What did lyric poetry and song tradition(s) mean for imperial audiences and writers? How did Greeks, and Romans, under the Empire encounter ancient song culture and the poets by then associated with the lyric genre? Above all, what position did lyric hold in imperial Greek culture, and what was at stake in engaging with and reusing lyric poetry? The imperial significance of lyric emerges in full only when we trace its presence in diverse but intersecting spheres of imperial culture: readership and education, visual representations, local traditions and contemporary performances. As heirs of Hellenistic scholarship on archaic and classical poetry, imperial readers of lyric were bound to encounter the genre through the lens of Alexandrian systematisation. Indeed, by the Roman period lyric poetry was primarily identified with a collection of nine canonical poets, whose works could be accessed in the form of Hellenistic editions and read with the help of critical tools such as commentaries and poetical biographies.Footnote 1 But the imperial view on lyric was much richer, and much more fluid, than the idea of a stable textual and scholarly tradition may suggest.Footnote 2 Outside of their texts, the lyric nine and the features attached to them were ‘reanimated’ in their artistic portraits,Footnote 3 while epichoric lore still fostered strong associations between specific singers and local communities, so that lyric traditions played a crucial role in the spatial dynamics of the local and the global activated by the Empire.Footnote 4 At the same time, performances of ancient and newly composed songs at symposia and musical festivals contributed to the creative reshaping of experiences and traditions of lyric.
The result is a phenomenology of lyric revealing how the complex system of lyric – made up of texts, anecdotes, distinctive poetic figures, performances and more generic imagery – moved in imperial culture along multiple vectors, covering a spectrum of activities and experiences which spanned from textual to broader cultural circulation and recognisability.Footnote 5 This imperial lyric system was active on multiple levels in sociological terms too: while my primary focus is on the imperial elite as an obvious consequence of my concern with sophistic literature, certain aspects of song culture and elements relevant to the characterisation of ancient poets as biographical figures circulated and persisted well beyond in-depth engagement with archaic and classical texts, and beyond elite circles too.
At this point, a word of caution is in order. To distinguish between different cultural and social experiences of lyric aids their analysis and interpretation, but such a clear-cut separation between these manifestations is always bound to be artificial. Different levels of interaction with lyric continuously overlapped. The way lyric texts were tackled in imperial elite paideia both influenced and was influenced by the wider cultural currency of lyric poets. Stylistic analyses contributed to the association of individual poets with labels (e.g. ‘sweet’ Sappho, ‘sublime’ Pindar), which then intertwined with the stories told about them. Similarly, anecdotes about poets filtered from specialised literary biographies into wider cultural lore, and vice versa. It is the breadth and cultural meaning(s) of this lyric spectrum that made up the imperial experience of lyric.
Reassessing the evidence on imperial readership, literary education and criticism will allow us to place the lyric canon in relation to the canon of imperial paideia, before zooming in on the distinctive characterisation and values attached to single poets and subgenres of lyric such as erotic and encomiastic poetry. I will then turn to integrating the literary discourse with the wider role of lyric in imperial culture, as reflected in the visual representations and local ties of poets, as well as in the continuous engagement with performed lyric, both among Greek communities and as appropriated by the Roman side of the Empire (Chapter 2). The background thus traced will be crucial for understanding what was at stake in Aristides’ reuse of lyric and how he reconfigured lyric tradition to construct his public voice(s) in different imperial settings.
1.1 A Canon at the Limits of the Canon
In Carm. 4.9.5–12, Horace articulates lyric poetry’s position with respect to Homer and to the immortal fame which epic both provides and enjoys. Even if Homer ‘holds first place’ (priores … tenet | sedes), the (very diverse) Muses of Pindar, Simonides, Alcaeus and Stesichorus are not ‘forgotten’, ‘time has not eclipsed’ Anacreon’s ‘fun’ (nec, si olim lusit Anacreon, | delevit aetas) and Sappho’s love ‘lives still’ (spirat adhuc).Footnote 6 The tension between the two genres seems to resolve itself into coexistence, or ‘co-persistence’; Homer’s primacy does not consign to oblivion the kaleidoscopic production of lyric poets; the popularity of epic does not equal effacement of lyric. There is of course a self-conscious, and self-serving, dimension to this brisk Horatian history of poetic genres. Horace’s point is precisely that his poetry, modelled as it is on Greek lyric, can compete with epic in its literary persistence and ability to bestow lasting fame.Footnote 7 And yet, when Horace ‘brought’ Greek lyric metres into Latium (Carm. 3.30.13–14), he was not selecting a genre that, in the panorama of Roman literary culture, was simply interchangeable with epic. Apart from some poems by Catullus, Horace’s appropriation of lyric rhythms, and personas, represented a unicum, a ‘staggering achievement’ which, Horace himself claims, had the potential to ‘set him apart from the crowd’ (Carm. 1.1.30–4).Footnote 8
Horace’s self-reflective reception of Greek poetry is useful to foreground two central aspects of the study of lyric as part of imperial literary culture. When describing the place of lyric tradition in imperial paideia, we may be tempted to paraphrase Horace’s observation about the genre’s tension/coexistence with epic. Even if elite paideia focused, besides Homeric epic, on Attic poetry and prose of the classical period, lyric still lived on in imperial literary society.Footnote 9 And yet, as we are reminded by Horace’s lyric project, the persistence of lyric alongside other genres requires qualification: just how did lyric poetry relate, in terms of diffusion and popularity, to the staples of imperial literary education? This is a central question in my exploration of imperial lyric phenomenology and will represent a key factor in my analysis of Aristides’ engagement with lyric. It is my contention that lyric poetry held a distinctive place at the upper, more specialised end of imperial literary culture. Lyric, that is, was not part of the primary selection of essential texts at the heart of elite paideia; it represented rather an optional (but) more refined level of readership. It is in this sense that, as foregrounded by the section title, the lyric canon sat at the (upper) limits of the imperial literary canon.
In order to contextualise lyric in relation to the system of readings and models most common to imperial Greek authors, I draw on three major interrelated areas and types of evidence: the availability, circulation and treatment of lyric texts as suggested by papyrus findings; the role of lyric in imperial education; and its place in rhetorical theory and practice, especially in relation to the Atticist trend of the period. Rather than attempting an all-encompassing account, the discussion will trace the significance of lyric authors and texts in imperial literary society, in order to define the literary and social implications, the cultural politics, attached to the reuse and appropriation of lyric.
1.1.1 Imperial Libraries
When reconstructing ancient reading trends and the circulation of literary genres and works, papyri are an inevitable starting point. From a chronological point of view, for the imperial period such an analysis is favoured by the fact that most of our papyri are dated between the first and third century.Footnote 10 This provides us with a volume of evidence allowing for meaningful consideration in statistical terms.Footnote 11 The number of papyri at our disposal, however, does not address the main objection (or call for caution) that might be advanced with regard to the use of papyri to trace overall literary trends of the Empire: can general imperial tendencies be reconstructed on the basis of evidence coming almost uniquely from Egypt, and mostly from Oxyrhynchus? As recently argued by Reviel Netz, both internal evidence – for example, the fact that Egyptian papyri do not appear to favour Egyptian authors over non-Egyptian ones – and different forms of cultural presence suggest that the selection of writers and works represented by papyri may be expected to be generally stable geographically, and thus to describe a literary canon which was Mediterranean rather than exclusively Egyptian.Footnote 12 This does not mean that local variations should be excluded or not expected; on the contrary, ‘it would be rather incredible had the typical Athenian library in the Hellenistic era not included a rather larger fraction of Athenian philosophy; if the cities of the islands and of Asia Minor did not display some preference, at least, for their native sons’.Footnote 13 As we have anticipated, and as we shall see in more detail (§2.2), the same traffic between local and supralocal discourses characterised the imperial phenomenology, and reception, of lyric.
At first glance, the panorama of imperial lyric papyri shows breadth and variety. Fragments of imperial papyri survive for all nine canonical poets.Footnote 14 From the extant papyri containing fragments of books of ancient lyric, we know that copies containing poems of Pindar, Alcaeus and Sappho were produced in large numbers between the first and third century and, in the case of Pindar and Sappho, all the way down to the sixth and seventh century; but some book-rolls with Anacreon’s and Simonides’ melic poems were also available during the same period.Footnote 15 As shown by PParis 71 and POxy. 24.2387, Alcman’s partheneia were still copied between the first century BCE and the first century CE, whereas the extraordinary PLond. 733, containing Bacchylides’ Epinicians and Dithyrambs, was produced in the late second or early third century. Finally, Stesichorus and Ibycus too seem to have experienced some diffusion during the High Empire.Footnote 16 The overall panorama of imperial lyric papyri was broad concerning the corpus of individual poets too. Besides the Epinicians, Pindar’s Hymns, Paeans, Dithyrambs, Hyporchemata and Threnoi were all still copied as late as the end of the second century.Footnote 17
Variety is also suggested by the availability of a range of interpretative tools. Stemming from the tradition of Hellenistic scholarship, lyric hypomnemata copied, and often composed, in the imperial period supplemented editions.Footnote 18 At the same time, biographical material – collected, once again, by zealous scholars from Hellenistic times onwards – contributed to imperial readers’ perceptions of ancient lyric poets and their poetry. This is the case with the Life of Sappho preserved by POxy. 15.1800 fr. 1 (late second century), where we find two of the traits marking her reception in imperial times.Footnote 19 As if in counterpoint to the nature and content of her poetry, POxy. 1800 describes Sappho as ‘contemptible in appearance and quite ugly’, with a dark complexion and very short (19–24), a depiction which recurs in writers like Ovid (Her. 15.33–6) and Maximus of Tyre (Diss. 18.7).Footnote 20 Even more strikingly, this imperial biography attests to a strain of tradition in which Sappho’s connection with female homosexuality was singled out as problematic or even reprehensible: the information that Sappho was accused by some of being ‘irregular in her ways’ and a ‘woman-lover’ (16–19: ἄτακτος οὖ[σα] τὸν τρόπον καὶ γυναικε[ράσ]τρια) resonates with Horace’s mascula Sappho (Hor. Epist. 1.19.28) and with its ancient exegesis (Porphyrio in Hor. Epist. 1.19.28 p. 362 Holder), as well as with the Suda notice about Sappho’s ‘impure friendship’ with her companions (Σ 107).Footnote 21
The (re)construction of poetic lives was not separate from the reading of poems but went hand in hand with their interpretation: first-person statements, central to lyric poems, were used to deduce personal events concerning the authors, and the biographical narratives so obtained further shaped the reading of single texts.Footnote 22 The importance of the synergy between lives and texts is supported, for example, by POxy. 29.2506. In this commentary on a collection of lyric poets, copied between the first and second century, Alcaeus’ troublesome involvement in politics – a staple of his reception in Roman times – is traced through quotations from his poetry.Footnote 23 Similarly, the only Life of Pindar so far discovered on papyrus (POxy. 26.2438, mid second–mid third century) exemplifies the poet’s outlook by quoting Pind. Ol. 2.86–8, a self-reflective and self-conscious statement in praise of inborn qualities, and one which enabled the construction of Aristides’ own Pindaric, and epinician, persona.Footnote 24
When we look more closely at the data provided by papyri, however, we realise that two crucial distinctions apply: one internal to the lyric canon, and one concerning the position of the lyric genre in comparison with other genres and their circulation among imperial readers. As we have mentioned, Pindaric poetry and Lesbian lyric dominate the papyrus count, suggesting the popularity of these lyric subgenres over other lyric readings. Clearly, such a selection was still representative of the genre’s variety, for it included central lyric discourses such as eros, politics and praise. What really stands out from a survey of imperial lyric papyri, however, is the clear gap marking off Pindar from the rest of the lyric nine: the number of papyri containing Pindar’s poems is almost double the number of copies of Alcaeus, and more than double the number of Sappho’s.Footnote 25 Once again, the trend described by papyri dovetails with what we know from literary sources. Pindar’s primacy is far from surprising when we recall his identification as regnator lyricae cohortis (Stat. Silv. 4.7.5) and lyricorum princeps (Quint. Inst. 10.1.61), definitions which were in turn re-enacting a critical topos of Hellenistic descent.Footnote 26 Pindar’s top position within the lyric canon was a direct function of his characterisation as a superior, almost divine voice, a perception rooted in Pindar’s own metapoetic statements as well as in the tradition concerning his extraordinary nature (cf. §1.3.2, §§2.1–2). In this context, the circulation of Pindaric texts reflected, and most probably helped secure, Pindar’s position as leader of the lyric canon.
Thus, Pindar dominated the imperial lyric reading list, followed by Alcaeus and Sappho. But how did lyric poetry fare on imperial shelves as a whole, in comparison with the authors most commonly identified with imperial elite paideia? When set against the evidence for the circulation of favourite classical models, the extent of the penetration of lyric in imperial readership is clearly on a smaller scale. Leaving Homer and his over 1,400 copies aside, even Pindar was never as frequent as Demosthenes (over 200), Euripides (almost 130) or Menander (over 90).Footnote 27 This result is perfectly in line with what we know from imperial writers’ preferences in terms of classical touchstones. Besides the ‘special, all-encompassing and superlative position of Homer’ in imperial education and culture,Footnote 28 Demosthenes’ popularity is easily gauged from the number of imperial sophists composing declamations in his style, as well as from his mention among the classics targeted for intensive (but, in this case, intellectually ineffective) collection by Lucian’s ignorant book collector (Luc. Ind. 4).Footnote 29 As for dramatic poetry, when advising a friend willing to acquire rhetorical training well into his adult life and with little time for his studies, Dio Chrysostom still selected Menander and Euripides as profitable reading matter (Or. 18.5; see further p. 38).
Yet the difference in papyrus numbers between core classical models and lyric authors does not mean that the presence of lyric in imperial readership was qualitatively irrelevant; quite the opposite. As far as we can see from the evidence available, reading and textual engagement with lyric represented a super-elite form of paideia. Lyric authors were less mainstream than the classical models foregrounded by imperial elite education, but as such they also represented a more specialised and niche tradition. A similar model for the circulation of non-core genres has been proposed by Theresa Morgan, who distinguishes precisely between ‘core’ and ‘peripheral’ authors.Footnote 30 Unlike Morgan’s ‘peripheral’, however, my notion of lyric poetry as super-elite paideia avoids the risk of implying marginality.Footnote 31 In the economy of imperial literary culture, lyric poets were not marginal but more refined readings. In this sense, the situation of Pindaric papyri versus copies of Homer or Demosthenes is better illuminated by the contrast between ‘small’ and ‘big’ libraries recently proposed by Netz. According to this model, while small libraries collected ‘just the canon in the narrow … sense’, bigger, and more specialised, collections featured ‘much larger holdings’, which included lyric poetry.Footnote 32 Furthermore, apart from a few big collections, papyrological material of the Roman period was ‘more heavily populated by the ideal-type small library’.Footnote 33 It is thus only to be expected that copies of lyric poets circulated in smaller numbers than copies of Euripides or Menander. This difference does not simply confirm the larger circulation and popularity of the Athenian playwrights over lyric poetry; instead, it marks the latter as specialised, more refined reading.
1.1.2 Between School and Scholars
However, statistical considerations about authors’ diffusion within ancient collections – the core of Netz’s quantitative model – are only part of the evidence supporting my argument about the distinctive status of lyric in imperial Greek literary society. Through the analysis of their features, papyri give us information about the types of readers tackling lyric texts, their reading contexts and strategies, and, supplemented by other sources on contemporary education and reading practices, they illuminate further the special and specialised place of lyric in the system of paideia shared by Aristides.
A first point to make is that lyric poetry is normally absent from school papyri documenting elementary exercises and students’ first steps into grammatical learning.Footnote 34 This indicates that lyric was not considered appropriate to the first stages of literate schooling, most probably also because of the difficulties posed by its dialectal variety (more on this later, pp. 40–2). A curious exception might be PBour. 1, a fourth-century notebook with syllabic wordlists including the name of Bacchylides, possibly that of Pindar and Ψαῦμις (Psaumis), the name of an athlete celebrated by Pindar in Ol. 4 and 5.Footnote 35 The inclusion of lyric poets per se does not need to indicate familiarity with their texts: their identity as lyric icons would suffice to explain their inclusion (cf. Chapter 2). The case of Psaumis is different though. Since, as stressed by Morgan, wordlists of this kind were the first educational texts where pupils met literary characters, the Pindaric trace in PBour. 1 might call to question the idea that lyric texts did not feature in elementary teaching at all.Footnote 36 There is however a simple and more plausible explanation: the author of the lists, most probably the teacher himself, could have found Psaumis ‘through a hypothesis or a table of contents’.Footnote 37 No direct or deep engagement with Pindar’s ode, then, is required to explain the Pindaric presence in the school wordlist of PBour. 1.
The absence of lyric from elementary school exercises is balanced by its extensive presence among copies that as far as we can see were destined for more advanced and sophisticated readers. As pointed out by Giuseppe Ucciardello, in contrast with papyri of classics like Euripides or Menander, who were often copied on the verso of discarded documents, the absence of copies of Pindar’s epinicians written on reused papyri indicates a ‘well-accomplished readership’.Footnote 38 Moreover, most lyric papyri present a variety of readers’ aids (e.g. punctuation, distinguishing marks) and marginal annotations, often derived from and ranging from variant readings to glosses and explanatory notes – features which suggest once more that reading lyric poetry required a certain degree of scholarly attention and preparation.Footnote 39 This point is made especially evident by the work of William Johnson on the reading context of annotated papyri. According to Johnson, literary papyri annotated by multiple hands may be telling of production and use within elite reading communities, where the analysis of challenging texts was ‘constructed as a collective endeavour’.Footnote 40 To support this claim, Johnson concentrates on variants added by distinct but contemporary hands, and more than half of the book-rolls he examines to this end contain lyric poets.Footnote 41 Some of the (numerous) imperial lyric papyri with marginalia may therefore be the result of hermeneutic activity by accomplished readers with a penchant for ancient lyric. Such a scenario dovetails nicely with what we know about the existence of lyric-focused caches: as stressed by George W. Houston, the Oxyrhynchus book-rolls forming Grenfell and Hunt’s second find included a substantial number of lyric volumes, revealing a strong lyric interest on the part of the owner(s).Footnote 42
Annotated lyric papyri, however, have also been connected to more advanced stages of literary education, where they could be used by both students and literature teachers (grammatici).Footnote 43 The prevalence of marginalia in volumes of Pindar and Alcaeus in particular has been read by Kathleen McNamee as proof of their use as readings in the grammaticus’ classroom.Footnote 44 I do not agree, though, with McNamee’s conclusion that Pindar and Alcaeus must have therefore ranked on the same level as Homer or Euripides, as part of the ‘minimal cultural package at grammatical level’.Footnote 45 It is true that, according to Sextus Empiricus, Pindar belonged alongside Homer, Hesiod, Euripides and Menander among the authors targeted by the hermeneutic efforts of the grammarian (Math. 1.58). But while familiarity with Homer or Euripides began with the very first stage of literacy, lyric texts were reserved for advanced students, so that no continuous textual engagement comparable to that with the core classics can be assumed for lyric authors.Footnote 46
Moreover, while Pindar’s leading position in the lyric canon may have made him the first lyric choice for grammarians, the inclusion of further lyric poetry in the teaching repertoire most probably depended on the choice and specialisation of individual grammatici.Footnote 47 In the portrait of some imperial grammarians, the ability to offer less mainstream poetry is marked out as a sign of scholarly distinction.Footnote 48 This is the case of Statius’ father, a grammaticus and poet himself celebrated in Silv. 5.3.146–59 for teaching, besides Homer, Hesiod and Pindar, a series of lyric authors not even limited to the canonical nine (uolucrumque precator | Ibycus et tetricis Alcman cantatus Amyclis Stesichorusque ferox saltusque ingressa uiriles |non formidata temeraria Chalcide Sappho; ‘and Ibycus, who prayed to birds, and Alcman, sung in austere Amyclae, and bold Stesichorus and rash Sappho, who feared not Leucas but took the manly leap’) but stretched to include the ‘secrets of gentle Corinna’ (tenuisque arcana Corinnae).Footnote 49 By closing the syllabus with the rhetorical question quid parva loquor? (‘But why speak of trifles?’) and turning to his father’s prose paraphrase of Homer, Statius himself appears to stress that the majority of the poets taught by the elder Statius were ‘not among the “core” authors’.Footnote 50 It was precisely their status as minor (cf. parva), though, that added refinement and literary sophistication to the syllabus.
Aristides himself ascribes a similarly extensive teaching repertoire to his literature teacher Alexander of Cotiaeum, in a letter sent to the grammarian’s fellow citizens after Alexander’s death. The passage is worth considering in detail (Or. 32.24):
καὶ μὴν εἰ Ὅμηρον Σμυρναίοις παρασχέσθαι καὶ Παρίοις Ἀρχίλοχον καὶ Βοιωτοῖς Ἡσίοδον καὶ Κείοις δὴ Σιμωνίδην καὶ Στησίχορον Ἱμεραίοις καὶ Θηβαίοις Πίνδαρον καὶ Μυτιληναίοις Σαπφὼ καὶ Ἀλκαῖον καὶ ἑτέροις ἑτέρους τινὰς φέρει φιλοτιμίαν – τὰς γὰρ Ἀθήνας ἐῶ τὰ νῦν –, ἦ που καὶ ὑμᾶς μέγα φρονεῖν εἰκὸς ἐπὶ τῷ τούτους ἅπαντας κοσμήσαντι καὶ δείξαντι.
Indeed, if the Smyrnaeans are proud of having given Homer to the world, and the Parians because of Archilochus, and the Boeotians because of Hesiod, and the Ceans because of Simonides, and the Himerians because of Stesichorus, and the Thebans because of Pindar, and the Mytilenaeans because of Sappho and Alcaeus, and other cities because of various other authors – for I now omit Athens – it is then likely that you are very proud of the man who adorned and explicated all of these.
In terms of prestige, Alexander’s exegetical activity is equated to the works of the great poets he used to teach; as a result, the grammarian’s legacy adds lustre to Cotiaeum just as the ancient poets did to their respective hometowns. In providing an overview of Alexander’s repertoire, however, Aristides foregrounds certain authors (five poets out of the lyric nine, besides Archilochus, himself a ‘lyric’ author in the broader sense, and the ever-present Homer and Hesiod) and glosses over others (ἑτέροις ἑτέρους τινάς), most notably the authors who enhanced the glory of Athens. As proposed by Leofranc Holford-Strevens, Aristides’ decision to avoid dwelling on Attic poetry, especially drama, may have been due to the risk that Athens ‘would overshadow Cotiaeum’.Footnote 51 But the choice is also justified by the ‘must-have’ status of Attic literature in imperial paideia: Aristides could bypass Alexander’s teaching and grammatical work on Athenian poets precisely because they were a standard component of literary education. On the contrary, the fact that lyric was associated with a more advanced and sophisticated stage of paideia justifies Aristides’ precision in detailing all the lyric poets included in Alexander’s expertise: as in the case of Statius’ father, the inclusion of lyric adds lustre to the teacher’s syllabus and teaching.
Specialisation on lyric texts continued to mark the profile of some grammarians well into late antiquity. The grammaticus Horapollon (fourth–fifth century) was credited with a commentary on Alcaeus, whereas a certain George (sixth century), possibly from Gaza, composed Anacreontic poems with Pindaric and Sapphic reminiscences.Footnote 52 The association between lyric learning and intellectual prestige was equally active in the late antique West: in his epithalamium for Honorius and Stilicho’s daughter Maria, Claudian lists Homer, Orpheus and Sappho among the Greek authors studied by the young bride under her mother’s guidance (232–5: Latios nec uoluere libros | Desinit aut Graios, ipsa genetrice magistra, |Maeonius quaecumque senex aut Thracius Orpheus | Aut Mytilenaeo modulatur pectine Sappho). As observed by Anna Lefteratou, here Claudian may be ‘trying to show Maria’s erudition across the East and the West, in Greek and Latin’, even ‘irrespective of whether she could read Sappho in the original or not’.Footnote 53
1.1.3 Rhetorical Practice and Atticism
For aspiring orators, at any rate, grammatical teaching was only a transitional step to the ultimate stage of elite paideia, when they were trained by a rhetor in public speaking. As anticipated in the Introduction, in the context of imperial rhetoric, lyric was perceived as discursively akin to epideictic oratory, as both genres ‘of presence’ centred on performance, embedded in well-defined performative contexts and negotiating similar discourses of praise and blame (cf. Menander Rhetor’s use of lyric comparanda, pp. 333, 340, 393 Russell–Wilson). On the other hand, the existence and recognition of generic kinship between lyric and sophistic oratory does not automatically demonstrate that imperial orators, established and not established, engaged with lyric texts extensively, so that questions concerning the use and utility of lyric poetry in rhetorical training still need to be addressed.
Lyric features prominently in some famous reading lists for orators-to-be. Both Dionysius of Halicarnassus (De imit. fr. 31.2.5–8 Usener–Radermacher) and Quintilian (Inst. 10.1.61–4), for instance, refer to Pindar, Stesichorus, Alcaeus and Simonides as models of style that may be useful for ‘modern’ orators.Footnote 54 Yet, as stressed by Morgan and Laura Miguélez-Cavero, the inclusion of specific authors in such lists does not mean that the authors in question were all equally used by all rhetoricians; catalogues of this kind point rather to ‘a desiderandum’ and tend therefore to be both broad and ideal.Footnote 55 In fact, models of rhetorical education different from, more abridged and/or more practical than Dionysius’ or Quintilian’s may have treated lyric as a more dispensable genre. According to the reading plan laid out by Dio, if a mature student wished to acquire rhetorical skills to bolster his public career,Footnote 56 then he should stay away from ‘lyric, elegy, iambi and dithyrambs’; all very valuable for the man of leisure, but time-consuming and impractical for a man of political and rhetorical action (Or. 18.8).Footnote 57
Despite the peculiar student profile considered by Dio, his exclusion of lyric seems compatible with the way the genre is treated (or not treated) in other sources on rhetorical training such as progymnasmata (‘preliminary exercises’ in declamatory tropes) and declamations.Footnote 58 As the initial stage of rhetorical education after the grammarian’s teaching, progymnasmata reflect the penchant for Homeric epic and Attic literature already observed for literary education.Footnote 59 Accordingly, in the extant exercises references to lyric poetry are difficult to find, and in the few cases where specific lyric elements can be identified, they point rather to broadly cultural reminiscences. References are usually to a famous myth, like the story of the golden rain on Rhodes told in Pind. Ol. 7;Footnote 60 to a well-known lyric gnome;Footnote 61 or to the figures of the poets themselves, mentioned for their poetic activity or as protagonists of exemplary anecdotes.Footnote 62 But the limited relevance of lyric to progymnasmata must depend also on the thematic nature of declamations, to which progymnasmata served as introduction. When they were not building on mythological characters and events, declamations (re‑)created deliberative debates (suasoriae) and court proceedings (controversiae) set in the context of the classical polis, mostly Athens: for such re‑animation of the classical past, Attic literature, especially oratory, was the main (and sufficient) point of reference.Footnote 63 On the contrary, lyric figures and themes were less immediately useful as sources for declamatory subjects, to the extent that even on the rare occasions when a lyric poet or poem was recalled, these were subordinated to declamatory fiction and lost their poetic specificity to be turned into court cases or deliberative issues.Footnote 64
Yet again, we should be wary of excluding any relevance of lyric to imperial orators, either established or aspiring. If we go back to Dionysius or Quintilian, it is clear that they both found it useful to refer to lyric authors and texts as models to illustrate the specific stylistic features available, if needed or wanted, to the (new) professionals of public speaking. The same approach is confirmed by other works concerning style and aesthetics, such as Pseudo-Longinus’ On the Sublime or the treatise On Types of Style by Hermogenes of Tarsus. While motivated by very different aesthetic and rhetorical criteria, both Pseudo-Longinus and Hermogenes felt the need to draw on lyric examples,Footnote 65 a choice that is all the more significant in the case of Hermogenes, considering his notion of prose as superior to poetry and the small number of poetic references he uses throughout.Footnote 66 Lyric was therefore not a regular but a possible reference genre for orators. What is more, its presence in sophisticated discussions of style suggests that lyric poetry mapped onto rhetorical education and practice in the same way as it mapped onto grammatical learning: that is, as an advanced model in comparison with core classics. It is not by chance that Pseudo-Longinus identifies central poets of the lyric canon like Sappho and Pindar with (different forms of) the sublime (Subl. 10.1–3; 33.5).Footnote 67 A standard, functional level of rhetorical expertise did not require orators to engage with ancient lyric poetry: declamations were rooted in Athenian oratory and historiography (including its reported speeches), and core poetic genres such as epic and drama already provided sufficient scope for engagement with classical poetry. Against the background of imperial readership and literary education and in the context of rhetorical practice, lyric was one of the least common and most refined models a sophist could pick for himself.
There is then another reason we need to factor in to understand why lyric texts did not represent common readings and models for imperial orators: language. The Doric dialect of choral poetry or the Aeolic forms found in Sappho and Alcaeus could not but stand out in the literary culture of Roman Greece, where Atticism was dominant, and where authors, and audiences, were thus especially self-conscious about the use of different Greek dialects.Footnote 68 When referring to Alcaeus as a useful stylistic model (De imit. fr. 31.2.8), Dionysius remarks that his poetry can be a model of ‘figures and clarity’ (τοὺς σχηματισμοὺς καὶ τὴν σαφήνειαν), ‘inasmuch as it is not damaged by his dialect’ (ὃσον αὐτῆς μὴ τῇ διαλέκτῳ τι κεκάκωται).Footnote 69 This caveat is all the more striking when we recall that Dionysius was anything but an extreme Atticist.Footnote 70 A similar observation is made by Pausanias about Alcman’s poetry, whose charm was not spoilt by the Laconian dialect, ‘the least musical of them all’ (3.15.2: ἥκιστα παρεχομένη τὸ εὔφωνον).Footnote 71
To remedy the ‘damaging’ peculiarities (or aberrations?) of lyric language, advanced readers made extensive use of dialectal glosses, found in large numbers on Alcaean papyri but attested also in book-rolls of Sappho, Pindar and Corinna.Footnote 72 At the same time, lyric authors provided grammarians like Apollonius Dyscolus and Herodian with a treasure trove of dialectal forms (cf. Sappho frr. 38, 45–6 Voigt; Alc. frr. 309–10 Voigt) and were the subject of dialectal treatises like Tryphon’s work ‘On the dialects of Homer, Simonides, Pindar, Alcman and the other lyric poets’ (Augustan age; Suda Τ 1115).Footnote 73 It is even probable that lyric texts formed the backbone of specialist lexica and glossaries, on which grammarians and papyrus marginalia drew.Footnote 74
If this hypothesis is correct, then such lyric-based lexica represented a dialectal counterpart to the Atticist lexicography followed, more or less dutifully, by sophists.Footnote 75 Besides posing issues of linguistic knowledge and preference, however, in a context where Attic Greek was used, and mandated, as the expression of a global Panhellenic identity, the non-Attic character of lyric added a regional aspect to the genre, which in turn must have contributed to placing lyric outside of the core canon. By this, I do not mean that in the imperial period lyric retained no Panhellenic value, but rather that its linguistic features favoured its reception as a more regionalised form of poetry in comparison with other genres central to elite paideia.Footnote 76 With its linguistic diversity, lyric served as a counterpoint to the globalising (and ‘oppressive’?) tradition of Atticism;Footnote 77 and this, at a time when, against the seemingly uniform backdrop of Attic Greek, ‘the persistence of local languages, dialects, and onomastics provided a space for the preservation and celebration of local identity’ in response to imperial uniformity.Footnote 78 Its regional aspect enhanced the unique potential of lyric to broadcast local identity within the spatial dynamics of the Empire, a cultural function which was active in parallel with, and to a certain extent even independently from, the circulation of lyric texts among elite readers and writers (see further §2.2).
1.2 Telling a Pepaideumenos by the (Lyric) Texts He Quotes
There is at least one more area from which we can get a fuller sense of the standing of lyric within the literary culture of the Roman period. In a world where social standing was coterminous with the display of literary education, what better way to understand the role played by lyric in paideia than by looking at what pepaideumenoi themselves quoted?Footnote 79 In doing so, my aim is precisely to trace some major quotation patterns in a selection of imperial prose authors, to assess how these patterns compare with my argument on the super-elite place of lyric in the imperial canon. Furthermore, defining habits of lyric quotation in other Second Sophistic writers proves essential to framing Aristides’ own take on lyric poetry.
A survey of this kind requires some preliminary remarks, though. First, what do we mean exactly by ‘quotations’ in this context? While my analysis of Aristides’ work will bring attention to a dense web of references, including possible allusions and the appropriation of lyric tropes and personas, here I will concentrate primarily on mapping the volume of references to specific lyric texts, in the form of either textual quotations or paraphrases of poems.Footnote 80 The reasons for this choice are both practical and methodological: besides the impossibility of surveying every lyric reference (assured or hypothetical) in all the writers I consider, the data I take into account are enough to assess the circulation of lyric poetry among pepaideumenoi, which is my focus in this section.
The second observation is more of a caveat: even the presence of a specific quote may not point necessarily to direct lyric knowledge. As argued by Ian Rutherford, at least some of the Pindaric quotations found repeatedly in imperial literature are in fact secondary quotes derived from Plato or drawn from other earlier prose authors, to the point that some recurrent Pindaric tags may be isolated.Footnote 81 Moreover, lyric snippets could be taken from anthologies without knowledge of or engagement with their context.Footnote 82 However, these are still not serious objections to what I plan to do here. We have evidence that the same phenomenon of ‘recycled’ quotations applied to authors more popular than lyric poets, and that this happened in parallel with the independent circulation of their works.Footnote 83 More importantly, whether direct or secondary, contextualised or not, the occurrence of references to lyric texts, much like their absence, gives us a sense of the relevance and function of lyric within the traffic of paideia performed by imperial texts – a measure, that is, of the role of lyric ‘knowingness’ in imperial society (cf. Introduction).
In the works of other sophists like Dio, Maximus of Tyre and Lucian, the volume of lyric quotations is overall quite limited and certainly nowhere close to the frequency of quotations from core poets like Homer, Hesiod or Euripides.Footnote 84 Applying a ‘do as I say and as I do’ approach (cf. Or. 18, discussed in §1.1.3), Dio uses lyric very rarely, and his apparent distaste for melic poetry even resurfaces in his ethopoetic imagination. In Or. 2, Dio’s Alexander explains to Philip that kings should avoid singing Sappho and Anacreon, and if they really (εἴπερ ἄρα) want to sing then they should prefer Stesichorus or Pindar, although ‘Homer is perhaps all one needs even to that end’ (28–9).Footnote 85 By the end of Or. 2, Alexander contradicts his own prescription by quoting one of Anacreon’s poems in full (62 = Anacr. fr. 357 PMG), but he does so only to exemplify once again what kinds of prayers (i.e. erotic) do not befit kings.Footnote 86
The same thematic selectivity is found in Maximus, who quotes a substantial string of fragments from Sappho (fifteen) and Anacreon (four), in positive terms this time, when defending Socrates’ homoeroticism (Diss. 18.9).Footnote 87 Apart from this erotic cluster, however, lyric does not have great relevance in Maximus. A Pindaric fragment does provide the formal starting point for a discussion on injustice (Diss. 12.1 = fr. 213), but for the sophist/philosopher, Pindar’s concern with words, metre and music makes his poetry as important as ‘children’s toys’ (χώραν ἔχει ὅσηνπερ καὶ τοῖς παισὶν τὰ ἀθύρματα).Footnote 88 The only other passages pointing to lyric in Maximus contain legends told (among others) by Pindar or generic references to poetic figures and their production, something that Maximus shares with the progymnasmata.Footnote 89 If we exclude an Anacreontic reference about love and old age and a quotation from Simonides, Pindar is also the only lyric author receiving (a little) attention in Lucian, where snippets from ‘beautiful Pindar’ add dainty diction to ecphrasis (Hipp. 7 = Ol. 6.3) or counterpoint humorously the down-to-earth tone of dialogues (Gall. 7; Tim. 41 = Ol. 1.1).Footnote 90
Less common than other models, hand-picked (and sometimes chastised) for specific themes or used as strokes of refined paideia: this pattern for the (re‑)use of lyric texts is confirmed by another staple of sophistic literature, Philostratus’ Imagines.Footnote 91 As a ‘big library’ type of reading, lyric suits Philostratus’ sophisticated exegetical game, and, though far less frequently than Homer, it is part of the literary erudition framing his paintings and their interpretation.Footnote 92 The mythical narrative of famous Pindaric epinicians underlies whole scenes (Imag. 1.17, 1.30 = Ol. 1; Imag. 2.24 = Ol. 7) through a combination of paraphrase and verbatim quotations helping (or challenging?) the educated reader to identify the lyric source, which is not named explicitly.Footnote 93 As in other sophists, then, in Philostratus too Pindar is joined by Sappho and Anacreon as main lyric references, and the deployment of their poetry is tied to different erotic contexts. A ‘sweet epithet’ (τὸ ἡδὺ πρόσφθεγμα) and other textual echoes from Sappho (e.g. frr. 2, 44.30, 185 Voigt) turn the description of a maiden chorus led by a skilled didaskalos into the imperial recreation of a Sapphic rite for Aphrodite (Imag. 2.1).Footnote 94 Anacreon’s lines, on the other hand, come in handy to describe Dionysus’ inebriating passion for Ariadne (Imag. 1.15.2 = fr. 376 PMG) or to portray courtship in the animal world (Imag. 1.16.4 = fr. 417 PMG). Yet other Philostratean works make less use of lyric. As shown by Bowie, for example, in the Life of Apollonius Pindar ranks among more niche authors like Empedocles and Iuba of Mauretania for number of quotes.Footnote 95 Again, a reference to the double nature of Sappho’s poetry, erotic and hymnic, is the only trace left of Philostratus’ apparently ‘good knowledge’ of Sappho in the Life (VA 1.30).Footnote 96
As far as we can see, then, quotation trends dovetail with the picture emerging from imperial papyri, schools and rhetorical practice.Footnote 97 Far from belonging among the standard go-to texts for literary references and display of paideia, lyric poems represented a specialised genre which could at times serve specific purposes of diction and content. The association of lyric poets with specific themes and values also had a role in their exclusion from the works of some sophists or their (rare) selection by others, according to each author’s arguments and self-presentation agenda (see §1.3). This conclusion reflects the status of lyric in imperial works that would not be immediately described as ‘sophistic’. One would expect at least some lyric to thrive amid the erotic twists and turns of the romance, but Daphnis and Chloe seems to be the only novel showing a sustained intertextual dialogue with lyric models, primarily with Sappho but with (some) Alcaeus too.Footnote 98
Among a few other strictly non-sophistic authors, however, lyric poetry was the object of (re‑)use and recasting loaded with literary, cultural and political significance. All the canonical nine feature in Plutarch’s extensive corpus, but with some significant differences in the extent of their presence. Alcman, Stesichorus, Ibycus and Bacchylides are cited only rarely.Footnote 99 Similarly, Plutarch does not seem to have found erotic lyric worthy of extensive consideration. Anacreon is scarcely quoted, and while Plutarch was of course aware of his erotic production (he pairs him twice with Sappho: Mul. virt. 243b; Quaest. conv. 711d), he draws on Anacreon’s words on passion only once (Amat. 751a).Footnote 100 As already observed by Bowie, this treatment was probably the result of Plutarch’s moral stance and philosophically oriented approach to literature (illustrated most forcefully in How a Young Man Should Listen to Poetry), which must also have shaped his attitude towards Sappho; her erotic poems too are rarely recalled by Plutarch, who ‘might have been especially concerned about the capacity of Sappho’s poetry to corrupt’.Footnote 101 Interestingly enough, though, Plutarch’s apparent caution towards, or rejection of, erotic lyric did not prevent him from working with, and drastically recasting, one of Sappho’s most famous poems: in De prof. in virt. 81d, the dramatic symptoms of passion so forcefully depicted in Sappho fr. 31 are ‘reassigned’ to ‘the young man who has tasted true progress in philosophy’. Even Sappho could hold some philosophical value, if her poetry was appropriately turned on its head.Footnote 102
Even reconfigured versions of Sapphic lyric, however, could not compete with Simonides and Pindar for Plutarch’s favour. Of the nine lyric poets, Simonides is the second most cited in the Lives and Moralia, where his poetry is deployed as a major source on the Persian Wars (e.g. Them. 15.2; De Her. mal. 869c) as well as the work of a ‘serious moral thinker’ able, among other things, to deal cleverly with power (De tranq. anim. 470d; De frat. amor. 485c; Quaest. conv. 743f).Footnote 103 Even more conspicuous, though, is the presence of Pindar, whose poetry is recalled by Plutarch more often than Menander’s.Footnote 104 From the content and context of quotations, it is clear that the Theban poet attracted the attention and approval of the Boeotian Plutarch for his aristocratic views on natural excellence (Arat. 1.2), as well as for his treatment of divine subjects, which went hand in hand with traditions about Pindar’s personal connection to the gods, especially Delphic Apollo (De sera 557f; Quaest. conv. 717d; Non posse 1102f–1103b; cf. §§1.3.2, 2.1–2).Footnote 105 What is less recognised but equally, if not more, interesting is how Plutarch recast Pindaric lines to comment on or tackle contemporary Roman politics. In the probably early epideictic speech On the Fortune of the Romans, two Pindaric definitions of Τύχα are quoted to celebrate Rome’s providential destiny to rule (De fort. Rom. 318a = fr. 40 Snell–Maehler, 322c = fr. 39).Footnote 106 In Marc. 21.2, on the other hand, Pindar’s description of Syracuse as ‘sanctuary of Ares mighty in war’ (Pyth. 2.1–2) is used to highlight, in less flattering terms, the exclusively military focus of early Republican Rome, which ‘filled full of barbaric arms and bloody spoils, and crowned round about with memorials and trophies of triumphs … was not a gladdening or reassuring sight’.Footnote 107 Pindaric lyric could equally enhance Plutarch’s political advice for his Greek contemporaries. In An seni 783b, a fragment with agonistic content, allegedly appreciated by the addressee (fr. 228: ‘when contests are instituted, excuse … casts excellence into sheer darkness’), is recalled to encourage him not to retire from local political duties, despite his old age. Yet Pindar also came in handy to stress the miseries of politics under Rome. In De exil. 602f, Pae. 4.50–3 is rephrased as a ‘mantra’ for an exiled friend; he should be glad to have been allotted ‘no sorrows, no civil strife’, nor, Plutarch adds, ‘commands from the governor (προσταγμάτων ἡγεμονικῶν) nor assistance to the needs of countrymen and public services that are difficult to decline’. The consolatory aim of the text justifies this perspective; and still the emphasis on the troubles of political life under an external power, buttressed by the rephrasing of Pindar’s lines, suggests just how tightly the rules of the imperial political game could be construed.Footnote 108
Political and cultural ramifications concerning the imperial present underlie another noticeable exception to the rarity of lyric among imperial writers: Pausanias’ Periegesis. What is eye-catching about the Periegesis is the range of lyric poets and texts it brings up: besides Pindar and Sappho,Footnote 109 Alcman, Alcaeus, Stesichorus and Ibycus are all among Pausanias’ literary guides to the Greek landscape. Such interest in lyric can be explained as a function of Pausanias’ focus on ancient cults and myths, especially those testifying to local traditions diverging from the homogeneous Hellenism prized at the time.Footnote 110 Pausanias resorts to Alcaeus’ hymnic production for alternative accounts of the actions and whereabouts of the gods (7.20.4 = Alc. fr. 308c Voigt; 10.8.10 = fr. 307d); whereas Stesichorus’ works provide epic narratives not covered by Homer (cf. 9.2.3 = Stesich. fr. 236 PMGF). The centrality of local lore to the Periegesis, and the regional relevance of lyric, is evident in particular in the use of Alcman’s poetry, deployed specifically in the description of Laconia (Paus. 3.18.6 = Alcman fr. 62 PMGF; 3.26.2 = fr. 23).Footnote 111 But Pausanias’ interest in the localism of lyric did more than merely preserve epichoric traditions and precious lyric snippets; it should rather be reconsidered as part and parcel of Pausanias’ reaction to the globalising gaze of imperial rule, which risked absorbing local diversity into a ‘seamlessly unified, non-diverse empire’.Footnote 112
1.3 Manners (and Style) Maketh Poet
The lyric range available to writers who, like Aristides, chose to engage with the genre was fairly wide. As we have just seen, specific lyric poets were recalled in connection to distinctive themes and contexts, a reception process which resulted in a mental map of distinctive lyric styles and personas active in imperial literary society. Such a map of lyric voices shaped the experience and view of lyric for imperial readers of the genre. But it is worth stressing once more that some of the features attached to these lyric figures and their poetry had a wider circulation beyond (super‑)elite paideia. Close acquaintance with her poems was not necessary to know about Sappho’s erotic poetry; Alcman’s connection with Laconia could live on separately from the text of his Spartan songs; while Pindar’s ‘sublime’ style had a direct correspondence with stories about his closeness to the gods (cf. Chapter 2).
When pitting lyric against epic in Carm. 4.9, Horace uses some precise attributes and images to distinguish between different poets: as leader of the canon, Pindar opens the sequence; Alcaeus’ and Stesichorus’ Muses are described respectively as ‘aggressive’ (l. 7: minaces) and ‘august’ (l. 8: graves); Anacreon is associated with playfulness, while Sappho is kept alive (poetically) by her amor and calores, ‘love’ and ‘passions’. Such a differential characterisation, however, is hardly the result of Horace’s personal taste. Compare for instance AP 9.184:
Despite being anonymous, the epigram is dated to the Hellenistic period (second/first century BCE) and is one of the few canon-epigrams which offer a miniaturised account of lyric poetry.Footnote 113 Just as in Horace, here too individual traits identify single poets, with distinctive features of their poetry reflecting back on, and conflating with, their personas. Of course, differences remain between the epigram and the ode as a result of variatio; most conspicuously, Horace focuses only on a further selection from among the canonical nine. Even so, from a comparison of the two texts it is apparent that poets are placed within the lyric system according to absolute coordinates followed by the anonymous author of the Hellenistic epigram as well as by Horace. Leaving aside the primacy of Pindar, marked by his position at the beginning of both lists (and by his identification as the sacred spokesman of the Muses in the epigram), Alcaeus’ ‘aggressive’ political poetry is similarly foregrounded in AP 9.184.7 through the image of his sword (ξίφος Ἀλκαίοιο); Stesichorus’ relation to Homeric poetry, expanded into a liquid metaphor by the epigrammatist, has left traces in Horace’s use of graves to qualify Stesichorus’ poetry (cf. Quint. Inst. 10.1.46: idem [i.e. Homer] … iucundus et gravis); and both the themes isolated by Horace (amor, calores) and the Sapphic ‘graces’ evoked in AP 9.184.2 match the erotic character of Sappho’s work.Footnote 114 Starting with the Hellenistic period, lyric poets underwent a process not only of canonisation but of crystallisation, whereby each of them came to be associated with specific themes, subgenres and stylistic traits which ended up defining their figures in terms of personal and moral traits.Footnote 115 Taking the cue from these two instances of lyric crystallisation, the following survey of lyric ‘types’ draws on epigrams, treatises on style, ancient biographies and other sources of critical (and personal) judgements on poets, in order to reconstruct the mental map of lyric styles and voices in its imperial configuration.
1.3.1 Lyric LSD: Love, Strife, Drunkenness
The distinction between monody and choral lyric is notoriously modern, and a risky one to make.Footnote 116 Leaving aside the difficulty of pinning down ancient performances, and taking account of the impossibility of tracking all the changes and adaptations which occurred between original and subsequent re‑performances, the same poets often composed both solo and choral songs: since it includes wedding songs, for example, at least part of Sappho’s poetry was originally sung by a chorus.Footnote 117 Nonetheless, this distinction – especially when strictness is avoided – is a convenient one to work with, if only because what modern taxonomies have described as monody and choral lyric map onto two different subsets of poetry already recognisable in antiquity: the former more personal and emotionally charged, as opposed to the more public and encomiastic profile of the latter.Footnote 118
Of the focus on personal emotions and passions traditionally associated with erotic monody, Sappho represents both the norm and the exception. On the one hand, the centrality of love to (a substantial portion of) her poetry, magnified by the stereotyping tendency just mentioned, made Sappho into a staple, if not the archetype, of erotic poetry.Footnote 119 On the other hand, Sappho’s female identity, paired with the strong homoerotic feelings expressed in her poems, also resulted in a markedly gendered reception of her figure, in which poetic glory and (potential for) sexual transgression coexisted in tension with each other.Footnote 120 Within the male-dominated lyric canon, Sappho was compared and eventually assimilated to the Muses: in a series of epigrams marking her off as the most famous female poet, Sappho is referred to as ‘the tenth Muse’ (AP 9.66, 506, 571) or as the ‘mortal Muse’ who achieved immortal fame through the ‘deathless gifts’ of poetry (AP 7.14). It is in light of this characterisation that we should read Lucian’s (mocking) remark about wealthy Roman ladies surrounding themselves with connoisseurs who would praise their songs as ‘not much inferior to Sappho’s’ (Merc. cond. 36).Footnote 121
As already mentioned when discussing papyrus biographies (p. 30), however, the erotic, and especially homosexual, content of her poetry exposed Sappho to moral issues stemming from her role as sexually active woman: as explained by Melissa Mueller, ‘the shameful and basically incomprehensible thing about lesbian affairs for the ancient reader was that they demanded that one woman take on an active role, a role that could only be properly occupied by the phallic male’.Footnote 122 Sappho’s controversial homoeroticism thus played a fundamental part in the (re‑)construction of her figure and poetry, with reactions ranging from sensual emphasis to sanitisation and sublimation.
Sappho is a titillating, eroticised presence in Roman poetry, where her reception as a wanton lover is pitted against chaste Roman matrons (cf. Mart. 7.69, 10.35) or appropriated according to new erotic agendas. She is the intertext of Catullus’ Lesbia, while Ovid makes her into ‘a proponent of his own eroto-didactic verse’.Footnote 123 Even the heterosexual anecdote of her deadly passion for Phaon becomes a chance to rehash her homoerotic ‘crimes’ (cf. Ov. Her. 15.15–20; 19: aliae centum, quas hic sine crimine amavi). However, at the other end of the spectrum from lesbian Sappho (Schol. Hor. Epist. 1.19.28: tribas diffamatur fuisse), Sappho’s erotic nature was tackled by emphasising the less problematic and safer aspects of her love poetry. Since her corpus included hymns and epithalamia, Sappho’s figure was thus sublimated into the priestess of sacred rites involving female choruses, or into the composer of songs marking weddings – events, that is, which represented socially acceptable manifestations of eros and female beauty. This normalised version of Sappho is found in epigrams from the Hellenistic period onwards but was mirrored by imperial literature too. Besides stressing her connection with the Muses, AP 7.407 portrays Sappho as standing ‘over bridal beds’ together with Hymen, or joining Aphrodite in lamenting Adonis. Similarly, in AP 9.189 Sappho leads a ‘delicate’ dance performed by Lesbian women at the local precinct of Hera, a description resonating with the Sapphic scene depicted by Philostratus in Imag. 2.1.Footnote 124
A soft, sanitised image of Sappho also dominated her reception by late Hellenistic and Roman-period critics, for whom Sappho is a prime example of ‘sweetness’ (γλυκύτης), ‘charm’ (χάρις) and ‘elegant style’ (γλαφυρὰ ἁρμονία), with no mention made of homoerotic topics (Demetr. Eloc. 132, 179; Dion. Hal. Comp. 23.45, 50–101; Dem. 40.65; Hermog. Id. p. 334 Rabe).Footnote 125 It is telling that most of the quotations used by Demetrius to illustrate Sappho’s charming style either refer explicitly to a wedding context or have been plausibly related to the epithalamia.Footnote 126 Of course, readers and critics could still prefer to foreground Sappho’s erotically charged nature: this is the case for Pseudo-Longinus, who praises her inspiration in selecting and combining the ‘symptoms that accompany erotic madness’ (Subl. 10.1: τὰ συμβαίνοντα ταῖς ἐρωτικαῖς μανίαις παθήματα) in fr. 31. But Sappho’s sublimated voice still proved more popular, and useful, among rhetorical theorists: in the late imperial period, Menander Rhetor still points to Sappho as a model for (prose) hymns and wedding pieces (pp. 333.9, 334.28, 402.17–18 Russell–Wilson), and ‘epithalamic Sappho’ continued to represent an acceptable model even for Christian authors into the Byzantine era.Footnote 127 The persistence of this safer characterisation, however, testifies all the more to Sappho’s potentially distressing eroticism. The alternative to a sublimated Sappho was no Sappho at all: both Dionysius’ On Imitation and Quintilian omit her from the lyric models appropriate for public speakers.
When rejecting Sappho’s poetry as unsuitable for kings, Dio’s Alexander extends his ‘ban’ to Anacreon (Dio Chrys. Or. 2.28). Naturally, the erotic element characterising Anacreon’s lyric, directed at both boys and girls, made him a fitting male counterpart to the erotic tradition represented by Sappho. Among other imperial sources, the popularity of this connection, in fact going back to classical times, is confirmed by Pausanias, who refers to Anacreon as ‘the first after Lesbian Sappho to have devoted most of his poetry to eros’ (1.25.1: πρῶτος μετὰ Σαπφὼ τὴν Λεσβίαν τὰ πολλὰ ὧν ἔγραψεν ἐρωτικὰ ποιήσας).Footnote 128 Accordingly, in Dionysius’ system of styles, Anacreon was grouped with Sappho under the ‘elegant style’ (Comp. 23.45, Dem. 40.66).
Yet in the differential mapping of lyric voices, Anacreon was also distinguished from Sappho through the link to wine and drunkenness. From the classical to the imperial period, symposia played a central role in the performance and reception of both Sappho’s and Anacreon’s poetry,Footnote 129 but the sympotic element was activated only in Anacreon’s characterisation, undoubtedly as a result of the references to drinking in his own poems (frr. 356, 389 PMG). Together with old age, intoxication contributed to mark off Anacreon as a humorous, light-hearted and salacious singer.Footnote 130 Epigrammatic portraits down to late antiquity depict him as an old drunkard singing of love and wine (cf. AP 7.33, sixth century: ‘you died from too much drinking, Anacreon’), an image that returns in Max. Tyr. Diss. 21.2, where a ‘drunk, garlanded and singing’ Anacreon bumps into a nurse holding baby Cleobulus, one of Anacreon’s future eromenoi. The same stereotype translated (literally) into Ovid’s vinosus senex (Ars am. 3.330; cf. Tr. 2.363–4: ‘What but the mixing of love and abundant wine did the Muse of the old Tean poet teach?’) and was even further crystallised in the Anacreontea.Footnote 131 Drunkenness is reflected in the criticism of Anacreon’s poetry too: while Athenaeus suggested that Anacreon’s focus on drinking and intoxication be read as a literary construction rather than as a biographical fact (10.429b), according to Demetrius the asyndeton in Anacr. fr. 396 (φέρ᾿ ὕδωρ, φέρ᾿ οἶνον, ὦ παῖ) is ‘plainly the rhythm for a drunk old man’ (Eloc. 5).Footnote 132
A further strain of tradition insisted on Anacreon’s presence at the court of Polycrates of Samos, where Anacreon’s lyric contributed to the refinement of Polycrates’ tyranny and where the poet shared with his patron a passion for beautiful boys (Strabo 14.1.16; Max. Tyr. 20.1, 29.2; Apul. Flor. 15.11–12; Ael. VH 9.4). Despite this potential for political implications, however, politics is not a prominent theme in Anacreon’s fragments (at least, as far as we can see), nor has it left any trace in the reception of his literary persona beyond the stereotypical characterisation of the court poet.Footnote 133
The situation could not be more different in the case of Alcaeus.Footnote 134 As mentioned in §1.1.1, ancient scholarship on Alcaeus’ figure and poetry tied them up with the political turmoil of Mytilene, where Alcaeus and the other members of his hetaireia were involved in a partisan fight against the tyrants Pittacus and Myrsilus.Footnote 135 In Roman times, the connection between Alcaeus and Mytilenaean politics is recalled by Strabo, among others, who testifies to the existence of a specific group of poems dealing with Alcaeus’ involvement in the civic conflict (13.2.3: τὰ στασιωτικὰ καλούμενα, ‘the so-called stasiotic poems’),Footnote 136 whereas Alcaeus’ songs on ‘battles and banished tyrants’ continue to captivate his human and mythological audience in Horace’s underworld (Carm. 2.13.26–40).
But the significance of politics to the imperial reception of Alcaeus’ persona emerges in particular when his civic voice is evaluated taking into account the other, erotic facet of his poetry. Even though no textual trace of Alcaeus’ love poetry remains, Alcaeus’ lyric probably shared with Anacreon’s erotic songs more than Anacreon’s poetry shared in terms of political involvement with Alcaeus’. In Carm. 1.32, Horace picks Alcaeus as ‘his lyric forebear’ precisely because, in the calm after the storm of political unrest, the Lesbian poet would sing ‘of Bacchus and the Muses and Venus and the little boy who always clings to her’ (9–10), a description which miniaturises the themes of Horace’s own lyric and its relation to the Roman civil war.Footnote 137 The same characterisation underlies Alcaeus’ inclusion, with Anacreon and Ibycus, in a hackneyed trio of erotic poets going back at least to comedy (Ar. Thesm. 159–63); but still, according to Sextus Empiricus, both Alcaeus and Anacreon inflamed ‘erotomanes’ and ‘drunkards’ (Math. 1.298).Footnote 138 In rhetorical theory, however, the erotic side of Alcaeus’ voice seems to have been the object, much like Sappho’s, of selective censure, which in Alcaeus’ case entailed the condemnation and sidelining of his literary identity as love poet. No mention of eros is made by Dionysius, who instead praises the ‘forcefulness’ and ‘character’ (δεινότης, ἦθος) of Alcaeus’ political lyric, marking it off as an example of ‘political rhetoric’ in metre (De imit. fr. 31.2.8). When offering an assessment similar to Dionysius’, Quintilian takes pains to specify that, while Alcaeus is commendable for the part of his poetry where he attacks the tyrants and where he often sounds like an orator (Inst. 10.1.63: plerumque oratori similis), unfortunately he still ‘stooped to frivolity and passions’ (et lusit et in amores descendit).Footnote 139
With their diverse links to emotions, desires and pleasures, Sappho, Anacreon and Alcaeus provided a set of potentially ambivalent personas: associated with charm and elegance, but also targeted by disapproval and/or sublimation. As we shall see, an analogous rejection of inebriation and everything erotic brought about rejection and recasting of erotic lyric by Aristides (§3.1). At the same time, like Dionysius and Quintilian, Aristides may have ascribed rhetorical ‘forcefulness’ to Alcaeus’ political songs, but he also recognised too well the dangers of civil strife and discordant political conduct under Rome (Chapter 5).
1.3.2 Between Human and Divine: Public Choral Voices
The case of Ibycus makes clear the need to keep the distinction between monody and choral lyric somewhat loose and flexible: his poetry shared the erotic themes of Sappho and Anacreon; but, as suggested by his poem in praise of Polycrates (fr. 263 PMG), Ibycus’ lyric also anticipated the encomiastic focus of poets traditionally linked to choral performances like Simonides and Pindar, their ‘interest in the construction of kleos’ both for their patrons and for themselves.Footnote 140 Furthermore, besides its traditional association with Anacreon and Alcaeus, Ibycus’ lyric shared substantial features with the poetry of both Alcman and Stesichorus: active between the late seventh and the late sixth century BCE, this trio showcased the development of archaic lyric in the area stretching from mainland Greece to the Greek West and had in common the use of the Doric dialect, albeit with some differences.Footnote 141 As in the case of erotic monody, however, Alcman and Stesichorus were identified with poetic and personal traits which distinguished them from Ibycus and from each other, and which were ultimately crystallised by their imperial reception.
In the ancient characterisation of Alcman, the main distinctive feature was the poet’s link to Laconia and in particular to Sparta, an association built on and enhanced by the Laconian elements of Alcman’s dialect, as well as by the apparent function of his songs within the context of Spartan ritual. In the biographical tradition, however, the Laconian connection brought about a debate concerning Alcman’s birthplace, probably based on references found in the songs themselves and possibly fostering the claim of different locales to the famous poet: according to the Suda, Alcman was born in the Spartan village of Messoa, but a line of interpretation going back to Pergamon made him a Lydian from Sardis (Α 1289). The issue is discussed in the same commentary on papyrus previously mentioned for the political interpretation of Alcaeus (POxy. 2506 (p. 31)) and underlies the portrait of Alcman in the epigram: as recalled in AP 7.18.6 (Antipater of Thessalonica, Augustan period), Alcman was ‘a reason of dispute for two continents’. Within the discussion of Alcman’s birthplace, POxy. 2506 also foregrounds a second element central to Alcman’s persona by noting that ‘the Spartans put him in charge of the traditional choruses of daughters and young men’. The role of the poet as official composer for Spartan choruses, now so central to our understanding of Alcman’s lyric, was already stressed by imperial sources, especially with regard to maiden choruses: Alcman is singled out as an author of partheneia in [Plut.] De mus. 1136f, whereas Aristides referred to him as ‘the praiser and counsellor of maidens’ (Or. 2.129; cf. Introduction).Footnote 142
When compared to Alcman’s local relevance, Stesichorus’ voice stands out for its Panhellenic character, which resulted from the Homeric content of his poetry and in turn shaped the way his figure was (re‑)constructed. This is illustrated above all by the anecdote about the Palinode, brought to fame by Plato (Phaedr. 243a) and alluded to some six times by Aristides (Orr. 1.128, 166; 2.234; 3.557; 4.8; 33.2; see §§3.1, 6.1.1), according to which Stesichorus’ Homeric subject, in the ‘person’ of Helen, came back to haunt him and exact a more favourable treatment of the myth. Except for his traditional association with the Sicilian city of Himera, Stesichorus’ biography thus absorbed and repurposed the engagement with Homeric material distinctive of his poetry. As a poetic figure, the boundaries of Stesichorus’ identity were never quite as defined as those of Alcaeus or Ibycus and tended to blur into the figure of Homer himself: in AP 7.75, Stesichorus is the vessel for the soul previously belonging to Homer.Footnote 143 Unsurprisingly, it is once more in connection with Homer that Stesichorus lived on in literary criticism: he is ‘Homeric in the highest degree’ ([Longinus], Subl. 13.3) and ‘sustains on his lyre the weight of epic poetry’ (Quint. Inst. 10.1.62).
Born (allegedly) the same year as Stesichorus’ death (Cic. Rep. 2.20), by the imperial era Simonides stood for two distinct aspects of the occasional/encomiastic lyric of the early classical period. As we are reminded by Horace’s reference to the ‘Cean dirge’ (Carm. 2.1.38; cf. Catull. 38.8: lacrimis Simonideis), more than any other poet of choral songs Simonides came to be associated with threnodic songs, an idea that might have also intersected with his popularity as the author of funeral epigrams. It is to Simonides’ threnoi that Dionysius alludes when he says that Simonides was better even than Pindar ‘at expressing pity not in a grand style but through emotions’ (De imit. fr. 31.2.7: μὴ μεγαλοπρεπῶς ἀλλὰ παθητικῶς), an ability which Quintilian considered his ‘chief merit’ (Inst. 10.1.64). While he still did not prefer him to Pindar, Aristides too found it appropriate to evoke Simonides in one of his funeral orations (Or. 31.2: ‘What Simonides will mourn this (θρηνήσει), what Pindar?’).Footnote 144
Perhaps in contrast with his pathetic voice, Simonides was also the lyric poet who was most frequently linked with issues of patronage and compensation in exchange for praising wealthy and powerful individuals. In particular, his characterisation as a hired, and often greedy, poet, going back to comedy (cf. Ar. Pax 699) and later picked up by the Alexandrians (Callim. fr. 222 Pfeiffer; Theoc. Id. 16), was made popular among imperial orators by an anecdote that combined the poet’s perceived greed with piety: after being refused full payment for a song in which the Dioscuri had a prominent place (according to the paying patron, more prominent than himself), Simonides was saved by the gods from the collapse of his patron’s hall and was later able to recognise the dead by remembering their places at the banquet (cf. Cic. De or. 2.86; Quint. Inst. 11.2.11; Fav. De ex. 8.2; Lib. Or. 5.53).Footnote 145 Even though it was probably its link to Simonides’ invention of mnemotechnics that attracted the attention of sophists and rhetoricians, the story illustrates Simonides’ importance for the reception of the ancient lyric economy and provided Aristides with the opportunity to appropriate and reshape the lyric discourse of patronage (§3.4.2).
Simonides may have been memorable for his laments and occasional divine patrons, but in the communis opinio of imperial literary society, neither Simonides’ nor any other lyric voice could compete with Pindar’s, the ‘Pierian trumpet’ (AP 7.34.1). The same ‘grand style’ that made Pindar less suitable for threnodic poetry was part and parcel of his unmatched primacy within the lyric canon. In Roman-era criticism, ‘magnificence’ (μεγαλοπρέπεια, magnificentia), ‘solemnity’ (σεμνότης) and inspiration (spiritus) are all tags applied to Pindaric style, a prime example of ‘austere harmonia’ according to Dionysius (Dion. Hal. Comp. 22.42, 57–195; Dem. 39.42; De imit. fr. 31.2.5; cf. Quint. Inst. 10.1.61–2). ‘Bold, electric Pindar’ (Elizabeth Barrett Browning, ‘A vision of poets’) incarnates the instinctive genius of nature unleashed: his poetry ‘boils and surges’ like a thunderous river ‘rushing down the mountainside swollen with rains’ (Hor. Carm. 4.2.5–8, echoed by Quintilian’s eloquentiae flumine); or it sweeps as fire would, leaving no escape to the elegant smoothness of a Bacchylides ([Longinus], Subl. 33.5). Such descriptions matched the notion of Pindar’s link to the divine, another field where Pindar outdid Simonides and his felicitous relationship with the Dioscuri. Famous for his piety (εὐσέβεια, cf. De imit. fr. 31.2.5), Pindar’s poetic skills took on supernatural significance through the anecdote of his investiture by (unusually delicate) bees (Philostr. Imag. 2.12.1, 4), while ancient biographies collected stories on Pindar’s close relations with a variety of gods: Pan performed one of Pindar’s songs, Demeter (or Persephone) ‘commissioned’ him to compose a hymn, Apollo guaranteed his Panhellenic success in return for his Apolline poems (Vit. Ambr. p. 2.2–9 Drachmann; Vit. Thom. p. 5.4–11; Vit. metr. p. 9.9–10; cf. Philostr. Imag. 2.12.2; Paus. 9.23.3).
Pindar’s proximity to the gods was similarly popular in the epigrammatic tradition (see AP 16.305, from the Augustan period), with the result that, according to Barbantani, Pindar’s figure ended up being too idealised to clearly reflect historical details or the content of his work.Footnote 146 While the relationship between poet and poetry is less explicit and transparent in literary portraits of Pindar than in those of other lyric authors, however, the crystallised figure of Pindar inherited by imperial authors in fact relates in complex ways to the themes and persona found in his poems. While recalling the bees anecdote, for example, the description of Pindar’s song as fashioned ‘in Cadmus’ bridal chambers’ by a swarm coming directly from the Muses in AP 7.34 alludes to the opening poem of the whole Pindaric collection (Hymn 1); whereas the notion of Pindar as ‘congenial to strangers and dear to his countrymen’ (AP 7.35) epitomises his involvement with foreign and Theban patrons.
What is more, while Barbantani is right in stressing that the ancient portrait does not include Pindar’s epinician production explicitly, the insistence on Pindar’s excellence in epigrams as well as in other sources is ultimately predicated upon Pindar’s own engagement with epinician values: in victory odes, the success of athletes and powerful patrons is directly proportional to the success of the poet celebrating them (cf. Ol. 1.115–16, addressed to Hiero of Syracuse: ‘May you walk on high for the time that is yours, and may I join victors whenever they win | and be foremost in wisdom among Hellenes everywhere’); so much so that Pindar’s sophisticated metapoetics presents the poet as an exceptional voice with divine endorsement.Footnote 147 It is not surprising, then, that the idea of superiority inherent in epinician poetry fed into Pindar’s characterisation. In combination with Pindar’s production of cultic songs and the resulting connection to the divine, the epinician notion of excellence produced a lyric persona with literary, political and supernatural ramifications – a voice which, as realised by Horace, was definitely more appropriate than Alcaeus’ poetics of strife for the lyric treatment of encomiastic and public themes.Footnote 148
As opposed to Sappho’s or Anacreon’s erotic poetry, and perhaps with the exception of Stesichorus’ re-elaboration of Homer, the lyric figures most frequently associated with choral performances were distinguished by their more official and public profiles. They embodied forms of song that merged the human and divine spheres on specific social and religious occasions, involving communities and groups, as in the case of Spartan dancing maidens or funeral songs, and dealing with the ruling elite as patrons. Now, the relationship between human and divine was similarly at the core of Aristides’ own public persona and underpinned his self-positioning in relation to Roman power, so much so that it is little surprise that a preference for choral voices, Pindar’s above all, characterises Aristides’ reception of the genre (Part II). The rhetorician Hermogenes attacked the ‘fake sophists’ (Id. p. 249 Rabe: τοῖς ὑποξύλοις τουτοισὶ σοφισταῖς) who, ruined by Pindar’s model, display a style too lofty and metaphorical. Far from deploying Pindaric poetry as a mere stylistic template, however, Aristides took on and refreshed the range of literary, socio-political and religious discourses mobilised by choral lyric.
1.4 Conclusions
As lyric-quoting writers, Aristides, Plutarch, Pausanias and, in part, Philostratus stand out precisely because the display of lyric in imperial literature was not as common as that of other archaic and classical genres. As I have reconstructed it here, the place of lyric in the literary society of Roman Greece is far from easy to pin down. In the shape of the canon of nine poets established by Hellenistic scholars, lyric was copied, read, commented on and interpreted throughout the imperial period; the lyric canon, that is, was still a part of the literary canon. Yet a closer look at the evidence available shows that lyric was the province of advanced readership and scholarship, which tackled, among others, the linguistic features that made the genre appear so complex and sound so distant from the Attic tradition. Lyric was different from ‘small library’ genres like epic, drama and Attic oratory: textual engagement with these genres was required from and defined elite Greeks; lyric, on the other hand, marked a more specialised and therefore refined, super-elite paideia. This is corroborated by the presence of lyric texts in imperial literature, especially, but not exclusively, in rhetorical and sophistic works. The epideictic practices of imperial orators were ideally related to lyric predecessors, but by looking at progymnasmata, declamations and the works of Dio or Lucian we would hardly imagine the majority of sophists bent over lyric volumes and with lyric references at their fingertips.
Besides affecting our understanding of imperial Greek culture as a whole, this conclusion has weighty consequences for the interpretation of Aristides’ lyric reception. With over eighty identifiable references, the volume of lyric poetry re‑used across the Orations singles Aristides out from the majority of imperial sophists and writers.Footnote 149 This number is even more impressive if compared to the volume of quotations from core authors like Euripides (twenty-eight) and Menander (six).Footnote 150 Even before we start to tease out the meaning of the individual lyric models deployed and appropriated by Aristides, his evident preference for lyric points to the construction of a refined literary identity distinctive from, and more consciously sophisticated than, that of other pepaideumenoi. Exposure to lyric authors through Alexander’s teaching is not enough to justify Aristides’ extensive appropriation of lyric texts; this suggests instead a carefully thought-out strategy of self-positioning, underpinned by the super-elite status of lyric in imperial paideia. In the performance of literary culture so central to imperial Greek society, displaying familiarity with lyric poetry was a statement of positionality laden with intellectual and social implications. As in Hor. Carm. 1.1.30–4 (‘the cool grove and the light-footed bands of Nymphs and Satyrs set me apart from the crowd, provided Euterpe does not cease to pipe and Polyhymnia does not refuse to tune the Lesbian lyre’), the choice of lyric sets Aristides ‘apart from the crowd’ of imperial pepaideumenoi – a strategy that aligns with his self-centred, almost egomaniacal literary persona.
The super-elite nature of lyric readership might be thought to have thwarted, rather than enhanced, Aristides’ self-positioning as a refined pepaideumenos. For this strategy to work, at least some among his audience must recognise the lyric references he appropriated. But who could do so, if lyric poetry was by no means part of the core canon of imperial elite education? As we shall see when discussing specific Orations, Aristides often makes his lyric references explicit by either naming the poets in question or at least hinting at their identity through details traditionally associated with their make-up as literary icons. On other occasions, unnamed quotations and allusions would have required from listeners and/or readers a familiarity with precise texts similar to that displayed by Aristides, so that the sophistication of Aristides’ lyric reception could be fully appreciated only by equally sophisticated readers. Recognising lyric references would enhance the audience’s sense of sharing the same refined paideia as the speaker/author, while adding depth and nuance to the text. And taking into account (the possibility or probability of) recognition is certainly necessary to assess and understand how Aristides constructed the lyric tradition and the Greek literary and cultural traditions more broadly. Finally, even when potentially going unrecognised by (most of) the audience and readers, the lyric elements found in Aristides’ diverse works played a foundational role in his arguments and self-fashioning.
While in general lyric authors stood for more refined paideia, however, not all lyric subgenres and poets activated the same thematic, ethical and aesthetic associations. Imperial readers and writers who did engage with lyric had a clear sense of how different poetic figures mapped onto distinctive voices, resulting in a traditional, crystallised system of personas; from the erotic voices of Sappho and Anacreon – themselves well distinguished from one another – to the politically laden poetry of Alcaeus or the politically implicated praise of Pindar and Simonides. Studying with Alexander, Aristides himself probably had the opportunity to read at least some of the diverse poets taught by his teacher. In fact, such variety of lyric voices, also reflected in the quotation patterns of other imperial authors, contrasts starkly with the fewer lyric models eventually adopted, and adapted, by Aristides. Comparing Alexander’s broad lyric syllabus with Aristides’ selection of lyric voices will throw light on the aims shaping Aristides’ own use and construction of lyric (Part II). But before we can do so, we need to assess the presence of lyric in imperial culture beyond textual circulation and literary education.
