Hostname: page-component-78c5997874-g7gxr Total loading time: 0 Render date: 2024-11-05T00:41:42.121Z Has data issue: false hasContentIssue false

Alcaeus, Pittakos ‘son of “Hyrras”’, and the Lesbian aristocracy, or ‘How to do things with words’

Published online by Cambridge University Press:  09 November 2023

Alexander Dale*
Affiliation:
Concordia University
Rights & Permissions [Opens in a new window]

Abstract

This paper offers a reassessment of the representation of Pittakos, tyrant of Lesbos, in Alcaeus’ verse. I begin by examining the textual evidence for Pittakos’ father, Hyrras, before progressing to re-evaluate the evidence for the aristocratic clans of Lesbos as attested in the Lesbian poets. Building on this, and with reference to the patronymic/gentilic Arkheanaktidās seemingly used of Pittakos in Alcaeus, I relate the preponderance of patronymic forms found in Alcaeus’ verse to the iambic and comedic use of ‘characterizing’ patronymics in -(ι)δᾱς. I then argue that both ῎Υρρας and Ἀρχϵανακτίδας are to be interpreted via a cross-cultural and bilingual rhetoric of kingship, with the latter being in essence a calque of a Lydian intermediary of the Luwian designation ura- handawati-, ‘great king’, with ὔρρας and its derivatives in Alcaeus a nominalization of the Luwian adjectival stem ura-, ‘great’. This argument is then related to the increasing evidence for Lesbos as a central locus for Graeco-Anatolian cultural exchange. The end result is a comprehensive reassessment of historical reconstructions emanating from the texts of Sappho and Alcaeus, as well as a reassessment of Alcaeus’ poetic objectives in his attacks on Pittakos, ‘son of Hyrras’.

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of the Society for the Promotion of Hellenic Studies

In memoriam: Annette Teffeteller (1944–2020)

I. Introduction

Pittakos, tyrant of Lesbos, is something of a chameleon. Already by the Classical period he was grouped together with the body of statesmen, lawgivers and all-round sophoi that included figures such as Thales, Solon and Periander. While the explicit identification of Pittakos as one of the ‘Seven Sages’ of Greece first occurs in Plato’s Protagoras (343a), Herodotus’ story of his intervention to advise Croesus against building a navy (1.27) already shows the groundwork for the later reception of Pittakos being laid. We are, however, in a somewhat unique position as regards the careers of the ἑπτὰ σοφοί, for in the poetry of Alcaeus we have a largish body of contemporary verse that mentions Pittakos directly. The disparity between this contemporary evidence and the later received tradition could not be more stark. For Alcaeus, Pittakos is the paradigmatic base-born oath-breaker, a treacherous one-time accomplice of Alcaeus and his faction who betrayed them and entered into a marriage alliance with the Penthilidai, before being elected as tyrant by the ‘cowardly’ Mytileneans. With invective flair, Pittakos is lambasted as ‘pot-bellied’ (φύσκωνα), ‘drag-foot’ (σαράποδα), ‘braggart’ (γαύρηκα), ‘son of a shadowy eater’ (ζοφοδορπίδαν), all preserved for posterity by Diogenes Laertius (1.81 = Alc. 429).

Much debate, in antiquity and the modern day, has revolved around Pittakos’ paternity. This was an evident avenue of attack in Alcaeus’ poetry: in fr. 348 he is ‘base-born’, fr. 68 appears to characterize his father and father’s father as guilty of crimes, while fr. 72 is usually seen as a reference to the ‘barbaric’ customs of his mother and father.Footnote 1 According to Diogenes Laertius 1.4.74, Douris of Samos (FGrH 76 F 75) related that Pittakos’ father was a Thracian, a statement that has achieved orthodoxy in many modern accounts. It should be noted, however, that this is unlikely to have been found in the one reliable witness we can identify, namely the poetry of Alcaeus himself. If Alcaeus had explicitly identified and/or attacked Pittakos or his father as Thracian, it would hardly have taken the detective work of Douris to ascertain the fact. I think it more likely that Douris or his source is extrapolating a Thracian pedigree for Pittakos from implicit attacks in Alcaeus’ verse, as well as from external evidence which I shall address below.

When it comes to the sociopolitical history of Lesbos in the Archaic period, we are of course largely dependent on the evidence of Sappho and Alcaeus themselves. The same was no doubt true for most of our ancient sources, from Herodotus to Aristotle to Douris to Diogenes. While local and oral tradition likely retained some refracted memory of Lesbos at the turn of the sixth century BC, it is evident from the nature of our later sources that the Lesbian poets themselves would have been the best, and in many cases only, source material on which to draw. There is a tenacious habit in scholarship on Alcaeus to take much of what we read at face value, and thus to construct historical narratives, relative chronology and prosopography on the basis of what is at best questionable evidence. Had we no contemporary textual sources for Roman history in the first half of the first century BC beyond Cicero’s De consulatu suo, our understanding of the period would be markedly different from what it is. Whether Alcaeus was a more successful statesman than Cicero could be an amusing topic for debate, but I think that few would argue with the proposition that Alcaeus was the better poet. Thus in examining the character of Pittakos as represented in Alcaeus’ verse, the poetic agenda of Alcaeus must be privileged, which necessitates contextualizing the political poems within the constituents that define the poetic tradition: language, myth and culture, both synchronic and diachronic. Much of what follows is concerned with what might be deemed minutiae: aspiration, nominal derivation, marginal scholia and the dynamics of language contact and the morphophonology of loanwords. This cannot be helped. Greek poetry was an ‘art of the word’,Footnote 2 and in order to understand how Alcaeus did things with words, we must start from the ground up.

II. ῎Υρρας

It is generally agreed that the name of Pittakos’ father was ῎Υρρας, though both Alcaeus and the later tradition provide us with at times conflicting evidence. The Suda (Π 1659 Adler) gives Κάϊκος as his father’s name, while the Suda (loc. cit.), Diogenes Laertius, and some witnesses for Callimachus give Ὑρράδιος (i.e. a second declension noun), and a marginal scholion in P.Berol. 9569 (= fr. 112.24) appears to identify Pittakos as (Ἀ)ρχϵανακτ̣ίδαν, ‘son/descendent of Arkheanax’. ῎Υρρας is usually rendered as ‘Hyrras’ or ‘Hyrrhas’, though the text of Alcaeus should of course admit no other rendering than Urras, a point to which I shall return in a moment. The name is most transparently recoverable at 129.13 τὸν ῎Υρραον δὲ πα[ῖδ]α, where ῎Υρραον appears to be a typical Lesbian patronymic adjective in -(i̯)os built to a first declension masculine a-stem ῎Υρρας.Footnote 3 Beyond this, we have ωυρραδον[ at 298.47 (= SLG 262.23).Footnote 4 Reinhold Merkelbach in the editio princeps articulated ὦ ῎Υρραδον, while Hugh Lloyd-Jones suggested ὦ ᾿Υρράδ⟨ι⟩ον, both problematic in their own right;Footnote 5 Carlo Gallavotti articulated ὦ ῎Υρραδ’ ον[, interpreting it as a vocative of a patronymic in -(ι)δᾱς;Footnote 6 Eva-Maria Voigt prints Bruno Snell’s ὦ ῎Υρρα δον[, interpreting it as a vocative of the simplex ῎Υρρας.Footnote 7 Lastly, we have Herrn Hofrath Seidler’s conjecture at Alc. 383.1 τὼι τ’ ᾿Υρραδήωι for τῶ τυρρακηω in the manuscripts of Hephaestion.Footnote 8 I have previously defended Seidler’s conjecture,Footnote 9 and interpreted the form as a relational adjective built to a patronymic in -(ι)δᾱς (although the correct form would almost certainly be ᾿Υρραδάωι). These are the only possible attestations of ῎Υρρας or a derivative thereof in the verse of Alcaeus. The secondary evidence from scholia, the grammarians and etymologica will be considered in due course.

It is evident that there was some misunderstanding about the form of the patronymic derived from Pittakos’ father’s name in antiquity. The Ars grammatica of Dionysius Thrax, as well as two derivative passages from the later grammatical tradition, provide an illuminating glimpse into the methods of ancient dialectology:

1) Dionysius Thrax, Grammatici Graeci 1.1 26.1–4 Uhlig

τύποι δὲ τῶν πατρωνυμικῶν ἀρσϵνικῶν μὲν τρϵῖς … καὶ ὁ τῶν Αἰολέων ἴδιος τύπος Ὑρράδιος· Ὕρρα γὰρ παῖς ὁ Πιττακός.

There are three types of masculine patronymics … and the appropriate form of Aeolic (patronymics) is Ὑρράδιος; Pittakos was the son of Hyrras.

2) Heliodorus apud Scholia Marciana in Dionysium Thracem, Grammatici Graeci 1.2 368.13–15 Hilgard

ὁ δὲ [sc. τῶν πατρωνυμικῶν] ϵἰς αδιος τῆς Αἰολίδος· Ὑρράδιος δέ ἐστιν ὁ τοῦ Ὕρρα υἱός, ῎Υρρας δὲ Μιτυληναίων ἐγένϵτο βασιλϵύς, οὗ υἱὸς ὁ Πιττακός ϵἷς τῶν ἑπτὰ σοφῶν φανϵίς.

The form of patronymic in -αδιος is from Aeolis; Hyrradios is the son of Hyrras, and Hyrras was the king of Mitylene, whose son Pittakos became one of the Seven Sages.

3) Ps.-Herodian, Πϵρὶ παρωνύμων, Grammatici Graeci 3.2 858.26–30 Lentz

τὰ διὰ τοῦ αδιος [sc. πατρωνυμικά] Αἰολικά … Ὑρράδιος δέ ἐστιν ὁ τοῦ Ὕρρα παῖς. Ὕρρας δὲ Μιτυληναίων τύραννος, οὗ υἱὸς Πιττακός ϵἷς τῶν ἑπτὰ φιλοσόφων.

The patronymics in -αδιος are Aeolic … Hyrradios is the son of Hyrras, and Hyrras was the tyrant of Mitylene, whose son Pittakos was one of the Seven Sages.

Despite its uniformity of opinion, the grammatical tradition concerning ‘Aeolic’ patronymics in -αδιος is irreconcilable with our evidence for Aeolic. No such form is ever found in extant Lesbian poetry or in inscriptions from Lesbos and Aeolis. The ‘vernacular’ Aeolic patronymic, amply attested in the Lesbian poets and in inscriptions from Lesbos and Aeolis, was an adjective in -(i̯)os in a-stems and -eios elsewhere.Footnote 10 In addition to patronymic adjectives, the Lesbian poets also employ the patronymic suffix -(ι)δᾱς, an inherited feature of the Kunstsprache. Lastly, Sappho and Alcaeus are not averse to forming patronymics with the adnominal genitive of the father’s name, as was the standard practice in most dialects of the historical period. In other words, as one might expect in the highly developed poetic diction represented by the Aeolian tradition, we find every sort of patronymic formation possible, except for those in -αδιος, the one form of patronymic that the ancient grammatical tradition identified for Aeolic.Footnote 11 Furthermore, the sole and universal example of such patronymic formations in the grammatical and lexicographical tradition is Ὑρράδιος. An argument in favour of the presence of a form ᾿Υρράδιος in Alcaeus’ text might come from Callimachus, Anth. Pal. 7.89 = Ep. 1 Pf. = 54 Gow–Page, GP. However, the manuscript evidence for Callimachus’ use of the form is inconclusive. While most modern editions print παῖδα τὸν Ὑρράδιον, taking Ὑρράδιος as a patronymic adjective as per the grammarians, this is only the reading of the initial text scribe of the Palatine Anthology. The Planudean Anthology, the MSS of Diogenes Laertius (who quotes the epigram in full) and the corrector of the Palatine Anthology all read Πιττακὸν … παῖδα τὸν Ὑρραδίου, taking Ὑρράδιος as a second declension noun.Footnote 12 As Gow and Page noted long ago, the grammarians’ insistence on ᾿Υρράδιος as an Aeolic form might suggest that it was found in Lesbian verse. However, given the absence of any comparanda for such a patronymic formation elsewhere in the Lesbian poets or in Aeolic inscriptions, its status as a patronymic must remain suspect.

We then have the issue of aspiration. Later sources give the name or derivatives thereof with an initial aspirate, thus Ὕρρας. Lesbian was, of course, psilotic, and there can be no question of an aspirated form either sung by Alcaeus or transmitted in the earliest texts of the poet. One would assume that the aspirated form we find later on was analogical with other attestations of the name that did contain an initial aspirate; we might compare Ἔρμαις vs Ἑρμᾶς/Ἑρμῆς. The problem here is that (H)yrras, with or without aspiration, is nowhere else attested in the onomastic inventory of Greek. The only entry in all the volumes of the Lexicon of Greek Personal Names is that for the father of Pittakos. Perhaps the aspiration was introduced through popular etymology on analogy with doublets such as σῦς/ὕς/ὔς or, to stay within the onomastic sphere of northwest Anatolia, Ὑρτάκος, as immortalized in the splendidly Anatolian line Ἀσιάδην τ’ Ἀδάμαντα καὶ Ἄσιον Ὑρτάκου υἱόν at Il. 13.759.Footnote 13 The Hesychean gloss (υ 810 Hansen–Cunningham) ὑρράδιος· ἀπό τινος τῶν προγόνων, ἄδοξος, ἢ ϵἰκαῖος (which is likely connected to σ 2786 Hansen συρραδ⟨ι⟩ος· νόθος. μικτός. ϵἰκαῖος) might provide evidence for a separate lexeme that Alcaeus could have been punning on, and which came to inspire the aspiration in ῎Υρρας. However, given the murky waters of Hesychius’ sources, and the questionable nature of the probative value of dialect glosses in Hesychius, it could well be the case that ὑρράδιος reflects a garbled interpretation of a possible ᾿Υρράδιος in Alcaeus’ text, with συρραδ⟨ι⟩ος a secondary back-formation. The only certain point is that, when talking about the textual evidence for the name of Pittakos’ father in Alcaeus’ text, we have a masculine a-stem urra- with a closed, and thus syllabically heavy, first syllable.

Greek affords no compelling candidates for a cognate for the nominal stem urra-, while those who have advocated for a Thracian pedigree for Pittakos have been unsuccessful in identifying anything in the exiguous body of evidence for the supposed (Dacio-)Thracian branch of Indo-European that might shed light on the name. However, when we turn our attention east, to the ancient soil of Anatolia, we find a nominal stem that would be a perfect formal match for urra-/῎Υρρας and, as I shall argue, an equally fitting semantic match. However, in order to make the case, it will be necessary to stay with the text of Alcaeus and examine further the evidence for both Hyrras and Pittakos, as well as the aristocratic clans of Lesbos at the turn of the sixth century.

III. Lesbian eupatridae

The fragments of Sappho and Alcaeus, along with the ancient exegetical tradition on their work, seem to reveal three clans that were active on Lesbos at the turn of the sixth century, the Kleanaktidai, Polyanaktidai and Arkheanaktidai. To these we can add the Penthilidai, the clan which had supposedly ruled Mytilene since the Aeolian migration, and whose overthrow set in motion the factional strife that is given voice to in the poetry of Alcaeus.Footnote 14 These clans loom large in discussions of Sappho and Alcaeus, and one or more are often identified as equating to the hetaireia to which Sappho and Alcaeus respectively belonged, and for which they composed.Footnote 15 Yet, despite some valiant attempts to delineate a prosopography of Lesbian society, there is little certainty about how any of these patronymics/gentilics relate to either Sappho and Alcaeus themselves, or to any of the individuals identified by name in their verse. It is noteworthy that there is not a single transparent use in extant Lesbian verse of any of these gentilics with any of the names that are commonly associated with them in scholarship, ancient or modern. Such identifications as we have, for instance that Gorgo is the Polyanaktid mentioned at Sappho fr. 155 πόλλα μοι τὰν Πωλυανάκτιδα παῖδα χαίρην, are due to secondary sources (in this case Maximus of Tyre (18.9d), who quotes the line). In addition to this line of Sappho, a form of a patronymic or gentilic derived from Polyanax occurs twice more in Lesbian poetry. The authorship of the poem in question is disputed; Edgar Lobel assigned it to Sappho, in which he is followed by Gauthier Liberman, while Voigt, following Snell, assigned it to Alcaeus.Footnote 16 The two relevant passages are:

1. P.Oxy. 2291 = Alc. 303A a 2 = Sappho fr. 99.2 LP:

δ[ ̣]οῖ Πωλυανακτιδα ̣ ̣[.

where Lobel suggests nom. sg. m. Πωλυανακτίδαι̣ς̣, and

2. P.Oxy. 2291 = Alc. 303A b 14–15 = Sappho fr. 99.23–4 LP:

δϵι̣χ̣ν̣υϲ[ ̣ ̣ ̣]ϵ δηὖτϵ Π ̣ωλ̣υ̣α̣ν̣ακτίδαν

τὸν μάργον ὄν̣δ̣ϵ̣ιξα̣ι̣ θέλω.

Accepting the attribution of P.Oxy. 2291 to Alcaeus, Franco Ferrari, comparing the reproachful tone of Π ̣ωλ̣υ̣α̣ν̣ακτίδαν τὸν μάργον ὄν̣δ̣ϵ̣ιξα̣ι̣ θέλω (‘I want to point out the rapacious Polyanaktid’) to passages of invective against Pittakos, and combining this with the identification of the Πωλυανάκτιδα παῖδα as Gorgo in Sappho 155, constructs a narrative in which Pittakos and Gorgo were brother and sister, both Polyanaktids, with their father Hyrras a Thracian émigré who had managed to establish himself amongst the Mytilenean aristocracy.Footnote 17 Even if the authorship of P.Oxy. 2291 were beyond a shadow of a doubt, this would be a remarkably flimsy foundation for such an elaborate prosopographic reconstruction. In truth, with the guarded exception of the identification of Gorgo as a Polyanaktid, we can say nothing categorical about this supposed aristocratic clan whatsoever. Furthermore, if we ascribe P.Oxy. 2291 to Sappho, as there are strong reasons to do, there is then no evidence for any Polyanaktid in Alcaeus’ verse.

As for the Kleanaktidai, there is a general consensus that Myrsilos and possibly Melankhros were members.Footnote 18 This is largely based on three witnesses:

1. P.Berol. 9569 (= Alc. 112.23), where a marginal scholion identifies the Κλϵανακτ̣ίδαν of line 23 as Myrsilos.

2. P.Oxy. 2733 (= Alc. S 263 SLG), an ancient commentary on Alcaeus, lines 11–12 of which read ] υἱὸς τοῦ Κλϵάνορος ὅτι [|] ἑξῆς τὸν Μύρσιλον ̣ϵγϵ ̣[.

3. Strabo 13.2.3 (= Alc. 468) Ἀλκαῖος … ἐλοιδορϵῖτο … Μυρσίλωι καὶ Μϵλάγχρωι καὶ τοῖς Κλϵανακτίδαις καὶ ἄλλοις τισίν. On face value, this would seem to suggest that neither Myrsilos nor Melanchros was a Kleanaktid. However, Wilamowitz’s deletion of the καί after Μϵλάγχρωι has proved popular, with Strabo then saying that Alcaeus abused ‘the Kleanaktids Myrsilos and Melanchros and certain others’.

Other hypotheses concerning the Kleanaktids have been put forward. In Sappho 98b, the speaker of the lines laments that she is unable to provide her daughter Kleis with a ποικίλαν μιτράν⟨αν⟩. The next sentence mentions ‘the Mytilenean (man?)’ (τώι Μυτιληνάωι), while a few lines later we have a reference to τὰς Κλϵανακτιδα̣[ | φύγας. Some have seen this as a reference to the period of Sappho’s supposed exile, while others have deduced an import ban on luxury items under Pittakos, whom many have identified as ‘the Mytilenean man’.Footnote 19 Denys Page guardedly reads the fragment as a reference to Sappho’s exile which, he surmises, came about as a result of the rule of the Kleanaktidai, that is Melankhros and/or Myrsilos. Liberman sees it as a reference to sumptuary laws enacted by Pittakos and a lament for the exile of the Kleanaktidai (which necessarily presupposes a different relationship between Sappho and the Kleanaktidai than that assumed by Page).Footnote 20 For Ferrari, Sappho was herself a Kleanaktid.Footnote 21 For scholars predisposed to such literal biographical readings of literary texts, relative chronology then becomes a problem. Antonio Aloni objected that Sappho, if her birth was to be placed ca. 650 BC,Footnote 22 would have been too old to have an adolescent daughter under the rule of Pittakos, traditionally dated to ca. 597 BCE.Footnote 23 Ferrari reconciles this by advocating for the chronology of Eusebius, which places Sappho’s floruit around 600–599 (thus born ca. 640 BC).Footnote 24 All of these readings presuppose a rigid biographical context for all fragments, from which a coherent internal chronology can be established, and reconciled with the relative chronology of the period generally.

Most ephemeral of all the supposed aristocratic clans of Lesbos are the Arkheanaktidai, for whom we have only two references:

1. Schol. Nic. Ther. 613 καὶ Ἀλκαῖός φησι †ἐν† τοῖς πϵρὶ Ἀρχϵανακτίδην κατὰ τὸν πρὸς Ἐρυθραίους πόλϵμον φανῆναι τὸν Ἀπόλλωνα καθ’ ὕπνον ἔχοντα μυρίκης κλῶνα.

Alcaeus says … that Apollo appeared to the men with Arkheanaktides during the war against the Erythraians in a dream holding a branch of tamarisk.

2. A marginal scholion in P.Berol. 9569 (= Alc. 112.24) appears to identify the (Ἀ)ρχϵανακτ̣ίδαν of line 24 as Pittakos:

××‒⏑⏑‒]ι̣γϵ Κλϵανακτ̣ίδαν

××‒⏑⏑‒⏑] ἢ (Ἀ)ρχϵανακτ̣ίδαν·

schol. ad v. 23 τ(ὸν) Μύρσιλ(ον), ad v. 24 τ(ὸν) Φιττακ(όν).

Regarding the Nicander scholion, the general sense seems clear, despite the mild corruption, for which Theodor Bergk suggested ἐν ᾱ (i.e. in the first book), while Friedrich Welcker deleted ἐν, both taking τοῖς with φανῆναι. Another possibility might be ἐν τοῖς πϵρὶ Ἀρχϵανακτίδην ⟨ποιήμασι⟩, ‘Alcaeus said in the <poems> about Arkheanaktides that Apollo appeared, etc.’. Strabo 13.599 records that a certain Archaianax (sic) of Mytilene fortified the walls of Sigeon with stones taken from Troy, and it would seem a safe assumption to relate this event to the Sigean War, with Alcaeus as the source.Footnote 25 Given the evidence above, it would not be unreasonable to wonder whether we should read Ἀρχϵάνακτα or Ἀρχϵανακτίδην for the Ἀρχαιάνακτα in Strabo’s text: we will see below another possible conflation of an Arkheanaktid and Arkhaianaktid in later sources.

Whatever the precise tradition reflected in Strabo, we are left with the question of the identity of the Arkheanaktid(es) in the Nicander scholion, and how to reconcile the statement of the scholiast in P.Berol. 9569 with the tradition that Pittakos was a son of Urras, and not Arkheanax. Ferrari, eager to have us believe that Pittakos was a Polyanaktid, takes Ἀρχϵανακτίδην in the Nicander scholion as a personal name.Footnote 26 As for P.Berol. 9569, it has been popular ever since Wilamowitz to assume that the scholiast’s τ(ὸν) Φιττακ(όν) refers to something that stood in the first half of line 24, and so obviate the seeming incongruity of Ἀρχϵανακτ̣ίδαν referring to Pittakos.Footnote 27 Yet none of the proposals is convincing. Hermann Diehls’ genitive plural Ὑρραδίων (necessitated by the metre) sits awkwardly with the two homeoteleutic forms (which could be genitive or accusative) at line end, while Santo Mazzarino’s Καϊκίδαν is clutching at straws. Nor is there any reason to posit a name/patronymic/gentilic standing in the first half of the line, but for the desire not to have Pittakos identified as Ἀρχϵανακτ̣ίδαν: the first member of the disjunctive pairing need be nothing other than Κλϵανακτ̣ίδαν at the end of the previous line. Ockham’s razor would suggest that the Arkheanaktid(es) of the Nicander scholion and P.Berol. 9569 are one and the same, while the only reason not to follow the scholiast’s identification of Ἀρχϵανακτ̣ίδαν as Pittakos is the entrenched desire to expunge apparent inconsistencies from our historical reconstructions, even though these inconsistencies have only arisen as a result of overly biographical readings in the first place. If we could allow ourselves the liberty of granting that Pittakos might be both ῎Υρραος and Ἀρχϵανακτίδας, our troubles would disappear.

IV. Patronymic formations in Alcaeus

As mentioned above, Lesbian Aeolic formed patronymics with the suffixes -(i̯)os in a-stems and -eios elsewhere. Sappho and Alcaeus also preserve a number of patronymics in -(ι)δᾱς, which likely reflect an inheritance within the Aeolic poetic tradition rather than a borrowing from the Ionian hexameter tradition; thus we find Αἰακίδα̣ι̣[ς at Alc. 42.5, Ἀτρϵΐδα[ν] at Alc. 70.6,Footnote 28 Λατο[ΐδ]α at Alc. 67.3 and Kronidas multiple times. In addition to names drawn from the mythical tradition, we have uncertain cases, where a form in -(ι)δᾱς could be either a personal name or a patronymic, such as Ἀγϵσιλαΐδα at Alc. 130b.4, and Αἰσιμίδα at Alc. 365 (both vocatives). As Gregory Hutchinson notes, both forms could well be patronymics in the strict sense; neither occurs anywhere as a proper name.Footnote 29 We also have the inscrutable Δαμοανακτίδ[ at Alc. 296b.1.

In addition to forming patronymics and gentilics, the suffix -(ι)δᾱς had another generically circumscribed function, to characterize people as belonging to certain categories or character types.Footnote 30 This use of the suffix finds its most florid development in Old Comedy. A couple of passages can serve as illustration:

Ar. Ach. 595–97

ὅστις; πολίτης χρηστός, οὐ σπουδαρχίδης

ἀλλ’ ἐξ ὅτουπϵρ ὁ πόλϵμος, στρατιωνίδης,

σὺ δ’ ἐξ ὅτουπϵρ ὁ πόλϵμος, μισθαρχίδης.

‘Who am I? A good citizen, not a place-seeker, but ever since the war began, a soldier’s soldier, while you, ever since the war began, are a wage-seeker’.

Ar. Ran. 840–44

ἄληθϵς, ὦ παῖ τῆς ἀρουραίας θϵοῦ;

σὺ δὴ ’μὲ ταῦτ’, ὦ στωμυλιοσυλλϵκτάδη

καὶ πτωχοποιὲ καὶ ῥακιοσυρραπτάδη.

‘Really, you son of the vegetable-patch goddess? You say this about me, you babble-gatherer, beggar-maker and rag-stitcher …’.

This type of comedic patronymic is rightly seen as akin to those that we find in archaic iambus, such as Archilochus’ Ἐρασμονίδη at fr. 168.1; Κηρυκίδη at fr. 185.1; συκοτραγίδης, ‘son-of-a-fig-eater’, ascribed to both Archilochus (fr. 250) and Hipponax (fr. 167).Footnote 31 I suspect we can include here Hipponax’s Ἐυρυμϵδοντιάδϵω at fr. 128.1. The development of this type of -(ι)δᾱς formation likely represents a confluence of both the patronymic and gentilic functions of the suffix. Certain examples, such as συκοτραγίδης, are insults based on social status, for which parentage is the determinant; cf. the frequent characterization of Euripides as the son of a cabbage-seller in Aristophanes. Others, such as στωμυλιοσυλλϵκτάδης in the Frogs, have a less patent patronymic force. That Euripides is the son of a ‘babble-gatherer’ would be of less importance than the fact that he is a, or belongs to the category of, babble-gatherer.

As it happens, we also find this type of -(ι)δᾱς formation in Alcaeus. As noted above, Plutarch and Diogenes Laertius relate that in Alcaeus’ invective directed at Pittakos he called him ζοφοδορπίδαν, which one usually sees translated as if a two-termination adjective: ‘dusky-diner’ (Campbell), or ‘dîneur-d’ombre’ (Liberman). Formally, ζοφοδορπίδας can only be a formation along the lines of those discussed above; there are no derivatives of the nominal stem δόρπ- or denominative verb δορπϵ- that have a dental stem *δορπίδ- from which an adjective in -ᾱς could be abstracted, and thus ζοφοδορπίδας can only reflect a (pseudo-)patronymic formation, ‘son of a dark-diner’. This could of course have a patronymic force in the strict sense, contextualized within the rhetoric of negatively characterized paternity that we see directed against Pittakos elsewhere in Alcaeus’ verse. Equally, it could have more of the non-patronymic characterizing force that we see most patently in Aristophanes. As to the underlying contextual meaning of ζοφοδορπίδας in Alcaeus, I will have more to say on this in section VI below.

We then have κακοπατρίδας, used of Pittakos explicitly at Alc. 348, and likely used of Pittakos at 75.12 and possibly at 67.4. Formally it is a patronymic built to the zero-grade stem of πατήρ, κακο-πατρ-ίδας, ‘son-of-a-bad-father’ (presupposing *κακοπάτωρ, cf. ϵὐπάτωρ: ϵὐπατρίδης),Footnote 32 though the definite article at 348 and 67.4 points towards an adjectival use.Footnote 33 Despite its formal structure, I think it would be foolish to disregard the prospect of polysemy in κακοπατρίδας, with a play on a putative κακο-πατριδ-ας, ‘bad for the πατρίς’, ‘enemy of the people’, etc.Footnote 34 Scholars have been concerned to ‘explain’ why Alcaeus calls Pittakos κακοπατρίδας. In the early 20th century, it was explained as being due to Pittakos’ supposed plebeian status and politics.Footnote 35 Latterly, it has been taken as proof of his foreign origins.Footnote 36 In modern English, I could call someone ‘son-of-a bitch’ without imputing any particular (let alone factual) knowledge of my target’s parentage. Likewise, Hector can be brother-in-law to a dog (Il. 6.344) without anyone supposing that Paris or even Helen were, in fact, dogs. We should not be so literal-minded when reading Alcaeus.Footnote 37 As an Alcaean coinage unattested elsewhere in Greek literature, κακοπατρίδας can be interpreted within the framework of ‘iambic’ patronymics in -(ι)δᾱς, and together with ζοφοδορπίδας furnishes us with another link between Alcaeus’ invective and the linguistic registers of the more generically defined genre of Ionian iambus.Footnote 38

As to the question of reconciling the evidence for Pittakos being both ῎Υρραος and Ἀρχϵανακτίδας, I think the answer lies in just this sort of patronymic formation. That is to say, rather than Ἀρχϵανακτίδας being a patronymic or gentilic in the strict sense, indicating that the historical Pittakos was descended from someone named Arkheanax, I suspect both patronymic forms in Alcaeus are to be contextualized within a framework of invective. As to how the semantics of ῎Υρραος and Ἀρχϵανακτίδας are to be reconciled with a supposed invective context, this aspect of Alcaean poetic discourse can be further illuminated with reference to the cross-cultural dynamics that characterize Lesbos in the Archaic period as an important locus for Graeco-Anatolian cultural exchange, to which we now turn.

V. Ex oriente dux

The Luwian adjectival stem ura/i-, ‘great’, is widely attested in both cuneiform Luwian of the second millennium BC and in Hieroglyphic Luwian of the second and first millennia. In first-millennium western Anatolia, reflexes of ura/i- are a common element in personal names,Footnote 39 for example, Carian urom-/wrm-, Lycian Ορας, Cilician Ουραμουτας,Footnote 40 Pamphylian and Pisidean Ουρος, Lycian Ορνπϵιμις reflecting *Urnepijẽmi.Footnote 41 The same is true in the second millennium, where in cuneiform documents we find m Ura-ḫattuša-, ‘Great Hattusha’; m Ura-walkui-, ‘Great lion’; m Maššana-ura-, ‘Great (one) of the gods’; m Ura-dU- (i.e. Ura-Tarhunda-), ‘Great Storm God’, or ‘Great (is) the Stormgod’.Footnote 42 In Hieroglyphic Luwian we find MAGNUS-LEO-, reflecting *Ura-walwi-, ‘Great lion’.

But there is one particular use of the adjectival stem ura/i- that is of special relevance for the present discussion. Throughout the attested use of Hieroglyphic Luwian, from Bronze Age inscriptions of the Empire period through to the late Neo-Hittite states in southern Anatolia and northern Syria in the Iron Age, ura/i-, written with the Luwian Hieroglyphic sign MAGNUS, is the designation for a ‘great’ king.Footnote 43 The Luwian title MAGNUS.REX, traditionally read as ura- handawati-, ‘Great King’, is equivalent to the Summerogram LUGAL.GAL used in cuneiform Hittite texts, which is traditionally read as salli- ḫassu-, ‘Great king’. Recent discoveries have revealed the use of this titulature in Late Bronze Age western Anatolia. An inscription discovered in 2007 in Torbalı, just south of the Karabel Pass, and first published in 2011 allows us to read the title as either MAGNUS.REX, ‘great king’, or perhaps MAGNUS.DOMINA, ‘great queen’.Footnote 44 Further south, a graffito discovered in 2000 and published in 2001 from Suratkaya in the Latmos Mountains provides the unique designation MAGNUS.REX.FILIUS, ‘son of the Great king’ or ‘Great son of the king’.Footnote 45 While the reading of the personal name is uncertain,Footnote 46 Rostislav Oreshko has argued that the designation MAGNUS.REX.FILIUS is equivalent to the sign PRINCEPS found in inscriptions from central and southern Anatolia which, it is generally agreed, designates the ‘crown prince’, heir apparent to the throne. Furthermore, in addition to the use of cognates of ura- in first-millennium western Anatolian onomastics discussed in the previous paragraph, we also find a continuation of the exact syntagm ura- handawati- in the Milyan/Lycian B personal name xñtabura-, ‘great king’.Footnote 47

Language pertaining to kingship has a remarkable tenacity, as well as a tendency to cross ethno-linguistic boundaries. Titles and even names of kings can become culture words and Wanderwörter, for which there are few better examples than the developmental history of the original cognomen Caesar. We can now be fairly certain that the Luwian word for king, handawati-, is continued in Kandaules, the name given by Herodotus for the last Heraclid king of Sardis.Footnote 48 Furthermore, Kandaules is quite possibly a throne name or title; the ‘standard’ Lydian word for a king is qaλmu-, which was borrowed into Greek as παλμύς.Footnote 49 There can be little doubt that the name transmitted as Kandaules in Herodotus reflects an underlying Lydian form *kandawle-.Footnote 50 Given that the phonology of handawati-/*kandawle- likely precludes an internal development within Lydian, Onofrio Carruba’s suggestion of a borrowing into Lydian from Luwian provides the most straightforward derivational path.Footnote 51 If the Luwian word for ‘king’ could survive to enter into Lydian as a personal name, throne name or title, there is no prima facie reason why ura-, the natural complement of handawati- to designate a ‘Great King’, would not have followed suit, just as it did in Mylian xñtabura-. Presuming that ura/i- entered Lydian from Luwian, the phonology would have remained largely stable, and thus we can posit a Lydian form *ura- or perhaps *wra-, ‘great’,Footnote 52 which could, furthermore, have maintained a certain regnal association given the widely dispersed syntagm ura- handawati-.

VI. A synthesis

As we saw above, Pittakos is explicitly identified as Ἀρχϵανακτίδας by the scholiast in P.Berol. 9569, while Ockham’s razor would suggest that the Ἀρχϵανακτίδης of the Nicander scholion is one and the same person. Now, Ἀρχϵάναξ is about as straightforward a calque of ura- handawati- as anyone could hope for,Footnote 53 while ὔρρας can be seen as a nominalization of the adjectival stem ura-. I thus propose in the first instance that Alcaeus’ designation of Pittakos as Ἀρχϵανακτίδας, a characterizing patronymic formation along the lines of those discussed above, plays upon a cross-cultural and cross-linguistic rhetoric of kingship. Pittakos is Ἀρχϵανακτίδας not because he is the son or descendant of Arkheanax, but because, within the rhetoric of Alcaeus’ invective, he is the ἀρχϵ-άναξ, the supreme ruler, the great king.Footnote 54 At fr. 348 Alcaeus castigates the dāmos for establishing Pittakos as τύραννος. Alcaeus’ choice of words is usually seen as sour grapes; smarting from his and his faction’s defeat in the Lesbian game of thrones, Alcaeus calls Pittakos τύραννος rather than aisumnētēs, the office to which Aristotle tells us he was elected.Footnote 55 In truth we know very little about the facts on the ground in Mytilene ca. 597 BC, and it might be naïve to accept at face value the later portrayal of Pittakos as a benign ruler who held a constitutionally demarcated office with a term limit and then retired into obscurity. Furthermore, while the semantic range of τύραννος is unclear at this early period, it is evident that it did not always have the pejorative connotations it developed later.Footnote 56 Fr. 348 is not the only passage in Alcaeus where Pittakos is associated with the concept of turannia. At fr. 75.12–13 we find κακοπάτριδ[ | τ]υραννϵύ-, while in the marginal scholia of several papyri we find references to Pittakos, Myrsilos, monarchia and turannos/turannia.Footnote 57 Given the near certainty of τύραννος being a loanword from Anatolia, the designation of Pittakos as τύραννος would resonate within the cross-cultural rhetoric of kingship and power that underlies the syntagm of Ἀρχϵανακτίδας/ura- handawati- set forth here.Footnote 58 In turn, I suggest that this invective rhetoric of kingship is a facet of what Dimitrios Yatromanolakis has described as ‘an interdiscursive rhetoric of power that articulates the poet’s and his comrades’ superiority in the hierarchy of traditional, sanctioned sociopolitical values’.Footnote 59 It is, in other words, a facet of the sociolect of Alcaeus and his hetaireia in Mytilene that reflects the cultural and geopolitical realities of Lesbos at a particular point in time.

As for ῎Υρρας, the name of Pittakos’ supposed father, we can entertain a couple of possibilities. With reference to the calque discussed above, ὔρρας could be nothing more than a nominalized form of ura-, ‘great’, abstracted with reference to the syntagm ura- handawati-.Footnote 60 Let us consider for a moment how this would work. As discussed near the beginning of this paper, the only securely attested forms in Alcaeus’ verse are the adjectival ὔρραος at 129.13 τὸν ῎Υρραον δὲ πα[ῖδ]α, and ωυρραδον[ at 298.47, which is best articulated as ὦ ῎Υρραδ’ ὀν[. We can add the uncertain but possible ᾿Υρραδήωι at fr. 383. Now, ὔρραος need not be a patronymic at all, but could simply be an adjective: the ‘great’ or ‘regal’ one, a sneering reference to the turrania that Pittakos holds or seeks to hold. However, in light of the patronymic form ᾿Υρράδας that likely lies behind 298.47 and 383, I think we could allow that ὔρραος is performing a similar function. However, ὔρραος and ὐρράδας need not be interpreted as patronymics in the strict sense, ‘son of Urras’, but rather along the lines of the invective patronymic formations discussed previously. Thus ὔρραος and ὐρράδας could be taken as ‘(son-of) the Great One’, ‘(son-of) Mr Big’.

Yet it might be rash to rule out the possibility that ῎Υρρας was the actual name of Pittakos’ father, or at least an appellative by which he was known: ‘the Great’. Here we might recall in passing the statements of Ps.-Herodian and Heliodorus (quoted above) that ῎Υρρας was τύραννος or βασιλϵύς at Mytilene. Both Mazzarino and Page warned against dismissing this out of hand, though one finds little appetite for it in recent discussions of Pittakos.Footnote 61 Though ura- is more common as an element in compound names, we do have the simplex Ορας from Lycia. Meanwhile, the structure of Pittakos’ name might point towards a hybridized Lydian/northwest Anatolian pedigree. We noted earlier the Trojan Ὑρτάκος. One could argue that the name contains a syncopated form of ura- (thus rejecting an etymological link with Hittite ḫartakka-, and positing an originally unaspirated form), and thus analysable as ur-takos, which would allow us to analyse Pittakos as pit-takos. If so, I would be tempted to relate the element pit- in Pittakos’ name to Lydian pid-, ‘give’, cognate with Luwian piya-, Lycian pije-, ‘give’, a particularly common element in personal names throughout the western Anatolian onomastic tradition, and functionally equivalent to -δοτος in Greek personal names.Footnote 62 Pittakos is also attested as the name of a king of Edonia in Thrace at Thuc. 4.107. However, given the continuum of Thracian population groups in the north Aegean and northwest Anatolia, it would not be overly surprising to find hybridized Anatolian/Thracian elements in the onomastic inventory.

Northwest Anatolia in the first half of the first millennium is something of an ethno-linguistic grey area. Strabo (12.8.3) famously relates that the Mysians spoke a language that was a mixture of Lydian and Phrygian. Given the exiguous epigraphic remains and the perils inherent in extrapolating evidence from personal names and toponyms, the prospects of our gaining a clearer understanding of the linguistic make-up of the area are slim. That being said, the peripheral nature of northwest Anatolia in the Archaic period, with Thracians to the north, Phrygians to the east, Lydians to the south, and Greeks to the west, would likely have resulted, at least at the peripheries, in a particularly dynamic Sprachbund, which could well lie behind Strabo’s characterization of the language of the Mysians. To this we can guardedly add a further datum. According to Diodorus Siculus 12.32.1, the kingdom of the Kimmerian Bosporus, centred at Pantikapaion on the Straits of Kertch on the Crimean Peninsula, was ruled for some 40 years by the Archaianaktidai (possibly of Milesian origin), before their overthrow in 438 BC by a certain Spartokos,Footnote 63 who established the Thracian Spartokid dynasty which ruled for over 300 years before the kingdom’s absorption by Mithradates VI in 108. It would be hazardous in the extreme to attempt to base any argument on such an ethereal onomastic mélange. All I will say is that the evidence surveyed could point towards a Thracian/northwest Anatolian element in the name Pittakos. Furthermore, onomastic comparanda, and the existence of an Arkhaianaktid dynasty in a region under Thracian hegemony, could have led Douris or his source to posit a Thracian origin for Pittakos ‘Arkheanaktides’.

The cross-linguistic wordplay presupposed in the readings of ῎Υρρας and Ἀρχϵανακτίδας argued for above is certainly not without parallel. Hipponax’s use of Lydian can at times be seen to have parodic effects on a linguistic level. His invocation of Ἑρμῆ κυνάγχα, Μηιονιστὶ Κανδαῦλα at fr. 3a is certainly more complex and polysemic than the syncretizing gloss it has often been taken as, and inter alia resonates on the phonetic level of approximation between καν-, κυν- and *kãn-, the predicted Lydian reflex of PIE *ƙṷón-, ‘dog’.Footnote 64 Though beyond the scope of the present paper, it seems likely that a number of Lydian words in Hipponax entail a certain amount of cross-linguistic punning, and thus presuppose some level of familiarity with Lydian on the part of his contemporary audience.Footnote 65 Staying with Alcaeus, I have previously argued that the homeoteleuton and 12-syllable lines of Alc. 383 (the latter unparalleled in Lesbian poetry) are a subtle metalinguistic referencing of Lydian verse structures, of which vowel assonance and 12-syllable lines are a defining hallmark,Footnote 66 and that this resonates in tandem with the reading Μυρσιλήωι in line 2 of the fragment, with Pittakos’ rule being given a decidedly Lydian, and pejorative, colouring.Footnote 67 I should furthermore note that the level of cross-linguistic awareness presupposed in the above discussion falls far short of full bilingualism. While it seems more than likely that there would have been Mytileneans who spoke Lydian, and Sardians who spoke Lesbian, what is at issue here is the likelihood of awareness of cultural Wanderwörter, and the very prevalence of terms such as πάλμυς, τύραννος and (arguably) μυρσίλος in eastern Greek sources would argue in favour of a sufficient level of awareness amongst internationalist elites in early sixth-century Lesbos.

VII. Blood will out

Having spent some time arguing against an overly historicist interpretation of the traditions concerning Pittakos and Urras and their portrayal in Alcaeus, I would like to conclude by considering how Alcaeus’ invectives might nonetheless be grounded in the historical circumstances of the familial connections of Pittakos. Aside from the explicit references to Urras as Pittakos’ (putative) father in Alcaeus, there is still a strong emphasis on familial affiliation and ancestral descent in Alcaeus’ invective. Explicit attacks on Pittakos’ ancestry or marriage alliance seem to occur in frr. 67, 68, 70, 72, while the preponderance of -(ι)δᾱς formations used of him, though not to be taken as patronymics in the strict sense, nonetheless operate on one level or another within a rhetorical framework of ancestry and identity. This negative characterization of the ancestry of Pittakos is counterposed to the frequent appeals to the ancestry and collected identity of Alcaeus and his hetaireia.Footnote 68

We can begin with ζοφοδορπίδας, which, as we saw above, is analogous in its formation and resonance to invective patronymics that are a recognizable feature of Ionian iambus and Old Comedy. The lemma is preserved in two sources. In the first, at Diog. Laert. 1.81, it is simply included in a list of abusive terms which Alcaeus used of Pittakos, and glossed as ἄλυχνον, ‘lampless’. In the second, Mor. 8.3, one of Plutarch’s interlocutors says that Alcaeus called Pittakos ζοφοδορπίδας not because he dined in the dark, but because he preferred the company of disreputable and common people, ἀλλ’ ὡς ἀδόξοις τὰ πολλὰ καὶ φαύλοις ἡδόμϵνον συμπόταις. This is a good guess, and Duccio Guasti has argued that ζοφοδορπίδας characterized Pittakos as being removed from the convivial sympotic space that was a defining feature of the aristocratic hetaireia.Footnote 69 However, I doubt this was the resonance the coinage had in Alcaeus, while the differing interpretations in our sources argue against first-hand knowledge of the context in which it was used. ζόφος is no ordinary word for ‘dark’, ‘dusk’, etc., but is lexically and semantically marked as the paradigmatic description of the murky gloom of the Underworld.Footnote 70 In Poseidon’s cosmogonic pronouncement at Il. 15.184–99, in which he describes the apportionment that he, Zeus and Hades received, he states that Ἀΐδης ἔλαχϵ ζόφον ἠϵρόϵντα, and cf. the formulaic phrase ὑπὸ ζόφου (ἠϵρόϵντος) at Il. 21.56 and 23.51, Od. 3.335, 11.57 and 155, used to describe the realm of the dead; at Od. 20.356 the shades of the dead hasten Ἐρϵβόσδϵ ὑπὸ ζόφον, ‘to Erebus beneath the gloom’ whence, at Hom. Hymn Dem. 334–37, Hermes is sent to retrieve Persephone. Now, with reference to the mythological tradition and the eschatology associated with ζόφος, it is hard not to associate a putative *ζοφοδορπός (nomina agentis) with Tantalus, whose abnegated feast in the gloom of the Underworld served as his eternal punishment for the infamous feast at which he served his own son to the gods. From his earliest mention in Greek literature, at Od. 11.582–92, it is this second, eternal, ‘feast’ of Tantalus that stands as the paradigmatic substantiation of the requital for his uniquely blood-soaked and unholy gluttony.

Tantalus was, of course, the ultimate progenitor of the house of Atreus, from which, via Orestes’ son Penthilus, the old ruling dynasty of Lesbos, the Penthilidai, claimed descent.Footnote 71 Pittakos supposedly married into the Penthilid clan, if we are to take the evidence of Alc. 70 at face value, and it is this marriage alliance with the ancestral ruling house of Mytilene that supposedly accounts for his treachery and broken oaths, rued by Alcaeus in fr. 129, where he invokes the Erinyes to pursue the transgressor. Yet in fr. 70.6–9, in the passage in which Pittakos is said to have married into the ancestral house of Lesbos, Alcaeus does not refer (as we might expect) to the house of Penthilus,Footnote 72 but rather to the house of Atreus:

κῆνος δὲ παώθϵις Ἀτρϵΐδα[ ̣] ̣ [

δαπτέτω πόλιν ὠς καὶ πϵδὰ Μυρσί̣[λ]ω̣[

θᾶς κ’ ἄμμϵ βόλλητ’ Ἄρϵυς ἐπιτ ̣ ύχϵ ̣ ̣[

τρόπην.

Let him, wedded to the house of Atreus, devour the city as also with Myrsilos … until Ares is minded to turn us to arms.

As Renaud Gagné has discussed, Alcaeus’ identification of the Penthilidai as the Atreidai here is highly significant.Footnote 73 With onomastic sleight of hand, Alcaeus entangles Pittakos with the intergenerational and internecine bloodshed of one of the paradigmatically cursed dynasties of the mythical tradition. The image of Pittakos, an Atreid, devouring the city no doubt plays on the resonance of feasting that recurs as an intergenerational motif in the Atreid myth, from Tantalus to Thyestes. The rhetoric of the bad ruler which Pittakos embodies, devouring the people and their wealth, is thus assimilated to a mythical paradigm.Footnote 74 Pittakos, the ‘shameless … hateful thing’ (fr. 65.5–6), is provided with a genealogy which can be used to damn him more powerfully than any accusation of Thracian pedigree could ever have achieved. He has become the living embodiment of an ancestral curse. Within this rhetoric of feasting and inherited guilt, ζοφοδορπίδας can only point back to the primordial crime of the Atreid line, of which Pittakos has become a living substantiation.Footnote 75

VIII. Conclusions

In the foregoing discussion I have tried, in the first instance, to reassess the evidence for the historical prosopography of Archaic Lesbos at the time of Sappho and Alcaeus, and to emphasize how tenuous the evidence is for the biographical constructs that in many respects we have inherited from antiquity. At the same time, I have tried to relate Alcaeus’ use of language, and in particular the many patronymic formations we find in his verse, to kindred generic typologies in archaic and classical Greek literature, and in turn to interpret Alcaeus’ use of language with reference to the historical realities of intercultural contact and transmission that were and remain a defining feature of the island of Lesbos throughout history.

Both the personal name Urras and the patronymic form Ἀρχϵανακτίδας are readily explicable via a cross-cultural and cross-linguistic rhetorical register, pointing towards awareness of and engagement with Anatolian, and particularly Lydian, population groups. This conclusion is in keeping with the archaeological record, which reveals Lesbos to have been a key node in East–West interaction from the Bronze Age into the Archaic period, as well as the documentary record which reveals increasing Lydian–Greek interaction at the turn of the sixth century BC.Footnote 76 The use of the patronymic formation Ἀρχϵανακτίδας, along with κακοπατρίδας and ζοφοδορπίδας, can furthermore be contextualized within the generic typology of invective, akin to the patronymic formations we find in Ionian iambus and Old Comedy. While the language and rhetoric of paternity figure prominently in Alcaeus’ invective against Pittakos, the evidence for this reflecting actual historical biographical reality is questionable.

Where then does this leave us with reference to our reconstructions of early Lesbian history? There is no reason to doubt that Lesbos witnessed great social, political and economic upheaval in the years of Sappho and Alcaeus. The rapid rise of Lydia in the east, along with the spread of Greek colonization and the growth of commodity trade, would have provoked challenges to the landed wealth of the earlier Archaic period, and these considerations were no doubt a factor in the factional divisions that erupted in Mytilene and elsewhere in the Greek world at the time.Footnote 77 It is likewise evident that Alcaeus’ political poems provide an ideologically consistent response to the anxieties of the age as filtered through a particular subset of aristocratic society. At the same time, the prevailing interpretive model that Alcaeus’ verse was largely occasional, and his audience a small hetaireia predicated almost exclusively on the symposium, with little to no public role,Footnote 78 is built largely on assumption and inference, and seems difficult to reconcile with the preservation and diffusion of the generically diverse body of verse that has survived. Yet, however one chooses to interpret the persona of Alcaeus and the identity of his audience, a degree of incredulity and resistance to the penchant towards biographical reconstructions and literal readings emanating out of the text of the Lesbian poets seems advisable. That the names of the three aristocratic clans of Lesbos should translate into, roughly, ‘the much/many lords’ (Polyanaktidai), ‘the famed lords’ (Kleanaktidai) and the ‘foremost lords’ (Arkheanaktidai), might give us pause. And if we allow for the possibility that Arkhaeanaktides is simply a calque on ura- handawati-, and is a snide speaking name applied to Pittakos, we must allow that the Polyanaktids and Kleanaktids too might be no more than convenient fictions, appellatives sarcastic in context, even if not derogatory in and of themselves, applied to whoever might be a target of invective in Alcaeus’ and Sappho’s verse at any given time. While the tangible personas of people such as Pittakos and even Alcaeus himself might, as a result of this reading, recede further from view, I hope that the arguments I have advanced will serve to shed new light on the cultural and poetic dynamics that yielded the verse that we have, and in turn suggest new ways of reading Alcaeus, Sappho and early Greek literature more generally.

Footnotes

1 Fragments of Sappho and Alcaeus are cited from Voigt (Reference Voigt1971) unless otherwise stated. Archilochus and Hipponax are cited from West (Reference West1989–1991). On Alcaeus’ attacks against Pittakos’ paternity and ancestry, see Page (Reference Page1955) 169–73; Gomme (Reference Gomme1957) 255–57; Davies (Reference Davies1985); Yatromanolakis (Reference Yatromanolakis2008); Gagné (Reference Gagné2009); (Reference Gagné2013) 210–26; Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010).

2 Cf. Watkins (Reference Watkins1995) 97–108.

3 In what follows I give the nom. sg. of masc. a-stems as -ᾱς, unless citing a specific form attested in the paradosis as -αις; see further Dale (Reference Dale2021) 522 n.18, following Hamm (Reference Hamm1958) 24–25, §49; Blümel (Reference Blümel1982) 74–75, §86 and 232–33, §247; Page (Reference Page1955) 83–84 n.1; and Liberman (Reference Liberman1999) 1.xliii n.137

4 The metre, last line of an Alcaic, requires ‒⏑⏑[‒, with ωυ in synezesis.

5 I’m not quite sure what ῎Υρραδον is meant to be, while ᾿Υρράδ⟨ι⟩ον (aside from the questionable status of the form ᾿Υρράδιος as a patronymic, which I address below) sits awkwardly with the vocative implied in ὦ, and requires a periphrasis such as ὦ ᾿Υρράδ⟨ι⟩ον [γένος. See further Liberman (Reference Liberman1999) 2.220–21.

6 The elision of -α in the vocative ῎Υρραδα should not cause concern. Masculine a-stems in the Lesbian poets show both (original) -ᾰ and (secondary) -ᾱ, for which see Hamm (Reference Hamm1958) 148, §233 and (for the development) Sihler (Reference Sihler1995) 273–75, §267.

7 Note that the metre excludes a genitive singular, which in the Lesbian poets is always either uncontracted -ᾱο or contracted -ᾱ. Cf. Hamm (Reference Hamm1958) 31, §61 and Sihler (Reference Sihler1995) 274–75, §267.

8 For the corruption we can compare Pamphila fr. 3 FHG apud Diog. Laert. 1.4.76, who identifies Pittakos’ son (attested only here) as Τυρραῖος, which can only reflect a misinterpretation of an original ΤϒΡΡΑΟϹ in Alcaeus’ text, that is τ’ ῎Υρραος.

9 Dale (Reference Dale2011) 19–20.

10 For Aeolic adjectival patronymics see in particular Morpurgo Davies (Reference Morpurgo Davies1968) and, in nuce, (Reference Morpurgo Davies, Hornblower and Matthews2000) 17–20; Blümel (Reference Blümel1982) 86; Killen (Reference Killen1983) and in particular p. 99 n.93; Wackernagel (Reference Wackernagel and Langslow2009) 486 with n.3 (Langslow).

11 The grammarians’ failure to recognize the patronymic function of adjectives in -(i̯)os/-eios is all the more surprising in the light of Theocritus 28.9, who evidently appreciated the use of possessive adjectives as a characteristic feature of Lesbian Aeolic. See further Gow (Reference Gow1952) ad loc.

12 The corrector of the Palatine Anthology went through the entire manuscript after its initial transcription, collating it with a manuscript that was, in part, the autograph of Michael Chartophylax’s apograph of the autograph of Constantine Cephalas’ Anthology. Thus, with respect to textual constitution, the readings of the Palatine manuscript’s corrector are not to be dismissed lightly. See further Cameron (1993) 111 and Lauxtermann (Reference Lauxtermann2003) 84–85. On the epigram, and its editorial context in the Anthology, see Livrea (Reference Livrea1995).

13 If the aspiration in Ὑρτάκος is genuine, that is. Watkins (Reference Watkins1986) 54–55 has suggested that the name reflects Hittite ḫartakka-, ‘bear’ (in which case the aspirate could be genuine), though as Watkins notes the vocalism of the initial vowel in Greek is problematic. For more on Ὑρτάκος, see section V below.

14 For discussion of Lesbian history in this period based on textual sources, see, for example, Mazzarino (Reference Mazzarino1943); Page (Reference Page1955); Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010); Dale (Reference Dale2011); Gagné (Reference Gagné2013) 210–26.

15 Thus, for example, Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) and Caciagli (Reference Caciagli2011), who both identify Sappho as a member of the Kleanaktidai, while Ferrari (see below) identifies Pittakos as a Polyanaktid. That Alcaeus’ poetry was intended for a narrowly defined group was famously advocated by Rösler (Reference Rösler1980), though without explicit identification with one of the named clans.

16 See Liberman (Reference Liberman1999) 1.xcii–xciv.

17 Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 81–98, especially 88–89 and 92–95.

18 See Page (Reference Page1955) 102, 174–75, though he expresses scepticism that Melankhros is to be included amongst them. On the Kleanaktids see also Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 9–12 and 18–19, for whom Melankhros, Myrsilos and Sappho were Kleanaktids. On Myrsilos see Dale (Reference Dale2011), where I raise the possibility that mursilos might have been a title applied to both Melankhros and Pittakos.

20 Page (Reference Page1955) 97–103; Liberman (Reference Liberman2014) 9. Mazzarino proposed that Pittakos had introduced some sort of sumptuary law that prevented the import of Lydian luxuries (cf. also Caciagli (Reference Caciagli2011) 210–11 and Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 8–9, who talks of ‘the period of austerity marked by the decade of Pittacus’ government’). There is no basis for such a law in the tradition concerning Pittakos, and Mazzarino’s argument is nothing more than an ad hoc attempt to relate this passage of Sappho to an already questionable historical reconstruction. Also unfounded is Mazzarino’s suggestion that in Sappho’s mother’s day there was no trade between Lesbos and Lydia, and that the Kleanaktidai began trading relationships with the Sardians, which Pittakos then proscribed upon coming to power.

22 As per the Suda’s statement (Σ 107 Adler) that her floruit (not birth, cf. Rohde (Reference Rohde1879)) was the 42nd Olympiad (612–609 BC).

23 Aloni (Reference Aloni1997) lxix.

24 Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 8–9 and (Reference Ferrari2014) 1. TEST 6 in Campbell. The Armenian version of Eusebius gives 595/4 for the floruit of Sappho and Alcaeus.

25 Thus, for example, Page (Reference Page1955) 175.

28 And conjectured at Sappho 17.3.

29 Hutchinson (Reference Hutchinson2001) 207.

30 See in particular Meyer (Reference Meyer1923) 140–46; on the origin of the suffix see Keurentjes (Reference Keurentjes1997). Guasti (Reference Guasti2017) has argued that -(ι)δᾱς patronymics were never a productive feature of the language, but were rather a means of making names ‘sound’ more aristocratic and to ‘make vile characters more noble’, with the patronymic function discernible in poetry a secondary feature of the poetic Kunstsprache. A number of oversights and false assumptions render Guasti’s work problematic. In particular, Guasti seems to be under the erroneous assumption that Mycenaean is simply an old form of Greek, and thus that the absence of -(ι)δᾱς patronymics in Mycenaean is damning (Mycenaean is of course simply one dialect, albeit the earliest attested. It forms patronymics with the adjectival suffix, which thus constitutes an isogloss between Mycenaean and Aeolic). Furthermore, Guasti’s ignorance of Keurentjes (Reference Keurentjes1997) is a serious flaw.

31 Meyer (Reference Meyer1923) 140–46; cf. Rosen (Reference Rosen1988) 39–40.

32 Chantraine (Reference Chantraine1968–1980) 864 s.v. πατήρ. Note that it is not built to the dental stem πατριδ- seen in πατρίς (though the -ιδ- suffix used in deriving areal toponyms is cognate with the -ιδ- in -(ι)δᾱς).

33 See Lobel (Reference Lobel1927) lxxvi and lxxxviii.

34 For the structure, with an active sense for κακο-, we might compare κακοδικία at Pl. Leg. 938b, ‘corruption of justice’, but literally ‘harmful to δική’.

35 Thus influentially Wilamowitz (Reference Wilamowitz-Moellendorff1914) 235–36.

36 Thus Mazzarino (Reference Mazzarino1943) 38–52; Page (Reference Page1955) 170–73; Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 90.

37 Cf. Gomme (Reference Gomme1957) 256–57 and Davies (Reference Davies1985) 33–34.

38 Cf. Yatromanolakis (Reference Yatromanolakis2008), who rightly argues that the variations in register for which Alcaeus has at times been castigated reflect a generic interdiscursivity that performs both a poetic and sociopolitical function within the hetaireia and larger community to which Alcaeus’ verse was directed. On commonalities between Alcaeus and iambus, see Andrisano (Reference Andrisano, Cavarzere, Aloni and Barchiesi2001).

39 See generally Houwink ten Cate (Reference Houwink ten Cate1961) 164–65; Zgusta (Reference Zgusta1964); Adiego (Reference Adiego2007) 338 and 430; Melchert (Reference Melchert and Parker2013).

40 ‘Having great strength’, the second element cognate with Luwian muwa-, ‘strength’.

41 See Schürr (Reference Schürr and Schuler2007) 36 and Melchert (Reference Melchert and Parker2013) 48. As Schürr notes, Lycian *Urnepijẽmi (the second element is a participle of pije-, ‘give’, cognate with Luwian piya-, Lydian pid-) is a calque on his father’s name Μϵγιστόδοτος.

42 See generally Laroche (Reference Laroche1966) 197–98 and Melchert (Reference Melchert and Parker2013).

43 See in particular Hawkins (Reference Hawkins1995) 27–28; Yakubovich (Reference Yakubovich and Mouton2017).

45 SURATKAYA graffito 2. For the inscription see Peschlow-Bindokat and Herbordt (Reference Peschlow-Bindokat and Herbordt2001) and Oreshko (Reference Oreshko, Mouton, Rutherford and Yakubovich2013) 346–68.

46 Oreshko (Reference Oreshko, Mouton, Rutherford and Yakubovich2013) 355–58 tentatively suggests ku-*324-i(a), which he interprets as ku-ku(wa)na/i-ia (i.e. Kukuni- or Kukuwani-), but the phonetic value of the hieroglyphic sign *324, and its possible presence in SURATKAYA 3, is uncertain.

47 Tituli Asiae Minoris I (Tituli Lyciae Linguiae) 103.2, 125b; cf. Melchert (Reference Melchert2004) 136, Yakubovich (Reference Yakubovich and Mouton2017) 46. The PN can be segmented as xñtab-ura-, with b a voiced fricative (/v/) continuing pre-Lycian *w. The name is rendered in Greek as Κϵνδαβυρα.

48 Hdt. 1.7, identified as Adyattes or Sadyattes by Nicolaus of Damascus FGrH 90 F 47. The connection between Kandaules and handawati- was first suggested by Szemerényi (Reference Szemerényi1969). In addition to the works cited below, see the discussions in Hawkins (Reference Hawkins2013) 167–82; Högemann and Oettinger (Reference Högemann and Oettinger2018) 70–71; and n.51 below.

49 Dale (Reference Dale2011) 17 with further references. Following Yakubovich (Reference Yakubovich, Adiego, Trabazo, Vernet, Obrador-Cursach and Rodríguez2019) 310–11, the underlying stem is likely qaλmu- rather than qaλmλu- as previously read, while the etymology proposed by Carruba (Reference Carruba, Bombi, Cifoletti, Fusco, Innocente and Orioles2006), from Luwian *kuwala(n)muwa-, ‘warlord’, ‘having the strength of the army’, would be another instance of Luwian vocabulary continued in Lydian.

50 Or perhaps *kãndawle- vel sim., depending on the position of the accent and the extent to which loanwords are subject to diachronic phonological change. Note the alternation between aλiksãntru-/aλiksantru- in renderings of Ἀλέξανδρος in Lydian (texts 3.1 and 50.2 in Gusmani (Reference Gusmani1964)).

51 Luwian handawati- is built to the Proto-Anatolian adjectival stem *Hant- ‘front’ (< PIE *h 2ent-), and there is conflicting evidence for the preservation of PA *H (a voiceless fricative reflecting the outcome of PIE *h 2) in initial position in Lydian; see Melchert (Reference Melchert1994) 361, §14.1.3.2. Thus Carruba (Reference Carruba, Giorgieri, Salvini, Tremouille and Vannicelli2003) 154 proposed Luwian handawati- as the source of Lydian *kandawle-. Yakubovich (Reference Yakubovich2010) 94–95 objected that the development of t > l in handawati-/Kandaules rules out a borrowing from Luwian, and instead proposed a Carian form as the source (though attested Carian kδou- is not a precise formal match). However, as Craig Melchert (personal comment) has pointed out to me, while Lycian A χñtawati- and Hieroglyphic Luwian REX-(wa/i)-ti- both reflect a true voiceless (fortis) stop /t/, the spelling with single t in Cuneiform Luwian (for example, ḫa-an-da-wa-te-eš KUB XXXV 123 iv 1) points towards the voiced (lenis) allomorph /d/, and in intervocalic position d and l (and r) fell together in Luwian by the first millennium; see Melchert (Reference Melchert2003) 179–82; Rieken and Yakubovich (Reference Rieken, Yakubovich and Singer2010). Thus the phonology of the form reflected in Cuneiform Luwian ḫandawata/i-, with /d/, would be a perfect formal source for Lydian *kandawle-. More recently, Högemann and Oettinger (Reference Högemann and Oettinger2018) 367–69 have argued for the preservation of both *H (<*h 2-) and *h (<*h 3-) in initial position in Lydian, given the growing list of presumed Lydian words with initial laryngeal reflexes that otherwise have to be explained as loans, including kuka- (→ Γύγης) < *HuHa-, *kandawle- (→ Κανδαύλης) < *Hant- and *kapala- (→ κάπηλος, καπηλϵία) < *hap-. The possibility is intriguing, and more evidence might tip the scales, though the absence of a reflex of initial *h 2 in forms such as e.g. eśa- ‘child’ <*h 2ónso- remains problematic.

52 There is some evidence for the reduction of unaccented initial *u- to /v/ in Lydian (e.g. *ustó- > wśta- with Melchert (Reference Melchert1994) 365 §14.1.5.1). While the position of the accent on a presumed Lydian/Pre-Lydian *ura- is indiscernible, we cannot rule out a realization as *wrá-. In light of this, it is tempting to see a reflex of ura- in wratoś (nom. sg. common gender, Text 13.9 and 11; possibly a personal name, cf. Schürr (Reference Schürr2016) 85). Furthermore, in Text 41.4 we find the form wrauλ (transparently a dative-locative noun), long recognized as problematic in that Lydian nowhere else writes two vowels together. Brandenstein (Reference Brandenstein1932) 68 n.3 suggested an error for wra[t]uλ (a variant of wratoś found in Text 11.3), but I would instead suggest wralλ (the signs 1 () and u () are similar). We could then read the line as wralλ artimulλ, identifying the recipient of the dedication: the ‘great Artemis’, perhaps a reference to Artemis of Ephesos (more precisely, wralλ artimulλ are the dat.-loc. of possessive adjectives in -li-, and I assume an ellipsis ‘in/for [the temple of] the great Artemis’). Even allowing for *wrá- as the Lydian reflex of ura-, the morphophonemic alternation of w and u in Lydian, as well as the frequent realization of Lydian w as υ in Greek, would make /ur-/ the most likely outcome of /vr-/ in the transmission from Lydian to Greek. See further Dale (Reference Dale2016) 162–63.

53 Note that Ἀρχϵάναξ is a determinative compound, ‘first, foremost anax’, and not a verbal governing compound such as e.g. Ἀρχιλόχος (‘he who leads the troop’).

54 It could be argued that, with reference to the parodic and polemic uses of patronymic formations in Alcaeus surveyed in section IV above, it is not necessary to posit that Ἀρχϵανακτίδας is a punning calque on ura- handawati-. I would counter by saying that within a cross-linguistic rhetoric of kingship (as further argued for below), Ἀρχϵανακτίδας cannot but resonate with reference to comparable Anatolian cognates, for all that it could equally stand alone within Alcaeus’ poetic lexicon.

55 See Liberman (Reference Liberman1999) 2.151 n.300 with further references.

56 Note especially Alc. 34 A 6, where the Dioscuri appear to be addressed as turannoi.

57 See the discussion in Dale (Reference Dale2011) 16–17 with further references.

58 Melchert (Reference Melchert, Avetisyan, Dan and Grekyan2019) and (Reference Melchert, le Feuvre and Pettit2020) demonstrates beyond doubt that Greek τύραννος is an Anatolian loan ultimately deriving from Luwian tarrawann(i)-, an honorific title meaning ‘the just/righteous one’.

59 Yatromanolakis (Reference Yatromanolakis2008) 180.

60 Regarding the geminate rho in ὔρρας, there is nothing in the (admittedly exiguous) evidence from cuneiform spelling conventions to suggest that the initial syllable of ura- was long. However, the varying outcome of the initial syllable in Greek renderings (cf. Lycian Ορας vs Cilician Ουρα- above) suggests that the articulation and syllabic weight of ur- did not correspond precisely to a short upsilon + rho, but could be interpreted as a heavy/closed syllable, approximated in Lesbian Aeolic with gemination of the consonant rather than lengthening of the vowel.

61 Mazzarino (Reference Mazzarino1943) 43; Page (Reference Page1955) 170.

62 Cf. Houwink ten Cate (Reference Houwink ten Cate1961) 175–77; Laroche (Reference Laroche1966) 141–42; Melchert (Reference Melchert and Parker2013) 41 and 47; n.41 above. (Lydian d, /đ/, is the regular outcome of *i̯ in intervocalic position, cf. Melchert (Reference Melchert1994) 364, §14.1.4.8.). Lydian pid- might be continued in Πιδύτης, a Trojan killed by Odysseus at Il. 6.30 (pace Kirk ad loc., the name is likely non-Greek, cf. Stoevesandt on 6.29–36), as well as Πιτθϵός (not to be confused with Πιτθϵύς) from Lydia (LGPN 5A s.v.). The second member of the names, -tako-, would appear to have more in common with the structure of Thracian personal names, for which see further below.

63 To which name we can relate Σπαραδόκος, father of Seuthis, king of Odrycian Thrace, at Thuc. 2.101.5 and, of course, Spartacus, who, Plutarch and Appian tell us, was of Thracian origin. For Pantikapaion and the Kimmerian Bosposus, see Hansen and Nielsen (Reference Hansen and Nielsen2004) 949–50, no. 705.

66 Dale (Reference Dale2011) 23; for Lydian verse structure see Eichner (Reference Eichner, Dobesch and Rehrenböck1993).

67 Typological parallels for cross-linguistic punning could be multiplied, particularly from the Roman world. Plautus, in particular, is a trove of cross-linguistic puns and wordplay (see Fontaine (Reference Fontaine2010), (Reference Fontaine, Fontaine, McNamara and Short2018)), while the Punic from Act V of Poenulus reveals a tolerance for languages that would not have been intimately familiar to a large segment of the audience.

68 Cf. Yatromanolakis (Reference Yatromanolakis2008) 174, who notes that the ‘lineage of the members of the hetaireia … is implicitly contrasted with the political manoeuvres of Pittakos and the κακοπάτριδαι … who wreak havoc on the old institutions of the city’.

69 Guasti (Reference Guasti2017). Cf. the characterization of Pittakos’ hetaireia at Alc. 68.23 (φιλώνων πϵδ’ ἀλϵμ[άτων) and 72, where someone drinks unmixed wine in an inversion of a ‘proper’ sympotic context. This is often seen as a reference to the barbaric practices of Pittakos’ Thracian father (cf. Page (Reference Page1955) 169–79 and the extreme biographical reading of Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 85–7). However, as Davies (Reference Davies1985) 35 aptly illustrates, characterizations of an adversary as a drunkard (or descendent of one) can frequently be made with little reference to reality.

70 Variously associated with Erebos, Hades and Tartaros; cf. West on Hes. Theog. 123 and Hom. Hymn Dem. 334–37 with Richardson.

71 For the line of Tantalus see Gantz (Reference Gantz1993) 531–65. For ancient sources on Penthilus and the Aiolian migration see Page (Reference Page1955) 149; Gantz (Reference Gantz1993) 685. For a critical reassessment of the evidence, documentary and archaeological, for an Aiolian migration to Lesbos and the Troad, see Rose (Reference Rose2008).

72 Explicitly named at frr. 75.10 and 302, as well as Sappho fr. 71.3. For the Penthilidai see Arist. Pol. 1311b = Alc. fr. 472, and previous note.

73 Gagné (Reference Gagné2009) and cf. (Reference Gagné2013) 210–26.

74 There can be no doubt that Alcaeus’ characterization of Pittakos as ‘devouring the city’ (also at fr. 129.23–24) draws on the same rhetorical typologies as Achilles’ denunciation of Agamemnon as a δημοβόρος βασιλϵύς at Il. 1.231 and Hesiod’s castigation of δωροφάγοι βασιλῆϵς at Op. 39 and 221. However, I think it highly unlikely that Alcaeus is directly evoking the Iliad, let alone positioning himself as Achilles to Pittakos’ Agamemnon, as Lentini (Reference Lentini2000) 9–12 and Ferrari (Reference Ferrari, Acosta-Hughes and Prauscello2010) 90–92 have suggested.

75 While this interpretation of ζοφοδορπίδας might appear riddling, it is in keeping with the metaphorical ambiguity that characterizes Greek poetry of all genres and periods, while context would likely have rendered it no more inscrutable than Hesiod’s ἡμϵρόκοιτος ἀνήρ at Op. 605, or ὅς τ’ ἔπϵιτ’ ἔφυ at Aesch. Ag. 171.

76 For Bronze Age Lesbos, see Houwink ten Cate (Reference Houwink ten Cate1983–1984); Mason (Reference Mason, Collins, Bachvarova and Rutherford2008); Teffeteller (Reference Teffeteller, Mouton, Rutherford and Yakubovich2013); for the Iron Age see Spencer (Reference Spencer1995) and Rose (Reference Rose2008). On Lydian expansionism in the later seventh and earlier sixth centuries see Kerschner (Reference Kerschner and Cahill2010); Dale (Reference Dale2016); Payne and Wintjes (Reference Payne and Wintjes2016); Högemann and Oettinger (Reference Högemann and Oettinger2018). For Anatolian elements in Sappho and Alcaeus, see Watkins (Reference Watkins2007); Dale (Reference Dale, Kim, Mynářová and Pavúk2020).

77 With reference to the prominence of Lydia in the broader growth of commerce and commodity trade in this period, note Oettinger’s recent argument (Högemann and Oettinger (Reference Högemann and Oettinger2018) 367–69) that the very terms κάπηλος (‘merchant’) and καπηλϵία (‘retail trade’) are borrowings from Lydian.

78 Thus influentially Rösler (Reference Rösler1980), and recently Caciagli (Reference Caciagli2011).

References

Adiego, I.-J. (2007) The Carian Language (Leiden and Boston)Google Scholar
Aloni, A. (1997) Saffo: frammenti (Florence)Google Scholar
Andrisano, A. (2001) ‘Iambic motifs in Alcaeus’ lyrics’, in Cavarzere, A., Aloni, A. and Barchiesi, A. (eds), Iambic Ideas: Essays on a Poetic Tradition from Archaic Greece to the Late Roman Empire (Oxford), 4164 Google Scholar
Blümel, W. (1982) Die aiolischen Dialekte (Göttingen)Google Scholar
Brandenstein, W. (1932) ‘Die lydische Nominalflexion’, Caucasica 10, 6794 Google Scholar
Caciagli, S. (2011) Poeti e Società: communicazione poetica e formazioni sociali nella Lesbo del VII/VI secolo a.C. (Amsterdam)Google Scholar
Cameron, A. (1993) The Greek Anthology from Meleager to Planudes (Oxford)Google Scholar
Carruba, O. (2003) ‘Λυδικὴ ἀρχαιολογία. La lidia fra II et I millennio’, in Giorgieri, M., Salvini, M., Tremouille, M.C. and Vannicelli, P. (eds), Licia e Lidia prima dell’ ellenizzazione. Atti del Convegno internazionale. Roma, 11–12 ottobre 1999 (Rome) 145–69Google Scholar
Carruba, O. (2006) ‘Il nome della lidia e altri problemi lidii’, in Bombi, R., Cifoletti, G., Fusco, F., Innocente, L. and Orioles, V. (eds), Studi linguistici in onore di Roberto Gusmani (Alessandria) 393411 Google Scholar
Chantraine, P. (1968–1980) Dictionnaire étymologique de la langue grecque (Paris)Google Scholar
Dale, A. (2011) ‘Alcaeus on the career of Myrsilos: Greeks, Lydians, and Luwians at the East Aegean-West Anatolian interface’, JHS 131, 1524 Google Scholar
Dale, A. (2016) ‘ walwet and kukalim: Lydian coin legends, dynastic succession, and the chronology of Mermnad kings’, Kadmos 54, 151–66Google Scholar
Dale, A. (2020) ‘Venus in furs: Sappho fr. 101 Voigt between East and West’, in Kim, R., Mynářová, J. and Pavúk, P. (eds), Hrozný and Hittite: The First Hundred Years. Proceedings of the International Conference Held at Charles University, Prague, 11–14 November 2015 (Leiden and Boston) 469–83Google Scholar
Dale, A. (2021) ‘Gyges at Delphi: Herodotus 1.14’, CQ 70, 518–23Google Scholar
Dale, A. (forthcoming) ‘Kandaules “the dog-throttler”, Greco-Lydian onomastics, and the poetics of iambus: Hipponax fr. 3a West (2 Dg.) in context’, in Cazzatto, V. and Prodi, E. (eds), The Limping Muse.Google Scholar
Davies, M. (1985) ‘Conventional topics of invective in Alcaeus’, Prometheus 11, 3139 Google Scholar
Eichner, H. (1993) ‘Problem von Vers und Metrum in epichorischer Dichtung Altkleinasiens’, in Dobesch, G. and Rehrenböck, G. (eds), Die epigraphische und altertumskundliche Erforschung Kleinasiens (Denkschriften der Österreichischen Akademie der Wissenschaften 236) (Vienna) 97169 Google Scholar
Ferrari, F. (2010) Sappho’s Gift: The Poet and Her Community (tr. Acosta-Hughes, B. and Prauscello, L.) (Ann Arbor)Google Scholar
Ferrari, F. (2014) ‘Saffo e i suoi fratelli e altri brani del primo libro’, ZPE 192, 119 Google Scholar
Fontaine, M. (2010) Funny Words in Plautine Comedy (Oxford)Google Scholar
Fontaine, M. (2018) ‘A cute illness in Epidaurus: eight sick jokes in Plautus’ Gorgylio (Curculio)’, in Fontaine, M., McNamara, C.J. and Short, W.M. (eds), Quasi Labor Intus: Ambiguity in Latin Literature. Papers in Honor of Reginald Thomas Foster, OCD (New York) 2753 Google Scholar
Gagné, R. (2009) ‘Atreid ancestors in Alkaios’, JHS 129, 3943 Google Scholar
Gagné, R. (2013) Ancestral Fault in Ancient Greece (Cambridge)Google Scholar
Gantz, T. (1993) Early Greek Myth (Baltimore and London)Google Scholar
Gomme, A.W. (1957) ‘Interpretations of some poems of Alkaios and Sappho’, JHS 77, 255–66Google Scholar
Gow, A.S.F. (1952) Theocritus (2 vols) (2nd edition) (Cambridge)Google Scholar
Guasti, D. (2017) ‘Pittaco ζοφοδορπίδαις e la storia dei sostantivi in -ίδης’, Graeco-Latina Brunensia 22, 5767 Google Scholar
Gusmani, R. (1964) Lydisches Wörterbuch (Heidelberg)Google Scholar
Hamm, E.-M. (1958) Grammatik zu Sappho und Alkaios (Berlin)Google Scholar
Hansen, M.H. and Nielsen, T.H. (2004) An Inventory of Archaic and Classical Poleis (Oxford)Google Scholar
Hawkins, J.D. (1995) The Hieroglyphic Inscription of the Sacred Pool Complex at Hattusa (SÜDBURG) (Wiesbaden)Google Scholar
Hawkins, S. (2013) Studies in the Language of Hipponax (Bremen)Google Scholar
Högemann, P. and Oettinger, N. (2018) Lydien: Ein altanatolischer Staat zwischen Griechenland und dem Vorderen Orient (Berlin)Google Scholar
Houwink ten Cate, P.H.J. (1961) The Luwian Population Groups of Lycia and Cilicia Aspera during the Hellenistic Period (Leiden)Google Scholar
Houwink ten Cate, P.H.J. (1983–1984) ‘Sidelights on the Ahhiyawa question from Hittite vassal and royal correspondence’, JVEG 28, 3379 Google Scholar
Hutchinson, G.O. (2001) Greek Lyric Poetry: A Commentary on Selected Larger Pieces (Oxford)Google Scholar
Işık, F., Atıcı, M. and Tekoğlu, R. (2011) ‘Die nachhetitische Köningsstele von Karakuyu beim Karabel-Pass. Zur kulturellen Kontinuität vom bronzezeitlichen Mira zum eisenzeitlichen Ionia’, in Schwertheim, E. (ed.), Studien zum antiken Kleinasien VII (Asia Minor Studien 66) (Bonn) 133 Google Scholar
Kerschner, M. (2010) ‘The Lydians and their Ionian and Aiolian neighbours’, in Cahill, N. (ed.), The Lydians and Their World (Istanbul) 247–65Google Scholar
Keurentjes, M.B.G. (1997) ‘The Greek patronymics in -(ί)δας/-(ί)δης’, Mnemosyne 50, 385400 Google Scholar
Killen, J. (1983) ‘Mycenaean possessive adjectives in e-jo ’, TPhS 81, 6699 Google Scholar
Laroche, E. (1966) Les noms des Hittites (Paris)Google Scholar
Lauxtermann, M.D. (2003) Byzantine Poetry from Pisides to Geometres 1 (Vienna)Google Scholar
Lentini, G. (2000) ‘Pittaco erede degli Atridi: il fr. 70 V. di Alceo’, SIFC 3rd series, 18, 314 Google Scholar
Liberman, G. (1999) Alcée: fragments (2 vols) (Paris)Google Scholar
Liberman, G. (2014) ‘Reflections on a new poem by Sappho concerning her anguish and her brothers Charaxos and Larichos’, Paper delivered at F.I.E.C Bordeaux, August 2014. http://www.papyrology.ox.ac.uk/Fragments/Liberman.FIEC.Bordeaux.2014.pdf Google Scholar
Livrea, E. (1995) ‘From Pittacus to Byzantium: the history of a Callimachean epigram’, CQ (n.s.) 45, 474–80Google Scholar
Lobel, E. (1927) Ἀλκαίου μέλη: The Fragments of the Lyrical Poems of Alcaeus (Oxford)Google Scholar
Mason, H. (2008) ‘Hittite Lesbos?’, in Collins, B.J., Bachvarova, M. and Rutherford, I.C., (eds), Anatolian Interfaces: Hittites, Greeks, and Their Neighbours (Oxford) 5762 Google Scholar
Mazzarino, S. (1943) ‘Per la storia di Lesbo nel VI secolo a.C’, Athenaeum 21, 3878 Google Scholar
Melchert, H.C. (1994) Anatolian Historical Phonology (Amsterdam)Google Scholar
Melchert, H.C. (ed.) (2003) The Luwians (Leiden and Boston)Google Scholar
Melchert, H.C. (2004) A Dictionary of the Lycian Language (Ann Arbor)Google Scholar
Melchert, H.C. (2013) ‘Naming practices in second and first millennium Anatolia’, in Parker, R. (ed.), Personal Names in Ancient Anatolia (Oxford) 3149 Google Scholar
Melchert, H.C. (2019) ‘Iron Age Luvian tarrawann(i)-’, in Avetisyan, P.S., Dan, R. and Grekyan, Y.H. (eds), Over the Mountains and Far Away: Studies in Near Eastern History and Archaeology Presented to Mirjo Salvini on the Occasion of His 80th Birthday (Oxford) 337–45Google Scholar
Melchert, H.C. (2020) ‘Hittite tarru- “firm, secure” and Luvian cognates’, in le Feuvre, C. and Pettit, D. (eds), ᾿Ονομάτων ἵστωρ: mélanges offerts à Charles de Lamberterie (Leuven) 541–53Google Scholar
Meyer, G. (1923) Die stilistische Verwendung der Nominalkomposition im griechischen: Ein Beitrag zur Geschichte der διπλᾶ ὀνόματα (Philologus Supplement 16.3) (Leipzig)Google Scholar
Morpurgo Davies, A. (1968) ‘Thessalian patronymic adjectives’, Glotta 46, 85106 Google Scholar
Morpurgo Davies, A. (2000) ‘Greek personal names and linguistic continuity’, in Hornblower, S. and Matthews, E. (eds), Greek Personal Names: Their Value as Evidence (PBA 104) (Oxford) 1540 Google Scholar
Oreshko, R. (2013) ‘Hieroglyphic inscriptions of western Anatolia: long arm of the empire or vernacular tradition(s)?’, in Mouton, A., Rutherford, I.C. and Yakubovich, I. (eds), Luwian Identities: Culture, Language and Religion between Anatolia and the Aegean (Leiden and Boston) 345420 Google Scholar
Page, D.L. (1955) Sappho and Alcaeus (Oxford)Google Scholar
Payne, A. and Wintjes, J. (2016) Lords of Asia Minor: An Introduction to the Lydians (Wiesbaden)Google Scholar
Peschlow-Bindokat, A. and Herbordt, S. (2001) ‘Ein hethitische Großprinzeninschrift aus dem Latmos’, AA, 363–78Google Scholar
Rieken, E. and Yakubovich, I. (2010) ‘The new values of the Luwian signs L 319 and L 172’, in Singer, I. (ed.), ipamati kistamati pari tumatimis: Luwian and Hittite Studies Presented to J. David Hawkins on the Occasion of His 70th Birthday (Tel Aviv) 199219 Google Scholar
Rohde, E. (1879) ‘Γέγονϵ in dem Biographica des Suidas’, RhM 33, 161220 Google Scholar
Rose, C.B. (2008) ‘Separating fact from fiction in the Aiolian migration’, Hesperia 77, 399430 Google Scholar
Rosen, R. (1988) ‘Hipponax, Boupalos, and the conventions of the psogos ’, TAPhA 118, 2941 Google Scholar
Rösler, W. (1980) Dichter und Gruppe: ein Untersuschung zu den Bedingungen und zur historische Funktion früher griechischer Lyrik am Beispiel Alkaios (Munich)Google Scholar
Schadewaldt, W. (1960) ‘Sappho: an die Tochter Kleis’, in Hellas und Hesperien. Gesammelte Schriften zur Antike und neuren Literatur (Zurich and Stuttgart) 7785 Google Scholar
Schürr, D. (2007) ‘Formen der Akkulteration in Lykien: Griechisch-lykische Sprachbeziehungen’, in Schuler, C. (ed.), Griechische Epigraphik in Lykien: ein Zwischenbilanz. Akten des Internationalen Kolloquiums, München, 24–26 Februar 2005 (Vienna) 2740 Google Scholar
Schürr, D. (2016) ‘Vermutungen zum Namen Mantaura’, Gephyra 13, 8388 Google Scholar
Sihler, A. (1995) A New Comparative Grammar of Greek and Latin (Oxford)Google Scholar
Spencer, N. (1995) ‘Early Lesbos between East and West’, ABSA 90, 269306 Google Scholar
Szemerényi, O. (1969) ‘Etyma Latina II (7–18)’, in Studi Linguistici in onore di Vittore Pisani 2 (Brescia) 963–94Google Scholar
Teffeteller, A. (2013) ‘Singers of Lazpa: reconstructing identities on Bronze Age Lesbos’, in Mouton, A., Rutherford, I.C. and Yakubovich, I. (eds), Luwian Identities: Culture, Language and Religion between Anatolia and the Aegean (Leiden and Boston) 567–89Google Scholar
Voigt, E.-M. (1971) Sappho et Alcaeus: Fragmenta (Amsterdam)Google Scholar
Wackernagel, J. (2009) Lectures on Syntax (tr. and edited with notes and bibliography by Langslow, D.) (Oxford)Google Scholar
Watkins, C. (1986) ‘The language of the Trojans’, in M.J. Mellink (ed.), Troy and the Trojan War: A Symposium Held at Bryn Mawr College, October 1984 (Bryn Mawr) 45–62 = L. Oliver (ed.) (1994) Selected Writings Volume II: Culture and Poetics, ed. (Innsbrucker Beiträge zur Sprachwissenschaft 80) (Innsbruck) 700–17Google Scholar
Watkins, C. (1995) How to Kill a Dragon: Aspects of Indo-European Poetics (Oxford)Google Scholar
Watkins, C. (2007) ‘The golden bowl: thoughts on the new Sappho and its Asianic background’, ClAnt 26, 305–25Google Scholar
West, M.L. (1989–1991) Iambi et Elegi Graeci ante Alexandrum Cantata (2 vols) (2nd edition) (Oxford)Google Scholar
Wilamowitz-Moellendorff, U. von. (1914) ‘Neue lesbische Lyrik (Oxyrhynchus Papyri X)’, Neue Jahrbücher für das klassische Altertum, Geschichte und deutsche Literatur 33, 225–47Google Scholar
Yakubovich, I. (2010) Sociolinguistics of the Luvian Language (Leiden and Boston)Google Scholar
Yakubovich, I. (2017) ‘The Luwian title of the Great King’, in Mouton, A. (ed.), Hittitology Today: Studies on Hittite and Neo-Hittite Anatolia in Honor of Emmanuel Laroche’s 100th Birthday (Istanbul) 3950 Google Scholar
Yakubovich, I. (2019) ‘The Lydian dating formulae’, in Adiego, I.-X., Trabazo, J.V.G., Vernet, M., Obrador-Cursach, B. and Rodríguez, E.M. (eds), Luwic Dialects and Anatolian: Inheritance and Diffusion (Barcelona) 299316 Google Scholar
Yatromanolakis, D. (2008) ‘Genre categories and interdiscursivity in Alkaios and Archaic Greece’, Σύγκριση 19, 169–87Google Scholar
Zgusta, L. (1964) Kleinasiatische Personennamen (Prague)Google Scholar