsterdam, 1958), 52–6Google Scholar; Thalmann, W., Conventions of Form and Thought in Early Greek Epic Poetry (Baltimore, MD, and London, 1984), ch. 1Google Scholar; Minchin, E., ‘Ring-patterns and ring-composition: some observations on the framing of stories in Homer’, Helios 22.1 (1995), 23–35Google Scholar.

⁶ West (n. 4), 16.

⁷ See Mayer (n. 5), who argues that the poem is critical, rather than consolatory in tone; Aloni, A. and Iannucci, A., L'elegia greca e l'epigramma dalle origini al V secolo: con un'appendice sulla ‘nuova’ elegia di Archiloco (Florence, 2007)Google Scholar, who find the celebration of Telephus at odds with the consolatory opening: and also Nobili, C., ‘Tra epos ed elegia: il nuovo Archiloco’, Maia 61.2 (2009), 229–49, at 231–2Google Scholar. For Barker, E.T.E. and Christensen, J.P., ‘Flight club: the new Archilochus fragment and its resonance with Homeric epic’, MD 57 (2006), 9–41Google Scholar, the issue is best resolved by reading the poem as deliberately counter-cultural: they argue that the poem rejects Homeric precedent by presenting flight as something to celebrate rather than to be ashamed of.

⁸ If we compare descriptions of retreat in Homer, we find that, although it might in principle be logical to stress the glory of one's opponent in order to lessen culpability for defeat, in fact the poet tends to avoid this strategy (presumably because a glorious opponent risks stealing the hero's thunder): see e.g. Il. 11.544–74, where Hector is kept well away from Ajax, who retreats before an anonymous crowd of Trojans, thus allowing our emotional focus to remain entirely on him. For a fuller discussion of this issue, see Swift, L.A., ‘Archilochus the “anti-hero”? Heroism, flight and values in Homer and the new Archilochus fragment (P.Oxy. LXIX 4708)’, JHS 132 (2012), 139–55CrossRefGoogle Scholar.

⁹ Aloni, A., ‘Storie di Telefo a Paro: a proposito di POxy. LXIX 4708: Archilochus, Elegy’, in Bernadini, P.A. (ed.), L'epos minore, le tradizioni locali e la poesia arcaica: atti dell'incontro di studio: Urbino, 7 giugno 2005 (Pisa, 2007), 73–90Google Scholar, at 79–80; Aloni and Iannucci (n. 7), 212–16.

¹⁰ For discussion of this idea see Strauss, M., ‘Frühe Bilder des Kindes Telephos’, MDAI(I) 40 (1990), 79–100Google Scholar; Scheer, T.S., Mythische Vorväter: zur Bedeutung griechischer Heroenmythen im Selbstverständnis kleinasiatischer Städte (Munich, 1993), 85–7Google Scholar; Stewart, A., ‘Telephos/Telepinu and Dionysos: a distant light on an ancient myth’, in Dreyfus, R. and Schraudolph, E. (edd.), Pergamon: The Telephos Frieze from the Great Altar. Vol. 2 (New York, 1996), 109–19Google Scholar, at 110–11.

¹¹ For the sources of the myth, see Gantz, T., Early Greek Myth: A Guide to Literary and Artistic Sources (Baltimore, MD, 1993), 428–31Google Scholar; Bauchhenns-Thüriedl, C., ‘Der Mythos von Telephos in der antiken Bildkunst’, LIMC 7.1 (1994), 856–7Google Scholar.

¹² See Collard, C., Cropp, M.J., and Lee, K.H., Euripides: Selected Fragmentary Plays. Vol. 1 (Warminster, 1995), 24Google Scholar; Hall, E., Inventing the Barbarian: Greek Self-definition through Tragedy (Oxford, 1989), 174–5Google Scholar.

¹³ Handley, E.W. and Rea, J., The Telephus of Euripides, BICS Supplement 5 (London, 1957), 33Google Scholar, suggest that fr. 719 (Ἕλληνες ὄντες βαρβάροις δουλεύσομεν; ‘shall we, who are Greeks, be slaves to barbarians?’) is Achilles' response to the news that Telephus will lead the expedition, and so is also connected to the theme of Telephus' ethnic identity.

¹⁴ Of course, if one did want to posit a chronological development in which Telephus becomes partially Hellenized at an early date and fully Hellenized in response to anti-Asian sentiment after the Persian Wars, this would still be compatible with what we see in Archilochus, where Telephus is of Greek descent yet associated with Mysia. However, as I argue below, Archilochus goes out of his way to stress Telephus' Greek connections, which suggests that perceiving the Persian Wars as a sharp cut-off point is ill-advised.

¹⁵ Nobili (n. 7), 246, suggests that the celebration of the Mysian Telephus fits with the mixed nature of Thasian society after colonization. Yet, as I argue below, Telephus' Mysian identity is presented as dwarfed by his Greek parentage. Nobili's reading of the poem differs from mine in seeing it as celebratory in nature, and hence suitable for public performance, whereas my interpretation places greater emphasis on the mistaken and embarrassing nature of the defeat in Mysia.

¹⁶ For early elegy as a forum for the dissemination of local history and genealogy, see Bowie, E., ‘Ancestors of historiography in early Greek elegy and iambic poetry?’ in Luraghi, N. (ed.), The Historian's Craft in the Age of Herodotus (Oxford, 2001), 45–66Google Scholar.

¹⁷ ᾽ρκ[σίδης is universally accepted as the ending of line 5; the only alternative offered was Obbink's original suggestion of ᾽ρκς ἐών, but as Obbink (n. 4), 6, notes, the reading μ]οῦνοὼ earlier in the line now makes this impossible.

¹⁸ Etymologically, Ἀρκασίδης is problematic, for, although Greek writers use the word to mean ‘descendant of Arcas’, the correct form would in fact be 'Αρκαδίδης; Ἀρκασίδης ought to mean ‘descendant of Arcasus’. Some scholars have therefore posited a Mysian founding figure called Arcasus, from whom Telephus was originally descended before he was given a Greek genealogy: for discussion of this issue, see Strauss (n. 10), 79–100; Scheer (n. 10), 71–94; Obbink (n. 4), 6. Yet no Arcasus is attested in Greek literature, or as a personal name in Asia Minor. Later Greeks certainly seem to have identified the patronymic with Arcas: thus grammarians, who were aware that the patronymic was problematic, try to find ways to maintain the connection with Arcadia: see Steph. Byz. 120.14–15 (s.v. Ἀρκαδία); also Fraser, P.M., Greek Ethnic Terminology (Oxford, 2009), 262CrossRefGoogle Scholar. However, since Archilochus and his audience were not trained grammarians, there is no reason to suppose that they would perceive Ἀρκασίδης as an illegitimate form, and since it is used elsewhere to indicate a connection with Arcadia (cf. [Hes.] fr. 129.16–22 M–W), we can be confident that this is what it would have meant to the original audience.

¹⁹ See West, M.L., The Hesiodic Catalogue of Women (Oxford, 1985), 2–11Google Scholar, and 42 on Ἀρκασίδης; for the importance which the Catalogue places on females and their desirability to men, see Osborne, R., ‘Ordering women in Hesiod's Catalogue’, in Hunter, R. (ed.), The Hesiodic Catalogue of Women: Constructions and Reconstructions (Cambridge, 2005), 5–24Google Scholar, at 13–15.

²⁰ On genealogical techniques in the Catalogue and in early mythography, see Bertelli, L., ‘Hecataeus: from genealogy to historiography’, in Luraghi, N. (ed.), The Historian's Craft in the Age of Herodotus (Oxford, 2001), 67–94Google Scholar, at 73–6; R.L. Fowler, ‘Early historiē and literacy’, in Luraghi (this note), 95–115, at 103–5. For the political importance of genealogy and marriage both within the polis and between communities, see Fowler, ‘Genealogical thinking, Hesiod's Catalogue, and the creation of the Hellenes’, PCPhS 44 (1998), 1–19Google Scholar; E. Irwin ‘Gods among men? The social and political dynamics of the Hesiodic Catalogue of Women’, in Hunter (n. 19), 35–98, at 60–5.

²¹ See Berranger, D., Recherches sur l'histoire et la prosopographie de Paros à l'époque archaïque (Clermont-Ferrand, 1992), 144Google Scholar; Aloni (n. 9), 80; Aloni and Iannucci (n. 7), 215; Katsonopoulou, D., ‘Telephos Arkasides in a new poem of Archilochus’, in Katsonopoulou, D., Petropoulos, I., and Katsarou, S. (edd.), Archilochus and his Age: Proceedings of the Second International Conference on the Archaeology of Paros and the Cyclades (Athens, 2008), 289–301Google Scholar, at 291.

²² As Nicolosi (n. 4 [2007]), 304–5, notes, there is a later tradition in Hellenistic and Roman literature of the bloodying of the waters of the Caicus during the conflict between the Achaeans and the Mysians (adesp. ep. Alex. fr. 3.15–16; Philostr. 23.24; Ov. Met. 12.111). As the motif does not appear elsewhere before the Hellenistic period, it is hard to judge whether these references attest to a widespread archaic tradition for which we have no other source, or whether Archilochus' poem could itself have been a source. In Ovid and Philostratus, it is Achilles who stains the water with Mysian blood; if they are drawing on an earlier tradition, then Archilochus' transference of the image to Telephus would be in keeping with his glorification of the hero.

²³ For myth and genealogy as a tool in local self-definition, see Hall, J.M., Ethnic Identity in Greek Antiquity (Cambridge, 1997), 67–106CrossRefGoogle Scholar.

²⁴ Although Heracles' name appears only in a supplement, it is virtually impossible to find an alternative, and the reading Ἡρακλ] has been accepted by almost all scholars who have studied the papyrus. Given the syntax of the line, a proper name is required, and whoever is named here is introduced abruptly and without preparation. The person concerned must be in the singular, and has a friendly relationship to Telephus, to whom he is shouting encouragement (τη[ε] βοῶ). Moreover, the person involved ought to make sense of the reference πατρὶ χαριζ[ενος (25), which is very abrupt if no previous mention of Telephus' father has been involved and we are to take it in general terms. The only logical alternative therefore would be a reference to Telephus' adoptive father, Teuthras, but this would bring problems of spacing on the papyrus and of metre. The only variant reading suggested is that of Luppe (n. 4), 3, who reinterprets the whole start of the line as ἀλλ' [σ][ὶ] ῆ' [ἐ]βόων, but this does not fit the traces well and has been rejected by other scholars who have worked on the poem.

²⁵ Berranger (n. 21), 191–3; Aloni (n. 9), 79–80; Aloni and Iannucci (n. 7), 214–16.

²⁶ See Malkin, I., The Returns of Odysseus: Colonization and Ethnicity (Berkeley, CA, 1998)Google Scholar, who shows (with particular reference to Odysseus) how myths of journeys and returns can provide authority for community identity, and for colonization. Malkin, I., Religion and Colonization in Ancient Greece (Leiden, 1987), 56–9CrossRefGoogle Scholar, also demonstrates how mythological and religious authority was used to justify Parian colonization retrospectively, through the use of oracle stories.

²⁷ A suggestion originally made by Gerber, D.E., Euterpe: An Anthology of Early Greek Lyric, Elegiac, and Iambic Poetry (Amsterdam, 1970), 41–2Google Scholar.

²⁸ The other myths that we know Archilochus to have told are Heracles fighting Achelous and Nessus for the sake of Deianeira (frr. 286–8 W) and Neoptolemus' killing of Telephus' son (and Heracles' grandson) Eurypylus (fr. 304 W). This is further evidence of Parian interest in Heracles and the Heraclids, and in particular in Heracles' Mysian descendants: see Aloni and Iannucci (n. 7), 212; also Bowie (n. 16), 51–2; Bowie (n. 3).

²⁹ As the anonymous reviewer points out, τη[ε need not mean that Heracles was actually involved in the fighting: it is that more likely he is present but not an active participant. It is unusual to find ἀντάω used absolutely as it mostly takes a genitive or dative. There is a Homeric formula which uses the verb without expressing the object (οὐ γὰρ ἔγωγε | ἤντησ' οὐδὲ ἴδον, Il. 4.375, Od. 4.201) and here the verb is used to mean ‘meet with’ or ‘encounter’ in a very general sense. This parallel perhaps suggests that we are meant to understand Heracles as making an appearance on the battlefield rather than physically joining in the fighting: nevertheless, his appearance is dramatic and acts as a spur to Telephus' martial prowess.

³⁰ Heracles' support does not contradict the earlier description of Telephus being ‘alone’, since divine assistance increases rather than lessens personal achievement (cf. Athena's support for Achilles at Il. 22.214–47); hence calling Telephus μ]οῦνος in Heracles' presence is less problematic than if we took the line to refer to a mortal character, and this is another reason to accept Heracles' name as the correct supplement in 22.

³¹ Scholars have noted this double perspective but have tended to argue for one strand or the other representing the poem's dominant function: e.g. for readings of the poem as a defence of flight, see Obbink (n. 4); Barker and Christensen (n. 7); for readings as critical of the Achaeans, see Mayer (n. 5).

³² The occasions on which ἀγάλλομαι is used in a military context in the Iliad all use it to describe arrogant folly: see Il. 12.114, 17.473, 18.132. On the negative presentation of the Saian, see Di Benedetto, V., ‘Archil, fr. 5 W.’, Eikasmos 2 (1991), 13–27Google Scholar, at 17–18; and, for the humorous gulf between epic and contemporary here, see Corrêa, P., Armas e varões: a guerra na lírica de Arquíloco (São Paulo, 2008), 127Google Scholar.

³³ The traditional reading of the lines, whereby the poet speaks of ‘Thracian dogs’ (σὶ Θρέϊξιν, 3) makes the reference to the non-Greeks much more pejorative in tone, but σ has recently been challenged following re-examination of the Sosthenes inscription: see Tsantsanoglou, K., ‘Archilochus fighting in Thasos: frr. 93a and 94 from the Sosthenes inscription’, Hellenica 53 (2003), 235–55Google Scholar, reprinted in Katsonopoulou, Petropoulos, and Katsarou (n. 21), 163–79; Owen, S., ‘Of dogs and men’, PCPhS 49 (2003), 1–18Google Scholar. Tsantsanoglou proposes a new reading ϕύσι, ‘Thracians by nature’ or ‘Thracians by birth’. While less negative, this also presupposes a ‘natural’ difference between Greeks and barbarians, and an awareness of a separation between ethnic groups in the region.

³⁴ West (n. 4) suggests the supplement [ὐδ' ἐγένοντο for the end of line 6, which would then qualify λκ[οι]; however, this is much less satisfactory than supposing that Archilochus did describe the Achaeans as ‘brave’, since the point of the opening gnômê is that even good men can be forced to retreat. As Tammaro (n. 4) notes, λκ[οι] and αἰχμη are connected, so Archilochus cannot be denying the Achaeans' bravery in this line only to reassert it in line 8; αἰχμηταί does not simply indicate a type of fighter but rather has normative overtones, implying bravery and strength: cf Il. 1.20, 5.602, 7.281, 22.269.

³⁵ I understand ἀδελϕε to go with θανἀτων, following the interpretation of West (n. 4), 14, and Bernsdorff (n. 4), 4. The other possibility, favoured by Obbink (n. 1) and Nicolosi (n. 4 [2006] and [2007]) is to take ἀδελϕε separately as referring to the pair of famous brothers Agamemnon and Menelaus. However, this makes little sense if one prints [οὓς ‘Αγαμέμνων at the end of the line, for how can Agamemnon be leading himself? Even if one prints a different supplement at the end of 14, it is a stretch to claim that Archilochus can expect his audience to interpret ἀδελϕε, with no further description or qualification, as unambiguously indicating the Atridae.

³⁶ Aloni and Iannucci (n. 7), 231–6, go so far as to suggest that the shifting between glorification of Telephus and defence of the Achaeans is evidence for two originally separate compositions: one celebratory and one consolatory. While they are right to note the two distinct tones within the poem, their proposed solution seems far too radical, and the fact that the poem contains competing perspectives need be no obstacle to reading it as a cohesive unit.

³⁷ For a detailed analysis of the tension between these two forms of discourse, see Swift (n. 8), 144–6.

³⁸ The effectiveness of this irony depends on West's supplement ϕύζαν at the end of 24 (the alternative supplements of Janko and Livrea printed in Obbink's apparatus [n. 1] would have μοῖραν not ϕύζαν described as κακή) but is for this reason a more poetically effective reading. In a poem with such marked ring composition, it would be surprising if κακή was not meant to echo κακότητα and so carry normative overtones. Nevertheless, a reading which makes κακή dependent on μοῖραν still fits into the poem's broader theme: in this case, the surviving section of the poem ends by shifting once again towards the consolatory and stressing divine intervention.

³⁹ A point I develop in more detail at Swift (n. 8), 151–3.

⁴⁰ See West (n. 4), 15. However, the reading οῖσι in 24 is not secure, and alternatives have been suggested; see D'Alessio (n. 4), 20.

⁴¹ Hall, J.M., Hellenicity: Between Ethnicity and Culture (Chicago, IL, 2002), 125–8Google Scholar, questions the sense of Hellenic self-consciousness found in Homer, because of the absence of a single term for ‘Greeks’ and the association of the three Homeric terms with different parts of the Greek world. However, as he notes (131), Archilochus fr. 102 W uses the term Πανελλήνες, which indicates that by his time (even if not in Homer's) there is evidence for a sense of Greek identity. On the development of concepts of Hellenic or other ethnic identities, see Konstan, D., ‘To Hellēnikon ethnos: ethnicity and the construction of ancient Greek identity’, in Malkin, I. (ed.), Ancient Perceptions of Greek Ethnicity (Washington, DC, 2001), 29–50Google Scholar; J. McInerney, ‘Ethnos and ethnicity in early Greece’, in Malkin (this note), 51–73.

⁴² See Tsantsanoglou (n. 33) on these fragments, and E.L. Bowie, ‘Sex and politics in Archilochus’ poetry', in Katsonopoulou, Petropoulos, and Katsarou (n. 21), 133–43, for an overview of political attack in Archilochus.

⁴³ For the Colophonian colonization at Siris, see Athen. 523c, citing Timaeus and Aristotle ; also Bérard, J., La Colonisation grecque de l'Italie méridionale et de la Sicile dans l'antiquité (Paris, 1957), 201–14Google Scholar. Some scholars have been troubled by Archilochus referring to a colony so far away from Thasos and so have attempted to emend the text but the proposed changes are problematic on metrical grounds: for further discussion see Braccesi, L., ‘Σύρος ποταμός (nota ad Archil. 18 Diehl)’, RFIC 101 (1973), 220–4Google Scholar; Bossi, F., ‘Archiloco e la Propontide’, RFIC 103 (1973), 129–35Google Scholar; Mosino, F., ‘Σῖρις ποταμός (nota ad Archil. 18 D.)’, QUCC 20 (1975), 157–8Google Scholar.

⁴⁴ Plutarch quotes fr. 21 in order to criticize Archilochus for overlooking Thasos' benefits and focussing on its negative qualities, and the donkey simile should therefore be interpreted as a negative one. Moreover, donkeys are associated with humility, poverty, and hard toil: see Griffith, M., ‘Horsepower and donkeywork: equids and the ancient Greek imagination’, CPh 101.3 (2006), 185–246Google Scholar, at 224–8. For Archilochus' presentation of Thasos as uncivilized, see also Corrêa, P., Um bestiário arcaico: fábulas e imagens de animais na poesia de Arquíloco (São Paolo, 2010), 260Google Scholar.

⁴⁵ Obbink (n. 4); Nobili (n. 4).