SECTION 2
CHAPTER 7
C. Michael Sampson
Introduction
The study of Greek lyric poetry has always been limited by the state of the extant corpus. Not only does it consist primarily of fragments, but apart from four books of Pindaric epinician and the elegies attributed to Theognis, it was until recently also transmitted indirectly, i.e., via quotations in sources such as Athenaeus, Stobaeus, or the scholia to other works (Phillips, this volume). But with the dawn of papyrology, which emerged as a discipline in its own right toward the end of the nineteenth century, a new era was ushered in. Although only a small number of the upwards of a million papyrus fragments recovered from the sands of Egypt preserve lyric poetry, and although many of those fragments are rather scanty, they nonetheless provide ancient witnesses to textual traditions that do not otherwise survive, frequently shedding light on the work of Alexandrian scholars. More often than not, they challenge received wisdom and long-held dogma: thanks to papyrology, the poetic output of Timotheus, Bacchylides, Alcman, Sappho, Stesichorus, Simonides, and Archilochus (to name a few) is understood more richly today than at any point since antiquity.
The corpus of Greek lyric remains meager and frustratingly incomplete, but papyri have added considerably to it, and because of the staggering amount of unpublished material, further additions may await discovery—especially in the largest institutional collections. Relatively few students and scholars of lyric will ever have to edit a papyrus, but even making use of these texts, fragmentary and riddled with specialized symbols, can be intimidating. This chapter therefore endeavors to provide a wide-ranging and demystifying introduction.
Using Papyrological Editions
On the printed page, papyrological editions look unusual: literary papyrology widely prefers the lunate sigma (ϲ, in place of σ/ς), and, in addition to the all-too-frequent gaps in the text that complicate its layout, a variety of symbols—including several varieties of brackets—can appear. Abbreviations and other conventions add to the potential for confusion.
Editorial Symbols
The standard editorial sigla, which are collectively known as the “Leiden conventions,” reflect a twofold imperative: to represent faithfully what appears on a papyrus and to indicate any interventions by the editor. For literary papyri, the two most common symbols are the sublinear dot and square brackets, but they are by no means the only ones:
Because these symbols are not easily transferred to a translation, the study of Greek lyric is by necessity demanding, specialized work.
For the purposes of illustration, we might consider the first column of P.Fouad inv. 239 (Figure 7.1) both as first published by Lobel and Page (1952) and subsequently reedited by Eva-Maria Voigt. The former tentatively attributed the poem to Alcaeus, but since Treu (1976: 161–164) the case for Sappho has been endorsed by many editors, including most recently by Battezzato 2018. The appearance of an asterisk on the papyrus, however, renders the case for Sappho problematic (de Kreij 2022).
Figure 7.1 P.Fouad inv. 239: lyrics by Sappho or Alcaeus. An asterisk is visible in the left margin of the second column, opposite line 2. (© Institut français d&38217;archéologie orientale du Caire.)
|
Lobel-Page 1952 |
Sappho fr. 44a Voigt |
|
|
]ϲανορεϲ̣ ̣ ̣[ ]μ̣αιτονετικτεκόω ̣[ ]ονιδαιμεγαλωνύμ ̣[ ]μεγανορκοναπωμοϲε ]λαν · ά⟦ε⟧ϊπαρθενοϲεϲϲομαι ] ̣ωνορέωνκορύφα̣ι̣ϲέπι ]δ̣ενευϲονέμανχαριν· ] ̣εθέωνμακαρωνπατηρ· ]ολοναγροτέρανθέο̣ι ] ̣ϲινεπωνύμιονμεγα· ]εροϲουδάμαπίλναται· ] ̣[ ̣] . . . .α̣φόβε[ ̣ ̣] ́ ̣ω· |
5 10 |
]ϲανορεϲ̣ ̣ ̣[ [Φοίβωι χρυσοκό]μ̣αι τὸν ἔτικτε Κόω ̣[ [μίγειϲ(α) Κρ]ονίδαι μεγαλωνύμω̣⟨ι⟩ [Ἄρτεμιϲ δὲ θέων] μέγαν ὄρκον ἀπώμοϲε κεφά]λαν· ἄϊ πάρθενοϲ ἔϲϲομαι ] ̣ων ὀρέων κορύφα̣ι̣ϲ’ ἔπι ]δ̣ε νεῦϲον ἔμαν χάριν· ένευ]ϲ̣ε θέων μακάρων πάτηρ· ἐλαφάβ]ολον ἀγροτέραν θέο̣ι ].ϲιν ἐπωνύμιον μέγα· ]εροϲ οὐδάμα πίλναται· ] ̣[ ̣] . . . .α̣φόβε[ ̣ ̣] ́ ̣ω · |
[ … golden-haired Phoebus,] whom [the daughter of] Koios bore, [mingling] with Kronos’ mighty-named son. [But Artemis] swore a great oath … “I will always be a maiden … on the peaks of mountains … grant my favor …” The father of the blessed gods agreed … the gods [call her deer-]shooter, huntress … a great title … Eros(?) never approaches her.
The relative paucity of dots indicates that the extant text is fairly secure, but it is by no means complete: the square brackets at left indicate that only the right side of the column survives. Lobel-Page’s text represents what the ancient scribe wrote: occasional accents on the papyrus mark the Aeolic dialect’s characteristic barytonēsis (= recessive accent); punctuation in the form of a raised dot appears in lines 5, 7, 8, 10, and 11. The scribe also deleted an epsilon in line five, which changes ἀεί into Aeolic ἄϊ. Voigt’s articulated edition, by contrast, includes the interventions of modern editors, namely, the third line’s addition of iota adscript (to make the dative case explicit) as well as several textual supplements in square brackets to the left of lines 2–4 and 8–9. Those supplements are not chosen at random (see further, on Dialect and Meter, below), but they must nonetheless be understood as additions to the extant text (on which, see further, The Readings, below).
Common Abbreviations
The abbreviations commonly used in papyrology can be subdivided into two groups: those for the many scholarly editions, series, and corpora that are frequently cited on the one hand, and those pertaining to digital resources on the other. There is some overlap; the same text can be cited in several different (and interchangeable) ways, depending on the particular resource or system that is employed.
Printed Editions and Corpora
The most common form of abbreviation in the papyrological world (and the one employed by this chapter) refers to a text’s principal publication, frequently within a larger series. P.Oxy. 32.2624, for example, refers to the 2624th papyrus in the Oxyrhynchus Papyri series, published in the thirty-second volume (N.B.: volume numbers can be indicated by Roman numerals). The Oxyrhynchus Papyri enumerates its papyri consecutively, in their order of publication, but some series reset the counter with each new volume.
Each series or individual volume of texts has its own designated abbreviation, which typically refers to an ancient site (e.g., P.Oxy. = Oxyrhynchus), a modern institution or collection (e.g., P.Ryl. = John Rylands Library, Manchester), an individual or archive either ancient or modern (e.g., P.Dryton, P.Turner), or some other unique identifier (e.g., P.Nekr.). A checklist of documentary editions with robust bibliographical information is now maintained online (http://papyri.info/docs/checklist), though exclusively literary editions are not included. Like the journal abbreviations that have been standardized by l’Année philologique, checklist abbreviations vastly simplify the process of citation and are therefore ubiquitous in papyrological scholarship.
But there are other ways of identifying a text, too. For one thing, every papyrus has a collection-specific inventory number under which it has been archived: when one is published outside of a series, therefore, it will continue to be identified by its inventory number (i.e., P.Fouad inv. 239, Figure 7.1) instead of a number assigned by the series. In some cases, a text published in this way can nonetheless be republished within a series (i.e., P.Köln inv. 21351 + 21376 = P.Köln 11.429). Additionally, when the works of a particular poet are collected in a standard edition, they are renumbered by their editor (the same is true when groups of poets are edited together). Not only can such systems of numeration be complicated by the addition, over time, of new discoveries to the corpus, but competing volumes or editions may assign different numbers to the same fragment! This companion has elected to use the enumeration of Campbell’s Loeb editions, but much has changed since they were first published: for Sappho and Alcaeus, scholars frequently cite the editions of either Lobel-Page (= LP) or Voigt (= V.); for elegy, the editions of West or Gentili-Prato compete. Sometimes it is not the editor but the title of an edition which provides an identifying acronym, e.g., Page’s Poetae Melici Graeci (= PMG) or Supplementum Lyricis Graecis (= SLG), Davies’ Poetarum Melicorum Graecorum Fragmenta (= PMGF), and Lloyd-Jones’ Supplementum Hellenisticum (= SH). For these, and other common abbreviations, see the Abbreviations and Standard Editions in this Companion.
Digital Resources
Papyrology has long been in the vanguard of digital humanities: the Duke Databank of Documentary Papyri (= DDbDP) was established in 1982 (i.e., before the creation of the World Wide Web!); the Heidelberger Gesamtverzeichnis der griechischen Papyrusurkunden (= HGV) was launched in 1988, and in the mid-1990s, the Advanced Papyrological Information System (= APIS) was conceived and developed out of the Duke Papyrus Archive. Papyrologists now collectively curate papyri.info (http://papyri.info), which aggregates data and metadata from these and other resources. An equivalent portal for literary papyri (dubbed DCLP) was launched in December 2017 and has since been incorporated into papyri.info. But there are other important digital corpora of data and metadata about literary texts, which warrant mention in this chapter inasmuch as papyri can also be identified according to their unique systems of enumeration:
The various systems of enumeration both digital and archival can be used in tandem: for the purpose of illustrating the overlap, consider the cases of a codex from Berlin containing Sappho and the famous Lille Stesichorus.
|
Berlin Sappho |
Lille Stesichorus |
|
|
Inventory Number |
P.Berol. inv. 9722 |
P.Lille inv. 111c + 73 + 76a–c |
|
editio princeps |
BKT 5.2, no. XIII 2 (pp. 10–18) |
• CRIPEL 4 (1976): 287–303 • cf. ZPE 26 (1977): 1–6 • cf. ZPE 26 (1977): 7–36 |
|
Other Critical Editions |
• Frr. 92–97 V. • Frr. 92–97 LP |
• PMGF 222b • Fr. 97 F |
|
TM |
62713 |
62787 |
|
M.-P.3 |
1451 |
1486.1 |
|
LDAB |
3901 |
3975 |
In the former case, six fragmentary poems of Sappho from a single codex have as many possible identifying numbers (LDAB 3901 = M.-P.3 1451 = TM 62713, etc.); so too in the latter case, where several separately inventoried fragments from a single poem are reunited under an individual number (PMGF 222b = fr. 97 F = LDAB 3975 = M.-P.3 1486.1, etc.). Encountering a TM or M.-P.3 number in the course of one’s research should not startle; knowing the different names for something is akin to mastering it—πολυωνυμία is not only a property of gods!
The Editorial Process
The editor’s most important tasks—namely, the production of an accurate transcription (i.e., “good readings”), as well as attributing the work to a particular poet—are the basis for all further research. But fulfilling these responsibilities depends on further wide-ranging, synthesizing analyses, including of dialect (de Kreij, this volume), of meter (Battezzato 2009; D’Angour, this volume), of palaeography, of diction, and of physical layout. The last can be particularly challenging: a group of small fragments is like a jigsaw puzzle, but with papyri there are inevitably missing or otherwise ill-fitting pieces! (Figure 7.2) In general, the larger the scrap(s) and the more extensive the amount of text available for analysis, the more richly it can be understood.
Figure 7.2 P.Oxy. 25.2430; fragments of Simonides (= fr. 519). (Courtesy of The Egypt Exploration Society and the University of Oxford Imaging Papyri Project.)
The order in which various aspects of the process are presented below in no way reflects a standardized technique: every papyrus is unique, and yields insights on its own terms as its readings are confirmed and its text stabilized. A papyrologist invariably wears many hats in the course of completing the job (Youtie 1963; Turner 1968: 54–73), but all responsible editions will include some combination of the following.
Origins/Provenance
Papyri are archaeological objects, and are therefore best understood in their archaeological and historical contexts, to the extent that these can be reconstructed: from archaeological find spot, to use (and reuse) in antiquity and modern ownership or collecting history. But due especially to philology’s tendency to privilege the text above all else, papyrology—and especially literary papyrology—has long been weak on archaeology. Early excavations such as those of the famed Oxford “Dioscuri” (Bernard Grenfell and Arthur Hunt), for example, prioritized the extraction and accumulation of papyri. While attempts were made to organize texts that were uncovered together where possible, the “torrent” of papyri Grenfell and Hunt were uncovering on a daily basis at al-Bahnasa precluded any sort of detailed inventory or accounting of find-spots, let alone a more scientific recording of stratigraphy. (Figure 7.3) We are not much better served on this front by the reality of a lively antiquities market: purchased papyri were accompanied by the dealer’s word, which could (but need not) be well informed. There are exceptions (see, e.g., Claytor and Verhoogt 2018) but the principal methods of acquisition during the heyday of discovery regularly make the analysis of a papyrus in its archaeological context impossible. Documentation is frequently frustrating or nonexistent.
Figure 7.3 Excavating for papyri at Oxyrhynchus (al-Bahnasa). (GR.NEG.048, courtesy of the Egypt Exploration Society.)
Such historical limitations notwithstanding, every editio princeps should include a frank and thorough accounting of a papyrus’ provenance, an all-encompassing term under whose umbrella the object’s history (from antiquity to the present) is meant: its ancient context, the circumstances of its discovery, and its modern ownership or collecting history. Where documentation exists in the form of receipt(s) for sale, archaeological notebook(s), or institutional acquisition/inventory report(s), it should also be reported. Transparency on this front is essential to responsible scholarship; in addition to the possibility of forgery, Egyptian law has since 1983 prohibited the domestic antiquities trade and has established definitively that all archaeological sites and objects are the property of the state. Any papyrus that cannot be proven to have been exported before that date therefore falls into a legal—not to mention ethical—grey area (see further discussion of ethics, below).
Physical Description and Layout
Before an editor begins to transcribe a text, all aspects of the physical papyrus are measured and described:
Among the insights potentially resulting from physical examination is the type of ancient book from which the fragment derives (i.e., bookroll or codex). But more significant findings are possible, as well. New analyses of the columns in P.Oxy. 10.1232, for example, have clarified the organization of Sappho fr. 44 (Sampson 2016; de Kreij 2020). Regularly recurring folds or wormholes, similarly, can permit the modeling of a bookroll, facilitating the placement of fragments (or columns) relative to one another. And where the fibers of the papyrus align, disparate fragments can be rejoined with some confidence.
Palaeography
A familiarity with ancient handwriting styles (and their evolution) is essential for the papyrologist: due to abrasion, dirt, holes, or other damage, letters can be quite unclear and educated guesswork is therefore inevitable. In order to produce good readings, it is often necessary to produce an alphabet of letter shapes drawn by a particular scribe, the recourse to which helps to narrow the possibilities for fragmentary letters.
Palaeography is also important for assigning a date to the text. Although this is true of both documentary and literary texts, documents can helpfully include a dating formula while literary texts lack such an internal indication (unless the papyrus has a document on its obverse side!). The differences between the two kinds of text are often obvious even to an untrained eye; unlike the cursive scripts of rapid documentary hands, whose analysis often requires specialized training, those of professionally produced ancient books can be remarkably elegant (Figures 7.4 and 7.5). Their features have been analyzed and categorized in several scholarly studies: in some cases, the handiwork of an individual scribe can even be identified (e.g., Johnson 2004: 16–37). But not all literary texts are the product of professional workshops: the text of PSI 13.1300, the so-called Sappho ostracon (= fr. 2), for example, is very corrupt, and the hand that transcribed it, while practiced, is less regular than that of a formal bookroll (Figure 7.6). This poem, coincidentally, continues to challenge its editors and interpreters.
Figure 7.4 P.Tebt. 2.684 (= Pindar, Ol. 9 and 10). The clear script is an example of biblical majuscule. (Courtesy of the Center for the Tebtunis Papyri, University of California, Berkeley.)
Figure 7.5 P.Tebt. 2.620 descr., a receipt for poll-tax. The rapid script is an example of a documentary cursive. (Courtesy of the Center for the Tebtunis Papyri, University of California, Berkeley.)
Figure 7.6 PSI 13.1300 (= Sappho fr. 2) (Florence, Biblioteca Medicea Laurenziana, inv. 22008. Su concessione del Ministero per i Beni e le Attività Culturali.)
The Readings
Literary papyri are often published in parallel, with a diplomatic edition (i.e., the raw, visible text) alongside an edited text (i.e., articulated, normalized, and reconstructed)—as with the two editions of P.Fouad inv. 239, Figure 7.1. Each is ideally accompanied by a critical apparatus: the former’s describes the traces of ink and the possibilities for individual letters where there is doubt, the latter’s the emendations or conjectures of other scholars, conventionally by way of a compressed, abbreviated Latin (Figure 7.7). Each apparatus is important: understanding what the editor saw on the papyrus is the necessary precursor to a reanalysis of the readings, and any second-guessing of the editor’s judgment regarding the articulated text is similarly facilitated by a catalog of alternatives. A good editor is therefore honest and humble in addition to being meticulous and learned.
Figure 7.7 Critical apparatus of the “new” Simonides elegy on Plataea (= P.Oxy. 22.2327 + 59.3965), in abbreviated Latin as per convention. (The New Simonides: Contexts of Praise and Desire, edited by Deborah Boedeker and David Sider, 2001, ca. 600w from p. 19. By permission of Oxford University Press, USA.)
The movement from individual letters to a fully articulated text is where editors shine. Ancient scribes, for one thing, wrote in scriptio continua (i.e., without spaces between the words). Holes or other gaps in legibility, moreover, obstruct the analysis of letters into separate words. But even where the letters are clear enough, scribal error, peculiarities of dialect, or (rarely) new additions to the lexicon can perplex, invite emendation, or otherwise hinder the production of recognizable, articulated Greek. Editorial intervention, in other words, is an inevitable part of the job, and everyone who makes use of texts preserved on papyrus must therefore appreciate the countless decisions that make up an edition’s artifice. But although they deserve our profound gratitude, editors’ judgment is not sacrosanct, and every intervention is subject to interrogation (and, potentially, revision). The study of papyri—and especially literary papyri—is perpetually work-in-progress.
Some best practices are commonly recognized: “Youtie’s law,” for example, advises against emendation in the vicinity of a lacuna—iuxta lacunam ne mutaveris. But editors occasionally disagree in their approaches to a text, the most significant case in point of which is the treatment of textual supplements. Where a supplement is obvious or where grammar demands a particular form, most casually supply it in the service of continuous text; when only a few letters are missing, such interventions are relatively benign. But with more substantial lacunae, the impulse to supplement the text—a papyrological horror vacui!—requires verse composition on a larger scale. For poetry (and lyric, in particular), such exercises are fraught with hazard: the corpus of some poets is so threadbare, for example, that both style and output are imperfectly understood. But even in relatively fulsome cases, caution is still warranted: the syntax and poetic imagery of Pindar can challenge even when the text is relatively secure (Brown, this volume). And even for less idiosyncratic poets, reconstruction and interpretation run the risk of reinforcing one another in an editorial feedback loop: this is how I understand the poem leads to this is what I think is missing, which can become this is what the poet must have written before concluding my intuition/analysis of the poem works! No matter how ingenious the scholar and how sensible the suggestion appears, it must still be remembered that the reasoning can be circular and that we are puzzling over something that is not there. Users of papyrological texts must therefore be sensitive to the extent of an editor’s interventions and (un)willingness to presume the poet’s thinking, lest they construct a larger interpretation upon conjectured or uncertain text. But that caveat notwithstanding, reasoned arguments for and against various supplements, as well as informed interpretations of individual fragments, are at the heart of scholarship on lyric: the goal, as always, is to maximize our understanding of poem, poet, corpus, and context. And the fragmentary nature of most texts means that some amount of speculation is unavoidable.
Where a publisher grants the space, alternative versions can be printed in tandem, illustrating the preserved text and acknowledging the possibility of diverging reconstructions: compare the two versions (Sider’s and West’s, respectively) of a passage from P.Oxy. 22.2327 + 59.3965 (the “new” Simonides elegy on Plataea). Although Sider endorses many of West’s supplements, his text makes a point of keeping restoration to a minimum (see, e.g., lines 24–26) (Figures 7.8 and 7.9).
Figure 7.8 Excerpt from the “new” Simonides elegy on Plataea (= P.Oxy. 22.2327 + 59.3965), as conservatively restored by David Sider. (The New Simonides: Contexts of Praise and Desire, edited by Deborah Boedeker and David Sider, 2001, 145w from p. 18. By permission of Oxford University Press, USA.)
Figure 7.9 Excerpt from the “new” Simonides elegy on Plataea (= P.Oxy. 22.2327 + 59.3965), as more comprehensively restored by Martin West. (The New Simonides: Contexts of Praise and Desire, edited by Deborah Boedeker and David Sider, 2001, 205w from p. 28. By permission of Oxford University Press, USA.)
In P.Köln 11.429 (the so-called “Tithonus” or “Old Age” song of Sappho), similarly, the opening doublet has been variously restored (in bold):
[φέρω τάδε Μοίϲαν ἰ]ο̣κ[ό]λ̣πων κάλα δῶρα, παῖδεϲ,
[λάβοιϲα πάλιν τὰ]ν̣ φιλάοιδον λιγύραν χελύνναν·
[I bear these] lovely gifts of the fragrant-bosomed Muses,
girls, [having taken up again the] clear melodious lyre.
Gronewald and Daniel (2004a: 7, printed exempli gratia)
[ὔμμεϲ πεδὰ Μοίϲαν ἰ]ο̣κ[ό]λ̣πων κάλα δῶρα, παῖδεϲ,
[ϲπουδάϲδετε καὶ τὰ]ν̣ φιλάοιδον λιγύραν χελύνναν·
[You for] the fragrant-bosomed Muses’ lovely gifts
[be zealous,] girls, [and the] clear melodious lyre.
West (2005: 5); cf. Janko (2017b)
[ὔμμιν φίλα Μοίϲαν ἰ]ο̣κ[ό]λ̣πων κάλα δῶρα, παῖδεϲ,
[πρέπει δὲ λάβην τὰ]ν̣ φιλάοιδον λιγύραν χελύνναν·
[The] lovely gifts of the fragrant-bosomed Muses [are dear to you],
girls, [and it is appropriate to take up the] clear melodious lyre.
Di Benedetto (2005: 18)
[νῦν (τ’) ἄδεα Μοίϲαν ἰ]ο̣κ[ό]λ̣πων κάλα δῶρα, παῖδεϲ,
[φίλημμί τε φώνα]ν̣ φιλάοιδον λιγύραν χελύνναν·
“[Now] the fragrant-bosomed Muses’ lovely gifts [are sweet], girls
[and I love] the song-loving [voice] of resounding lyres.”
Yatromanolakis (2008: 243), cf. Lidov (2009: 93–94).
While the supplement Μοίσαν is generally accepted, other crucial information (i.e., the verb!) is lacunose (Figure 7.10). The extant text only includes beautiful gifts, a tortoise-shell lyre, several adjectives, and what most scholars agree is a vocative addressing a chorus of young girls. Inasmuch as each reconstruction reflects a particular interpretation, the problem is, at a certain level, intractable: whether the doublet originally contained an exhortation (as West would have it), or a values-statement by the poem’s first-person persona (as elsewhere in Sappho—so Gronewald/Daniel and Yatromanolakis) depends upon individual scholars’ perceptions and projections of Sappho and her work. For a poet whose corpus was said in antiquity to have totaled nine books (AP 7.17; see further Prauscello 2021), even reasonable inferences are based on only the small sample that survives. Such hazards are inherent to papyrological work and the study of lyric. The most recent edition by Neri-Cinti (2017) conservatively declines to supplement the Greek text.
Figure 7.10 P.Köln inv. 21351 fr.a. The last four lines preserve the beginning of the “Tithonus Song” or “Old Age Song” (fr. 58). (© Institut für Altertumskunde an der Universität zu Köln. Used under Creative Commons Attribution 4.0 International License.)
[× ‒ ⏑ ⏑ ‒ ‒ ἰ]ο̣κ[ό]λ̣πων κάλα δῶρα, παῖδ⌊εϲ⌋,
[× ‒ ⏑ ⏑ ‒ ‒] ̣ φιλάοιδον λιγύραν ⌊χελύνναν⌋ ·
(For an up-to-date critical apparatus of this poem see Benelli 2017: 2.268, and for discussion of the interpretive possibilities 2.278–81).
Marginal Notes, Sigla, and Corrections
In addition to the text itself, lyric papyri occasionally contain additional material of practical and exegetical value: critical symbols such as the paragraphus or the diplē; punctuation; diacritical signs (i.e., breathings, accents); marginalia (including corrections or annotations); and metrical, stichometrical, or colometrical marks (Figure 7.11). Some are the work of the original scribe, while others were added later by one or more users. Accents assist in reading, scansion, as well as the analysis of dialect (as in P.Fouad inv. 239, above; see further, Dialect and Meter, below). Critical symbols, by contrast, tend to mark a division of one sort or another: in lyric papyri one finds especially the asterisk (※), paragraphus ( ⸏ ) coronis (⸎, essentially a paragraphus with decorative curlicue), and diplē obelismenē (˒–, or “forked paragraphus”) (Figure 7.12). The coronis and asterisk most commonly mark the end of a poem, but can do so in conjunction both with one another and with other symbols, which are predominately used for metrical divisions (i.e., distinguishing stanzas or triads). Other sigla can indicate textual variants, omissions, or marginal notes, though their purpose is sometimes opaque.
Figure 7.11 P.Köln 2.59 (= Alcaeus fr. 298), with accents as well as long and short quantities marked. (© Institut für Altertumskunde an der Universität zu Köln. Used under Creative Commons Attribution 4.0 International License.)
Figure 7.12 P.Oxy. 26.2441 (= Pindar, Paeans 14-15 S-M), with accents, marginal comments, coronis, and asterisk. (Courtesy of The Egypt Exploration Society and the University of Oxford Imaging Papyri Project.)
Sometimes practical aids have exegetical value, too. Thanks to the stichometrical indicator ΧΗΗΗΔΔ in the colophon of one fragment of P.Oxy. 10.1231 (Figure 7.13), we know that the first book of the Alexandrian edition of Sappho contained 1,320 verses (= 330 Sapphic stanzas); the numeral Ν (= 13) in the margin of P.Oxy. 33.2617 similarly marks the 1,300th line of Stesichorus’ Geryoneis, a scale of narrative that confirms the ancient attribution of multi-volume works to his oeuvre (see Finglass, this volume). Other marginalia are significant for providing a glimpse of ancient scholarship (Figure 7.14). Such research was originally produced as independent hypomnemata (= commentaries), but was subsequently incorporated into texts by users in the form of notes. Sometimes, as was previously mentioned, textual variants or omissions are marked, but twenty-first-century students of lyric will appreciate that there were aspects of the poetry which required commentary and exegesis in antiquity, as well—especially technical matters (e.g., dialect, meter) and interpretive ones (e.g., historical context). All such ancient scholarship preserved on papyrus is now in the process of being collected in the multi-volume Commentaria et lexica Graeca in papyris reperta (= CLGP). To date, the scholarship on Alcaeus, Alcman, Anacreon, and Bacchylides has been published; that on Pindar, Simonides, Stesichorus, and lyric adespota is expected in future volumes.
Figure 7.13 P.Oxy. 10.1231, fr. 56 (= Sappho fr. 30), now in the Bodleian Library MS. Gr. Class. c. 76. This fragment preserves the final column of the first book of the Alexandrian edition of Sappho, with coronis and stichometrical colophon. (Courtesy of The Bodleian Libraries, The University of Oxford.)
Dialect and Meter
Lyric poetry admits a variety of dialectal forms: depending on the poet, a poem’s genre, or its meter, Doric, Aeolic, or Ionic features might predominate over one another. The emergence of a lyric koinē, furthermore, means that the relationship between a poet’s vernacular and Kunstsprache is neither obvious nor straightforward (see de Kreij, this volume). For the papyrologist’s purposes, dialect is a particularly important basis for attribution: in the case of the first fragment of P.Oxy. 35.2735, for example, dialect and Doric accentuation alone are sufficient to whittle the authorial possibilities to two—Stesichorus or Ibycus (Finglass 2017c: 21) (Figure 7.15).
Figure 7.14 P.Oxy. 21.2295, frr. 18 and 28 (= Alcaeus frr. 157 and 167), with marginalia, including metrical observations (fr. 18.3) and a variant reading (fr. 18.8) from the grammarian Apion. (Courtesy of The Egypt Exploration Society and the University of Oxford Imaging Papyri Project.)
Figure 7.15 P.Oxy. 35.2735, fr. 1 (= Ibycus fr. 282A), with Doric accentuation. (Courtesy of The Egypt Exploration Society and the University of Oxford Imaging Papyri Project.)
Conclusions regarding dialect are often buttressed or refined by analyses of diction and of meter, but certainty on all fronts is uncommon. The combination of Ionic dialect and meter in P.Oxy. 22.2321 (Figure 7.16), for example, means that ascription to Anacreon “will hardly be disputed” (in the words of its editor, Edgar Lobel), even though there is no exact parallel in his extant oeuvre for its particular metrical arrangement—three-line stanzas in varieties of anaclastic ionics. The previously mentioned supplements to P.Fouad inv. 239, similarly, build upon that fragment’s Aeolic dialect by way of a metrical hypothesis, namely, that the poem’s meter was a glyconic with dactylic expansion (gl2d). Since the second book of the Alexandrian edition of Sappho consisted of poems in this meter, and since the line-beginnings in the papyrus’ second column are also consistent with it, the hypothesis was especially attractive until the appearance of an asterisk rendered the case for Sappho unlikely (de Kreij 2022). And lastly, a recent analysis of P.Oxy. 32.2624 suggests that at least some of its 56 fragments derive from a fifth-century work of choral lyric in dactylo-epitrite, more likely to have been composed by Simonides than Pindar (Ucciardello 2017).
Figure 7.16 P.Oxy. 22.2321, fr. 1 (= Anacreon fr. 346), with accents. (Courtesy of The Egypt Exploration Society and the University of Oxford Imaging Papyri Project.)
In truth, part of what makes the study of lyric papyri so exciting is that expectations are regularly defied: because so little survives, new data are almost inevitably surprising. The publication of the “new” Archilochus elegy on Telephus (= P.Oxy. 69.4708) in 2005 was revolutionary because of that poem’s mythological narrative, something not previously found in early elegy; that of the “new” Simonides elegy on Plataea (= P.Oxy. 22.2327 + 59.3965) in 1992, similarly, bore witness to the possibility of elegy on the subject of recent history. But these are relatively tame compared to cases such as P.Dryton 50 (see Morrison, this volume), an extensive lyrical monologue from the Hellenistic period (also known as the fragmentum Grenfellianum). Its frequently prosaic language, for one thing, is at odds with its metrical complexity: the lyrics are of a polymetric and non-strophic type characteristic of late Classical “New Music” (LeVen, this volume). They are also rife with dochmiacs, a colon (= metrical unit) whose extensive use had previously been thought peculiar to the lyrics of Classical tragedy and which appears for the last time in this fragment. It remains a work largely without parallel, though the presence of a loan dating to 174 BC on the papyrus’ obverse is consistent with the hypothesis that it is a Hellenistic composition.
As with that of dialect, the analysis of meter can be daunting, technical work (Battezzato 2009; D’Angour, this volume). For papyri, the matter is complicated not only by the fragmentary state of most texts but also by the fact that, prior to the second century BC, lyric poetry was transcribed as prose, i.e., without any indication of metrical division. The aforementioned Sappho ostracon (PSI 13.1300, III/II BC, Figure 7.6), P.Dryton 50 (circa 174 BC), and the Berlin fragment of Timotheus (P.Berol. inv. 9875, IV BC, Figure 7.17) are all examples of early texts of lyric that lack a meaningful colometry: such papyri place the onus of metrical analysis squarely on their editors. In the first case, the result is the familiar Sapphic stanza, but the latter two poems, by contrast, are vastly more complicated for their polymetry, with an array of cola to identify and distinguish. Luckily, such a layout is the exception rather than the rule: the division of lyric into cola is often ascribed to the scholar Aristophanes of Byzantium (circa 255–180 BC), and from the second century BC on, poetry divided colometrically into stanzas and triads is the norm (Barbantani 2009: 301).
Figure 7.17 P. Berol. inv. 9875 (= Timotheus fr. 791). (bpk Bildagentur / Aegyptisches Museum und Papyrussammlung, Staatliche Museen, Berlin, Germany / Photo: Sandra Steiß / Art Resource, NY.)
Papyrological Ethics
Some readers may be surprised to find a section devoted to ethics in a chapter like this, which aims principally to demystify the papyrology of Greek lyric. But some background in the history of both papyrology and papyrological research is no less important than an understanding of the technical aspects of the field. Indeed, it is precisely because ethical issues are not aired frequently enough that their discussion here is essential.
A historical example will help set the stage: in mid-November 1896, E.A. Wallis Budge, Keeper in the British Museum’s Department of Egyptian and Assyrian Antiquities, began to transact the purchase of a papyrus roll from an Egyptian dealer with whom he denied a previous relationship, but who was probably Ali Farag of Giza (Sayce 1923: 334). Budge was no specialist in Greek, but knew enough to recognize that the text was early and literary, that he needed to procure it for the Museum, and that it would be eagerly sought by other European collectors as well as by the Egyptian Antiquities Service, which would claim it for the Egyptian Museum. His account of the acquisition (1920 ii: 345–355) is fascinating: with pressure being applied by officials in the Antiquities Service as well as by the British Consul-General in Egypt, Budge was forced to purchase the papyrus personally before eluding the authorities via an elaborate ruse involving a crate of oranges and a midnight rendezvous with a P&O mail steamer anchored near Suez. The papyrus (= P.Lond.Lit. 46; British Library Papyrus 733) was published the following year: it turned out to contain 39 columns of poetry by Bacchylides—a truly significant find (Figure 7.18).
Figure 7.18 An excerpt from British Library Papyrus 733 (= Bacchylides). (Image by permission of the British Library, London, UK.)
The clandestine export of the papyrus (and Budge’s candor in describing it) is shocking today, but it is both typical of him (Ismail 2011: e.g., xvii) and representative of his era, generally. In the late nineteenth century, a complex market for antiquities in Egypt operated in parallel with both formal archaeological excavations and the informal looting of ancient sites. A variety of players—private individuals and State representatives, both Egyptian and foreign—were active. In addition to the professional salesman with physical storefront, anyone who might have occasion to sell an antiquity in his possession qualified as a dealer. No less engaged in the market were consular officials of foreign governments, foreigners working in Egypt (including professional archaeologists), and even the Egyptian Antiquities Service, the State body which oversaw archaeological work and which would eventually be charged with regulating the market. The distinction between a licit and illicit acquisition is often, regrettably, somewhat blurry in this period.
Further lessons can be gleaned from the tale of Budge and Bacchylides. Compared to the Bacchylides papyrus, concerning which little contextual information is known beyond the dealer’s assertion that he found it in a tomb at Meir, we might compare the Berlin Timotheus papyrus (= P.Berol. inv. 9875, Figure 7.17), which was excavated at the Greek necropolis at Abusir by Ludwig Borchardt in 1902. Archaeological evidence indicates that the Timotheus papyrus was deposited at the site before the Ptolemaic period, at which point the Greek cemetery was not in use. The papyrus, curiously, was not found in one of the wooden sarcophagi from the Greek cemetery, but on the ground to the north of a sarcophagus from the period of the Egyptian Middle Kingdom, with which it is unlikely to have had anything other than an accidental relationship. Although it is tempting to posit a straightforwardly funereal context for the find, in other words, the history of its deposit is potentially far more complex (Hordern 2002: 62–65; cf. van Minnen 1997). The contrasting tales of the Bacchylides purchase and the excavation of the Timotheus papyrus emphasize the benefits of a richer contextual understanding, but there is also an important similarity between the two stories which underscores a crucial detail about the period, namely, the important role of foreigners, foreign money, and foreign museums. This is part of a larger trend dating back to the Napoleonic era, in which the cultural heritage of Egypt has been exported, looted, or despoiled by occupying or colonial powers.
In the twenty-first century, the landscape has shifted somewhat. After relatively unsuccessful attempts to regulate its antiquities market via legislation in 1912 and 1951, Egyptian law has since 1983 effectively outlawed the domestic trade in antiquities and has established definitively that the State owns its archaeological heritage. Internationally, the UNESCO Convention on the Means of Prohibiting and Preventing the Illicit Import, Export and Transfer of Ownership of Cultural Property came into effect on April 24, 1972, though individual countries have only gradually accepted or ratified its terms (see further Mazza 2019: n. 13). Much like the legal environment, scholarly attitudes toward the acquisition, curation, and publication of ancient objects have hardened, for a combination of academic and ethical reasons. It is a maxim of archaeological science, for example, that an object is best understood in context: responsible archaeologists therefore condemn any activity that damages, destroys, or renders irretrievable the history of an object or its archaeological context. That rationale is echoed in the Society for Classical Studies’ Statement on Professional Ethics and the American Society of Papyrologists’ Resolution Condemning the Illicit Trade in Papyri. Such public documents are important for attempting to limit and qualify the scholarly imperative to acquire, preserve, study, and publish the material culture of antiquity. Indeed, there are good reasons for such restrictions: cultural heritage continues to be destroyed instead of professionally excavated, and the material culture of antiquity continues to be looted and sold on the antiquities market rather than properly studied in context.
Such concerns are especially germane to papyrology because of its historical weakness in archaeological matters. A brief example is again illustrative: although the cartonnage from which the famous Lille Stesichorus originates was excavated by Jouget and Lefebvre in 1901, for example, the location of the find remains unclear (Meillier: 1976: 339; Turner 1971: 124). There is more we would like to know about this text that, sadly, cannot now be ascertained: such gaps in paperwork or record-keeping are all too common when dealing even with legally acquired papyri. A further consideration raised by this text involves the excavated object itself: the physical reality of mummy cartonnage is that extracting papyrus from it long came at the cost of destroying the cartonnage. Papyrologists and archaeologists are unlikely to agree on which item is more important, but the dismantling of cartonnage is no longer especially common in responsible collections.
Regardless of evolving scholarly attitudes and a more stringent legal framework, there are significant challenges in the present moment. Not only does the market for antiquities (both licit and illicit) remain lively—particularly following the turmoil of the Arab Spring—but the boundary between the two remains imperfectly defined. Just because a papyrus is consigned and purchased through a major auction house, for example, does not mean that its legal provenance has been established with any certainty (Whitesell 2016), a lesson that art historians and archaeologists are all too familiar with. The case of the most recent Sappho discoveries indicates the lengths to which an owner might go in constructing a legitimate provenance for an object with an eye to lucrative resale (Sampson 2020). The anonymity granted to collectors by the market (and, occasionally, by scholars) further impedes transparency in research. And when an object’s origins are mysterious or unknown, all sorts of problematic questions are stirred: in addition to the specter of looting or illegal exportation, papyri can be forged—even ones written in ancient languages (Sabar 2016).
Between the institutional collections whose acquisition histories are relatively transparent and the black market, therefore, lies a considerable grey area to which the various ethics statements respond. Practically the only check the academic community can place on the free market in antiquities is to be diligent in the course of conducting research in the present, by demanding proof of an object’s provenance before authenticating or publishing it, and by reporting that provenance in full after verifying it to the best of one’s abilities. No matter how exciting a new text may be for the addition it potentially makes to the corpus of lyric, it is also inherently problematic, and therefore should be presented with a high bar to clear before receiving a scholarly audience. The academic argument concerning the preservation and dissemination of knowledge must be reconciled to an ethical one that reflects an unsavory reality: one’s participation in activities that destroy data or knowledge—wittingly or unwittingly, directly or indirectly—effectively encourages them. Profit is a powerful incentive for criminals, and when a scholar identifies, authenticates, or publishes an object that was acquired in contravention of the law or outside of a controlled archaeological excavation, both the value of that object and the incentive for the perpetrator(s) to continue are thereby increased. Such practices are antithetical to the scholarly imperative to recover, study, preserve, curate, and disseminate knowledge of antiquity.
Further Reading
Because lyric is but a small segment of the corpus of literary papyri, and literary papyri but a small segment of the corpus of papyri, many of the following suggestions consider aspects of papyrology more broadly. Turner 1968 remains a standard introduction, now richly supplemented by Bagnall 2009; on archaic poetry in particular see Haslam 1994. The history (and future) of the discipline are addressed by van Minnen 1993 and 2007; field reports from Oxyrhynchus are reproduced in Grenfell and Hunt 2007. On digital papyrology see Reggiani 2017 and 2018. Comprehensive studies of book hands of different periods are Turner 1971 as well as Cavallo and Maehler 1987 and 2008; on the bookrolls and scribes from Oxyrhynchus, specifically, see Johnson 2004. An illustrated database of palaeography, from papyri with internally established dates, can be searched online at www.pappal.info. Scholarship is a complex topic: McNamee 1992 and 2007 treat sigla and annotations, respectively; the corpus of ancient scholarship on individual authors is now in the process of being published in the CLGP; on accentuation see Probert 2003 (esp. §301–319, pp. 158–168). Ethics, the law, and the antiquities trade are subjects of Hagen and Ryholt 2016, Davoli 2008, Mazza 2015, 2018, and 2019. Fearn 2010 contextualizes the specific case of Bacchylides in terms of the imperialism and colonialism of the age. For an example of museum archaeology illuminating a group of papyri see Nongbri 2017. On criminal activity in the modern antiquities trade, see especially Watson and Todeschini 2006. For signatories to the UNESCO Convention (with dates), see http://www.unesco.org/eri/la/convention.asp?KO=13039&language=E.asp?KO=13039&language=E . The various statements regarding professional ethics should also be listed: the Archaeological Institute of America (https://www.archaeological.org/pdfs/AIA_Code_of_EthicsA5S.pdf), the Society for Classical Studies (https://classicalstudies.org/about/scs-statement-professional-ethics), the American Society of Papyrologists (http://www.papyrology.org/resolutions.html), and recommendations of the Association Internationale de Papyrologues’ Working Party on the Commerce in Papyri (https://www2.ulb.ac.be/assoc/aip/recomcommerce.html). On the history of the Lille Stesichorus see Jouget 1901 and Meillier 1976.
Note
* Andrew Bresch and Amber Leenders assisted in the research for this article.