Poems without Poets: Approaches to anonymous ancient poetry

The canon of classical Greek and Latin poetry is built around big names, with Homer and Virgil at the center, but many ancient poems survive without a firm ascription to a known author. This negative category, anonymity, ties together texts as different as, for instance, the orally derived Homeric Hymns and the learned interpolation that is the Helen episode in Aeneid 2, but they all have in common that they have been maltreated in various ways, consciously or through neglect, by generations of readers and scholars, ancient as well as modern. These accumulated layers of obliteration, which can manifest, for instance, in textual distortions or aesthetic condemnation, make it all but impossible to access anonymous poems in their pristine shape and context.

The essays collected in this volume attempt, each in its own way, to disentangle the bundles of historically accreted uncertainties and misconceptions that affect individual anonymous texts, including pseudepigrapha ascribed to Homer, Manetho, Virgil, and Tibullus, literary and inscribed epigrams, and unattributed fragments. Poems without Poets will be of interest to students and scholars working on any anonymous ancient texts, but also to readers seeking an introduction to classical poetry beyond the limits of the established canon.

• Language: English
• Publisher: Cambridge Philological Society
• Release date: Mar 31, 2021
• ISBN: 9781913701413

Book preview

INTRODUCTION

Boris Kayachev

Every ancient poem had an author (either individual or collective), and we either know, or do not know, their name and identity; when we do not know who wrote a particular poem, it can be described as anonymous. This is the sense in which the word anonymous is used in the volume’s title, and it is important to stress that it refers, in the first place, not to an intrinsic feature of a text, but to the state of our knowledge about it. Our (lack of) knowledge can of course have degrees: for instance, when we do not know for certain, sometimes we may still be able to make a more or less informed guess about the identity of a poem’s author. Or we may actually know the author’s name but nothing else, in which case the poem may not be technically ‘nameless’, but will still be all but anonymous for most practical purposes.¹

When we do not know who was a poem’s author, our lack of knowledge can have two causes: either the author intended to remain unknown, or was forgotten in consequence of some external circumstances.² It could be argued that the two kinds of anonymity (which may be labelled ‘intentional’ and ‘accidental’, or ‘primary’ and ‘secondary’) are essentially different, but there are reasons for looking at both categories of texts together. On the one hand, as a rule we cannot be certain from the outset, or even at all, whether a poem is anonymous intentionally or by accident.³ On the other, both categories of anonymous poems can share the same concomitant circumstances.

These, in turn, are of two kinds. First, the lack of a known author severs the vital link to the poem’s historical context: most crucially, when the author’s identity is unknown, we have less certainty about the poem’s period and circumstances of production, and it is all but impossible to relate it to any other works by the same author (even if we know them). Second, anonymous poems tend to be neglected and/or maltreated throughout the history of transmission and reception: anonymous poems often have less reliable textual traditions, which in itself is a serious obstacle to their appreciation, but perhaps no less damaging is the stigma of aesthetic condemnation that often attaches to anonymous, especially ‘spurious’, texts.⁴ The cumulative effect of such uncertainties and lacunae – concomitant with, if not necessarily brought about by, anonymity – is difficult to overestimate, and difficult to counteract.

It would be naive to suggest that a single method can suffice to solve all the different issues of all the different anonymous poems. Indeed, each text poses a unique set of individual problems which require special skills and techniques. If there is, however, one methodological principle that is valid for all of anonymous ancient poetry, it is the need to be meticulously aware of the distance that separates the modern reader from the original text: not only in temporal and cultural terms (which would apply to any ancient text), but also, and more crucially, in terms of the actual circumstances of transmission and reception. While it is true that our hermeneutic optics must be specially adjusted to view any particular text, in the case of anonymous poetry it is essential to bring sharply into focus not only the text itself, but also all the vicissitudes it has passed through before reaching our desks. This is the sense in which the present volume presents an assortment of approaches to anonymous poetry: it pays special attention to the distance that has to be covered on our way to the texts whose authors remain to us unknown.

Recent years have seen an increase of interest in anonymous poetry: not only in individual texts, but also, and perhaps more significantly, as a category. Niklas Holzberg’s (2005) and Irene Peirano’s (2012) research has paved the way for our appreciation of pseudepigraphy in the domain of Latin poetry as an intentional stance rather than merely an incident of transmission.⁵ More recently Tom Geue (2019) has explored anonymity in Latin literature (both poetry and prose) ‘not as a paralyzing lack but as a constitutive effect of the text, an enabling force fundamental to the way it works’.⁶ Geue successfully demonstrates that, whether or not we happen to know the author’s name, certain modes of discourse are especially compatible with, or even benefit from, anonymity. The difference between Geue’s approach and that of the present volume is one of perspective: while Geue views anonymity as an intrinsic feature of the text, our focus is on anonymity as an historical condition, with its concomitant circumstances, in which the text survives. This is a book not just about a number of anonymous texts taken in isolation, but also about what happened to them in the hands of readers and scholars, ancient as well as modern, when they lacked the protection of a powerful author.

Without attempting comprehensive coverage of all anonymous ancient poetry, our volume offers a panoramic overview of some of the more prominent groups and types of anonymous texts both Greek and Latin, from Homeric hymns to Virgilian pseudepigrapha, from inscribed poetry to literary fragments. Rather than having its contents arranged by formal criteria such as genre or chronology, the volume is organised in three parts that take into account the texts’ circumstances of transmission and reception. Part 1 deals with collections of anonymous poetry and pays close attention to the role of the compiler (as either opposed to, or identical with, the original author). Part 2 looks at anonymous fragments, exposing the process by which their fragmentary state and anonymity jointly produce an increased lack of contextualisation, and exploring the ways to counteract this. Part 3 finally views anonymous texts, literary or epigraphical, as integral artistic wholes, in an attempt to overcome existing negative preconceptions and to engage with such texts on their own terms. This arrangement is of course only one way to create a meaningful overarching narrative, and the reader should not be surprised to find chapters from different parts in dialogue with each other.

Part 1 opens with Alexander Hall analysing the evolution of the Homeric Hymns as a collection (Chapter 1). By paying close attention to individual Hymns’ conclusions, Hall is able to uncover three main stages in the chronological stratigraphy of the collection as we have it: an original proto-collection of twenty-four hymns (Hymns 1–26, excepting 6 and 8), sectioned into four six-poem groups; the addition of a seven-hymn appendix (Hymns 27–33), along with the insertion of an additional hymn into the first group (Hymn 6); the accretion of Hymn 8. The first two stages (and possibly the third as well) provide clear evidence of intentional design, which opens the possibility that the anonymous compiler(s) did not merely collect the Hymns, but actually revised, or even composed, at least some of them. The question of their authorship and anonymity turns thus out to be a complex affair: the original compiler may in fact have played the role of a co-author, whereas the Homeric ascription appears to be a generic marker rather than an act of forgery, especially as the groupings created involve an intentional interweaving of Homeric and non-Homeric material.

Jane Lightfoot explores the corpus of Greek astrological poetry ascribed to Manetho (Chapter 2), and comes to somewhat similar conclusions. The extant six-book collection consists of an original kernel (books 2, 3, 6), composed in the second century AD, and later accretions by different authors (1, 4, 5). Although the manuscript ascription to Manetho, which does not appear in the earliest books but seems to be assumed in books 1 and 5, may not go back to the original author, Lightfoot shows how it reflects the Hellenistic convention of attributing astrological poetry to Egyptian authority figures, such as Petosiris. Manetho’s corpus as we know it appears in fact to be just one manifestation of a highly fluid textual tradition, which also gave rise to parallel recensions (sometimes very similar to the one we have) transmitted under other names. Not unlike the Homeric Hymns, the Manethonian collection turns out to be the product of a number of anonymous authors, revisers and compilers, and its pseudepigraphic ascription may to a greater extent be a generic convention than a malevolent misattribution.

The next two chapters focus on Latin poetic collections. Robert Maltby uncovers authorial design in the structure of the third book of the Tibullan corpus, which he interprets as evidence for unitarian authorship of the anonymous collection (Chapter 3). Contrary to the communis opinio, Maltby argues that not only has the author of the Sulpicia Cycle (poems 8–12) based his collection on the elegies of Sulpicia herself (13–18), but also the Lygdamus poems (1–6) show knowledge of the Sulpicia Cycle. The author of the Messalla panegyric (7), impersonating a young Tibullus, seems in turn to be aware of the Lygdamus poems, and the Tibullan impersonation is then taken up, before the short closural epigram (20), in the penultimate elegy (19). Besides these intratextual links permeating the different sections, Maltby also detects significant continuities in poetological concerns and engagement with intertexts throughout the collection. The differences in style and persona between the sections can thus be seen to reflect the anonymous author’s strategy of putting on different masks, in a project of creative engagement with the literary landscape of the first century BC.

A comparable approach is applied by Tristan Franklinos to the Catalepton, with a focus on the two epigrams (4 and 11) addressed to Octavius (and) Musa (Chapter 4). While putting poems 14 and 15 aside as later accretions, he sees in the remaining thirteen pieces an original libellus consisting of two symmetrically arranged six-poem sections (1–6 and 7–12), supplemented by a thirteenth poem. Within this structure, poems 4 and 11 form a pair, and Franklinos examines their connection, inter alia, by exploring the possibility of their addressees being one and the same person, the historian Octavius Musa. Such analysis makes it possible to read the libellus as an artistic whole, retrospectively engaging with the literary milieu of the mid-first century BC, rather than as a random assemblage of loose pieces or a collection of Virgilian forgeries. For both Maltby and Franklinos, the detection of authorial design in the arrangement of [Tibullus] 3 and the Catalepton builds a crucial piece of argument for seeing the anonymous figures of the poet and the compiler as one and the same person, which in turn is a prerequisite for their reading of the collections as chronological fictions.

Part 1 is rounded off by Michael Tueller, advocating a more holistic approach to the Greek Anthology, in particular Metrodorus’ collection of mathematical epigrams (Chapter 5). One line of argument simply notes that, by collecting and arranging anonymous epigrams and furnishing them with detailed technical commentary, Metrodorus created an artistic whole, a mathematical workbook, which is worth studying in its own right and for its own sake. More fundamentally, Tueller also argues that, historically speaking, epigram is a genre for which authorship is in some ways a much less important category than context, first material and later literary: Metrodorus’ collection thus provides individual anonymous poems the most authentic kind of context, technical as well as literary, a scholar can hope for.

If Part 1 makes a case for paying close attention to the role played by collectors of anonymous poetry, Part 2 opens with a contribution by Hannah Čulík-Baird who similarly argues for the need to understand the logic underlying the use of anonymous fragments of Latin drama in Cicero’s works (Chapter 6). As she demonstrates, ‘anonymous’ fragments were not as a rule anonymous to Cicero and his audience; by contrast, fragments whose authors we do know were often quoted by Cicero with no indication of authorship. There is thus no functional difference between attributed and unattributed fragments, and Čulík- Baird explores the uses to which dramatic fragments are put by Cicero. It transpires that the majority of fragments were not excerpted by Cicero directly from their original sources, but had a textual currency of their own, grounded in their appearance in theatrical performance, rhetorical discourse or literary scholarship. This secondary context is often the main point of reference in Cicero’s use of dramatic fragments, and the anonymous mode of quotation appears to reflect that fact.

Patrick Finglass turns to anonymous Greek tragedy, and one category in particular: interpolations, including but not limited to those made by actors (Chapter 7). In contrast to papyrus fragments or fragmentary quotations, interpolations are ‘fragments’ by design, and likewise they intentionally suppress their authors’ identity in an attempt to fit seamlessly into the surrounding context. While most critics consider their job done once they have isolated such alien accretions, Finglass discusses how they could be collected and studied as instances of creative engagement with their host texts. Taken together with other kinds of anonymous tragic texts (including the complete plays Prometheus Bound and Rhesus), actors’ interpolations can provide crucial evidence for a better understanding of conventions and expectations underlying Greek tragedy as a performative tradition.

Another case of suspected interpolation is treated by Mikhail Shumilin who offers a novel solution for the longstanding problem of the Helen episode in book 2 of Virgil’s Aeneid (Chapter 8). While most earlier research has oscillated between defending the HE as genuinely Virgilian and rejecting it as a later anonymous interpolation (two positions supported, inter alia, by opposing views on the text’s literary qualities), Shumilin evades the rigid dichotomy of authenticity and anonymity by developing a compromise hypothesis that takes into account arguments advanced on both sides of the debate. A plausible solution is that the HE is an editorial supplement by Varius (marked as such in the editio princeps) intended to fill in a lacuna in Virgil’s unfinished poem. The HE thus turns out to share features of both authorial and anonymous poetry: while it merely performs a technical function and aims to blend in with the surrounding text, it was arguably composed by an outstanding poet and never intended to be taken for Virgil’s.

In the final chapter of Part 2, Stephen Heyworth revisits the material discussed earlier by Maltby, and offers a different take on [Tibullus] 3.19 and 20 (Chapter 9). If the central argument in support of Maltby’s case was provided by the reading of book 3 of the Tibullan corpus as an artistic whole, for Heyworth the crucial step consists in recognising the fragmentary nature of 3.19 and 20: both pieces are likely to be incomplete poems, which also means that they may have little to do with the rest of the book. Paradoxically, by severing the ties to what appears to be their only extant context, Heyworth is able to read poems 3.19 and 20 as genuinely by Tibullus (named at 3.19.13) – fragmentary remains of the lost end of his second book.

While Parts 1 and 2 explore anonymous poems in their relation to a variety of contexts within which they are transmitted, Part 3 views them as artistic wholes and deals with issues of their reception and interpretation. Matthew Hosty examines the ways in which the uncertain attribution of the Batrachomyomachia has affected both its textual transmission and modern editorial practice (Chapter 10). Unlike many other anonymous and pseudonymous poems, the BM enjoyed popularity with mediaeval readers, but, despite the significant number of surviving manuscripts, this failed to secure a stable transmission, as the poem’s uncertain authorship status invited creative interventions on a massive scale. Such attitudes to the text of the BM can be seen to continue in modern scholarship, culminating in Arthur Ludwich’s 1896 edition in which he printed, among a mass of conjectures, nine whole lines of his own composition.

By contrast, in twentieth-century editions of the Appendix Vergiliana excessive conservatism prevails, as Boris Kayachev attempts to demonstrate taking the Moretum as an example (Chapter 11). While the longer poems of the Appendix (the Culex, the Ciris, the Aetna) are often condemned as second-rate, which editors sometimes take as an excuse for ignoring even the most glaring textual problems, the Moretum enjoys a better reputation, but its editions are still full of passages in need of correction. As a collective body, textual critics have spent far less effort and time on texts that lack the incentive of a famous name, and the one way to counteract this relative neglect is by giving such texts time and attention. With this end in mind, Kayachev offers possible solutions for a handful of textual problems in the Moretum.

Finally, Richard Hunter looks at Greek verse inscriptions (Chapter 12). Besides being mostly anonymous, inscribed poems pose serious challenges to scholars versed in the study of literary epigrams – simply by virtue of not being ‘literary’. Inscribed poetry operates with a different set of conventions, and Hunter explores elements of such an epigraphic poetics. Not only do inscribed poems have a different notion of authorship, but they also exist in a different medium and aim at a different audience. In visual terms, this difference can be seen in the interaction of inscribed poems with the iconographic depictions they accompany. Metre and language are further areas in which standards of verse inscriptions differ from those followed in literary poetry. Lastly, several cases of allusive intertextuality come into focus: this is a feature usually associated with elite literary production, but, as Hunter shows, it is not altogether absent in inscribed poetry, even if it manifests itself differently.

This overview, and the actual chapters perhaps even more so, may leave the reader wondering about the point of bringing together studies that explore such widely different texts and from such disparate perspectives: even anonymity does not always stay unchallenged (Chapters 8 and 9). This very diversity of material and approach is the point: rather than focusing on a specific kind of problem and promoting a particular methodology (as some earlier works have done), our volume aims to bring out the fundamental complexity of issues associated with anonymous ancient poetry and, in doing so, also to highlight similarities between cases that otherwise might have had little chance of being connected.

A salient illustration of the complexity of issues involved is provided by the radically different treatments of the same material in Maltby’s and Heyworth’s contributions (Chapters 3 and 9). Tibullus (?) 3.19 is either genuine (Heyworth) or intentionally pseudepigraphic (Maltby); in order to substantiate their positions, both carry out not only stylistic and intertextual analyses of the controversial poem itself, but also seek to endow it with an original context: Maltby presents 3.19 as an integral part of a singleauthored ps.-Tibullan book 3, Heyworth sees in it a fragment of the lost end of Tibullus’ second book. It may be tempting to ask how their different approaches are underpinned by different theoretical and methodological traditions, and Maltby’s analysis can indeed be seen to develop recent trends in the research on Latin pseudepigrapha (see above), but it must also be noted that in fact both accounts turn standard expectations upside down (a conservative unitarian approach is usually associated with defence of authenticity, a sceptical analytical one with suspicion of spuriousness). Readers will decide for themselves which of the two scenarios is the more convincing, and indeed whether a decision is possible, but both treatments make it clear that meticulous study of evidence is the one way to achieve progress, while also offering a salutary reminder that even then conclusive answers may not be easily attainable.

Even if in and by itself the comparative approach may not be able to generate definitive solutions, in the midst of uncertainties affecting individual anonymous texts it can often provide a valuable point of orientation. An example of a non-obvious but illuminating point of affinity can be found between Euripides’ Iphigenia in Aulis and Virgil’s Aeneid, both left unfinished by their authors: Finglass’s discussion of Euripides Minor’s editorial activity on the Iphigenia may provide additional support for Shumilin’s case for viewing the Helen episode in Aeneid 2 as an editorial supplement by Varius (Chapters 7 and 8). Less straightforward but nonetheless important insights can be gained from a comparison of editorial approaches to the ps.-Homeric Batrachomyomachia (Hosty, Chapter 10) and the ps.-Virgilian Moretum (Kayachev, Chapter 11). While, on the surface level, in the editions of the BM and the Moretum their anonymous status has led to opposite results – an excessive interventionism as opposed to an excessive conservatism – the underlying condition is the same in both cases: a severe lack of hard facts about the original text. In the case of the BM it is manifested in an extreme fluidity of the tradition, in the case of the Moretum the tradition is fixed but likewise unreliable; the solution is fundamentally the same in both cases: to scrutinise every transmitted word with the help of all available philological tools, so as to decide whether or not it represents the original (compare Shumilin’s conclusion in respect of the Helen episode in Chapter 8, as well as Heyworth’s approach to the text of Tibullus 3.19 in Chapter 9). While it may be understandable that anonymous texts have attracted less editorial attention than texts with secure authorship and a reliable paradosis, objectively they are in need of more, not less, scrutiny.

These are just a couple of examples of potentially fruitful but often non-obvious intersections between the different treatments of different texts collected in this volume. Its ambition is in fact to initiate such a dialogue and exchange of experiences between scholars whose work on individual anonymous texts might otherwise suffer from excessive compartmentalisation. Though some anonymous poems have received more attention than others over generations of scholarship, in many cases essential groundwork still remains to be done: comprehensive collections, reliable critical editions, thorough commentaries, not to mention literary and historical assessment and interpretation. We hope that the discussions put together here, and further ones that may follow, will both stimulate and inform such future work. To echo the conclusion of Chapter 12 by Hunter, which seems equally applicable to the topic of the volume as a whole: ‘No one should complain that there is nothing left to do’.

Bibliography

Geue, T. (2017) Juvenal and the poetics of anonymity, Cambridge.

— (2019) Author unknown: the power of anonymity in ancient Rome, Cambridge, MA.

Holzberg, N., ed. (2005) Die Appendix Vergiliana: Pseudepigraphen im literarischen Kontext, Tübingen.

Hunter, R. (2002) ‘The sense of an author: Theocritus and [Theocritus]’, in R. K. Gibson and C. Shuttleworth Kraus (eds), The classical commentary: histories, practices, theory (Leiden) 89–108.

Kayachev, B. (2013) Review of Peirano 2012, BMCR 2013.11.52.

Peirano, I. (2012) The rhetoric of the Roman fake: Latin pseudepigrapha in context, Cambridge.

1For instance, Juvenal’s Satires have been interpreted as a quasi-anonymous text, both because close to nothing is known about their author from external sources and because the poems themselves say very little about him (Geue 2017). Clearer examples are Italicus’ Ilias Latina and Calpurnius’ Eclogues (cf. Geue 2019: 164-97): we (apparently) know the authors’ names, but are unable to tie them to any known individuals. Many inscriptions fall within this category: we may know the name of the person who composed (or at least commissioned) an inscription, without knowing their identity.

2To sharpen the distinction, it seems worth mentioning the possibility that the author’s intent to remain anonymous may not be realised.

3Intentional anonymity can be presumed in two cases: when we have an unsigned autograph (or something close to it), as e.g. in the case of Pompeian graffiti; or when we have a pseudonymous attribution within the text itself, as e.g. in the case of the Sibylline Oracles .

4For a criticism of such attitudes towards ‘authorless’ texts in classical scholarship, see Hunter 2002.

5Pseudepigraphic poetry can be considered a subset of anonymous poetry, and just as anonymity can be intentional or accidental (‘primary’ or ‘secondary’), so can pseudepigraphy. Both Holzberg and Peirano are mainly interested in intentional pseudepigrapha, and elsewhere I have expressed certain reservations about their approach, as (in my view) it often fails to justify the ascription of individual texts to that particular category (Kayachev 2013).

6Geue 2019: 5; he avoids the pitfall of Holzberg’s and Peirano’s approaches by declining to assume intent as a necessary precondition: ‘What if we were to treat texts as deliberately or, better, autonomously anonymous - not always in the strict sense that they were designed that way by their primary authors […] but that the coauthors of history, time, and accident have together licked them into a shape where the anonymity is a nonnegotiable part of them needing to be critically fondled like all the other aspects of the text?’

PART I

collections

1

The evolving arrangement of the Homeric Hymns

Alexander E. W. Hall

The Homeric Hymns were not composed by Homer. Despite the attribution of the Renaissance manuscripts and the testimony of some early citations of the collection, the judgment of the Hymns as ‘inauthentic’ was already commonplace in the ancient world. In the modern world it has become a unanimous verdict, even as the notion of authorship itself has been broadened and problematised, especially in application to orally composed works. Who might have composed the poems and when remains an unanswered (and probably unanswerable) question. A scholiast on Pindar identifies Cynaethus of Chios, a prolific interpolator, as the true author of the ‘hymn to Apollo attributed to Homer’. Even if that is true, however, establishing the composer of this one Homeric Hymn tells us nothing about the origin of the others. It is virtually certain that the single name ‘Homer’ conceals many different authors, perhaps as many as there are Hymns. This anonymity (pseudonymity) is the result both of the involvement of many generations of editors and also of the very different notions of ‘authorship’ at play in an oral literary culture, and the impact of both these factors is magnified by the diversity of the collection itself.

Indeed, what principally unites the Homeric Hymns – apart from their shared attribution to Homer, of course – is their diversity. These thirty-three poems vary tremendously in length, language, tone. This is not to say, however, that the Hymns are merely a hodgepodge or miscellany. They evince a great deal of thematic coherence, as demonstrated most clearly by Clay (2006). They are also, though it may not seem so at first glance, very carefully organised. It will be the work of the chapter that follows to outline that organisation, and to explore its implications for our understanding of the ‘Homeric’ nature of the Hymns.

An examination of the conclusions of the Homeric Hymns reveals a pattern. This pattern is connected with the arrangement of the collection: groups of similar Hymns begin and end with poems which deploy identical or nearly identical closing formulae. These formulae are peculiar to the paired Hymns in which they appear, not only in that they are not found in other Homeric Hymns, but also because their content is appropriate to the poems in which they appear, as well as the larger group those poems bracket. This suggests that the appearance of these parallel endings is not merely coincidental, but rather is intended to emphasise the connections between the paired Hymns, and thereby to mark the beginnings and endings of distinct groups within the larger collection.

The usefulness of this pattern is not mostly in decoding the arrangement of the Hymns, however. For one thing, the groups marked by the pattern are essentially identical to those observed by Torres 2003, based on the narrative content of the poems. For another, the pattern is not present for the whole length of the collection: we can be certain of its presence for three groups of Hymns (1–7, 9–14, 15–20) and perhaps a fourth (21–6), though for this last the indicators are less clear. In the final seven Hymns in the collection (27–33) the pattern breaks down entirely, with five of seven poems using the same or similar conclusions.

It is actually from this breakdown that useful insights arise, and they concern not the arrangement of the collection but its evolution. I believe that our collection of thirty-three Homeric Hymns was created from an earlier group of twenty-four poems, a collection that was not only smaller but more tightly organised than the one we possess. The parallel endings described above were part of the organisation of this ‘proto-collection’. The remaining nine Hymns were added at some later date or dates, by a compiler or compilers who either did not notice the existing system of groups marked by parallel concluding formulae or chose not to continue it, leading to the breakdown mentioned earlier.

In what follows, I will first outline how the pattern works in those groups where it is readily observable and explore why such a system might have been employed. Next, I will turn to the later groupings of Hymns where the pattern first becomes hazy and then disappears entirely, making the case that this section contains the end of the earlier protocollection and the beginning of a later ‘appendix’. By way of a conclusion, I will offer a sketch of this evolutionary process and turn in earnest to the issue of authorship. Indeed, I will argue that one of the organising principles of the structure outlined is a sorting of poems into markedly ‘Homeric’ and ‘non-Homeric’ groups, while at the same time creating a collection that fused them into a ‘Homeric’ whole. This is consistent with the nature of ‘Homeric’ authorship, at least as it was understood in some periods in the ancient world.

The pattern in practice

To demonstrate the system of parallel concluding formulae acting as markers for the beginning and end of a group of Homeric Hymns, I would like to begin not with Hymn 1 and the start of the collection, but instead with Hymn 9, a nine-line poem dedicated to Artemis. The reasons for this are two. First, the group of poems which this Hymn to Artemis inaugurates – Hymns 9 through 14 – is probably the most coherent group in the collection, no matter the organisational principle applied: all six poems are short (3–9 lines);¹ all are attributive, with central sections describing their dedicatees in the present tense;² and