On the Track of the Books: Scribes, Libraries and Textual Transmission 3110632594, 9783110632590

Table of contents :
[9783110632590 - On the Track of the Books] Frontmatter......Page 1
Preface Cupis Volitare Per Auras Books Libraries and Textual Transmission......Page 5
Contents......Page 7
Acknowledgments......Page 9
Introduction......Page 11
Figured Books: Horatian Book- Representations......Page 23
Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20......Page 35
Fake Intellectuals, and Books of Unquestionable Authority in Aulus Gellius’ Noctes Atticae and Lucian’s aduersus Indoctum......Page 57
Martialis Epigrammaton liber decimus: Strategies for a Second Edition......Page 69
Poetic Quotation in 4th Century BC Attic Oratory......Page 83
Jerome’s Two Libraries......Page 101
Some remarks on P. Lit. Lond. 63, a riddle epigram of an anthology?......Page 115
Textual Tradition and Reception in Theocritus......Page 121
Eratosthenes’ studia Aristophanica......Page 135
Eratosthenes’ Πλατωνικός between Philosophy and Mathematics......Page 153
Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus......Page 177
Latin Epigraphy and Literary Texts in 4th Century AD Rome......Page 199
The Scribal Habits of Codex Sangermanensis in Greek and Latin in Light of its Exemplar......Page 211
The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes......Page 249
The Library and the scriptorium of the Abbey of Montevergine in the 12th and 13th Century: Presences and Absences......Page 279
Apocolocyntosis, codex V and the manuscript of Hadrianus Junius......Page 287
The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age: the Example of Germany......Page 303
Bibliography......Page 317
Index......Page 349

Citation preview

On the Track of the Books

Beiträge zur Altertumskunde

Herausgegeben von Susanne Daub, Michael Erler, Dorothee Gall, Ludwig Koenen und Clemens Zintzen

Band 375

On the Track of the Books Scribes, Libraries and Textual Transmission Edited by Roberta Berardi, Nicoletta Bruno and Luisa Fizzarotti

ISBN 978-3-11-062288-1 e-ISBN (PDF) 978-3-11-063259-0 e-ISBN (EPUB) 978-3-11-063016-9 ISSN 1616-0452 Library of Congress Control Number: 2019937004 Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2019 Walter de Gruyter GmbH, Berlin/Boston Printing and binding: CPI books GmbH, Leck www.degruyter.com

Preface: Cupis Volitare Per Auras – Books, Libraries and Textual Transmission The present volume is conceived as the result of a discussion began during the two–day conference (Bari, 27–28, October, 2016) organised by the cultural association Prolepsis as its First International Postgraduate Conference. It is indeed from this meeting (Cupis Volitare Per Auras – Books, Libraries and Textual Transmission from the Ancient to the Medieval World) that this book takes its name. It is a title derived from Martial, who in Ep. 1.3 (‘Cupis volitare per auras’) addresses his book and blames it for wanting to fly away and become public instead of staying home. Thus, this sophisticated poetic quotation, offered us the hint for a wide reflection on ancient textual transmission and editorial practices in antiquity. In the vast and culturally lively scenario that characterised the reflection sprung on such a broad topic during the varied and multifaceted moments of discussions, emerged during this two–day conference, three main themes1 stood out as deserving of deeper investigation and further study: the concept of ‘book’ in antiquity and its development; the crucial connection between text, paratext and scholarly work; and the various aspects that characterise materiality as means of survival of texts. Therefore, these are the points on which the three sections of this volume are constructed and shaped. Section one Writers at Work: Books, Figured Books, and Ancient Authorial Strategies, through its five contributions, will constitute an exploration on the role of books both as a material element and as a mental concept, but also on ancient editorial processes; section two, composed of five papers and entitled Following the Routes of Textual Transmission: Corpora, Text and Paratext, aims at investigating deeply into the mechanisms of formation of corpora and circulation of texts, alongside their paratextual apparatus; finally, our third section, entitled ‘One More Link in the Chain’: Scribes, Stones, Codices, Libraries and formed of six contributions, will show how the history of texts is also the history of the materials and the people that made them. We would like to imagine the three sections of this volume as the representation of a path that keeps broadening and enlarging, starting from a simple object – the book and its text – , that becomes bigger and richer through its paratext and its mechanisms of circulation, in order to eventually reach a final advanced stage in which it is part of wider contexts, of intellectual communities, scriptoria, and libraries.

 1 Sections and related part of the Intoduction have been edited in the following order: Nicoletta Bruno, Roberta Berardi, Luisa Fizzarotti. https://doi.org/10.1515/9783110632590-202

Contents Preface   V Acknowledgements    IX Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti  Introduction  1 Stephen J. Harrison  Figured Books: Horatian Book-Representations  13 Georgios Taxidis  Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  25 Katherine Krauss  Fake Intellectuals, and Books of Unquestionable Authority in Aulus Gellius’ Noctes Atticae and Lucian’s aduersus Indoctum  47 Ambra Russotti  Martialis Epigrammaton liber decimus: Strategies for a Second Edition  59 Antonio Iacoviello  Poetic Quotation in 4th Century BC Attic Oratory  73 Giulia Marolla  Jerome’s Two Libraries  91 Daniela Immacolata Cagnazzo  Some Remarks on P. Lit. Lond. 63, a Riddle Epigram of an Anthology?  105 Leonor Hernandez Oñate  Textual Tradition and Reception in Theocritus  111 Federica Benuzzi  Eratosthenes’ studia Aristophanica  125 Sara Panteri  Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  143

VIII  Contents Nicola Reggiani  Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  167 Rosa Lorito  Latin Epigraphy and Literary Texts in 4th Century AD Rome: The Case of Vettius Agorius Praetextatus   189 Alan Taylor Farnes  The Scribal Habits of Codex Sangermanensis in Greek and Latin in Light of its Exemplar  201 John Bradley  The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  239 Veronica De Duonni  The Library and the scriptorium of the Abbey of Montevergine in the 12th and 13th Century: Presences and Absences  269 Olivia Montepaone  Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  277 Cristiana Roffi  The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age: the Example of Germany  293 Bibliography  307 List of Contributors  339 Index  343

Acknowledgments First of all, we would like to express our gratitude to Elena Barile and Martina Filosa, without whom the initiative that led to the assembling of this volume would have never taken place. Then, we would like to thank Antonio Stramaglia for his constant support and the precious academic and editorial advice. Finally, a special thank goes to our scientific committee, composed of Nunzio Bianchi, Mirko Canevaro, Guglielmo Cavallo, Irma Ciccarelli, Daniela Colomo, Andrea Cucchiarelli, Emanuele Dettori, Maria Jennifer Falcone, Antonio Enrico Felle, Paolo Fioretti, Guillermo Galán Vioque, Francesca Maltomini, Elena Merli, Rosa Otranto, Francesco Panarelli, Chris Pieper, Massimo Pinto, Francesca Schironi, Sonia Schönauer, Anna Usacheva. Roberta Berardi Nicoletta Bruno Luisa Fizzarotti

https://doi.org/10.1515/9783110632590-204

Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti

Introduction

 Writers at Work: Books, Figured Books and Ancient Authorial Strategies De los diversos instrumentos del hombre, el más asombroso es, sin duda, el libro. Los demás son extensiones de su cuerpo. El microscopio, el telescopio, son extensiones de su vista; el teléfono es extensión de la voz; luego tenemos el arado y la espada, extensiones de su brazo. Pero el libro es otra cosa: el libro es una extensión de la memoria y de la imaginación. (Jorge Luis Borges, El libro in Borges oral, 1979)

In the first section, the collected papers retrace the first steps of the process of ancient writing, editing and publishing.1 The reader will discover how the book is both a material object and a metaphorical personification, as is stands for something else, material or immaterial – a person, an object, and a literary motif. The book roll, its production, use and circulation, could be an instrument of cultural authority and the act of publishing could be also dangerous, especially if an author was accused of plagiarism.2 Is it, however, correct to talk about plagiarism in ancient Greece and Rome? What are the strategies of quotations and allusions used by ancient authors? What is the reader’s response? Are the readers able to understand the cross references within and beyond a generic–code system or a rhetorical strategy?3 The analogical relationship between materiality and symbology of books and libraries overcomes the literary game and becomes the mirror of the ancient society, in all its cultural, political and economic aspects and differences.4

 1 See in particular on this issue Cavallo (1989) 307–41, Fedeli (1989) 343–78, Blanck (1992), Dorandi (2007), Pecere (2010). On the design of Greek and Latin books of poems in the light of papyri, including recent discoveries see Hutchinson (2008). 2 On the theme of plagiarism in Latin literature see McGill (2012). 3 A good starting point for the reader–response theory in Ancient Rome are Citroni (1990) 53– 116 and (1995). On generic interactions in Latin poetry see Conte (1974) and (2017), Harrison (2007a). For the application of modern critical theory to Classics see Hexter–Selden (1992), de Jong–Sullivan (1994), Harrison (2001), Schmitz (2007). 4 On books, genres and society see Depew–Obbink (2000) and Winsbury (2009). https://doi.org/10.1515/9783110632590-001

2  Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti The first two essays concern Horace and the ways in which Horace’s own poetic books are personified and treated symbolically in his works. The first contribution is written by a distinguished Horatian scholar, Stephen Harrison (University of Oxford), and is devoted to the analysis of three examples of the metapoetical personifications of Horace’s book, Carm. 1.38, Epist. 1.13 and 1.20. The figures of personification and metaphors used by Horace seem to reflect both appropriate literary and cultural traditions and the physical form of ancient book–rolls. In Carm. 1.38, the poetic book is compared with a garland: the Horatian metaphor is clearly inherited by Greek epigrammatic tradition, and, according to Harrison, it looks back to Callimachus (Ep. 28 Pfeiffer) and Meleager (Anth. Pal. 4.1). As Harrison rightly points out, this is not the only Callimachean manifesto at the end of the Horatian poetry book, suffice to mention Carm. 2.20. The literary tone deeply changes in two examples picked from Epistles’ Book 1, according to the realistic ‘generic’ code of satire. The key point of humour in Epist. 1.13 is the ironic contrast between the lightness of the burden (the three papyri book–rolls of the Odes) and the melodramatic instructions given by the poet for their transport to Augustus. Harrison argues that the analogy is between soldiers and beasts of burden, and the poetic books are figured as military baggage. In the last example, the analysis of Epist. 1.20, the book is the personification of a boy, or better, a runaway slave, explored in hilarious ways which combine both Callimachean aesthetics and the nasty details of Roman prostitution.5 A detailed analysis of Horace Epist. 1.20 is the core of the essay by Georgios Taxidis (Universität Hamburg). Methods and perspectives used by Taxidis aim at a deep and new investigation on the use of Horatian lexicon of writing materials and the literary practice of sphragis. The main purpose of this paper is to shed light on the ways in which the technical terms of writing material items are used, according to a perception that is distant from any relation to the book’s image as a slave boy. Moreover, Taxidis compares the mentioned use of this specific lexicon with similar examples in other authors and other works of Horace. The second part of the paper deals with the comparison between two types of sphragis: one claiming poetic immortality (Carm. 3.30.1 exegi monumentum aere perennius) and the other one, the final lines of Epist. 1.20, characterised by a ‘self–deprecating humor’. In the third chapter, Katherine Krauss (University of Oxford) explores how the material text works as a literary motif in self–consciously learned writings of the 2nd century AD. Krauss focuses on the function and significance of the book roll in Aulus Gellius’s Noctes Atticae and Lucian’s adversus Indoctum, and she  5 Cf. Harrison (1988) 473–6.

Introduction  3

demonstrates how both the works share the same cultural milieu. She offers a treatment to the symbolic values of the book roll per se, which concern both Latin and Greek thought in the 2nd century AD. Both Noctes Atticae and adversus Indoctum reveal a preoccupation with the settings in which books can and cannot be employed as reliable symbols of authority. She traces the development of two related motifs, the use of the book as an implicit source of denigration of intellectuals, associated with the pecuniary aspects of education, and the limitations of the book’s ability to confer such an authoritative status to Gellius and Lucian as erudite narrators. The paper by Ambra Russotti (Alma Mater Studiorum – Università di Bologna) summarises the complex question of the second edition of Book 10 of Martial’s Epigrams, providing a useful overview, and highlighting three interesting key points (epigrammata longa; the theme of plagiarism; Martial’s return to Spain). The case of Book 10 of Epigrams is not isolated: this fate was not only peculiar of Martial’s book. The first problem deals with the epigrammata longa: according to Russotti, Martial inserted them for a political reason, with the aim to substitute and delete the praise of Domitian; in fact, Martial, during Domitian’s reign had flattered and celebrated the emperor, therefore after Domitian’s death, he fell in disgrace. In the second part of the contribution, Russotti analyses the issue of plagiarism in Martial, starting her reflection from the Roman cultural and social meaning of this phenomenon: even if in the age of Martial divulged work could potentially be stolen, and even if plagiarism were a constant problem for the poet, Martial speaks about the topic only in some particular moments of his career (especially in Book 1 and 10). In the last part of her paper, Russotti briefly deals with the topic of Martial’s return to Spain, that is one of the dominant topics of Book 10. Although it is impossible to precisely reconstruct Book 10’s first edition, the results reached with this study clarify the nature of the modifications that Martial did, by considering the features of this book in comparison with those of others. The last two essays of the first section aim at studying the practice of literary quotation and allusion to other ancient texts, according to different genres, languages and time. This common practice can also help the ancient and modern reader to reconstruct the libraries and the most widely read and quoted books. Antonio Iacoviello (University of Edinburgh), in his essay, considers the use of poetical piéces, especially from epic and tragic poetry (Homer and Euripides), in Attic oratory since the end of the 5th century BC. He focuses on three orators, Aeschines, Lycurgus and Demosthenes (in particular, on these speeches, Aeschin. 1– 3, Lycurg. Leoc., Dem. 18–19) and he shows how Homeric poems and Euripides’ tragedy were a powerful instrument to strengthen the orator’s political reasoning.

4  Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti Iacoviello’s aim is not limited to the comprehension of the context of the quotes and the interpretation the orator gives to his quotations for his own speech. These orations, in fact, are important testimonia both for the reconstruction of the pre– aristarchean Iliad, known in Athens in the 4th century BC, and for some fragments of lost Euripidean tragedies, such as long rheseis. Jerome’s bibliophilia is renowned and the author of the last chapter of the first section, Giulia Marolla (Università della Repubblica di San Marino), is well aware of the impossibility, for a modern reader, to recreate Jerome’s library. His massive quotations of Latin classics, Cicero and Virgil in the first place, especially in the enormous corpus of his Epistles, are ascribable to a ‘mental library’, one which probably Jerome refers to more often than it is necessary, to justify his quotations from detractors. It seems clear, according to Marolla, that Jerome’s purpose is to make use of the literary tradition of the past to enrich and embellish Christian writings with an excellent style. Marolla focuses on Jerome’s use of Cicero’s model in Ep. 123, that is an effective example of his recurring to a mental library and furthermore, his borrowing of classical echoes is a conscious choice. Though the content of his works reflects Christian models, the Scriptures, Paul and Tertullian, his style and language is riddled with classical references and reminiscences of his juvenile years as a student of Aelius Donatus. The study demonstrates that Jerome’s libraries are two: one physical in Bethlehem, impossible to recreate, but enriched by manuscripts of pagan authors copied for him, and one mental library, a sort of mental archive of his classical reminiscences.

Introduction  5

 Following the Routes of Textual Transmission: Corpora, Text, and Paratext Il paraît que les érudits arabes, en parlant du texte, emploient cette expression admirable: le corps certain. Quel corps? Nous en avons plusieurs; le corps des anatomistes et des physiologistes, celui que voit ou que parle la science : c’est le texte des grammairiens, des critiques, des commentateurs, des philologues (c’est le phéno–texte) (R. Barthes, Le Plaisir du Texte, 1970)

The second section of this volume will focus on corpora of Greek texts, their formation, and their paratextual apparatus. Throughout the five contributions included, the reader will explore various issues dealing with some of the mechanisms that are at the basis of the assembling of ancient Greek texts 6 belonging to different literary (and not only literary) genres, but great attention will also be given to the role that marginal annotations and ancient scholarly work7 have had in relation to these texts. The section is therefore mainly structured in two parts: in the first one, we will see how certain environments can strongly influence the process of formation of corpora and their circulation, but also how even some individual textual choices can be the result of specific cultural conditions.8 Then, in the second part, the focus will move from the text to the paratext and the exegetical material (transmitted by various means), and on the different approaches that scholars can employ to work on it. Finally, the discussion on paratext will shift towards some newly emerging editorial issues,9 arising from the necessity to establish a clear text for some rather complex cases where text and paratext are intrinsically connected. The first two essays deal with corpora of texts, the way they were put together, how their text was established, but also with the historical and cultural reasons that influenced their circulation. The first contribution is the one provided by Daniela Cagnazzo (Università degli Studi di Bari), it is entitled Some remarks on P. Lit. Lond. 63, a riddle epigram of an anthology?, and it contains the  6 About the formation of corpora of Greek authors, see relevant examples in Canfora (1995) 95– 250. 7 As regards scholia and scholarly activity, among the vast literature on the topic, it is worth mentioning some recent works: Dickey (2006); Montanari–Pagani (2011) and Montanari–Matthaios–Rengakos (2014). 8 On textual variants see Pasquali (19522) in particular chapters 6, ‘Varianti antiche e antiche edizioni’, and 7 ‘Edizioni originali e varianti di autore’. 9 On editorial issues see Grant (1989). For digital editions, see also Driscoll–Pierazzo (2016).

6  Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti author’s re–examination of a relatively understudied papyrus, P. Lit. Lond. 63, an example of riddle epigram from the 2nd century AD, which is to be attributed to the environment of the school. It is studied by Cagnazzo in the context of the lack of witnesses of epigrammatic anthologies in Egypt in a period that covers the 2nd and the 3rd centuries, and is seen as emblematic (along with a number of other additional instances) of the discrepancy in that period between the attested presence of epigrams in schools (and broadly speaking in private contexts), and the absence of witnesses of proper anthology books – a typically more suitable place to arrange sets of epigrams (and for this the author also provides some general information). Thus, this contribution aims at giving a specimen case of a possible investigation on what a difficult and delicate task it can be at times to draw definite conclusions on the circulation and spreading of a certain literary genre in a period characterised by a striking lack of documentation. Cagnazzo’s work, therefore, hints at the phenomenon of the formation of corpora of texts as a product of social and cultural changes (which often, as in this case, are almost impossible to determine due to the scarcity of the information at our disposal). Then, the analysis on the formation of corpora proceeds on the same track and becomes more specific with the second contribution by Leonor Hernandez Oñate (Universidad Autónoma Metropolitana). Her chapter, entitled Textual Tradition and Reception in Theocritus. The Case of αἰπολικὸν (Theoc. 1.56) constitutes a detailed investigation on the mechanisms of formation of the Theocritean corpus, and more specifically on textual choices that became part of the vulgata at a very early stage of the process. Social and cultural phenomena, indeed, as Hernandez Oñate argues, are not only at the basis of the fortune and survival of a genre (as we have seen for the case of epigrams in Cagnazzo’s paper) or of a single author, but they can also be intrinsically connected to textual choices. In this paper, in particular, Hernandez Oñate shows us how the early reception of the Theocritean text might have ended up strongly influencing the constitutio of his text; therefore, the author convincingly shows us this way as reception and textual transmission are much more related than they might seem at a first glance. This issue is carefully explained through the aid of a relevant example, the one of αἰπολικὸν (Theoc. 1.56), mentioned in the title, a lectio that is interestingly unanimously present in the manuscript tradition, but for which the scholia show traces of more complex textual possibilities and interpretations. This is explained in the light of the strong cultural influence of the ‘pastoralisation’ on the collective imagination about bucolic poetry, that ended up wiping out certain textual variants, regardless of the intentions of the authors, in favour of more normalised ones. This perfectly shows how the tradition of a work within a literary genre can have a huge influence on the manuscript tradition.

Introduction  7

Hernandez Oñate’s contribution, with its attention to the role of scholiastic and exegetical material as proof of lost branches of the tradition, works as a perfect link between the reflection on corpora and the one on paratext, which, declined in various ways, constitutes the core of the last three contributions. The third and fourth contributions included in this section are closely related, as they both deal with Erathostenes’ exegetical work: in the first case, the one illustrated by Federica Benuzzi (Università Ca’ Foscari Venezia) in her paper, a small fragment of Eratosthenes’ studia Aristophanica is transmitted by the scholia to Ar. Nu. 967 (and perhaps by a papyrus fragment), while in the other, the one illustrated by Sara Panteri’s work, a quotation of Erathosthenes’ Πλατωνικός is preserved through indirect tradition, thanks to a quotation transmitted by Theon of Smyrna’s Expositio rerum mathematicarum ad legendum Platonem utilium. These two papers well exemplify how different approaches can be employed when studying exegetical material. A first angle, the one adopted by Benuzzi, consists in studying the scholium with the aim of clarifying some obscurities in the information it preserves, rather than drawing conclusions on the kind of exegesis Eratosthenes did on the Artistophanic text. In particular, Benuzzi identifies three separate annotations in the scholium (for which we are given a critical text and translation), that can be reduced to two main redactions, α and β (while the third is just an abbreviated version of one of the others). She compares the two redactions, giving particular attention to one of the several points on which they differ consistently, namely the mention of Eratosthenes’ position on the interpretative issue dealing with the attribution of the hymn Παλλάδα περσέπολιν (the fact that he ascribed it to Lamprocles, on the ground of the evidence provided by Phrynichus) in α, in contrast with the initial ascription of the ode Παλλάδα περσέπολιν to Phrynichus by Eratosthenes according to β. The author’s work on the text allows her to provide emendations that make the sequence of the scholium more logical, and to prove that the two redactions can refer to a common ancient annotation. But Benuzzi’s work also goes beyond this specific individual issue and analyses in depth the whole content of the two redactions of the text, and studies it also in the light of a papyrus fragment, P. Oxy. 13.1611, that undeniably contains a text related to that of the scholium. A second possible approach to exegetical material is, then, the one adopted by Sara Panteri (Humboldt Universität zu Berlin), who does not simply use the extremely interesting testimony of Theon of Smyrna about Eratosthenes’ Πλατωνικός (one of the only two fragments of this lost work where the title is explicit – and of which the author offers us a critical edition and a translation) to shed new light on Eratosthenes’ exegetical work, but also – and most importantly

8  Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti – she interrogates the text in order to see how this fragment can be used to better interpret the text of Plato, in the light of the considerations of his most important exegetes (Eratosthenes was indeed named the second Plato). In particular, Panteri presents us with the problem of the nature of Eratosthenes’ commentary to Plato’s philosophy, showing us how the exegetes dealt with a problem that was both linguistic and factual, namely the difference in meaning between the terms διάστημα and λόγος (which was a distinctive point of disagreement between the main Greek schools of Harmonics). Moreover, the author also clearly explains how the typology of commentary was both mathematical and philosophical. Lastly, thanks to a deep analysis of the content of this fragment and the cross reference to other sources, Panteri is able to confirm the passage to which the ancient commentator was referring to in this bit of commentary (i.e. Timaeus 34b–36d, the one to which Theon’s section of music is about). Finally, this section is concluded by a paper that develops the reflection on the work on paratext, placing it in the context of contemporary editorial issues, specifically that of the crucial role that the digital humanities are gradually acquiring in the editorial process of some typologies of texts. In his chapter, Nicola Reggiani (Università di Parma) presents us the fairly complex and challenging case of fragments of Greek receipts and receptaria on papyrus. These texts constitute a quite singular and intricate case of textual transmission – which the author of the contribution himself does not hesitate to define ‘liquid’– as they can be seen fragments of an oral discourse, always needing to be updated or modified. Consequently, the paratextual and marginal annotations cannot here be seen as a separate set of exegetical material as it happens for other literary genres, but do represent an essential part of the papyrus, which is in constant dialogue with the main text. Hence, a crucial editorial problem: a traditional approach, as Reggiani rightly argues, would end up undervaluing the importance of annotations and ‘paratext’ (if we can still define it paratext for this rather singular case) and cancelling the dialogical nature of the different layers of the text in the papyrus. This problem can be solved through the aid of a digital edition, that could certainly give the right space to the fluidity of these kinds of texts, conferring the right importance to each stage of the transmission.

Introduction  9

 ‘One More Link in the Chain’: Scribes, Stones, Codices, Libraries Nel caso dei testi antichi vi è comunque sempre un anello in più. C’è sempre un altro che li ha scritti: ma chi fu il copista? (L. Canfora, Il copista come autore, 2002)

One of the peculiarities of ancient texts consists in having two levels of authorship: on the one hand, the author of the text, and on the other, the scribe who copies the text word by word. The scribe is not merely a medium, but he plays a fundamental role in presenting and often in changing the text, whether consciously or not. This section will focus on the protagonists of some interesting cases of textual transmission, but also on the books they manufactured or kept in the libraries, and on the words they engraved on stones as κτῆμα ἐς αἰεί. The history of textual transmission is also a history of men, objects, centres of librarian production in their simple materiality, their birth, death, and sometimes survival. This is why when we approach an ancient text, we should always consider that there is ‘one more link in the chain’, ‘un anello in più’, as Luciano Canfora writes in his work Il copista come autore.10 This section will take the reader from funeral dedicatory Late–Antique verses, to the scribal habits of a 9th– century codex, to a collegiate tomb of professional scribes in Rome, to a Southern–Italian Abbey with its library and scriptorium, to a Senecan book in the hand of a Dutch humanist, to a medieval remake of Ovid’s Metamorphoses. The first contribution is Latin Epigraphy and Literary Texts in 4th Century AD Rome: the Case of Vettius Agorius Praetextatus by Rosa Lorito (Università degli Studi di Palermo). The author analyzes the poem engraved on the funerary monument of Vettius Agorius Praetextatus, praefectus urbi in 367–368 and praefectus praetorio in 384. The long inscription (CIL VI 1779 = ILS 1259) reflects a singular twist of Epigraphy and Literature, because the verses are written in iambic senarii – a meter typical of Plautus and Terentius –, and because the eminent Roman politician is a character of Macrobius’ Saturnalia. The author starts depicting the profile of this member of the Senatorial aristocracy, from the cursus honorum, to his otium litteratum to the religious office. Then, Lorito describes the funerary  10 Canfora (2002) 14. For a selective and fundamental bibliography see also Becker (1885); Pasquali (1949) 942–7; Reynolds–Wilson (19684); Rizzo (1973); Cavallo (1975); Gamillscheg– Harlfinger–Hunger (1981); Turner (19872); Detienne (1988); Cavallo (1989) 307–41; Cavallo (1994) 613– 47; Cavallo (2002); Casson (2003); Blanck (1992); Buonopane (2009); Canfora (2018).

10  Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti marble ara now kept in the Musei Capitolini of Rome, and the disposition of the text on the grave. She transcribes the inscription and prints her new – original – English translation of the poem. The originality of this text lies indeed in the combination of funerary inscription and epigram features. The exaltation of pagan aristocratic values testifies to – as the author writes –, ‘the use of epigraphy as an instrument to make the memory of these values eternal’. With the second contribution, we consider again a writing process, but, in this case, executed on a Byzantine parchment manuscript as material support, i.e. the renowned Codex Sangermanensis. Alan Taylor Farnes (University of Birmingham) makes this book the subject of his work, The Scribal Habits of Codex Sangermanensis in Greek and Latin in Light of its Exemplar. The Codex Sangermanensis is a 9th–century bilingual book with Pauline Epistles, and it is now stored in the National Library of Russia in St. Petersburg. The author convincingly demonstrates the dependence between the Codex Sangermanensis and another bilingual majuscule parchment manuscript: the Codex Claromontanus (5th century). After a critical introduction on the debatable ‘Colwell method’, the author analyzes the Codex Claromontanus and its relatives, and then he compares the Claromontanus and the Sangermanensis, concluding that the second is a copy of the first. He enquires into the relation between the two manuscripts and gives strength to the commonly accepted hypothesis. In fact, he dedicates a long section to the different scribal habits of Codex Sangermanensis, firstly for the Greek text, and secondly for the Latin text. He finds out the habits of the A and B Scribes, and lists ‘insignificant variant readings’, ‘errors due to corrections’, ‘errors due to graphical confusion’, ‘orthographic variant readings’, ‘nonsense variant readings’, ‘additions’, ‘omissions’, ‘transpositions’, and ‘substitutions’. After his accurate analysis, Alan Taylor Farnes is able to conclude that the scribes were not fluent in Greek: ‘scribes who know the language make a different set of errors’, he writes in the conclusions. His contribution is undeniably an example of the kind of analysis to carry out if we want to reconstruct the personality of the scribe. John Bradley’s (Royal Holloway, University of London) article, The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes, takes us again to Rome, and to the year 240 A.D.; a fundamental link with Alan Taylor Farnes’ article is the attention devoted to the profile of the scribes. In fact, here the author proposes a new, original interpretation of a fresco found in Rome in 1919 and excavated by Goffredo Bendinelli in 1920 and 1921. After a short synopsis, the author describes the different interpretations of the previous scholars, generally in the light of the Gnostic attributions. Then he transcribes the text of the inscription and gives his English translation. One of the core points of the article is the conjecture fratrs for fratri(bus) and the proposal of a collective subject for the

Introduction  11

tomb of the Aurelii. Bradley goes on with an association between decoration and deceased, and a reflection on fratres as a clue for the existence of a collegium. He analyzes different literary sources in order to understand what kind of collegium this one could be, and he proposes that the fresco reproduces a collegium of scribes. In this case, the pictures of the hypogeum would be the first auto–celebration of the work of a group of people, united by the same profession of scribae or librarii. Veronica De Duonni (Università degli Studi di Salerno) is the author of the fourth contribution of this section: The Library and the Scriptorium of the Abbey of Montevergine in the 12th and 13th Century: Presences and Absences. Here comes again the word fratres as scriptores, but in a different context: De Duonni takes us to a medieval abbey in Southern Italy. Her contribution, in particular, focuses on the existence of a library and a scriptorium in this monastic center. In the building, in fact, the books are incredibly few and there is no extant catalogue. The only documents that can help us are some manuscripts: the ms. 1 (De vita et obitu sancti Guilielmi Confessoris et heremite), the mss. 2, 3 and 4, and the Vat. Lat. 5100 and Vat. Lat. 7606. Unfortunately, none these documents explicitly mentions the existence of a scriptorium, but it is noteworthy that sometimes the fratres sign themselves as scriptores, so the presence of a scriptorium cannot be completely excluded. Through those old parchments, we can glimpse ‘the whole life of a distant world’, as the author writes quoting the words of a fundamental study on the archive of the abbey of Montevergine. The contribution of Olivia Montepaone (Università degli Studi di Milano) also talks about the possession of books, but some centuries later and in the context of Humanism. The title of her article makes it immediately clear: Apocolocyntosis, Codex V and the Manuscript of Hadrianus Junius. Hadrianus Junius was a Dutch Humanist; he lived in the 16th century and he is mainly known for having been the first Scholar able to identify the Greek title (Apocolocyntosis) of Seneca’s work otherwise known as Ludus de morte Claudii. The author of the article investigates which was the manuscript possessed by Hadrianus Junius, generally identified with the Valentinianensis 411 (V in modern critical apparatus), a 9th–10th– century codex kept in the monastery of Saint Amande near Valenciennes. In order to conduct the investigation, Montepaone convincingly analyzes Junius’ Annotationes (1557) to Celio Secondo Curione’s edition of Seneca’s Opera Omnia. She compares codex V readings with those that emerges from Iunii Annotationes. After a thorough cross analysis, she concludes that the manuscript in the hand of the Dutch Scholar was not the Valentinianensis 411, but a copy of it, in the hand of Junius thanks to the Humanist and burgomaster of Utrecht, Johan van Cuyck.

12  Roberta Berardi, Nicoletta Bruno, Luisa Fizzarotti With the last contribution of this section, the one by Cristiana Roffi (Universität zu Köln; Alma Mater Studiorum – Università di Bologna), we remain in the context of 16th – century Northern Europe, but in Germany; the title of her article is The Textual Transmission of Ovid’s Metamorphoses During the Medieval Age: the Example of Germany. Cristiana Roffi focuses on the rescript of the episode of Narcissus and Echo (Ov. Met. 3.339– 510) in Jörg Wickram’s work published in 1545. There is something more: Wickram’s text is a remake of Albrecht von Halberstadt’s 12th –13th –century corrupted work. Of this text, only five fragments survived, from the so–called manuscript of Oldenburg (probably 12th –13th century). The author firstly presents the reception of Ovid’s Metamorphoses in the context of medieval Germany. Secondly, she analyzes Wickram’s remake from the perspective of the change of meter (from Latin hexameters to octosyllabic verses), exegetical inserts and omissions, and tendency to moralization. After her analysis, she concludes that the German author had a basic knowledge of Latin and that he worked on his Ovid’s remake thanks to a glossed manuscript and perhaps thanks to the aid of a schoolteacher.

Stephen J. Harrison

Figured Books: Horatian BookRepresentations  Introduction This paper1 considers some ways in which Horace’s own poetic books are personified and treated symbolically in his work, and how the particular images of personification and metaphors used reflect both appropriate literary and cultural traditions and the physical form of ancient book–rolls. Two of the three instances discussed here appear in the last poems of poetry–collections, a traditional and appropriate location for such meta–poetical elements.

 Odes 1.38: book as garland The linear reader who has reached the end of Odes 1.38 has already been presented with the image of poetry as garland in the brief ode to Lamia (1.26): Musis amicus tristitiam et metus tradam protervis in mare Creticum portare ventis, quis sub Arcto rex gelidae metuatur orae, quid Tiridaten terreat, unice securus. o quae fontibus integris gaudes, apricos necte flores, necte meo Lamiae coronam, Piplei dulcis. nil sine te mei prosunt honores: hunc fidibus novis, hunc Lesbio sacrare plectro teque tuasque decet sorores.





 1 My thanks to Nicoletta Bruno and the committee of Prolepsis for their kind invitation to a most enjoyable conference in Bari. https://doi.org/10.1515/9783110632590-002

14  Stephen J. Harrison As friend to the Muses let me consign Sadness and dread to the violent winds To carry to the Cretan sea, uniquely unconcerned What king of some frozen shore up under the Bear Is to be feared, or what may terrify Tiridates. Sweet Muse of Pimpla, who joys in untouched fountains, Weave together sunny flowers, Weave a garland for my dear Lamia. My achievements are nothing without you: It is fitting for you and your sisters To consecrate him with a new lyre, With the plectrum of Lesbos.

Here interpreters largely agree that the garland (coronam) of line 8 represents the current poem;2 Lamia is being celebrated with the ‘Lesbian plectrum’ of a lyric poem in Alcaic metre, and the image of the ‘crown of song’ is a well–established one in previous literature, especially in Pindaric lyric.3 Likewise, in Odes 1.7, where the praise of Athens is compared to placing a Panathenaian garland of olive–leaves on the brow of Pallas (1.7.5–7 ‘sunt quibus unum opus est intactae Palladis urbem / carmine perpetuo celebrare et / undique decerptam fronti praeponere olivam’), it is possible to see a metapoetic metaphor4, perhaps even a reference to a poem on Attica by Euphorion5 or perhaps the various poems on Athens by Pindar (fr.74–77 S/M). It has similarly been argued that the garland of Odes 1.38 represents ‘a general statement of aesthetic principles’6 and indeed the qualities of the poetic book to which it forms the finale.7 Thus, the linear reader of Book 1 is well prepared for this kind of interpretation of its last poem:

 2 See e.g. Davis (1991) 114–18. 3 See Nünlist (1998) 215–21. 4 See Davis (1991) 191–93. 5 So Nisbet and Hubbard (1970). 6 Davis (1991) 118 following Fraenkel (1957) 292–99 and Cody (1976) 15–44. 7 See especially Cody (1976) 15–55 and Lowrie (1997) 164–75 (and in summary Fedeli [2012] 731); more sceptical are Nisbet and Hubbard (1970), West (1995) 190–94, Syndikus (2001) 332–34, and Mayer (2012) 206–7.

Figured Books: Horatian Book-Representations  15

Persicos odi, puer, apparatus, displicent nexae philyra coronae, mitte sectari, rosa quo locorum sera moretur. simplici myrto nihil allabores sedulus curo: neque te ministrum dedecet myrtus neque me sub arta vite bibentem.



Persian paraphernalia, my boy, I disapprove; No pleasure to me are bast–fastened garlands; Don’t bother to pursue the select places where The late rose lingers. My mandate: take care to add no elaboration To simple myrtle – for you as server Myrtle is fully fitting, and for me too, as I drink Under the dense–woven vine.

As just noted, the symposiastic preparations of the poem (a common epigrammatic ‘topos’) have been examined in great detail for meta–poetical content. Here I would like to add further arguments in this direction, suggesting in particular that Greek epigram has more of a role here and that this is appropriate for such a short poem. It seems plausible that this poem looks back to Callimachus Epigram 28 Pf. = Anth. Pal. 12.43: Ἐχθαίρω τὸ ποίημα τὸ κυκλικόν, οὐδὲ κελεύθῳ χαίρω τίς πολλοὺς ὧδε καὶ ὧδε φέρει, μισῶ καὶ περίφοιτον ἐρώμενον, οὐδ᾿ ἀπὸ κρήνης πίνω· σικχαίνω πάντα τὰ δημόσια. Λυσανίη, σὺ δὲ ναιχὶ καλὸς καλός—ἀλλὰ πρὶν εἰπεῖν τοῦτο σαφῶς Ἠχώ, φησί τις “ἄλλος ἔχει.” I hate the poem that belongs to the normal round, nor do I Take pleasure in the road which carries many here and there. I loathe too the lover who goes around, and I do not drink from the spring; I loathe all things that are public property. Lysanias, you are indeed handsome, handsome – but before Echo has clearly Repeated this, someone says ‘Another possesses him’.

The verbal link odi / Ἐχθαίρω has indeed been noted,8 but the epigram’s parallel between poetic and erotic preference may also influence the ode; both poems are  8 But only by Cody (1976), not taken up by later commentators.

16  Stephen J. Harrison addressed to a young male who is a potential sexual partner and whose looks are commented on, and some recent scholarship has stressed the homoerotic colour of 1.38 (plausibly in my view).9 The ode is much the same brief length as the epigram, often a sign in Horace that an epigrammatic model is in play [1.5, 1.20, 1.23, 1.30, 3.13, 3.18, 3.20, 3.22, 3.26, 4.10]. Plausible too is the view that Persicos … apparatus looks back to the rejection of the Persian chain (measurement of length) as the criterion for poetic quality in the polemical preface to Callimachus’ Aetia (fr.1 Pf.17–20):10 αὖθι δὲ τέχνῃ κρίνετε,] μὴ σχοίνῳ Περσίδι τὴν σοφίην· μηδ᾿ ἀπ᾿ ἐμεῦ διφᾶτε μέγα ψοφέουσαν ἀοιδήν τίκτεσθαι· βροντᾶν οὐκ ἐμόν, ἀλλὰ Διός. Now judge poetry by its art, not by the Persian chain, And do not seek from me the birth of a song Which makes a loud noise; thundering Belongs not to me, but to Zeus.

Horace’s poem joins Callimachus’ in rejecting ‘Persian’ literary quantity; it is not impossible that there is an allusion here to Choerilus’ epic Persica, written at the end of the 5th C BCE, which would represent the kind of lengthy poetry here targeted; perhaps the rare term philyra (lime–bast, a fibrous membrane used as a framework for elaborate garlands) alludes to another poetic symbol in some lost Hellenistic text to match Callimachus (its prominence is otherwise strangely inexplicable). The references to the rose and myrtle can also be argued to show meta–poetical colour, again from a Hellenistic source. Echoes of the late Hellenistic epigrammatist Meleager have been suggested elsewhere in the Odes11, and it has been very fleetingly proposed12 that 1.38 alludes in its meta–poetical character to Meleager’s famous introduction to his Garland anthology of poets, Anth. Pal. 4.1. This hint can be followed up in more detail; it is not unlikely that Meleager’s prefatory poem might influence Horace’s concluding ode if both are programmatic, as argued here. I cite its opening (1–4):

 9 See especially West (1995), neatly linking this notion with Horatian imitation of Alcaean pederastic poetry. 10 Noted by (Cody) 1976, Davis (1991) and Lowrie (1997) [see n. 7 above]. 11 So La Penna (1997). 12 Davis (1991) 126.

Figured Books: Horatian Book-Representations  17

Μοῦσα φίλα, τίνι τάνδε φέρεις πάγκαρπον ἀοιδάν; ἢ τίς ὁ καὶ τεύξας ὑμνοθετᾶν στέφανον; ἄνυσε μὲν Μελέαγρος, ἀριζάλῳ δὲ Διοκλεῖ μναμόσυνον ταύταν ἐξεπόνησε χάριν … Dear Muse, to whom do you offer this full–fruited song? Or who is he who wrought this garland of singers? Meleager accomplished it, and laboured at this graceful object As a keepsake for the distinguished Diocles …

Here there are a number of elements which seem to be picked up in the Latin poem. Both poems are addressed to a male friend, potentially an erotic partner, and both involve a meta–poetical garland where particular types of flowers are woven in. The Horatian garland, composed simply of myrtle, is evidently different from its multifloral Meleagrian counterpart (Meleager’s 58–line poem, enormous for an epigram, goes on to list twenty–five poets, all with their counterpart flower), but may pick up an element in the Greek epigram. As has been noted,13 later on in Meleager’s poem we find myrtle as the flower linked with Callimachus (21–2 ἡδύ τε μύρτον / Καλλιμάχου), and given the double Callimachean allusion argued for above this is likely to be significant: a recommendation of the flower of Callimachus for a garland in this poem is consistent with its function as a Callimachean manifesto for the Odes. Thus Meleager’s famous prefatory poem is echoed in Horace’s concluding poem, and both reflect the nature of their collections: Meleager’s multi–author anthology incorporates a wide variety of flowers, Horace’s the myrtle of Callimachus, pointing to the key importance of Callimachean aesthetics for the polished and compressed texture of the Odes. This is not the only occasion on which we find a symbolic Callimachean manifesto at the end of a Horatian poetry–book. In Odes 2.20 (shortly to be discussed below) the poet’s swan flies on a wing which is unusual (1–2 non usitata … / penna), echoing Callimachus’ programmatic rejection of the commonplace in poetry in Epigram 28 (see above). and the poet will be inuidiaque maior (4), ‘superior to envy’, translating Callimachus’ self–description at Ep. 21.4 Pf. κρέσσονα βασκανίης, while the last poem of the first book of Satires presents an ironic version of Apollo’s intervention in the Aetia–preface, in which Quirinus intervenes to prevent the poet from writing Greek poetry rather than Latin (1.10.31–35).

 13 Davis (1991) 126.

18  Stephen J. Harrison

 Book(s) as military baggage . Epistles 1.13 – delivering the Odes to Augustus Vt proficiscentem docui te saepe diuque, Augusto reddes signata uolumina, Vinni, si ualidus, si laetus erit, si denique poscet; ne studio nostri pecces odiumque libellis sedulus inportes opera uehemente minister. Si te forte meae grauis uret sarcina chartae, abicito potius quam quo perferre iuberis clitellas ferus inpingas Asinaeque paternum cognomen uertas in risum et fabula fias. Viribus uteris per cliuos, flumina, lamas. Victor propositi simul ac perueneris illuc, sic positum seruabis onus, ne forte sub ala fasciculum portes librorum, ut rusticus agnum, ut uinosa glomus furtiuae Pyrria lanae, ut cum pilleolo soleas conuiua tribulis. Ne uolgo narres te sudauisse ferendo carmina quae possint oculos aurisque morari Caesaris; oratus multa prece, nitere porro. Vade, uale, caue ne titubes mandataque frangas. As I instructed you often and long as you set off, Vinnius, please render up my sealed volumes to Augustus, If he is healthy, if he is happy, if lastly he asks for them; Please don’t err in your support of me and by being over–attentive Bring distaste on my books as my helper by violent effort. If the heavy pack of my page happens to chafe you, Throw it away rather than fiercely dumping your pack–saddles In the place where you are bid to go, and turn your father’s name Of Ass into ridicule and a funny story. Use your strength through hills, rivers, bogs. When you achieve your plan and arrive there, Please preserve your burden so as not to carry my bundle of books Under your armpit, like a peasant with a lamb,







Figured Books: Horatian Book-Representations  19

Like drunken Pyrria with stolen balls of wool, Like a diner with fellow–tribesmen with his cap and shoes. Do not state in public that you sweated carrying poems Which could occupy the eyes and ears of Caesar. Press on after this volume of requests; Go on, farewell, take care not to stumble and break your assignment.

The key point of humour in this poem is the ironic contrast between the lightness of the burden (the three papyrus book–rolls of the Odes,14 no doubt carried in a capsa or book–box) and the melodramatic instructions given by the poet for their transport to Augustus, presumably in Rome, from some non–Roman location, perhaps the poet’s villa mentioned in the next epistle (1.14); distance and hardship are stressed in the poem, but are unlikely to have been great in practice. As Nisbet has attractively argued,15 this ironic contrast is reinforced if the Vinnius of Horace’s poem is the Augustan centurion Vinnius Valens mentioned by the elder Pliny (HN 7.82), famous for outstanding feats of strength such as lifting up waggons; there would then be considerable humour in the contrast between his likely massive physique and the lightness of the burden with which he is here entrusted. His military character would also explain the repeated military metaphors of the poem; commentators have noted victor propositi (11),16 but I would here like to add to these the general element of carrying baggage in the poem. I argue that this ironically reflects the large packs traditionally carried by Roman legionaries as well as playing on the common analogy between soldiers and beasts of burden. At Vegetius 1.19 in an account of military training we find an account of the heavy packs carried by the imperial Roman army: Pondus quoque baiulare usque ad LX libras et iter facere gradu militari frequentissime cogendi sunt iuniores, quibus in arduis expeditionibus necessitas imminet annonam pariter et arma portandi. Nec hoc credatur esse difficile, si usus accesserit; nihil enim est, quod non adsidua meditatio facillimum reddat. Quam rem antiquos milites factitauisse Uergilio ipso teste cognoscimus, qui ait [G.3.346–8}: non secus ac patriis acer Romanus in armis iniusto sub fasce uiam cum carpit, et hosti ante exspectatum positis stat in agmine castris.

 14 It is clearly the Odes – see Mayer (1994) 4, Cucchiarelli (2015) 150. 15 Nisbet (1959) 75–76 = (1995) 4–5. 16 E.g. Mayer (1994) 202.

20  Stephen J. Harrison Younger soldiers should very often be made to carry a weight of up to sixty pounds and make a march at military pace; for they have an impending need to carry both their corn and weapons on strenuous expeditions. And let this not be thought to be difficult, if the habit comes to them; for there is nothing which continual practice cannot render easy indeed. We know that the soldiers of old did this repeatedly through the testimony of Vergil himself, who says: Just as when the fierce Roman in his ancestral arms Makes his way under an oppressive burden, and stands In ranks having made camp before the enemy expects.

Cicero (Tusc. 2.37) and Horace elsewhere (Epod. 9.12) similarly punt the stress on the bulky rations and fortification – stakes carried by Roman legionaries alongside their weapons. This emphasis on the amount carried by soldiers led to their nickname of ‘Marius’ mules’, presumably coined not long after Marius’ major army reforms of 107 BCE. Marius himself is said to have introduced the relief of the furca, the forked staff on which some items were tied to make them easier to carry on the march (Frontin. Str. 4.7): C. Marius recidendorum impedimentorum gratia, quibus maxime exercitus agmen oneratur, vasa et cibaria militis in fasciculos aptata furcis imposuit, sub quibus et habile onus et facilis requies esset: unde et proverbium tractum est ‘muli Mariani’. Gaius Marius, in order to cut back the baggage, with which the army was especially laden, placed the cooking vessels and rations of the soldier on forked sticks, tied in bundles, with which the burden was handily carried and rest easily taken: this is the origin of the proverb ‘the mules of Marius’.

The proverb seems to have been current in the Augustan period (Festus p. 134 Lindsay), and can certainly inform the interpretation of Horace’s poem. We are told that Vinnius’ paternum cognomen is Asina (8), ‘female ass, jenny’, a common enough cognomen at Rome:17 whether this was a name which actually belonged to his father only or to Vinnius himself,18 it is clearly played on in the way that Vinnius is compared to a beast of burden, especially the idea that he will ‘dash down his panniers’ (8 clitellas impinges), a more violent version of the usual way in which mules are allowed to lay down their burdens which we see at Satires 1.5.47 hinc muli Capuae clitellas tempore ponunt, ‘from this point the mules put down their panniers at Capua in good time’. Vinnius the military son of the Ass is appropriately and wittily characterised as a mulus Marianus.  17 Cf. e.g. Graeus Cornelius Scipio Asina, cos. 260 BCE. 18 See the discussion of McGann (1963).

Figured Books: Horatian Book-Representations  21

But let us return to the presentation of Horace’s poetic books as military baggage. This is strongly suggested in sarcina (6), the standard word for the Roman military pack (OLD s.v. 1b) and a technical term found only here in Horace; the weight it implies (see above) specifically and amusingly contrasts with the light charta (6) of Horace’s papyrus rolls. The comic instruction to Vinnius to jettison this baggage (7 abicito) en route if it rubs him rather than make a faux pas in presenting it to Augustus in the wrong way is a parodic version both of the real discomfort of military packs and arms (cf. Plaut. Trin. 595–6) and real military emergencies where such action might be strategically necessary, for example, in Marius’ surprise attack on Capsa in Numidia (Sall. Iug. 91.2) where he requires packs to be dropped, omnibus sarcinis abiectis, or a Pompeian retreat at Ilerda (BCiv. 1.159.2), where packs are thrown away, proiectis sarcinis. Even the characterisation of the poetry–books as fasciculum (13), ‘little bundle’, might be a version of the legionary pack, called fascis by Vergil at Georgics 3.347 above and Quintilian19 and fascisculus by Frontinus (above again). Overall, we can see how much the idea that poetry–books are a strange species of military pack for the mulus Marianus Vinnius contributes to the interpretation and humour of this poem.

 Book as a boy . Epistles 1.20.1–18 Vertumnum Ianumque, liber, spectare uideris, scilicet ut prostes Sosiorum pumice mundus. Odisti clauis et grata sigilla pudico, paucis ostendi gemis et communia laudas, non ita nutritus. Fuge quo descendere gestis; non erit emisso reditus tibi: 'Quid miser egi? Quid uolui?' dices, ubi quid te laeserit; et scis in breue te cogi, cum plenus languet amator. Quodsi non odio peccantis desipit augur, carus eris Romae donec te deserat aetas; contrectatus ubi manibus sordescere uolgi coeperis, aut tineas pasces taciturnus inertis

 19 Quint. Inst. 11.3.26 militare iter fascemque et uigilias.





22  Stephen J. Harrison aut fugies Vticam aut uinctus mitteris Ilerdam. Ridebit monitor non exauditus, ut ille qui male parentem in rupes protrusit asellum iratus; quis enim inuitum seruare laboret? Hoc quoque te manet, ut pueros elementa docentem occupet extremis in uicis balba senectus.



Book, you seem to be contemplating Vertumnus and Janus’ temples, So that, I suppose, you can go on sale cleaned up by the pumice of the Sosii. You reject the keys and seals which are dear to the chaste, You complain at being shown only to a few and praise what is common to all, Quite against your upbringing. Go then where you want to on the way down, There will be no return for you once you leave. ‘What have I done, poor me? What did I mean?’, you will say, when some hurt comes to you; and yet you know How to roll yourself up when your lover has had enough and is tired. But if the augur is not without wisdom owing to his dislike of your offence, You will be dear at Rome until your youth deserts you; When you begin to become shabby, pawed by the hands of the mob, Either you will fall silent and feed the ill–educated lice, Or you will run away to Utica or be sent bound to Ilerda. He who warned you without your heeding will laugh, just like the man Who pushed the disobedient ass over the cliff in his anger; For who would struggle to save someone against his will? This too lies in store for you, that stammering old age Will overtake you teaching boys their letters at the ends of city blocks.

Since Fraenkel’s analysis of this poem, it has been clear that the poetry–book of the Epistles is here being compared to a runaway slave and his subsequent career of prostitution; this was indeed one of the few avenues open to a fugitive, in contrast with the more respectable business opportunities available to a dutiful freedman who had waited for his owner to free him and become his patron. In the coy 1950s Fraenkel could only bring this out subtly; Mayer’s 1994 edition is more forthright. I think this is a persuasive interpretation and would like to add further details which reinforce it, bringing out the continuing analogy in this section between boy (*lber) and book (liber). As Mayer notes in general terms, the play in this poem partly goes back to Callimachus Epigr. 28 Pf., already cited above, where popular poetry and promiscuous lovers are critically compared – cf. lines 1–4: Ἐχθαίρω τὸ ποίημα τὸ κυκλικόν, οὐδὲ κελεύθῳ χαίρω τίς πολλοὺς ὧδε καὶ ὧδε φέρει, μισῶ καὶ περίφοιτον ἐρώμενον, οὐδ᾿ ἀπὸ κρήνης πίνω· σικχαίνω πάντα τὰ δημόσια.

Figured Books: Horatian Book-Representations  23

I hate the poem that belongs to the normal round, nor do I Take pleasure in the road which carries many here and there. I loathe too the lover who goes around, and I do not drink from the spring; I detest all things that are public property.

In Horace’s poem odisti (3) surely specifically picks up Ἐχθαίρω like odi at Odes 1.38.1 (see above), while communia echoes δημόσια, down to its neuter plural nominalisation: the boy/book has been produced by the exclusivist Callimachean Horace but is rebelling against his background. A similar ambiguity can be found in the phrase non erit emisso reditus tibi (6). On the surface this looks to the fugitive slave who cannot return to his master’s house, but it also suggests the literary publication of the poetry–book. At Ars Poetica 386–90 Horace gives his famous advice about not publishing too soon: Siquid tamen olim scripseris, in Maeci descendat iudicis auris et patris et nostras, nonumque prematur in annum membranis intus positis; delere licebit quod non edideris; nescit uox missa reuerti. But if you ever complete some writing, Let it descend into the ears of the critic Maecius, And your father’s, and mine, and let it be suppressed Till its ninth year, all parchment put away; you can erase what you have not published, but a word once dispatched knows no return.

Both passages evoke the Homeric image of the hasty word which passes the boundary of the mouth and cannot be recalled,20 found again in this same book at Epist. 1.18.71 et semel emissum volat irrevocabile verbum, but transfer the quality of the spoken word to that of the written word once divulged to the public. Like the escaped slave, the published book cannot return to its previous status. On in breve te cogi (8) Mayer sees the relationship to the slave as unclear; as he says, the book will clearly be able to contract itself as a papyrus roll rolling up, useful for transport in a capsa as we saw above in Epistles 1.13; when the reader has had enough (plenus), the book can be conveniently put away. But given the post–coital context, one could suggest that this is the young rent–boy submissively making himself smaller in the bed when his lover wants the space to sleep after gaining sexual satisfaction (cum plenus languet amator); the small size of the stone beds in the Pompeii lupanar might suggest that this was a common

 20 Cf. Iliad 4.350, 14.83, Odyssey 1.230, 5.22, 19.492, 21.168, 23.70.

24  Stephen J. Harrison issue in Roman brothels. Similarly, ambiguous is 10 donec te deserit aetas. Ostensibly this is the traditional idea that no–one will find the boy attractive after he has lost his youthful looks, but there is also a possible application to the book. In Epistles 2.1.39–42 we find a debate on whether age by itself is a measure of poetic quality: ‘est vetus atque probus, centum qui perficit annos.’ quid, qui deperiit minor uno mense vel anno, inter quos referendus erit? veteresne poetas an quos et praesens et postera respuat aetas?



‘He is old and good, who has endured a hundred years’. What of the poet who perishes a month or year short of that, With whom is he to be classified? With the old poets, Or with those that both the present and future age will reject?

The aetas of Epist.1.20.10 can cover the literary fashion of the book’s own time as well as the boy’s physical attractions: both are ephemeral and impermanent.

 Conclusion I hope to have shown that these three Horatian poems (Odes 1.38, Epistles 1.13 and Epistles 1.20) deploy metaphors for poetry–books, whether their own books or others, which are rich and significant in a number of ways. In Odes 1.38, the last poem of a book within the collection, apparently referring merely to appropriate headgear for a symposium, in fact appropriately takes up the metaphor of the garland which had been used by Meleager in the opening poem of his epigrammatic collection, showing the affinity of this Horatian poem and Horatian brief lyric as a whole with the Hellenistic epigrammatic genre; this affinity is reinforced by the use of Callimachean metapoetic language. In Epistles 1.13, the three books of the first collection of Odes are comically described as a form of military baggage, a very light and fragile load carried by a messenger who is mostly likely a soldier of enormous physique; this image and the messenger’s father’s name Asina recall the proverbial reference to Roman legionaries as ‘Marius’ mules’ for their pack–carrying capacities. Finally, in Epistles 1.20 the image of the book of Epistles as a runaway slave is explored in amusing ways which rely both on Callimachean aesthetics and the sordid details of Roman prostitution. In all cases we find an awareness of the poetic book as physical artefact combined with symbolic description of its quality and poetic affinities, set in contexts of Roman realism.

Georgios Taxidis

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20 Horace’s Epistles 1.20,1 the closing epistle of the first book of the author’s latest work, has received significant critical attention by scholars. The readings that prevail tend to personify the book as a slave boy, and then in joint consideration with the sphragis at the end of the poem, relate this personificatiοn to Horace himself.2 This relation has been read – mostly as a result of Horace’s humble self– presentation – on the one hand as expressing ‘self–deprecating humor’, which fits the philosophical nuance of the Epistles (Harrison [2007] and [1988]; Oliensis [1995]) especially in comparison to the grand claims of the poet’s immortality in Odes 3.30 (Horace’s sphragis), while on the other hand, the poet of the Epistles achieves immortality by transforming himself into his own text. This process is similar to the one followed in Odes 2.20, where there is a swan–transformation, making him a curricular author (Pearcy [1994]). Moreover, Epistles 1.20 has been read with regard to the poet’s interplay of different ideas such as poetry, ethics and philosophy.3 This paper however attends to the writing–material–items used to describe the book in the poem’s opening lines (vv. 1–6), an aspect which – as such – has been to date neglected or only partially treated. Our purpose is firstly to elucidate the ways in which these terms function per se, regardless of the book’s image as a slave boy. For that reason, the linguistic, semantic and contextual attributes of these terms in Epistles 1.20 will be discussed, critically assessed and compared to the uses made by other authors and other works of Horace. Then this analysis will proceed with a more focused comparison between these terms and Horace’s own self–presentation at the end of the poem (signum–σφραγίς), based on a close reading of the poem itself. Overall, this critical reading proposes to illustrate the ways in which these terms function in relation to the seal of Horace, which is also

 1 I would like to deeply thank the organizers of the Prolepsis’ First International Postgraduate Conference ‘Cupis volitare per auras’, as well as the reviewers for their useful notes. Special thanks to Professor Dr. Claudia Schindler (Hamburg University) for the fruitful discussion and her advice on the subject of Horace’s sphragis and the Associate Professor of Latin Dr. Andreas Michalopoulos (University of Athens) for his comments on my draft talk. 2 Harrison (2007) and (1988), Oliensis (1995), Pearcy (1994), Kilpatrick (1986). 3 Trinacty (2012). https://doi.org/10.1515/9783110632590-003

26  Georgios Taxidis a material device that is used in documents and aims to contribute to a new interpretation of the poem as a whole. Vortumnum Ianumque, liber, spectare videris, scilicet ut prostes Sosiorum pumice mundus. odisti clavis et grata sigilla pudico; paucis ostendi gemis et communia laudas, non ita nutritus, fuge quo descendere gestis. non erit emisso reditus tibi. ‘quid miser egi? quid volui?’ dices, ubi quid te laeserit, et scis in breve te cogi, cum plenus languet amator. quodsi non odio peccantis desipit augur, carus eris Romae, donec te deserat aetas: contrectatus ubi manibus sordescere volgi coeperis, aut tineas pasces taciturnus inertis aut fugies Uticam aut vinctus mitteris Ilerdam. ridebit monitor non exauditus, ut ille qui male parentem in rupes protrusit asellum iratus; quis enim invitum servare laboret? hoc quoque te manet, ut pueros elementa docentem occupet extremis in vicis balba senectus. cum tibi sol tepidus pluris admoverit auris, me libertino natum patre et in tenui re maiores pinnas nido extendisse loqueris, ut, quantum generi demas, virtutibus addas; me primis urbis belli placuisse domique, corporis exigui, praecanum, solibus aptum, irasci celerem, tamen ut placabilis essem. forte meum siquis te percontabitur aevum, me quater undenos sciat implevisse Decembris, conlegam Lepidum quo duxit Lollius anno.4











You seem, book, to have eyes only for Vertumnus and Janus; you’ re longing (it’s obvious) to stand on the corner tidied up by the pumice of the Sosii brothers. You resent the lock and seal decent books are grateful for; you groan at being shown to a few, you sing the praises of the public, though you weren’t brought up that way. Run away where you’re so eager to do; once I let you out, there’s no returning. ‘What have I done, wretch that I am? What did I want?’ you’ll say, when something hurts you, or you feel the pinch when a sated lover’s interest droops. If my prophetic powers aren’t diminished be my disgust at your misconduct, you’ll be treasured in Rome until youthful bloom deserts you;

 4 The presented text derives from the edition Q. Horatius Flaccus opera edidit Friedrich Klingner, Leipzig 1959.

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  27

once you start to show signs of wear fingered by vulgar hands, you ‘ll feel feed lazy booklice, or run off to Utica, or be tied up and shipped to Ilerda. The adviser you ignored will have the last laugh, like the man who pushed his disobedient donkey onto the rocks; in anger why struggle to save someone who doesn’t want to be saved? This too awaits you: babbling old age will overtake you in some remote corner teaching boys their ABC’s. When the evening5 sun has drawn more ears your way, tell them that I, born by a freedman father and of slender means, spread my wings wide beyond my nest– crediting to my character what you subtract from my birth; tell them I found favor with the city’s leaders in war and in peace, that I was short, prematurely gray, suited to sunshine quick to anger, but readily placated. If someone should happen to ask you my age, let him know that I completed 44 Decembers the year Lepidus was brought in to be consul along with Lollius. (tr. Oliensis)

The book (liber)6 is the addressee of the epistle. Therefore, it is being addressed according to Horace’s epistolary style at the beginning of it. According to Thesaurus Linguae Latinae this lemma is to be recorded under the category: species externa, ornatus (= outer surface, decoration).7 So, this means that the word does not refer to a book as a piece of literature in general, but only to a book in its pure material form, as an item of specific dimensions. Moreover, it refers to the book’s outer presence. The word liber is used in this sense for the first time by Cicero, in De oratore 2.61: in philosophos vestros si quando incidi, deceptus indicibus librorum, qui sunt fere inscripti de rebus notis et inlustribus, de virtute, de iustitia, de honestate, de voluptate, verbum prorsus nullum intellego.

 5 For the discussion about the precise time of sol tepidus cf. Fedeli (1996) 18–19 and Ciccarelli (1997) 69 (n. 25). 6 At this point we would like to make the analysis’ course clear. Each time there will be a writing–material’s term picked out from the text and been set as liber here. The discussion about it will immediately follow. 7 Cf. TLL 7.2.1273.63–78, s.v. liber. The pure material form of the book at this point is further enhanced by the subdivisions of species externa at TLL’s article into following meanings of liber: a) capsae, fasces, sigilla (TLL 7.2.1273.78–1274.3), b) materia (TLL 7.2.1274.3–11), c) scriptura (TLL 7.2.1274.11–20) and d) partes voluminis (TLL 7.2.1274.20–28). Thus, the distinction that we propose, between liber as material item and as piece of literature should not be viewed as general, but within the developing argument.

28  Georgios Taxidis if I ever fall in with the philosophers, deluded by the titles to their books, as they generally profess to be written on well–known and plain subjects, as virtue, justice, probity, pleasure, I do not understand a single word of them. (tr. Watson)

The verb inscribere (= to entitle a book), the adjective illustris (= well known) as well as the established title–forms (de virtute etc.) indicate the book’s cover (indicibus). Antonius admits to have been deceived (deceptus) by the very interesting issues that philosophical books usually address. His deception is based on the fact that he does not understand a word of what he has read (verbum prorsus nullum intellego). The use on the one hand of vocabulary relating to the book’s material features (indicibus librorum) and the usual title forms (de virtute etc.), and on the other hand the use of the diction verbum, which refers to language expression, establishes a close relation between the book’s materiality and its content. As titles are inscribed on the book cover, the external–material–book form proves to be just a fallacy, since the philosophical treatments do not meet the expectation created by the title. But what Antonius was really deceived by and what prompted him to open the book and read it was the fact that the issues were illustris (= well known). So, the material artefact, brought thus in relation with its remarkable content, proves also to be deceiving. postea vero, quam Tyrannio mihi libros disposuit, mens addita videtur meis aedibus (…) nihil venustius quam illa tua pegmata, postquam mi sillybis libros inlustrarunt. (Cic. Att. 4.8.2) and now that Tyrannio has put my books straight, my house seems to have woken to life … Those shelves of yours are the last word in elegance, now that the labels have brightened up the volumes. (tr. Bailey)

In this exchange of letters between Cicero and Atticus it is clear how the grammarian Tyrannio helped Cicero order his library. Once again, we come across the topic of books, which are prominently placed on the shelf (pegmata), so their titles, written on the sittybae or sillibae (= thistles) can be read without difficulty. In Cicero’s letter the emphasis on the book’s outer presence is being enhanced by a reference to the shelves, where the book is going to be placed. The pegmata (= shelves) are elegant (nihil venustius quam illa tua pegmata). So, the high quality of the book shelf itself allows the book placed on it to shine (postquam mi sillybis libros inlustrarunt). This seems to cover the same semantic field as in Epistles 1.20. Horace's book will stand on the shelves in the same manner as Cicero's. It is going to present itself pumice mundus (‘well illustrated’ Hor. Epist. 1.20.2), ready to be

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  29

sold in the bookstores of the Sosii.8 The reputation of this publishing house correlates with the high quality of Cicero’s book shelves (nihil … venustius). In both texts it is again the book’s outer presence that defines the diction. chartae regiae, novi libri, novi umbilici, lora, rubra membranae, derecta plumbo et pumice omnia aequata. (Catull. 22. 6–8) new rolls of royal papyrus, new bosses, scarlet thongs and parchment covers, the whole lead–ruled and levelled off with pumice. (tr. Lee)

Catullus appeals to Varus in order to inform him about Suffenus. He is homo venustus et dicax et urbanus (‘charming, sarcastic and fine’, Catull. 22.2) and a very productive poet (vv. 4–5 puto esse ego illi milia aut decem / aut plura perscripta). Now Suffenus no longer uses a palimpsest in order to write down his poems, but writing materials of better quality such as royal papyrus (v. 6 chartae regiae), fresh brand–new books (v. 6 novi libri) with a new wooden cylinder around which the papyrus is wrapped (v. 8 novi umbilici). These upgraded writing materials are being accompanied by equivalent ornament, such as parchment paper and leather strings (v. 6 lora rubra, membranae). A pumice–stone (v. 7 et pumice omnia aequata), which was used to smoothen the papyrus’ sheets completes the description of the high quality of the material used. However, Suffenus’ poetry is not being described in a similar positive manner. According to Catullus’ taste Suffenus as a poet is just a ‘countryman’ (v. 10 caprimulgus) and a ‘common laborer, a clown’ (v. 10 fossor).9 Once again, we come up with the juxtaposition between writing materials and literary content they express. Like the illustrious titles which inclined Cicero positively towards the content of the book the same occurs with Suffenus, since the reader who holds his fresh, polished book would expect a content of such value. So, it becomes clear that the book as material object does not necessarily guarantee good poetry. Proof to this is the distinction made by Catullus between the writing–material–terms used by Suffenus in the present and those used in the past. In the past he used a decadent parchment or papyrus, since palimpseston is the kind of paper sheet ‘from which the old writing has been erased for the

 8 Cf. Holder, Keller (1894). ad loc.: ‘Sosi illo tempore fratres erant bibliopolae celeberrimi’. 9 In Catull. 14 Suffenus was included in a series of bad poets (Catull. 14.19).

30  Georgios Taxidis purpose of writing upon it again’.10 The change in the writing material does not imply a change in his poetic manner. He will remain a poet of low quality, sic ut fit (‘as he was’, Catull. 22.5) regardless of the materials he uses to express his poetry. On the contrary Horace does not yet make a distinction between the past and the present regarding the quality of his papyrus sheets, but he clearly opposes the public position of the book to its prior state enclosed in the bookcase: scilicet ut prostes Sosiorum pumice mundus odisti clavis et grata sigilla pudico paucis ostendi gemis et communia laudas (Hor. Epist. 1.20.2–4) you’ re longing (it’s obvious) to stand on the corner tidied up by the pumice of the Sosii brothers you resent the lock and seal decent books are grateful for; you groan at being shown to a few, you sing the praises of the public (tr. Oliensis)

The book will seek publicity placed on the shelves of famous publishing houses of Rome, well–polished and ready to expose itself to the market (v. 2 prostes), while Horace kept it enclosed from the eyes of strangers, which could damage its yet unfinished nature. As the book is being referenced as enclosed in the bookcase,11 it is worth observing one more lexical item in Horace’s own poetry. In Satires 1.10 Cassius appears, a brilliant poet who has been burned with his own scripts and bookcases: quale fuit Cassi rapido ferventius amni ingenium, capsis quem fama est esse librisque ambustum propriis. (Hor. Sat. 1.10.62–64) Such a gift was Cassius’, who made rapid rivers seem slow. At the end, you remember, his funeral pyre was made of his books and bookcases. (tr. Bovie)

 10 OLD s.v. palimpsestus. 11 TLL 7.2.1273.78 capsae, fasces, sigilla. This entry of the Thesaurus Linguae Latinae functions in a supplementary way to the one mentioned above (p. 4, n. 7).

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  31

According to the rumour (fama) Cassius and his book reach a common end, that is physical extinction. Cassius as a person dies and the book as a material item has been burnt out. The word propriis underlines the fact that the books were his own and still in place, enclosed in the capsae. So, we can trace a strong correlation developing between the poet and his scripts in terms of materiality and corporality. Comments on Cassius’ poetic talent (v. 63 ingenium) complete information which the reader gets in verses 62–64. It was more impetuous than a rapid river (v. 62 rapido ferventius amni). Fire, which is such a central theme, causes Cassius’ physical death, the book’s material extinction together with his talent in poetry. But it also helps distinguish one from the other, as the river (that is his poetic talent) despite its streaming nature cannot be destroyed by fire, unlike the dry surface of the papyrus–roll or the wooden bookcase. Horace’s book in Epistles 1.20 on the contrary is no more in his possession. It hated being enclosed and available solely for Horace and seeks public approval. The difference between keeping the book sealed and publishing it is reflected in the different uses of verb tenses. So, the book hated (v. 3 odisti) the seals of its bookcase, when Horace used to show it (v. 4 ostendi) only to a few people, while in the present he seeks out the market (v. 1 spectare), so that it can expose itself to public affairs (v. 2 prostes; v. 4 communia laudas). At the same time, Horace criticizes the behavior of his book as it was raised to be modest (v. 3 pudico) and not ostentatious (v. 5 non ita nutritus). The Lexikon Horatianum explains pumex as genus lapidis quo chartam poliebant et membranam complanabant librorum (= stone which was used to smoothen the book’s surface).12 Due to this fact it usually relates specifically to the book cover.13 As soon as a book was ready to be sold, it would be smoothened out with a pumice–stone. Horace’s book is now being smoothened and is getting ready to be published. Cui dono lepidum novum libellum arida modo pumice expolitum? (Catull. 1.1–2) Whom do I give a neat new booklet polished up lately with dry pumice? (tr. Lee)

 12 Cf. Bo (1966) vol. 2 (L–Z), s.v. pumex. 13 TLL 10.2.2634.64 librorum frontes.

32  Georgios Taxidis Once the poetic persona has completed writing the book as a gift to an addressee, he is now paying particular attention to its external decoration. Catullus’ book is to be given out to another person whilst in epistle 1.20 it is to be given out for publication. Not only does the use of pumex (= pumice–stone), which references the polishing procedure, define the final form of the book before it would be given away, but also adjectives ascribed to the polished book surface do define it both in Catullus and in Horace. In the first case the book is already expolitum (‘polished’, Catull. 1.2) by an arida … pumice (‘wet pumice–stone’, Catull. 1.2), while in the second it is mundus (= ‘neat, elegant’, Hor. Epist. 1.20.2). So, the material identity of pumex highlights the fact that the item, which the pumex has shaped, has to be given away. This stresses the fact that, at least at the beginning of Horace’s poem, the external characteristics of the book are important and not its content.14 nec fragili geminae poliantur pumice frontes (Ov. Tr. 1.1.11) let no brittle pumice polish your two edges (tr. Wheeler)

At the beginning of Tristia Ovid addresses his book: Parve—nec invideo—sine me, liber, ibis in urbem (‘I am not jealous, little book, that you go to the city without me’, Ov. Tr. 1.1.1), which is now ready to inform the public about its poet’s condition during the exile. Apart from the collection’s title (tristia = sad) Ovid also determines the external characteristics that his book should have in order to represent his inner world, which is full of grief due to his exile (v. 3 qualem decet exulis esse). According to this the book must lack any material ornament, which identifies the exteriority of the book with its pleasing content (v. 9 felices ornent haec instrumenta libellos). To achieve this, its cover must not be – among other things – smoothened by a pumice stone (v. 11). Thus, the material artefact is being adjusted accordingly to its content and not according to the usual procedures of publishing a book, in order to better serve poetic expression. In epistle 1.20 the book is being evoked in a similar mode to Ovid’s. As Horace’s book is going to be published, it has to be well smoothened and presentable, in order to sell. Compared to Ovid’s book of exile, the Epistles book is free of adjustments that refer to the poem’s content. The smoothening with the pumice– stone is simply mentioned as a part of the publishing procedure. So, in this case  14 However, the terms novus, libellus, expolitus have also poetological connotations in Catullus’ text.

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  33

the focus is being placed on the book’s material substance, having absolutely no relation to the content it carries. So, what we see is the fact that the use of the pumice–stone15 differs according to the purpose set up by the book. The writing material item is used thus in order to highlight the poet’s intentions. Ovid’s persona is sad, so it is crucial that the book does not look too perfect and beautiful. The materiality alludes to the person, contributing to a symbolic use of the materiality of the book. One more aspect of the symbolic use of pumex can be seen in its correlation to the poematum stilus politus (= composition of poems).16 In this sense the pumice–stone which creates a more appealing (smooth) effect than the initial raw papyrus–surface should be read as parallel to the diction lima (= ‘a file, as applied to literary compositions, i.e. polishing, revision’, OLD s.v. 2). quemque poetarum limae labor et mora. (Hor. Ars P. 291) If a single one of her poets could endure the effort and time–consuming slow discipline of the file. (tr. Bovie)

Horace displays the point that Roman writers are not willing to insist on a better poetic expression and do not examine their verses before editing them although they have addressed a series of major literary topics (nil intemptatum nostri liquere poetae – ‘our poets have not left anything untouched’, Hor. Ars P. 285). According to this lima is used metaphorically, i.e. not as a writing material identical with pumex, but as a way of critically examining the practice of the poet’s writing itself. So, the metaphorical use of the materiality of the file (examine procedure) corresponds with the metaphor of the papyrus’ materiality as a written text. Horace complains about his yet unpublished book because it is in a hurry to get to know the outer world and become famous. Through this phrase we are being informed about the exact material form of the addressed book. It is a papyrus  15 According to TLL 10.2.2634.73–74 the following explanation to this verse by Horace’s scholiast (probably Porphyrius?) has been proposed: ‘mensam bibliopolarum poetica licentia pumicem voca’. Despite the fact that the proposed meaning ‘changer’s counter’ of the bookstore seems plausible it cannot be easily adopted because in that way it would exclude an important part of the publishing as well as selling procedure. Moreover, the adjective mundus derives more naturally as a result of the used pumice–stone on the papyrus–sheet than simply ‘standing neat on the table’. 16 TLL 10.2.2635.7–9.

34  Georgios Taxidis since cogi means ‘to bring together’ and in breve means ‘so that you will be shortened’. Thus, we are being also informed about the way of reading17 an ancient book in papyrus–roll, that is by unrolling it. More interesting is the invocation of the book (te), which highlights its material form. So, the book, no longer in use, will end up enclosed in its capsa.18 Its (shortened)19 material form corresponds with Horace’s personal preoccupation about the book towards the lost interest of the plenus amator, who just leaves the papyrus–roll unrolled lying in front of him. As soon as Horace finished reading or working on it, he used to put it back in its bookcase in order to protect it and keep it away from the eyes of strangers’ (v. 4 paucis ostendi gemis). Hence, we may claim that the way of using the book as a material artefact may also indicate an attitude towards the poetic text. The book which sets off in search of publicity is presented as hating the keys and the seals (sigilla) that kept it closed in the bookcase (capsa). The diminutive diction sigilla (= little images, seals),20 which is semantically equivalent to signum (both deriving from signare = to set a mark upon, to mark)21 entails primarily the meaning of ‘little figures or images’.22 These figures (mostly images) are being set up on specific objects or stand somewhere (little figures) in order to represent something that it is absent. While once being inscribed on objects they establish and clarify the dependent relation of those objects to the sigilla. So, these objects become in some way reminiscences of the person or item that each specific signum resembles. Therefore, the absence turns into presence. This symbolic function has its concrete actualization in the semantic field of signare–signum–sigillum (sigilla) especially in respect to writing materials. Hence signum or sigillum defines ‘an image or device on a seal–ring; a seal, signet’23 which is used in order to ‘to mark with a seal; to seal, seal up, affix a seal to a

 17 Cf. Van Sickle’s (1980) detailed approach to the ancient book. 18 Cf. Bo (1966) vol. 1 s.v. brevis: ‘sc. vel in angustum locum abici’ vel potius „involvi et complicari’; cf. also Fedeli (1994) 1308. 19 Although Horace’s book is addressed with liber its description as a short papyrus–roll reveals similarities with Ovid’s Tristia book 1 (Ov. Tr. 1.1.1 parve … liber) as well as Catullus’ ‘little book’ (libellus– Catull. 1.1). For libellus cf. Mart. 3.2. 20 For the following lexical approach I considered the words signare–signum–sigilla as a consistent semantical field. Hence the dictionary entries (mostly from OLD) are not observed as different word references but as different lexical appearances of the same diction. 21 OLD s.v. signo 1. 22 With this meaning the word comes up in Hor. Epist. 2.2.180 Tyrrhena sigilla; Cic. Verr. II.4.22 apposuit patellam, in quā sigilla erant egregia, and elsewhere. 23 OLD s.v. signum II D.

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  35

thing’.24 The practice25 of applying this material artefact upon writing materials could be explained by the following parallels. Ut proficiscentem docui te saepe diuque, Augusto reddes signata volumina, Vinni (Hor. Epist. 1.13.1–2) Vinius, as I repeated to you when you started, you will hand these volumes of odes, still sealed, to Augustus (tr. Bovie)

Horace’s book–rolls (volumina) are given to Vinnius in order for him to deliver them to Augustus. It is important to stress the difference of the tenses that refer to the sealing, signing of the books, and those referring to them being sent to Augustus through Vinnius. So, he is taking off now (te … proficiscentem) in order to give the books to Augustus (ut … reddes). These books however have–already– been signed (signata). This time difference emphasizes the fact that signing the book was a procedure which takes place before a book was to be given out to another person. After having been smoothened with a pumice–stone the book had to carry the mark of its possessor.26 'inprimat his cura Maecenas signa tabellis.' (Hor. Sat. 2.6.38) ‘Do have Maecenas affix seal to these papers.’ (tr. Bovie)

This verse helps us understand the exact way in which a document, a book or a writing–table (tabellis) was sealed, that is by pressing (imprimere) a signet ring upon its surface. The previous word–references from Horace’s poetic corpus describe the semantic field of the term signare in its material aspect, that is in the  24 OLD s.v. signo I B 1 ‘usually obsignare’. 25 Peirano (2014) 224 defines signum as follows: ‘The sphragis or seal (Lt. signum) was the most common device through which Romans simultaneously signaled a text as complete and attested to its author. Such seals were impressions made of clay or wax and employed on letters and personal documents through gemstones or signet rings furnished with unique personal marks. The seal imprinted on a document was thought to endow it with trustworthiness (fides), in that it bore witness to its being a faithful representation of its owner’s words and simultaneously prevented the text from being tampered with’. 26 cf. Cic. Cat. 3.5.10 ostendi tabellas Lentulo, et quaesivi, cognosceretne signum. Annuit. Est vero, inquam, notum signum, imago avi tui; Plaut. Amph. 421 obsignata signo est. On the writing materials of Epist. 1.13 cf. Oliensis (1998) 188–90 (and n. 68, 69).

36  Georgios Taxidis literal sense of ‘pressing a signet ring upon a paper–sheet’ or more freely to mark a text by attaching one’s signature. However, it is also worth considering how this term may function in a symbolic way, in many ways echoing the function of the sigilla as ‘little figures or images’ capable of conjuring up something or someone, who is physically absent through the mode of a 'resemblance'. quid autem Caecilio Plautoque dabit Romanus ademptum Vergilio Varioque? ego cur, adquirere pauca si possum, invideor, cum lingua Catonis et Enni sermonem patrium ditaverit et nova rerum nomina protulerit? licuit semperque licebit signatum praesente nota producere nomen. (Hor. Ars P. 53–59) Will the Roman refuse the license to Vergil and Varius And grant it to Plautus and Caecilius? And why should I be refused the right to put in my bit, if I can, when the language of Ennius and Cato enriched the speech of our native land and produced some new terms for things? (tr. Bovie)

In his discourse about the production of poetry Horace picks up the fact of displaying the meaning of certain words or phrases throughout the literary tradition. According to this thought a Roman poet of his time could have picked up a diction so old and traditional as Ennius or Cato and adjusted to it a new meaning (vv. 57– 58 nova … nomina protulerit). Thus, signare is used in this abstract way in order to clarify that a diction has been loaded with a specific meaning and that this specific meaning was established by the literary use in a certain time. For this reason, a signatum nomen (‘signed diction’, Hor. Ars P. 59) expresses the idea of the stability and finished action on a word, since this has been ‘sealed’ with a specific semantic expression in a specific time. Proof of this is on the one hand the reference to poets such as Vergil and Varius, who enriched the Latin vocabulary with a spectrum of ideas attached to specific dictions, and on the other hand the (new) Roman poet who has to interact with these ‘sealed’ words, established through the literary tradition in order to create new poetry. Hence the function of signare – signum– sigilla in this literary context allows us to turn to another literary function of the seal, that is the literary seal of the poet, in this case Horace’s at the end of Epistles, through which the poet presents himself to his audience. As mentioned in the introduction, the main purpose of this paper is to support the connection between the vocabulary referring to

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  37

writing materials mentioned in the text’s opening lines (vv. 1–6) and Horace’s own description (σφραγίς) in the poem’s closing lines (vv. 20–28). Therefore, the observations made here about sigilla should be taken as a basis for the contrasting juxtaposition that will take place in the second part. It should however be said here that at the beginning of the poem the book wants to be unsealed (odisti clavis et sigilla – ‘you hated the keys and the seals’, Hor. Epist. 1.20.3), freeing itself thus both from its author’s possession as well as from his authority. Its ‘liberty’ will consist further in the fact that it could be read by everyone who could buy it (et communia laudas – ‘you praise public affairs’, Hor. Epist. 1.20.4) something that was not possible with Horace’s signet rings upon it. These confined the book to only a few people chosen by Horace to have this privilege (paucis ostendi gemis – ‘I have shown you to few fingers’, Hor. Epist. 1.20.4). Nevertheless, at the end of the poem the book becomes sealed,27 signed again through Horace’s literary seal. Τhe diminutive form of the seals (sigilla) could perhaps be explained by the fact that the book will be brevis. As soon as the book gets published, there is no way back anymore (v. 6 non erit emisso reditus tibi). In order for Horace to emphasize this irrevocable situation, he uses a linguistically ambiguous diction. On the one hand, emitto28 carries the meaning of ‘release’,29 on the other hand emitto is terminus technicus for publishing a book. With the same meaning we come up with the word in Ars poetica: quis tamen exiguos elegos emiserit auctor (Hor. Ars P. 77) Who wrote these first little couplets? (tr. Bovie)

The direct reference to the auctor as well as to the poetic content (elegos) makes it more obvious than it is in Epistle 1.20. The irrevocability of Horace’s warning towards the book is stretched – again – by the different tenses of the words emisso and erit. So, from the very moment the book is published there will be no way back. For this reason, the book’s sealing is the final procedure before its publication. Both terms (signum and emittere) signify the written text as complete once and for all. No further suggestions or corrections are possible. Hence it is obvious that these uses of the terms present the book and its surrounding objects as pure material items, as they focus on the outer surface of the  27 Cf. Epist. 1.13.2 signata volumina. 28 TLL 5.2.508.43 strictiore sensu: de publicatione, divulgatione a) libros, scripta sim. 29 Cf. Bo (1966) vol. 1 s.v. emitto: foras mitto.

38  Georgios Taxidis book. Their function reveals, as we have seen so far, a dialogue between the book’s material substance – with the emphasis placed on the book’s cover – and the literary content written in it. According to this, a finely–illustrated book does (and should) not indicate a good poet nor does a title foreshadow a well–articulated topic in the contents of the book. Furthermore, as we have noted the literary content also bears some parallels to the author himself, providing thus autobiographical information (for example the book’s humble decoration acts as a mirror of the poet’s bad living conditions during his exile). Hence the author himself becomes the content of his own poetry enriching in this way the dialogue with the book as material artefact. These facts allow us, as this analysis proposes, to imply a comparison of the writing–material–words found in Epistles 1.20 (vv. 1–6) with the seal of Horace at the end of the poem (vv. 20–28), since Horace, as we are going to see, describes himself by emphasizing his outer (i.e. physical) presence. The word sigillum and more generally the semantic field of signare as mentioned previously will now form the basis for the further development of this argument. So, the book hates the seals that were keeping it tight and closed as part of Horace’s property. To add to this interpretation, it might be noteworthy to stress the double significance of the word liber (= book/free), a topic to which we will return further on in this paper. But still focusing on the current point, we can see the word sigilla at the very beginning of the text and then we observe that a ‘real’ literary seal (signum) follows. How are the sigilla of verse three to be compared, read and understood in relation to the signum of Horace’s? In an attempt to further explicate this issue, I will proceed with a close–reading of the poem and I will draw a comparison between the writing–material– words mentioned earlier and Horace’s own description. The very first word uttered by Horace to the book is liber. In a similar tone, the first piece of information we receive about Horace relates to his origin from his freedman father (libertino patre). Not only does the word libertinus etymologically derive from the word liber = free,30 but the ambiguous meaning of the second term allows us the following observation. If the book (liber) is sold, it ceases to be part of its author’s property it becomes free (liber).31  30 The terms liber and libertinus are linguistically coincident with each other as shown by TLL (7.2.1280.53–57 s.v. liber): ‘i.q. ἐλεύϑεϱος (…) negatur servitus singulorum hominum, strictius respicitur is, qui dominio eri subiectus non est (de eis, qui sub potestate patris, hostium non sunt)’. 31 On the pun liber=book/ free cf. Williams (1992) 180 (and n. 11) as well as in terms of publication cf. Kenney (1982) 19 where he observes: ‘Once a book was released in this way the author had no rights in it whatever no control over its fate, and no secure prospect of being able to correct it’; 19 (n. 12): ‘Publication […] was less a matter of formal release to the public than a

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  39

A libertinus on the other hand is a former slave who bought his freedom.32 So, as we can see, Horace himself is a free man having been born from a freedman father, which means that both the book and Horace have become or are becoming free.33 In my opinion this statement is supported by two facts. On the one hand we see that both terms occupy the same position (first) on both sides of the descriptions (book–Horace). On the other hand, it is an established locus within sphragis–compositions34 or otherwise termed autobiographical compositions to start with an address to the god, person or thing to whom or which the composition was devoted. The application of this motive to this poem and to the Epistles in general can also be explained and enriched by the epistolary composition style of the beginning of the poem by addressing the person to whom it was referring to. So, Horace addresses his liber at the beginning of the poem to result in his own liber–origin at the end of it.35 The lexical item pumex also presents us with interesting interpretive possibilities. In this case we have an item that although not a writing material itself, is included in the current analysis as a term and an item used to enhance the book’s surface, the book’s outer presence, as shown earlier. So, the emphasis is on the book’s surface. Reading Horace’s self–presentation/sphragis we note that he also puts a clear emphasis on describing his body, especially in this case, its surface. Hence, as the reader draws his/her attention to the surface of the papyrus–roll being smoothened, s/he also reads that Horace’s hair is grey before its time (praecanum) and his body tanned by the sun or sunburnt (solibus aptum). Thus, we see that Horace’s self–description at the end of the poem displays similarities with the vocabulary of writing materials employed at the beginning of the poem. The materiality presented through the words that accompany the book–address meets the materiality of Horace’s self–presentation, that is not of himself as a

 recognition by the author that his work was now, so to speak, on its own in the world: the word usually translated ‘publish’ (edere = Greek ἐκδιδόναι) connotes the resignation of rights and responsibilities’. 32 Quint. Inst. 5.10.60 qui servus est, si manumittatur, fit libertinus. On Horace’s and –mostly– his father’s state of a freedman cf. Anderson (1995) 151–64. 33 On this topic cf. Mordine’s (2010) 534 comment on Tristia 1.1.1 about the book’s (slave’s) deliverance from its (his) dominus (poet), as well as n. 29, where he states: ‘Hor. Epist. 1.20 provided a model for the equation poet: bookroll = master: slave. There, too, the bookroll is sent forth like a freed slave (liber, line 1), although unlike Ovid’s it is nicely polished (pumice mundus, 2)’. 34 On the sphragis cf. Kranz (1961), for Horace’s self–presentations in specific 117–20. 35 Thus, the book becomes Horace’s surrogate, like the case in Tristia 1.1 (Tr. 1.1.57 tu tamen i pro me, tu, cui licet, aspice Romam), cf. Hardie (2002b) 298.

40  Georgios Taxidis poet, but as pure human–corporal being, that suffers the passage of time, and the impact of nature with its effects on the human body and especially the skin.36 This last aspect becomes more obvious if we consider the colour of the papyrus’ sheet. When the development of the papyrus’ leaves as a writing material was completed, its surface would have been pale and cream–coloured and with the passing of time it would turn into a more brownish colour.37 In addition, Horace’s sunburnt body must have been dry due to its extended exposure to sunlight. This fact corresponds with the dry climatological conditions which were necessary both for the papyrus’ plant to grow, and also for it to be sustained in order to be preserved. So, Horace’s and the book’s surface display a substantial similarity. The main stress of the analogy between materiality and corporality acts as a warning of the poet towards the book; that there is no material object that outlasts the passage of time: carus eris Romae donec te deserat aetas / contrectatus ubi minibus sordescere / volgi coeperis (‘you’ll be treasured in Rome until youthful bloom deserts you; once you start to show signs of wear, fingered by vulgar hands’, Hor. Epist. 1.20.10-12, tr. Bovie). Proof of this is the poet’s self–presentation. After the book has been smoothened it is ready to be published. The poet’s warning of verse six sounds severe: non erit emisso reditus tibi (‘There is no coming back once you will be published’, Hor. Epist. 1.20.6). The function of this term as a terminus technicus for publishing a book as well as its meaning of ‘bringing something out’’ (ex–mitto) offers yet more grounds for comparison between the book and Horace himself. Once the book is published, it will fly away from its author’s nest, it will introduce itself to the public and become famous. Horace’s σφραγίς at the end of the composition functions in an equivalent way. That means that when Horace introduces himself to the public, he also becomes emissus. Here we could suppose that Horace develops a word–play with the semantic possibilities of emittere, since it also carries the meaning of ‘release a person from one’s potestas, set free’.38 According to this, emittere stresses the common fate of Horace and the book, since Horace or his father39 in order to enjoy the free–man’s

 36 Fedeli (1996) 21 (n. 32) presents two parallels which describe Horace’s premature skin ageing: Hor. Epist. 1.7.5–26, Hor. Epist. 1.17.23, stating further that praecanus ad loc. consists a hapax legomenon. 37 On the papyrus as ancient writing material cf. Blanck (1992) 56–62. 38 OLD s.v. emitto I B manu emittere aliquem. 39 libertinus is used both for the person set free as well as for his children, OLD s.v. libertinus II A 2.

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  41

status should be set free from one’s (sc. their previous master’s) authority– (potestas), thus manu emittere aliquem.40 So, as we see there is a comparable function between the word emissus and more specifically due to its prefix ex– and the literary sphragis of Horace. Both elements want to reveal something. This correlation could be re–enforced, according to this analysis, by two further facts. Firstly, not only does the prefix ex– as a reference of bringing something to its end/present something stress the fact that this poem is the last of the first book of the Epistles, but we can see a contrasting juxtaposition with Epistles 1.1, where we find again self–references to the poet. The poet announces his intention to compose quid verum atque decens (‘everything right and proper’, Hor. Epist. 1.1.11). If we focus on the opposition between the prefixes con– (= to bring something together, to compose, further to seal) and ex–(= to bring something out, to free, to reveal), we could claim that Horace, according to this position of the prefixes, after having composed his entire first book of the collection reveals himself through the literary seal, becomes emissus, just like his book, which is now finally ready to be published. Moreover, the programmatic statements which Horace makes about his writing–intentions on the Epistles (Hor. Epist. 1.15–19) could be summed up by et mihi res, non me rebus subiungere conor (‘I try to subdue the theme, I do not let the theme to subdue me’, Hor. Epist. 1.1.19). Thus, Horace’s intention to adjust his poetry’s content at his own prescriptions41 at the beginning of the collection finds its equivalent in the book’s description which corresponds to his own at the end of it. Secondly, as we are dealing with a sphragis poem of Horace, the argument presented here could be further strengthened by looking at another poem by Horace that contains autobiographical information, which is carmen 3.30. In this also concluding poem we find again the prefix ex– at exegi monumentum aere perennius (‘I fulfilled a monument which will last more than bronze’, Hor. Carm. 3.30.1) and then autobiographical information about Horace. The verbs exigo and emitto do not have the exact same meaning42 but their similarity that is based on the prefix ex–, of the reference to an outer activity is more than obvious. The word that contains the prefix ex– is part of the literary sphragis of  40 Should one want to take the analysis further the expression does serve moreover the image of Horace’s book being handed from Horace’s hand over to the book store. 41 Oliensis (1998) 156: ‘This inward turn is supported in Horace's first epistle by an insistent rhetoric of self–reliance and self–sufficiency. If Horace was once a pleasure–loving grasshopper, he has matured into an industrious ant, putting up a supply of wisdom for his wintry age’. 42 OLD s.v. exigo: to drive out, push forth, thrust out, take out/emitto: to send out, send forth and TLL 5. 2.1464. 53 s.v. exigo: opera arte facienda (fere de carminibus, libris).

42  Georgios Taxidis Horace in carmen 3.30. We could make the same supposition about emisso on the basis that a sphragis reveals (emittit) the creator of a composition. In addition to this comes the meaning of exigo as ‘completing a piece of art, in reference to poems or books’ (cf. n. 42). So, the similarity of the two verbs according to their prefixes is being enhanced by their common semantic field. The different appearance of diction consists however in the fact that the lyric vates of the Odes constructs a literary monumentum (metaphorical denotation),43 whereas Horace the human being publishes his materially completed work. In this case we observe that the revealing function of the sphragis is being accompanied by verbal expressions which have the same purpose according to their meaning and their position. Furthermore, there is major similarity between Odes 3.30 and Epistles 1.20 to be observed; 3.30 is also based on materiality, in order for Horace to claim poetic immortality,44 i.e. the bronze (Hor. Carm. 3.30.1 aere perennius) and the Aegyptian pyramids (Hor. Carm. 3.30.2 regalique situ pyramidum altius) which define his poetic monument. By contrasting Horace’s material terms (Odes: monument–Epistles: book) we discover a similarity which sharpens the difference between Odes and Epistles. This similarity is Egypt. On the one hand Horace’s work is compared to the famous Egyptian pyramids while on the other hand Horace’s book is a papyrus. But their value defines their juxtaposition, since the pyramids are a monument of immense importance which represents a whole culture, something that Horace expresses clearly (Hor. Carm. 3.30.2 regalique situ). The papyrus, an everyday object is compared to the grandeur of the royal pyramids. This shift in chronological periods can be seen as reflecting the juxtaposition of materials. So, Horace’s ‘lyric’ materials which stand symbolically for his poetry will last for an immense number of years (Hor. Carm. 3.30.4–5 aut innumerabilis / annorum series et fuga temporum) insusceptible to any form of natural erosion (Hor. Carm. 3.30.3 non imber edax, non aquilo impotens). The sphragis of Epistles 1.20 on the contrary develops clearly the matter of the limitation of time (me quater undenos sciat implevisse Decembris – ‘I completed 44 Decembers’, Hor. Epist. 1.20.27). Like the poet’s body which is subordinate to certain time–conditions, according to which it is young or old, so the book as material artefact is restricted by constraints of time, which result in its corruption. Hence according to these observations, the poet’s sphragis of the Odes deliver a clear statement. That he will die, but his poetry will last forever, for it will even  43 For monumentum as literary work: OLD s.v. I B 2 cf. also Mordine (2010) 536 and n. 37. 44 Cf. Oliensis’ (1998) 180 connection of the two poems in relation to Horace’s dedicatee Maecenas.

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  43

surpass great human monuments. From this point of view the sphragis at 1.20 seems contra expectationem due to the fact, that Horace does not claim immortality. In contrast he insists on the perishability of his body and the book’s materiality. In addition to this, it must be noted, that Horace’s biographical information is extraneous to the understanding of the poem; it is almost banal, it belongs to the sermo humilis, having nothing to do with Horace’s greatness in carmen 3.30. Last but not least, we would like to propose a close reading of the phrase in breve te cogi (‘I rolled you up in short form’ Hor. Epist. 1.20.8). With this phrase Horace meets an emphatic description of the material–papyrus–form of the book. It will be kept closed and tight. This microscopic view of the papyrus finds its equivalence in Horace’s short body (Hor. Epist. 1. 20.24 corporis exigui). So, both the book and Horace are being described as short. At this point it might be helpful to refer to Suetonius’ Life of Horace, which displays similitudes with Epistles 1.20 both in respect of Horace’s self–description, as well as in the style of diction used. So, Vita Horati provides information –amongst other things– about Horace’s physical appearance (habitu corporis fuit brevis atque obesus – ‘as for his appearance he was short and fat’, Suet. Vita Hor. 3.7). Therefore, it becomes clear that we have a basic equivalency between Suetonius’ description of Horace and that of Horace himself at Epistles 1. 20. Furthermore, the historian goes on describing Horace’s body by saying that even Augustus noticed this characteristic within an epistle sent from him to Horace (Suet. Vita Hor. 3.9 et ab Augusto hac epistula). Suetonius quotes this epistle, from which we learn that Augustus was making fun of Horace due to his unappealing figure. But the interesting fact in this instance is the way that Augustus chooses to laugh at Horace. The emperor’s persona compares Horace with his books: vereri autem mihi videris ne maiores libelli tui sint quam ipse es. sed tibi statura deest, corpusculum non deest. itaque licebit in sextariolo scribas, quo circuitus voluminis tui sit ὀγκωδέστατος, sicut est ventriculi tui. (Suet. Vita Hor. 3. 7–9) But you seem to me to be afraid that your books may be bigger than you are yourself; but it is only stature that you lack, not girth. So you may write on a pint pot, that the circumference of your volume may be well rounded out, like that of your own belly. (tr. Rolfe)

Firstly, the words breve (about the book) and corporis exigui (Horace himself) do meet the semantically identical (in the first case), same word (in the second case) brevis. Secondly, Augustus uses diminutive forms to talk about the book (libellus,

44  Georgios Taxidis quantuluscumque) and about Horace (corpusculum, ventriculi tui), a fact that goes along with Horace’s way of presenting his book and himself in the Epistles. Moreover Augustus, as cited by Suetonius, meets a more particular comparison between Horace and his writings. The size of Horace’s book–rolls (circuitus voluminis) is as big as Horace’s stomach (ventriculi tui), that is huge (ὀγκωδέστατος), creating thus quite a humorous image of Horace, who is short with a very big stomach. Apart from the writing materials themselves there is one more decisive similarity between the two texts. After having warned his book, Horace says that he will laugh at the end, when the book will end up filthy and useless (Hor. Epist. 1.20.14 ridebit monitor non exauditus). Augustus’ epistle on the other hand has a sense of humor and laughter if one considers that the diminutive words for ‘little book’, ‘tiny body’ do come from the mouth or the hand of an emperor and especially Augustus. This fact is apparent from the comparison that Augustus draws, i. e. between Horace’s libelli and corpusculum. Thus, both figures, that place themselves in the position of authority in relation to their opponents end up laughing at them about their physical and material presence.45 While Augustus having obvious authority does not need to state that in his epistle, Horace clearly needs to by presenting himself as monitor. After the comparison of Horace’s and his book’s description based on close reading of singular dictions and their parallel interactions, we would like to consider one last parallel, that is Odes 2.20, which also contains self–references of the poet. Pearcy relates the sphragis of 1.20 with that of carmen 2.20 (by commenting on the fact that Horace spread his wings farer than his nest allowed–Hor. Carm. 2.20.21) where Horace transforms into a swan. ‘The poem (Carm. 2.20) began with a metamorphosis of poet into swan, just as Epistles 1.20 ends with a transformation of author into text’ and concludes by saying that ‘for Horace the author, immortality can be gained by transformation, and transformation entails becoming a curricular author’.46 The idea of the poet’s transformation into a medium which leads him to immortality can be in my view enhanced also by again referencing materiality and corporality, which reveals further correlations between Odes 2.20 and Epistles 1.20. In this way we can read the obvious emphasis placed  45 The disapproval (along with expressed laughter and critic) due to features of figure occurs also in Epistles 1.1, where Maecenas is upset about Horace’s unhealthy condition (Hor. Epist. 1.1.101–5). 46 cf. Pearcy (1994) 463.

Horace’s Book and Sphragis. Writing Materials in Horace’s Epistles 1.20  45

to Horace’s body in his sphragis as corresponding to his physical description in Odes 2.20: iam iam residunt cruribus asperae pelles et album mutor in alitem superne nascunturque leves per digitos umerosque plumae. (Hor. Carm. 2.20. 9–12) Already the rough skin is forming on my ankles; metamorphosis into swan moves up my body; downy plumage springs on arms and elbows; shoulder–blades sprout wings. (tr. Michie)

The poet’s transformation into a swan is being accompanied by a description of its body (vv. 9–12: skin, wings, color, finger and cheeks). Here the emphasis on the surface of the body follows a different perspective to that in Epistles 1.20. The main difference lies in the fact that the swan’s external surface (i.e. the poet’s skin) is described positively especially at the moment when different body parts grow (Hor. Carm. 2.20.11 nascuntur), while Horace’s (or book’s) body (cover) in 1.20 is presented at its moment of decadence, in a state of natural, physical corruption. Nevertheless, Horace’s book will travel covering huge geographical expanses through the empire (Hor. Epist. 1.20.18 extremis in vicis) a fact that comes along with Horace’s own life experience since he spread his wings further than his nest (poor financial and social family conditions) allowed (Hor. Epist. 1.20.21 maiores pinnas nido extendisse loqueris). Horace’s bird metaphor of v. 21 recalls the swan of Odes 2.20 which will expand his wings and fly to (any) distant place of the known world (v. 13–20). Furthermore, by adding the use of the bird’s image to the functional frame of the poet’s sphragis in Epistles 1.20 we could suppose that the poet hereby, in addition to the statements made about him and his book, also assigns himself to his audience as writer of the Odes, repeating an established metaphor used in his previous work (and especially in a sphragis poem just as in this case). In support of this there is a famous counterpart which should be considered, that of Vergil’s. The epic poet includes in his sphragis at the end of the fourth Georgics’ book the first verse of his Eclogues’ collection: carmina qui lusi pastorum audaxque iuventa, Tityre, te patulae cecini sub tegmine fagi. (Verg. G. 4. 565– 6)

46  Georgios Taxidis I–who toyed with shepherds’ songs, and, in youth’s boldness, sang of you, Tityrus, under the canopy of a spreading beech. (tr. Fairclough)

However, despite the fact, that the transformation of Horace in his own text is persuasive, we cannot be sure if Horace will become a ‘curricular author’ because we are not given any information regarding the content of the book. Its outer presence is well presented, but we do not know what the contents are; and we are not told in Epistles 1.20 what kind of text is sealed by Horace’s sphragis. According to this reading we could make a case that it is through the materiality itself that Horace claims immortality. Horace proves his ephemeral nature as a human being by describing himself in a similar manner to the book; the book, as material item and Horace, as human being can definitely not outlive the passing of time and its impact on the physical and the human body. However, Horace’s transformation into his text allows us to see the book as a medium, through which the text as author and the author as text will last (live) forever. This analysis hopes to have shown that there are specific equivalences between the materiality of the vocabulary used to describe writing materials and the ‘materiality’ (= corporality) of Horace. The writing materials found at the beginning of the poem foresee the literary σφραγίς of the poet at the end, both in terms of diction and function, since the book will become liber (= free) and emissus (= sent out) in ways similar to the freeing of Horace himself, according to the revealing character of the seal. The close relation between Horace and the book on the basis of materiality allows us an alternative interpretation of Horace’s literary seal, especially if we recall his use of the seal in Odes 3.30. As the poet gains immortality there by representing himself through his poetry, his lecture, so does his persona in Epistles 1.20 gain immortality, by transforming itself into the text. Horace, the person will fade; as will the book, the material object. But Horace the poet, just like the book as medium will last forever. Testament to this is the fact, that the reader, who perceived and perceives the text throughout the literary tradition, just as we do now, holds an open book in his/her hands; and probably one book, which is not yet well–thumbed. It is the material corruption, which enacts the symbolic use of the items. In this way Horace’s literary sphragis, which at first appeared extraordinary, proves to function in accordance with other well–known seals of Horace and other poets; it declares its author’s immortality. This emphasis on materiality can be read as leading ex negativo to immortality, in such a way that fuses materiality and symbol.

Katherine Krauss

Fake Intellectuals, and Books of Unquestionable Authority in Aulus Gellius’ Noctes Atticae and Lucian’s aduersus Indoctum This paper1 will explore how the material text2 operates as a literary motif in the self-consciously learned writings of the second century CE. By focusing on the function and significance of the book roll in Aulus Gellius’ Noctes Atticae and Lucian’s aduersus Indoctum, I will demonstrate how both works partake in a shared set of concerns about the physical book arising from their shared cultural milieu, rather than any direct intertextual relationships. Building on recent scholarly work on imperial reading practices,3 I hope to give a fuller treatment to the symbolic values of the book roll per se, which inform both Latin and Greek thought in the second century CE.4

 1 This paper emerged from a graduate seminar at Columbia University taught by Professor Joseph Howley in the spring 2016 and was subsequently presented at the Prolepsis Association’s First International Postgraduate Conference: ‘Cupis volitare per auras’: Books, libraries and textual transmission from the Ancient to the Medieval World’. I would like to thank Professor Howley, the members of the seminar, the participants of the conference, and the reviewers for their invaluable comments. 2 I refer to the material text as it is understood in Price (2012). 3 Scholars have most recently tended to concentrate on the experiences of reading and the community created by these acts of reading. On reading and reading communities in Lucian see Johnson (2010) 157–78. On reading, and reading communities in Gellius Vardi (2001), Keulen (2008), Gunderson (2009), Johnson (2010) 98–136, Howley (2018). 4 While scholarship such as Zadorojnyi (2013) and Johnson (2010) include both Greek and Latin sources in their analyses, the implications of this choice remain undertheorized. While Woolf (2003) 205, (2013) 12, and Reardon (1991) 161 point to the cultural similarities between texts produced in Greek and Latin in this period, their cursory remarks both overlook the extent of these parallels and fail to illuminate the significance of these shared literary tropes for scholarly approaches to Imperial Greek and Latin literature. Similarly, Anderson’s (1993) consistent treatment of Greek literary production at a greater length than Latin texts ultimately shifts the focus of his work away from his assertion of ‘cultural and social mix’ (12) in the High Empire. Bowersock (1969) by contrast frequently juxtaposes Latin and Greek texts but mentions the overlap between ‘Greek’ and ‘Latin’ intellectual spheres only briefly. Furthermore, his evidence for this claim of affinity, that the archaizing movements in both Greek and Latin shared a common ‘mood’ (16), is tenuous at best (Swain [2004] 17–18). On the interactions between Greek and Roman culture in Gellius, see Astarita (1993) 65–82, 175–99. https://doi.org/10.1515/9783110632590-004

48  Katherine Krauss What follows will explore how both Noctes Atticae and aduersus Indoctum reveal a preoccupation with the settings in which books can and cannot be wielded as reliable symbols of authority. It will do so by tracing the development of two, related motifs—the use of the book as an implicit source of authority in the denigration of intellectuals associated with the pecuniary aspects of education, and the limitations of the book’s ability to confer such an authoritative status to Gellius’ and Lucian’s pose5 as erudite narrators. In this way, the book roll takes on a function which extends beyond, even if closely related to, its literary contents.6 Gellius’ and Lucian’s reflection on the cultural capital of the physical text not only recalls the hesitancy about the written word in Plato’s Phaedrus,7 but also forms part of a heightened awareness of textual materiality and production in the literature of the High Empire. In fact, the self-conscious reflection on the production, use, circulation, and destruction of the material texts was widespread among second-century authors.8 Elucidating the implications of Gellius’ and Lucian’s shared treatment of the book roll thus adds a dimension to our understanding of the conceptual history of the material text in the High Empire. It reveals the ways in which specific images and uses of the book had resonances for second-century scholarly thought in both Latin and Greek.

 Bookshops and Booksellers in Gellius and Lucian Gellius and Lucian both expose false pretensions to learning by contrasting the value of the book roll handled by the supposed literary expert to their intellec-

 5 This paper will use Lucian to refer both to the author of the text and to the (otherwise unnamed) first person narrator of his satire, unless otherwise noted; on the issue of narrator vs. author in Lucian, see Winkler (1985) 275. 6 See similarly Price (2012) 19–44. 7 See further Zadorojnyi (2013) 400. 8 E.g. Folwer (1995) and Seo (2009) on Martial, Johnson (2010) 32-73 on Pliny the Younger, Howley (2017) on inter alia Tacitus, Suetonius, and Cassius Dio. It is important to note that this self-consciousness about the materiality of texts is prominent in, but not unique to, the literary culture of the second century; see (with further bibliography) Williams (1992) and Krevans (2010) on Ovid; Butler (2002) and (2011), Gurd (2010), and Frampton (2016) on Cicero; Roman (2006) on Catullus and Propertius; Frampton (2019) on Horace, Gallus, and Virgil.

Fake Intellectuals, and Books of Unquestionable Authority  49

tual integrity.9 In the thirteenth book of his Noctes Atticae, an eclectic collection of anecdotes about his time studying in both Athens and Rome, Gellius uses a book of high value to expose the emptiness of a supposed intellectual’s claim to expertise on Varro.10 Gellius recounts: tunc aliis etiam, qui ibi aderant, compluribus idem comprobantibus desiderantibusque accipit a me librum ueterem fidei spectatae luculente scriptum. accipit autem inconstantissimo uultu et maestissimo. sed quid deinde dicam? non audeo hercle postulare, ut id credatur mihi. pueri in ludo rudes, si eum librum accepissent, non hi magis in legendo deridiculi fuissent; ita et sententias intercidebat et uerba corrupte pronuntiabat. 11 (Gell. NA 13.31.6-10) Then, since several others who were there also agreed and asked for the same thing, he accepted from me an old book of proven reliability, beautifully written. He received it however with a very gloomy and faltering expression. But what should I say next? By Hercules I don’t dare ask that you believe me. If ignorant schoolboys had taken up the book, they would not have been more ridiculous in reading it; he butchered the thoughts so much and pronounced the words so badly. (tr. adapted from J.C. Rolfe)

In describing the book he hands over as a librum ueterem fidei spectatae written in a refined script, luculente scriptum, Gellius highlights the book’s high value. Both the accuracy and elegance of the text copied in the book lend it great worth, which Gellius contrasts to the poor quality of the ‘expert’s’ mind. According to Gellius, the ‘scholar’ of Varro is less sophisticated than even a schoolboy, and apparently unable to manipulate or ‘cut’ the scriptio continua on the bookroll properly, sententias intercidebat.12 While the anecdote in 13.31 falls within Gellius’ broader interest in the Noctes Atticae of undermining the authority of grammarians,13 in this particular narrative Gellius calls attention to the man’s ignorance through a tension between the prestige of the book he lends him and his poorly trained mind. Lucian opens his aduersus Indoctum, a lively tirade against a book collector, with a similar accusation:

 9 On a shared dislike of fake philosophers in Lucian and Gellius see Keulen (2008) 311. 10 On Gellius’ use of the book more generally see Johnson (2010) 111. 11 All translations are my own unless otherwise noted. The text used is Marshall’s OCT. 12 On scriptio continua see Saenger (1982) and McCutcheon (2015); on its significance to the intellectual elite, Johnson (2010) 201. 13 See Vardi (2001) 41.

50  Katherine Krauss οἴει μὲν γὰρ ἐν παιδείᾳ καὶ αὐτὸς εἶναί τις δόξειν σπουδῇ συνωνούμενος τὰ κάλλιστα τῶν βιβλίων· τὸ δέ σοι περὶ τὰ κάτω χωρεῖ, καὶ ἔλεγχος γίγνεται τῆς ἀπαιδευσίας πως τοῦτο14. (Lucian Ind. 1) Since you think that you too will be held in repute for your learning by eagerly buying the best books; but your actions fail and in a way become proof of your lack of learning.

The book collector’s taste for the most beautiful books implies that his books are of high financial value. This costliness stands in opposition to the book collector’s mind, whose intellectual crudeness Lucian points out in his reference to the book collector’s lack of paideia, τῆς ἀπαιδευσίας.15 Lucian uses this difference between the value of the books and the value of the book collector’s intellect even more explicitly in the later part of his censure. Comparing the book collector with his books to Thersites wearing Achilles’s armor, Lucian writes: τὸ αὐτὸ δὴ καὶ σὺ πάσχων οὐχ ὁρᾷς, ὁπόταν τὸ μὲν βιβλίον ἐν τῇ χειρὶ ἔχῃς πάγκαλον, πορφυρᾶν μὲν ἔχον τὴν διφθέραν, χρυσοῦν δὲ τὸν ὀμφαλόν, ἀναγιγνώσκῃς δὲ αὐτὸ βαρβαρίζων καὶ καταισχύνων καὶ διαστρέφων, ὑπὸ μὲν τῶν πεπαιδευμένων καταγελώμενος, ὑπὸ δὲ τῶν συνόντων σοι κολάκων ἐπαινούμενος, οἳ καὶ αὐτοὶ πρὸς ἀλλήλους ἐπιστρεφόμενοι γελῶσι τὰ πολλά; (Lucian Ind. 7) Do you not see that you in fact suffer the same thing too whenever you hold in your hand a very beautiful book-roll, having a purple slipcover and a golden knob and you read it committing barbarisms, dishonoring and distorting it, laughed at by the learned, but praised by your flatterers who, when turning to each other, also laugh a lot?

In this later derision of the book collector’s intellect, Lucian emphasizes the high cost of the book collector’s books, elaborating on his general description of the books as πάγκαλον by relating that they have πορφυρᾶν … τὴν διφθέραν and χρυσοῦν … ὀμφαλόν. Lucian thereby establishes a marked gap between the expensive books and the book collector’s lack of sophistication which causes him to disgrace the text inside his books.16 Lucian exposes the book collector’s lack of learnedness by contrasting the quality of his mind with the quality of the books he owns.

 14 All translations are my own, with reference to Harmon’s Loeb. The text used is Macleod’s OCT. 15 Similarly Johnson (2010) 170. 16 The use of αὐτὸ preserves an ambiguity that points to Lucian’s consideration of the material form on which the literary work is inscribed and the literary work itself as indistinguishable; see similarly Johnson (2010) 170.

Fake Intellectuals, and Books of Unquestionable Authority  51

In their exposés, Gellius and Lucian characterize pseudo-intellectual they deride with monetary language. Gellius opens his ridicule of the Varro ‘scholar’ in the bookshop thus, laudabat uenditabatque se nuper quispiam in libraria sedens homo inepte gloriosus (‘Recently some senselessly pompous man, sitting in a bookseller’s shop, was praising and selling himself,’ Gell. NA 13.31.1). Gellius’ quasi-metaphorical description17 of the man as selling himself, uenditabatque se, locates the pseudo-intellectual in the same pecuniary world as the bookshop, in which money is exchanged for a transferal of knowledge. In short, by presenting the pseudo-intellectual in the bookshop and by highlighting the intellectual’s and bookshop’s shared endeavor to sell literary knowledge, Gellius associates the intellectual he mocks with sale and financial exchange.18 Like Gellius, Lucian links the target of his satire with the transfer of money for knowledge. When comparing book sellers to the book collector, Lucian writes that if the acquisition of books imparted knowledge: τίς δὲ τοῖς ἐμπόροις καὶ τοῖς βιβλιοκαπήλοις ἤρισεν ἂν περὶ παιδείας τοσαῦτα βιβλία ἔχουσι καὶ πωλοῦσιν; ἀλλ᾿ εἴ γε διελέγχειν ἐθέλεις, ὄψει μηδ᾿ ἐκείνους πολύ σου τὰ εἰς παιδείαν ἀμείνους, ἀλλὰ βαρβάρους μὲν τὴν φωνὴν ὥσπερ σύ, ἀξυνέτους δὲ τῇ γνώσει, οἵους εἰκὸς εἶναι τοὺς μηδὲν τῶν καλῶν καὶ αἰσχρῶν καθεωρακότας. (Lucian Ind. 4) And then who would rival in learning the book traders and dealers who have and sell so many books? But if you want to investigate, you will see that they are not much better than you in learning, but barbaric in voice just like you, and foolish in mind, people of the sort who probably cannot discern good from bad.

Since the booksellers, like the book collector, possess books but still lack the ability to judge the good from the bad Lucian explains that the booksellers exude a mere appearance of learnedness through the material objects they have acquired, just like the book collector, ὄψει μηδ᾿ ἐκείνους πολύ σου τὰ εἰς παιδείαν ἀμείνους, ἀλλὰ βαρβάρους μὲν τὴν φωνὴν ὥσπερ σύ. As a result, Lucian suggests an affinity between the book collector and the booksellers, whose  17 See 13.31.13 in which the man refuses Gellius education without a fee, implying that Gellius’ description at the beginning of his anecdote may be more literal than metaphorical. See also Vardi (2001) 49. 18 Amiel Vardi (2001) connects Gellius’ attention to denouncing ignorance (especially in individuals like the man in 13.31 or 18.4 who demand a fee for their teachings) with an anxiety about the rise of the social status of the grammarian in second century Rome. Vardi thus locates Gellius’ critique as part of a broader desire to differentiate between those who impart knowledge for money with those who engage in academic studies purely as a leisure activity; see also Keulen (2008) 30.

52  Katherine Krauss involvement in financial transactions he makes explicit, τοῖς ἐμπόροις καὶ τοῖς βιβλιοκαπήλοις … τοσαῦτα βιβλία … πωλοῦσιν. Lucian thus highlights the similarities between the book collector and the pecuniary dimensions of the intellectual world. Gellius’ and Lucian’s shared use of books to expose baseless claims to knowledge, and connection between superficial learning and the presence of money in the intellectual sphere indicate their participation in a shared scholarly culture.19 By revealing the falsehood of claims to learnedness, the material book becomes a symbol of authority in both Noctes Atticae and aduersus Indoctum. In both texts, the bold assertions of erudition made by the aspiring scholars are closely linked the involvement of money in academic pursuits, and more specifically the buying and selling of books. Furthermore, the employment of books of great worth as a means to expose pretensions to learnedness in both Noctes Atticae as well as aduersus Indoctum illustrates that Gellius and Lucian not only explore the ridicule of pseudo-intellectuals in their texts but that they do so using the same techniques. Through their similar employment of the book as a tool central to the derision of empty boasts of knowledge, Gellius and Lucian both reveal the centrality of books to public displays of knowledge.20 In Noctes Atticae, the book explicitly becomes central to an epideictic display of the Varro scholar’s intellectual capacities. Gellius’ copy of Varro reveals the ignorance of the ‘expert’ in a highly public setting, the bookshop. Gellius records a similar exposure of a boastful grammarian in 5.4, in which the grammarian reveals his inability to recognize the faultlessness of a valuable copy of Fabius’ Annals in the public space of the bookshop. Gellius therefore points to the importance of the material text in public demonstrations of learning, endowing it with the ability to confirm or dislodge the status of the supposed literary expert. The role of the book in the exhibitions of the book collector’s knowledge (or lack thereof) in aduersus Indoctum21 parallels Gellius’ documentation of the book as central to epideictic displays of learning characteristic of second-

 19 While both Vardi (2001) 43–44 and Baldwin (1975) 48 have noticed the overlapping interests of Gellius and Lucian, their general characterizations ignore the extent to which Gellius and Lucian share similarities beyond the broad contours of their interests. 20 Galen’s de libris propriis 1.14 reinforces this centrality of the material text to oral displays of learning (Johnson [2010] 91). Johnson attributes to both the Galen scene the characteristics of a specifically sophistic or oratorical display (89), thus implying an important presence of the manner book in the realm of the rhetorical as well. 21 Johnson (2010) 115 makes a similar, more general, point about Gellius.

Fake Intellectuals, and Books of Unquestionable Authority  53

century intellectual life. In one of his first demonstrations of the insufficiency of the book collector’s intellect Lucian writes: ἢ πόθεν γάρ σοι διαγνῶναι δυνατόν, τίνα μὲν παλαιὰ καὶ πολλοῦ ἄξια, τίνα δὲ φαῦλα καὶ ἄλλως σαπρά, εἰ μὴ τῷ διαβεβρῶσθαι καὶ κατακεκόφθαι αὐτὰ τεκμαίροιο καὶ συμβούλους τοὺς σέας ἐπὶ τὴν ἐξέτασιν παραλαμβάνοις; (Lucian Ind. 1) For indeed how is it possible for you to discern what book-rolls are old and worth a lot, and what are of low quality and worn out, unless you should judge them by the extent to which they are corroded and cut up and should use the moths as advisers in the examination?

By opening his critique of the book collector’s ignorance with a mockery of his inability to distinguish between old books of high value and material texts in poor repair, Lucian points to the cultural capital carried by the ability to reveal familiarity with material texts. The description of the book collector’s book buying habits immediately before this comment implies that displaying such a critical judgment of material texts is especially important in the public setting of the bookshop. When foregrounding the book as their primary tool for the exposure of empty boasts of knowledge, Gellius and Lucian engage with a theme averted in both Greek and Latin Imperial sources. Due to their anxiety about the mutability of the material text, second century Greek and Latin texts often purposefully elide the presence of the book.22 Gellius’ and Lucian’s portrayals of the material text as an important means of determining status suggest, however, that the erasure of the role of the material book in Imperial Greek and Latin sources is a rhetorical trope, designed to perpetuate a specific literary agenda.23

 The Authority of the Text While Gellius and Lucian point to the importance of rewriting the book as a site of authority into Imperial intellectual culture, they also reflect upon the limits of

 22 Zadorojnyi (2013) 400. 23 In essence, Lucian and Gellius corroborate Zadorojnyi’s (2013) arguments for a philosophical significance behind the anxiety about and refrain from engaging with the material text (400). It is also important to note that Gellius’ and Lucian’s use of the material text in their confrontations of pseudo-intellectuals also functions as a rhetorical trope but one that stands in contrast to the anxieties of the texts examined Zadorojnyi.

54  Katherine Krauss their ability as narrators to use the material text to bolster their own credibility.24 Both Gellius and Lucian root this reflection in their exploration of the inaccessibility of the knowledge contained within books. Gellius’ emphasis on the inability of books to convey their learning, however, calls his textual analyses into question, whereas Lucian’s claim to intellectual status remains unaffected as he refrains from intertwining it with the material text and the knowledge it contains. When defending his interpretation of the archaic case endings of the fifth declension Gellius writes: quod autem supra scriptum est in Q. Claudi uerbis: ‘Propter magnitudinem atque inmanitatem facies,’ id nos aliquot ueteribus libris inspectis explorauimus atque ita esse, ut scriptum est, comperimus. (Gell. NA 9.14.1) Now the words of Quintus Claudius which were written out above, ‘On account of his greatness and fierceness in appearance (facies),’ we have investigated by examining old books and we have learned that the quote goes thus, as it was written above. (tr. adapted from J.C. Rolfe)

Gellius’ consultation of old copies of texts implies that the books have knowledge to impart on the correct archaic ending of fifth declension genitives. However, by omitting where he looked at these old book rolls, Gellius renders the material texts and their attestations of the archaic genitive endings inaccessible to his contemporary readership.25 Gellius continues his exploration of archaic fifth declension endings by detailing his findings on Cicero: Ciceronem quoque adfirmat Caesellius in oratione, quam pro P. Sestio fecit, ‘dies’ scripsisse pro ‘diei,’ quod ego inpensa opera conquisitis ueteribus libris plusculis ita, ut Caesellius ait, scriptum inueni. (Gell. NA 9.14.6–7) Caesellius also asserts that Cicero wrote ‘dies’ in place of ‘diei’ in the oration which he made on behalf of Publius Sestius, a word which I have found written thus as Caesellius said after I spent labor and sought out several old books. (tr. adapted from J.C. Rolfe)

 24 Gellius and Lucian thus engage in a discourse not unlike the anxiety about the material text discussed by Zadorojnyi (2013). 25 It is worth noting as well that although the texts Gellius consults are probably copies of Quintus Claudius, Gellius is not explicit on this matter either here or in his description of his research on Cicero explored below, thereby creating further ambiguity. On the general inaccessibility to texts, especially in libraries, Johnson (2010) 92–93, and Woolf (2013) 6; on restricted access to texts as a trope in Gellius more generally, Johnson (2010) 131–6.

Fake Intellectuals, and Books of Unquestionable Authority  55

As in the opening of his anecdote, Gellius’ use of old books to examine the veracity of Caesellius’ statement implies that the old books are able to teach him about the details of archaic grammar. Once again, by remaining silent on the location of the texts he consulted, Gellius prevents contemporary readers from verifying the learning contained within these book rolls. The juxtaposition of both Gellius’ research on Quintus Claudius and on Cicero with his description of a book attesting to the use of both facies and facii as fifth declension genitives in Tiburti bibliotheca (‘in the library at Tibur,’ Gell. NA 9.14.3) throws his silence on the location of his other sources into sharper relief. Since Gellius’ argument rests on the texts he is consulting, his obscurity on the location of these texts undercuts the integrity of his conclusions.26 In both his analysis of Quintus Claudius and Cicero, Gellius supports his statements with his research conducted on the old books. Since Gellius trains his readers throughout the course of the Noctes Atticae to become active and critical readers,27 Gellius’ silence on the source of the texts that form the basis of his arguments leaves his readers wondering why he does not provide them with the resources to check his analyses. The inaccessibility of the texts Gellius consults and the learning contained therein paradoxically both helps to ensure the monopoly of Gellius’ interpretation on the issue of archaic fifth declension genitives while simultaneously inviting the readers to question the methodology with which Gellius constructs this authoritative argument.28 Lucian alludes to the elusive nature of the knowledge contained within the book just as Gellius does. Comparing the book collector to his circle of learned friends, Lucian writes, τί οὖν; φὴς, καὶ ταὐτὰ μὴ μαθὼν ἡμῖν, εἰδέναι (‘Come on, are you saying that you know the same things as we do, although never having learned them?’ Lucian Ind. 3), and later answers his rhetorical question thus: καὶ σὺ τοίνυν βιβλίον μὲν ἔχεις ἐν τῇ χειρὶ καὶ ἀναγιγνώσκεις ἀεί, τῶν δὲ ἀναγιγνωσκομένων οἶσθα οὐδέν, ἀλλ᾿ ὄνος λύρας ἀκούεις κινῶν τὰ ὦτα. (Lucian Ind. 4) Moreover, you hold a book in your hand and are always reading it, but you understand none of what you read; you listen instead like a donkey moving its ears as it listens to the lyre.

 26 While this is not a trope that occurs frequently in Noctes Atticae, Gellius similarly subverts his own authority by not citing his sources in 2.3, 2.14, 18.5, and 20.10. 27 Howley (2018) 76-82. 28 On the possibility that Gellius’ authority is further undercut by the fact that many of the supposedly ‘ancient’ book rolls in the second century were fake see Zetzel (1973), Timpanaro (1986), and McDonnell (1996).

56  Katherine Krauss Lucian’s description of the ignorant book collector as always having a book in his hand to read yet never learning anything implies that the books the he owns are an insufficient means to becoming a true intellectual. Lucian underscores the inadequacy of these texts through his rhetorical question at Ind. 3, which infers that the book collector cannot possibly know as much as his own circle of friends since he has not received instruction.29 By highlighting the collector’s inability to understand the content he reads and to attain the level of learnedness of Lucian and his peers, Lucian emphasizes the inaccessibility of the knowledge in his books. Lucian’s metaphor likening the book collector to a donkey moving its ears to the sound of the lyre further confirms that the information contained within the books is completely unattainable for him. Since Lucian grounds his scholarly integrity in the experience of reading texts in a certain type of educational setting, his exploration of the unobtainability of the learning found in physical books does not undermine his authority as an intellectual. In his explanation of the book collector’s inability to read properly the books he purchases, Lucian emphasizes the importance of formal education: καίτοι οὐδέ, εἰ πάνυ ἀναίσχυντος εἶ καὶ ἀνδρεῖος τὰ τοιαῦτα, τολμήσειας ἄν ποτε εἰπεῖν ὡς ἐπαιδεύθης ἢ ἐμέλησέ σοι πώποτε τῆς ἐν χρῷ πρὸς τὰ βιβλία συνουσίας ἢ ὡς διδάσκαλός σοι ὁ δεῖνα ἢ τῷ δεῖνι συνεφοίτας. (Lucian Ind. 4) Indeed, if you are completely shameless and brave in matters of this sort, you would never dare to say that you were educated or that you ever took interest in close acquaintance with books or that so-and-so was you teacher or that you went to school with so- and-so.

Since the book collector never had a formal education, under the supervision of a teacher, he understands nothing when he reads his books ἀναγιγνώσκεις ἀεί… οἶσθα οὐδέν (‘you always read…but understand nothing,’ Lucian Ind. 4). As Johnson (2010) 168 points out, Lucian’s attribution of the book collector’s poor reading skills to his lack of education thus suggests that he cannot to learn the same things as Lucian’s circle of friends. Conversely, Lucian implies that his erudition rests on his learning through formal instruction.30 In other words, Lucian attributes his scholarly capabilities to his participation in a formal educational setting, and does not rely on the material texts themselves for proof of his intellectual dexterity as Gellius does. Lucian’s discussion of the inaccessibil-

 29 Similarly Johnson (2010) 163–8. 30 Johnson (2010) 168.

Fake Intellectuals, and Books of Unquestionable Authority  57

ity of the learning within the books purchased by the book collector thus does not impact his portrayal of his knowledgeability.31 By expressing this reserve about building narratorial credibility upon the authority of the material text and the learning contained therein, Lucian signals his interest in a set of thematic concerns also shared by Gellius’ Noctes Atticae. The adversus Indoctum’s approach to such an anxiety, however, is fundamentally different from that of the Noctes Atticae. Lucian cements his claims to erudition by privileging formal education over the knowledge contained within the book roll whereas Gellius constructs an image of authority from the books whose learning he obscures. In other words, the Noctes Atticae and aduersus Indoctum produce their own variation on a larger shared theme.

 Conclusions While Gellius and Lucian differ in their methods of articulating the book’s inability to augment their authority as narrators, the significance of the material text in imperial intellectual culture lies at the heart of the literary projects of both authors. Gellius and Lucian both uncover the superficiality of selfproclaimed ‘experts,’ described in pecuniary terms, through their inability to manipulate the material text. Furthermore, both Gellius and Lucian consider the capacity of the book roll to endow them as narrators with authoritative status. In so doing, they point to a shared cultural conception of the place of the book in second-century intellectual life. This paper has argued that Gellius’ and Lucian’s interest in the authority of the book casts light upon a larger network of ideas which transcends linguistic boundaries. Recent scholarship on Imperial Greek literature, however, has overlooked this shared rhetoric Greek and Latin sources from the second century. Instead, it has tended to divide elite literary and intellectual culture of the High Empire along these linguistic lines.32 Both the common nexus of concerns and shared tropes in Noctes Atticae and aduersus Indoctum call into question the meaningfulness of this separation of Latin and Greek intellectual culture in the second century.  31 As Johnson (2010) 161 rightly points out, however, the authority of the narrative voice is derided by Lucian (the author) who portrays the narrator as an exaggerated caricature of a certain type of second-century intellectual. 32 E.g. Whitmarsh (2001) 2, Gleason (2006) 12. See further Reardon’s (1971) 17–21 discussion on the absence of Rome in Greek texts.

58  Katherine Krauss This intersection of elite intellectual interests across Latin and Greek texts does not, however, solicit a consideration of these texts that elides all distinction. In fact, the differences between Noctes Atticae’s and aduersus Indoctum’s treatments of the breakdown of the authority of the book roll reflects in microcosm how such a totalizing assimilation does injustice to the diversity of thought within these works.33 Thus, while Gellius and Lucian both explore a pool of concerns that traverses Greco-Roman literary production they also mold these themes to fit their interests. In so doing, Gellius and Lucian reveal the affinity of ‘Latin’ and ‘Greek’ intellectual culture that underlies their ostensible differences.

 33 Swain (2004) 11–12, Bowersock (1969) 16.

Ambra Russotti

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  Introduction Book 10 of Epigrams is a very particular one in Martial’s literary production, as it went through a twofold publication1 handled by the author himself in AD 98, three years after the first edition. The poet takes on the issue of the second edition at the very beginning of the volume2 in epigram 10.2: Festinata prior, decimi mihi cura libelli elapsum manibus nunc revocavit opus. Nota leges quaedam, sed lima rasa recenti; pars nova maior erit: lector, utrique fave.3 In composing my tenth little book, too hastily issued earlier, I have now recalled the work that then slipped from my hands. Some of the pieces you will read are already known, but polished with a recent file, the greater part will be new. Reader, wish well both. (tr. Shackleton Bailey)

 1 It was not the only one to experience a second edition. Book 1, which was likely published for the first time in AD 86, must have had a second edition: the evidence is the opening section, in which epigrams like 1.1 (ll. 1–2: Hic est quem legis ille, quem requiris / toto notus in orbe Martialis […]) and 1.2 (ll. 1: Qui tecum cupis esse meos ubicumque libellous […]) were likely inserted after Martial’s great success rather than in his very first published book. See Schneidewin (1842) III, Dau (1887) 77, Immisch (1911) 485 and Citroni (1970) 82 (n. 1). According to Dau (Ibid.) Martial published a second edition of the whole batch of books 1–7, as we can argue by reading epigrams like 7.11 and 7.17, where Martial openly refers to an edition of this group of libelli. Dau’s hypothesis was embraced by Immisch (1885) and Lehmann (1931). Nevertheless, it is to be observed that in 7.17 the poet offers the collection of books 1–7 to the library of his friend Iulius Martialis, which had just been inaugurated. For this reason, it is quite likely that the group of libelli mentioned had been conceived as an occasional gift; see Nobili (2008) 356. 2 After one introductory epigram (10.1) in which the book itself directly invites his public to read just a few pages, in order not to get bored by its length. On epigrams 10.1 and 10.2 see Buongiovanni 2009. 3 The critical edition in use is the one by Shackleton Bailey (1990). https://doi.org/10.1515/9783110632590-005

60  Ambra Russotti In the first lines, Martial blames the festinata cura he had in composing his first version of the book for the necessity of publishing a second one. As he says, the epigrams which have been kept from the second edition have been subjected to a strict labor limae (nota leges quaedam, sed lima rasa recenti), but the most conspicuous part of the collection is made up of new epigrams (pars nova maior erit). The actual reason why the poet had to work on a second edition is a less ‘literary’ one. In September 96 Domitian had been killed in a conspiracy and Martial, who had made flattery and celebration of the tyrant a crucial part of his work, found himself in a very difficult situation, since everybody saw and remembered him as Domitian’s court poet. The first attempt he made to change his reputation was the publication of Book 11 during the Saturnalia of the same year, in which the mood of wildness is particularly accentuated by the death of the hated emperor.4 Nevertheless, the strategy of flattery carried out with Book 11 probably did not work – or had to be reinforced – if he decided to publish a second edition of his Book 10, in which all the positive references to Domitian had been removed. Besides, in January 98 the emperor with whom Martial had at least some familiarity, Nerva, died, and the poet had to deal with Trajan, a man he did not know at all – in spite of the fact that they both came from Spain – and whose reactions he could not predict. Martial knew that his time in Rome had come to an end, and only a few after the publication of editio altera of Book 10 he travelled back to Bilbilis. He spent the last years of his life in his hometown, where he wrote his last collection of epigrams and finally took a rest from the exhausting life he led in Rome.  4 See Fearney (2006) 622–5. In attempt to gain some credit, Martial also sent Nerva an anthology of his poems chosen from Book 11 and from the first edition of Book 10. Some of the additional poems written to be included in the anthology flew into the manuscript tradition, as they are inserted – we do not know whether by Martial or by an editor – in Book 12: they are 12.4. 12.5. 12.11 (numeration by Shackleton Bailey) and maybe 12.15, addressed to an anonymous emperor. The fact that the third family of manuscripts does not include those epigrams, in addition to the fact that it reports in general a reduced version of the book, led the scholars to believe that the text in family β and the one in family γ represented two different version of the text, modified either by Martial or by some editor. By considering the lack of epigrams for Nerva as a fundamental clue, Citroni (1988) 30 (n. 50) assumed that the recensio of γ could correspond to a less updated version of Epigrams; Sara Sparagna (2014) 5 observed that, despite the metrical homogeneity the more well finished appearance of family γ's text, also the thematic unity of the book in β's version could be a hint of an edition handled by Martial himself. It is a sure thing that by the time Book 12 was published Martial was already back in Spain and could not control directly the work of the librarii in Rome.

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  61

Getting back to 10.2, the real message of the epigram will now result clearer, even though the real motives are concealed behind a vague admission of formal negligence. It must not have been easy for a proud poet like Martial, absolutely aware of his talent, to simulate regret for his negligence,5 but he was pretty much forced to work on the revised edition of the book. The form in which Martial’s Book 10 of Epigrams first appeared in AD 95 is not known, since all the families of manuscripts agree in reporting the editio altera of 98. Nevertheless, there are some features of this last version which can become clues and help us to imagine how Martial adjusted his work. The flattering epigrams for Domitian Martial had to remove were probably a conspicuous amount, if we expect the poet to be coherent with the tendency started in Book 4 to grant more and more space to the celebration of the emperor. By taking into account the considerable increase of emperor’s celebration in Book 8 and then in Book 9, Citroni observed that likely the first edition of Book 10 included even more flattery, as Martial was more and more tied to official backgrounds.6 In this case, he had to eliminate a considerable amount of lines and, more importantly, to substitute them with others which were usable at the same way. In order to study the dynamics of Martial’s revision, it will be worthwhile to examine three fundamental aspects that from this point of view constitute integral part of Martial’s strategies for his second edition. First, the epigrammata longa. In our version of Book 10 there are nine of them: as it will be discussed, they present some interesting common features and were probably all tied to the second edition of the book. Secondly, the topic of plagiarism: Martial speaks quite often about the risks of ancient editorial world, where copyright was completely absent, yet there were some particular moments in his career in which this problem became more urgent. The last aspect is the topic of Spain, as by the time he published the second edition of the book Martial had already decided to go back to his hometown. The representation of Spain then becomes a crucial theme, which Martial links with another fundamental point of his poetry: the relationship with his public.

 5 It is remarkable, in the first line, the use of the verb festinare, which normally refers to the sphere of occasioning poetry and expresses the necessity of the court poet to be promptly present to celebrate the emperor. See Citroni (1988) 5–6 and Buongiovanni (2009) 520; for the use of this verb in Flavian poets in general and in Statius in particular, see Hardie (1983) 78–85, Johannsen (2006) 316–22, Newlands (2009) 232 (n. 1). 6 Citroni (1988) 27.

62  Ambra Russotti

 The epigrammata longa In the second edition of Book 10 there are nine poems that – according to the majority of the scholars – can be defined epigrammata longa:7 10.5 (19 lines), 10.20 (21 lines), 10.30 (29 lines), 10.35 (21 lines), 10.37 (20 lines), 10.48 (24 lines), 10.87 (20 lines), 10.92 (18 lines) and 10.104 (19 lines). What was their rule in the setting up of the second edition? Did they belong to the quaedam nota or to the pars nova? Four of them are addressed to important political personalities: 10.20 is the quite well-known epigram dedicated to Pliny the Younger;8 10.30 describes Domitius Apollinaris’ villa at Formia;9 in 10.48 Martial mentions Arruntius Stella and Iulius Frontinus;10 10.87 was written to celebrate Restitutus’ birthday.11  7 Likely, the typology of epigramma longum was already under debate in the age of Martial, given that in epigrams like 2.77, 6.65 e 10.59 the poet himself upholds the presence of such poems in his books. In modern studies, the discussion was opened by Hanna Szelest’s fundamental study in 1980, starting from which scholars have wondered when an epigram should be considered longum. The first criterion to be used is obviously a quantitative one. We ought to define as longum an epigram which has more than 11 lines according to Puelma (1997), more than 15 lines according to Szelest (1980), more than 20 according to Classen (1985) and more than 22 according to Canobbio (2008). More recently, new criteria have been proposed. Canobbio (2008), for example, suggested that we could consider the presence of apologetic poems as a clue for the fact that the longest preceding epigram was perceived as longum and for that reason requested apology. Buongiovanni, author of a very specific and rich study about the epigrammata longa in Book 10 (2012), simply suggested to be cautious about the choice of a univocal criterion: it is more appropriate to abide by the internal dynamics of the collection in object and to evaluate each case separately. Since in Book 10 the medium length of the epigrams is 8,6 lines, according to Buongiovanni we must consider longa all the poems that have more than 17–18 lines (that is, the double of the medium length), even if he admits that even this threshold is purely approximate, (2012) 15. For a formal classification of the epigrammata longa see Morelli (2008) 28–40. 8 The poet appoints the Muse Talia to bring Pliny a copy of his new collection, recommending to visit him only in the evening (cum regnat rosa, cum madent capilli, l.20) in order not to bother the magistrate during his diurnal activity. Pliny’s friendship with Martial is attested by the well-known Ep. 3.21, in which he honours the poet, recently deceased, and reports the lines that he had written for him; see Nauta (2006) 78. On Pliny’s career see Syme (1958) 75–85. 9 Domitius Apollinaris was a rich patron who Martial used to address in particular about literary and poetic topics (see 7.26 and 11.15), maybe the same man mentioned by Pliny the Younger in Ep. 9.13.13, consul suffectus in AD 97, lawyer and governor in Lycia and Pamphylia. On 10.30 see in particular Fabbrini (2007) 117–80. For the relationship with Martial, see Nauta (2006) 64 and 74, n. 124 and Nobili (2008). 10 Lucius Arruntius Stella was an important political man whose rise was not impeded by the fall of Domitian; see Sartori (1985). Mentioned by Martial in 17 more epigrams of his corpus, he

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  63

Martial wrote all those epigrams in order to flatter some of the most important personalities of the political élite (some of them already had power under Domitian and were not affected by the political change, like Arruntius Stella). Since the only reason why Martial could have written those epigrams was that he hoped to keep some relations with the dominating élite, we can attribute all of them to the editio altera.12  was probably one of his most important patrons and likely a sincere friend; cf. Buongiovanni (2012) 252. Sextius Iulius Frontinus – who is not between the friends invited by Martial but simply mentioned as Martial wants to serve his guests de Nomentana vinum sine faece lagona / quae bis Frontino consule trima fuit, ll. 19–20 – was a very influential figure at the end of the first century, praetor urbanus in AD 70, consul suffectus in 73 (and again in 98), legatus Augusti pro praetore in 77/78, proconsul in Asia between 85 and 85, curator aquarum in 97, consul ordinarius in 100. See Nauta (2006) 58; 61; 74 n. 124; 78–80 and 87–90. For the identification of the other personalities to which Martial addresses his invitation cf. Buongiovanni (2012) 252–9. 11 Restitutus should probably be identified with the Claudius Restitutus mentioned by Pliny the Younger in Ep. 3.19.16 (Solet dicere Claudius Restitutus, qui mihi respondit, vir exercitatus et vigilans et quamlibet subitis paratus); see Nauta (2006) 105–7. As Buongiovanni (2012) 308 properly observed the fact that Martial chose to celebrate a birthday which fell in October could be significant for two reasons. First of all, Domitian was born in October too, and in AD 85 he even changed the name of the month in Domitianus (September became Germanicus, to commemorate the fact that the emperor took the power in this month). Secondly, in October 97 Nerva adopted Trajan, and gave to this month a new significance: the passage to a new regime and the birth of a felicitas temporum (see Buongiovanni [2012] 309). 12 Epigram 10.48 does deserve some extra observations. As it has already been said, the important personalities here addressed are Arruntius Stella and Iulius Frontinus, two of Domitian’s officials whose career was not affected by the abolitio memoriae of the tyrant. Consequently, this poem could appear also in the first edition of the book. Developing the traditional subject of uocatio ad cenam, the poet addresses his invitation to six of his friends, he accurately describes the dishes and he concludes by saying that no risky conversations are going to be made at his table: de prasino conviva meus venetoque loquatur / nec facient quemquam pocula nostra reum. In line 23 uenetoque is reported by T, representing here the first family of manuscripts. In the second family we read de prasino conviva meus scutoque loquatur and in the third family de prasino conviva meus scipioque loquatur. By developing a suggestion made by Gruterius, Lindsay wittily supposed that the text reported by the second and the third family should come from a previous de prasino conviva meus Scorpoque loquatur. Scorpo was a very famous charioteer who rode for the Green faction: he was still alive when Martial published the first edition of Book 10, but he was already dead by the time the poet worked on the editio altera. Lindsay (1903) 14–15 supposed that Martial noticed it while working on the second edition and changed Scorpoque in uenetoque as the reference to the charioteer no longer sounded current. About 50 years later, Pasquali (1952) 420 supported this view in his discussion about author’s variants in Martial’s Epigrams. Fusi (2011) 273–5 has recently pointed out some important arguments against Lindsay’s hypothesis and proposed that the only right lectio was scutoque (that is, what family β reads), since it is well attested the metonymic use of the word scutum to

64  Ambra Russotti 10.35 is a celebration of the poetess Sulpicia, whose life and work cannot be fully reconstructed:13 Martial praises her modesty and her grace, lauds the fact that she used to criticize excessive passions, admires her loyalty to her husband Calenus. We should attribute 10.35 to the editio altera for two reasons: firstly, like Buongiovanni pointed out, the tradition attributes to Sulpicia a very harsh criticism of Domitian’s regime, and it is not very likely that Martial wrote this praise when the tyrant was still alive. Besides, if Sulpicia and her husband Calenus had been somehow unpleasant for Trajan, probably Martial would have avoided praising them. The only one epigram that according to Buongiovanni could have been ‘recycled’ from the edition of 95 is 10.5.14 It is a long invective against one poeta clancularius who wants to attribute to Martial some insulting lines. This polemic shares the theme of 10.3, 10.33 and 10.100, which can be considered all together as a small ‘cycle’ against plagiarism. This epigram should be attributed to the editio altera just like all the others epigrammata longa. In fact, as the following paragraph should prove, all the epigrams about plagiarism were published in AD 98 (see infra). To conclude, Martial probably inserted all the epigrammata longa to substitute and cancel the praise of Domitian. The first purpose was to preserve the length of the book because, as we have already seen, Martial probably accorded a lot of space to flattery in the first edition. Secondly, the poet used them to gain  designate gladiator games; this suggestion was already in Schmid (1980) 406–12. Through the nexus de prasino … scutoque, Martial would then allude both to gladiators and to charioteers. Consequently, family γ’s lectio, scipioque, would be a mechanical corruption, while the venetoque of the first family would be a ‘normalizing interpolation’. On the other hand, Shackleton Bailey, the last editor of the Epigrams (for Teubner in 1990), significantly prints in his text Scorpoque, which was in his opinion the only authentic reading. What is important, here, is the fact that if we really read an author’s variant in the manuscripts, 10.48 was the only epigramma longum to be included in both the editions of the book. 13 Not to be confused with the other – more famous – Sulpicia who lived during the Augustan age. Martial names the poetess again in 10.38, that is addressed to her husband Calenus (O molles tibi quindecim, Calene, / quos cum Sulpicia tua iugales / indulsit deus et peregit annos! ll. 1–3). Later mentions of her poetry are in Sidonius Apollinaris in a long list of poets whose style he intends not to imitate in his poetry (Carm. 9.259–264), in Ausonius (Cent. nupt.) and Fulgentius (Myth. 1), but the one by Martial is the most conspicuous and important one (besides, it is not to be excluded that Martial was the direct source of the other three authors). The tradition attributes to Sulpicia two lines included in the Probus Vallae, and the so–called Conquestio Sulpiciae, a complaint on the age and the state of Domitian; see Lana (1949), and Giordano Rampioni (1982). 14 See Buongiovanni (2012) 21, even though the scholar himself admits that also this epigram quite surely belonged to the second edition.

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  65

some prestige and to get some approval in Trajan’s entourage. The poet eventually achieved some consideration, even it was not the expected result: Pliny the Younger, addressed in 10.20, paid the viaticum to let the poet go back to Spain.

 Plagiarism In ancient Rome, the word plagium indicated the kidnapping of runaway slaves and indicated that their masters had a legal right of property over them, but it was not a widespread term with regards to literary activity. Since his epigrams were normally read and recited per convivia cuncta, per theatra (2.6) and consequently exposed to the risk of plagiarism, Martial was the very first poet to use the term plagiarius (1.52) referring to literary theft.15 As Mira Seo (2009) has pointed out, there are at least three categories in which we can divide Martial’s poems about plagiarism. A first category is made up of epigrams about ‘standard plagiarism’, in which Martial simply claims that someone is stealing his poetry and taking credit for his work. A second group is constituted by poems about forgery, that is, in Martial’s case, the insertion of insulting or unworthy lines in his published books. The last typology is represented by epigrams that denounce the fact that some anonymous detractor attributed to the poet from Bilbilis some denigrating carmina he would never have written.16 To the second  15 The evidence for Martial’s concerning for this type of accuses is already in the prefatory epistle of Book 1: Spero me secutum in libellis meis tale temperamentum ut de illis queri non possit quisquis de se bene senserit, cum salva infimarum quoque personarum reverentia ludant. (‘I hope to have struck a balance in my little books such that nobody can complain of them who has a good opinion of himself; their jesting is with respect of persons, even the humblest’; tr. Shackleton Bailey). 16 Even if Martial was the only one poet to deal with the topic of plagiarism in his work, Roman intellectuals used to give to it some consideration. For example, Pliny the Younger refers to plagiarism in 2.10, addressed to his friend Octavius, who procrastinates the publication of his poetry: Enotuerunt quidam tui versus, et invito te claustra sua refregerunt. Hos nisi retrahis in corpus, quandoque ut errones aliquem cuius dicantur invenient (‘Some few poems of yours have already, contrary to your inclinations indeed, broke their prison and escaped to light: these if you do not collect together, some person or other will claim the agreeable wanderers as their own’; tr. Melmoth). Even without the presence of real plagiarism, works could be spread without the authorization of the author: Pliny himself reports, in Ep. 4.26, that some of his orations did circulate against his will; Quintilian, in the prefatory letter to his editor Triphon – who was also editor of at least one part of Martial’s work, see Xen. 3 – at the beginning of Institutio oratoria, denounces the runaway circulation of some notes that some of his student took during his lessons and published without being authorized.

66  Ambra Russotti and the third category belong the epigrams dealing with the topic of plagiarism: 10.3; 10.5, which is also an epigramma longum; 10.33; 10.100. It is to be cleared that it is perfectly normal, in Martial’s work, for a single topic to be ‘divided up’ in two or more poems.17 For this reason, we are allowed to consider a group of epigrams regarding the same theme and localized in a single book as a ‘cycle’ which can be studied, if necessary, on the whole. 10.3 and 10.5 are absolutely complementary, as they are both written in scazons and integrative in attacking the same anonymous poet who spreads offensive epigrams under the name of Martial. 10.3 is strategically placed in the opening part of the book, following the poet’s warning about the second edition, as he wanted to immediately clarify that calumny and insinuation were not part of his work. The contrast between Martial’s authentic verses and his detractor’s verses is developed by following a chromatic criterion: the fama nigra which comes from libel is opposed to the alba pinna that is meant to lead Martial’s poetry towards true glory. In 10.5 we discover that the poeta clancularius wrote insulting verses against important personalities, such as matrons and magistrates. As it has been said, in AD 98, when he had to work on the editio altera of Book 10, Martial position in the eyes of the élite was already compromised, and he really had no need to be discredited further. The whole poem is in fact made up by a series of curses and it is closed by a terrible wish (ll. 13–19): Nec finiantur morte supplicis poenae, sed modo severi sectus Aeaci loris, nunc inquieti monte Sisyphi pressus, nunc inter undas garrulis senis siccus delasset omnis fabula poetarum: et cum fateri Furia iusserit verum, prodente clamet conscientia ‘Scripsi!’ And let not his punishment end with a simple death, but let him weary all the fables of the poets, now lashed by the thongs of stern Aeacus, now crushed by the mountain of restless Sisyphus, now dry amidst the waters of the old chatterbox. And when the Fury commands him to confess the truth, let conscience betray him and let him cry out: ‘I wrote it! (tr. Shackleton Bailey)

 17 In the twelve books of Epigrams there are various ‘cycles’ which have been studied separately. The most famous examples are the cycle of the lions and hares (1.6, 1.14, 1.22, 1.48, 1.51, 1.60, 1.104) or the cycle of Postumus in Book 2 (2.10, 2.12, 2.21, 2.22, 2.23), on which see Borgo (2005).

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  67

10.33 is more moderate. Addressing the friend Munatius Gallus, Martial defends his verses from the accusations and expresses the real sense of his poetry through a formula which would become quite well-known: parcere personis, dicere de vitiis. It is one of the epigrams in which Martial defends himself from the charges by asking his influent friends and patrons to guarantee for his poetry.18 The last epigram about plagiarism, 10.100, directly addresses to Martial’s rival an exasperate question: Quid, stulte, nostris versibus tuos misces? Cum litigante quid tibi, libro? Quid congregare cum leonibus vulpes aquilisque similes facere noctuas quaeris? Habeas licebit alterum pedem Ladae, inepte, frustra crure ligneo curres. Fool, why do you mix your verses with mine? What do you want, wretch, with a book at odds with itself? Why seek to herd foxes with lions and to make owls resemble eagles? Though you have one foot like Ladas, it’s no use, you silly man, trying to run with a wooden leg. (tr. Shackleton Bailey)

We should now make some considerations about plagiarism in Epigrams. It has been reminded that in the age of Martial any work which have been divulged – even not officially – could be stolen, and even if plagiarism must have been a constant problem for the poet, he speaks about the topic only in some particular moments of his career. In the first book of Epigrams is included a small cycle against the poet Fidentinus (1.29, 1.38, 1.53 and 1.72), blamed by Martial for repeatedly stealing and reciting his poems19. What we can deduce from the data  18 Di Giovine (2003) 84 identified three main categories of Martial’s poems against plagiarism: the direct attack of the detractor, the guarantee offered by the poet himself that he did not write the epigrams in question and the guarantee offered by the poet’s friends and protectors that he did not write the epigrams in question. 19 According to Barwick (1958) 308, the fictitious name Fidentinus alludes to the man’s insolence, while Citroni (1975) 97 thinks that the nomen hints to his lack of fides. In 1.29 accuses Fidentinus of stealing his epigrams and ironically invites him to buy them, in order to be allowed to recite them in public (on this theme see also epigram 2.20). In 1.38 Martial notices that even if his rival steals his epigrams to read them in public and take credits for other’s work, his recitation is so bad he is easily unmasked. Readers find out from the following two pieces against him that Fidentinus was not only a thief but also a (mediocre) poet himself: just a few of his poems are enough to clarify his poor talent even if they are inserted in a whole libellus stolen from Martial (Una est in nostris tua, Fidentine, libellis / pagina, sed domini signata figura, 1.52.1–2). Finally, 1.73 exploits some contents typical of skoptic Greek epigrams: just like the numerous expedients devised by human vanity to hide physical flaws are useless, Fidentinus

68  Ambra Russotti given by the poet himself is that, at the beginning of his career, the poet’s fame was not big enough allow audience to recognize his lines as soon as they hear them, and for that reason almost everyone had the possibility of stealing and even changing the epigrams without consequences.20 In the other books of Epigrams there are a few more poems21 about plagiarism but it is remarkable that the highest concentration is in Book 1 and in Book 10. It is very likely that plagiarism was a concrete danger for Martial only in some particular moments of his activity, that is to say when is reputation was more fragile. This dynamic is well expressed by the two different typologies of plagiarism to which the poet was subjected. At the beginning of his career people could steal his epigrams and recite them with impunity because he was not known enough,22 and that justifies the presence of poems about plagiarism in Book 1; after the death of Domitian, as a consequence of Martial’s loss of credit among the élite, detractors used his celebrity to spread offensive verses, causing the presence of the cycle against his detractors in Book 10. It will result clear, then, that all epigrams which deal with plagiarism must be attributed to the editio altera of AD 98.

 cannot expect to be successful in hiding his lack of poetic talent. Two more epigrams in Book 1.1.52 and 1.66, may be referred to some attempts of plagiarism by the same man. 20 Martial was not completely unknown by the time he published Book 1: he had already sent some of his libelli to the emperor, as in 1.101, mourning the death of his scribe Demetrius, he says that his hand was nota Caesaribus. Besides, in 1.103 the poet mentions the fact that his iuvenilia could be bought in Pollio’s taberna. Nevertheless, it is likely that his fame was not big enough to protect him from plagium. The fact that the majority of epigrams about plagiarism are included in Book 1 received more than one explanation: Lehmann (1931) 18 observed that this kind of complaint supposed a published work because we can’t speak about plagiarism if a book has not been divulged. On the other hand, Citroni (1975) 96 gave a more likely explanation, according to which Martial was trying to defend his poems from the risks they were exposed to during the recitations: those risks were probably the reason why he started publishing books. 21 For example, 7.62, addressed to the poet’s friend Paulus, and, later, 12.63, written against a plagiarius that pesters the poet even in Spain. 22 See the explanation in Citroni (1975) 96. This hypothesis was already in Friedländer (1886) 243.

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  69

 The return to Spain Martial normally comes across very proud of being Spanish, and more precisely Celtiberian.23 Citroni (2002) has rightly observed that he shows a singular attitude in comparison with other Spanish authors in Latin literature, who normally tend to hide their peregrinitas. Seneca, for example, tells us a lot about himself and his personal life, yet he never alludes to his origins even when we would expect him to do so;24 Quintilian does not mention the fact he is Spanish, and when he speaks about his homeland he even shows off ignorance about its linguistic common use: ‘gurdos’ quos pro stolidis accipit vulgu, ex Hispania duxisse originem audivi. I have heard that gurdus, which is colloquially used in the sense of ‘stupid’, is derived from Spain. (tr. Butler)

This shared refusal for their foreign origins could be interpreted as a stylistic issue: authors like Seneca and Quintilian – also Lucan, even though in his case the lack of personal information is to be explained with the fact he was an epic poet – connected their origins with the literary features (but it would be more appropriate to define them ‘flaws’) that at their age were considered typical of Spanish writers: we know from Seneca the Elder (Suas. 6.27) that Cicero, – speaking of poets from Cordova, defined them as pingue quiddam sonantes atque peregrinum and Seneca the Elder himself mentions the surda vox of the rhetor Latron as a typically Spanish feature.25 By refusing their peregrinitas,

 23 Of all Spanish natives, Celtiberians represented the most distant culture for Romans; see for example Strabo 3.4.20 and Cicero Planc. 84. Martial proudly states his belonging to the primitive tribe especially in 4.55, 10.65, 10.78. 24 For example, in two parts of the Consolatio ad Helviam matrem. The first one is 19.2, where he recalls the time in which he was brought to Rome in his aunt’s arms, and the lacking mentions of Spain is strange enough to look intentional; cf. Citroni (2002) 284. The other is chapter 6, where Seneca mentions the incredible quantity of strangers living in Rome for their own choice: among these voluntary exiles there was his family, but he does not say a single word about it. On the other hand, Seneca the Elder often speaks about other rhetors from Spain and about his – often close – relationship with them, but openly admits to come from Spain only twice (Suas. 2.18 and 6.27). 25 Even in modern criticism the Spanish origin of the majority of imperial authors was interpreted as the actual reason for some recurring characteristics in poetry and literature. The following words by Muret, in the praefatio to his commentary to Catullus (Venice 1554), are emblematic: Hispani poetae praecipue et Romani sermonis puritatem contaminarunt, et, cum

70  Ambra Russotti Spanish writers and poets intended to refuse those ‘flaws’. It is probably not for a coincidence that Mela and Columella, both authors of technical works, do not seem to have problems with their origins: style and formal elegance were not their primary concern.26 Martial, however, is the only one to exploit his foreign origins as a literary theme, often presenting his homeland to the public in a very personal depiction and, what is more, often presenting himself as a typical Spanish.27 The epigrams in which Martial deals with this topic are nearly sixty, while the ones in which he stresses the fact he is himself a Spanish are twenty–three.28 Most of the epigrams concerning the theme of homeland are gathered in the last three books,29 that is to say in Book 11 (published in AD 96), in the editio altera of Book 10 (published in AD 98) and in Book 12, published when the poet was already in Spain (around AD 101). The second edition of the Liber decimus was the last work that Martial published in Rome: by the time he edited it, he had arranged his travel back home, and the whole book is full of nostalgic epigrams which depict Spain as an ideal place to live, in contrast with Rome’s chaotic life. The poet’s imminent return is announced in the first section of the book by epigram 13 and the theme is constantly recalled by six more poems (10.37, 10.78, 10.92, 10.96, 10.103 and

 inflatum quoddam et tumidum, et gentis suae moribus congruens invexissent orationis genus, averterunt exemplo suo ceteros a recta illa et simplici, in qua praecipua poetarum sita laus est, et in quam superiores omni studioincubuerant, imitatione naturae. 26 Mela refers to his hometown in 2.96 (atque unde nos sumus … Tigentera), while Columella repeatedly alludes to Spain as his country of origin (see in particular 8.16.9 and 10.185). 27 See for example epigram 10.65: Cum te municipem Corinthiorum / iactes, Charmenion, negante nullo, / cur frater tibi dicor, ex Hiberis / et Celtis genitus Tagique civis? / An voltu similes videmur esse? / Tu flexa nitidus coma vagaris, / Hispanis ego contumax capillis; / levis dropace tu cotidiano, / hirsutis ego cruribus genisque; / os blaesum tibi debilisque lingua est, / nobis filia fortius loquetur: / tam dispar aquilae columba non est, / nec dorcas rigido fugax leoni. / Quare desine me vocare fratrem, / ne te, Charmenion, vocem sororem (‘Since you boast yourself a fellow townsman of Corinthian bronzes with none to gainsay you, Charmenion, why do you call me ‘brother’, born as I am of Iberians and Celts, a countryman of Tagus? Is it that we look alike? You go around looking smart with your hair in curls, mine is stubborn and Spanish. You are smooth with daily depilatory, my shins and cheeks are hairy. Your mouth lisps and your tongue are feeble; my daughter could speak bolder than you do. A dove is not so different from an eagle or a fugitive doe from a stark lion. So stop calling me ‘brother’, Charmenion, in case I call you sister’; tr. Shackleton Bailey). 28 See Citroni (2002) 287, n.13 for a complete list; see also Merli (2006b). 29 For the epigrams about Spain which are included in the first books of Epigrams, see Citroni (2002) 288.

Martialis Epigrammaton liber decimus: Strategies for a Second Edition  71

10.104); other six are dedicated to some friends who were about to leave (10.12, 10.30, 10.44, 10.51, 10.58, 10.78, 10.93). The topic of return is developed in two ways: on the one hand Martial exalts the calm life of the small town (as in 10.37, 10.78, or 10.96) and on the other hand he shows his resentment against Roman readers,30 that he thought to be partly responsible for his choice to go back to Bilbilis. It must be clear that Martial’s ideal representation of Spain is nothing but an illusion and that when he depicts his homeland like the perfect place he is actually describing the life of his Spanish friends and protectors, who were perfectly integrated in that reality. Nevertheless, the poet knows that his caustic poetry could be appreciated only in a metropolis like Rome.31 Consequently, in epigram 103, it is quite clear that Martial feared his return because he was sceptical about the reaction if his new public.32 Even before leaving, the poet knew he would feel the desire to go back to Rome: Exipitis placida reducem si mente, venimus; aspera si geritis corda, redire licet. If you receive me back in kindly mood, I come; if the hearts within you are ungentle, I can return. (tr. Shackleton Bailey)

 Conclusions It will never be possible to reconstruct precisely Book 10’s first edition: we do not know how extended the praise of Domitian was, how many ‘dangerous’  30 Which was already present in Book 11, for example in 11.108: Quamuis tam longo possis satur esse libello, / lector, adhuc a me disticha pauca petis: / sed Lupus usuram puerique diaria poscunt. / Lector, solve. Taces dissimulasque? Vale; see 10.74 for the presence of this topic in Book 10. 31 Once in Spain, the poet would immediately regret Rome and the life he used to lead, in spite of all the complications he had to face as a precarious cliens. In Book 12 it is possible to identify a new tension between Spain, once represented as the typical locus amoenus and then a symbol of cultural isolation in this very book, and Rome, once despised and now regretted; see Parroni (1984), Merli (2006a) and Sparagna (2014). 32 It has been optimally illustrated by Buongiovanni (2011) that the real addressee of 10.103 was not Bilbili’s potential new audience but Martial’s roman public. This epigram is in fact the poet’s last claim of dignity and literary authority, developed through a particular kind of sermo figuratus.

72  Ambra Russotti topics Martial had to wipe out and in general how different it was from the editio altera of AD 98. The aim of this study was to define at least the nature of the modification that Martial operated, by considering its features also in comparison with other books. Firstly, it has been clarified that all the epigrammata longa should be attributed to the second edition. The topic of these epigrams, in addition to the fact that Martial had to fill a lot of space that was previously dedicated to the praise of Domitian, should make us conclude that Martial used them both to keep intact the size of the book and – in part – to approach the new élite.33 Also the theme of plagiarism is likely a feature of the editio altera because, as suggested above, even if the lack of copyright was a constant problematic in his entire career, Martial mostly deals with this topic when his reputation and position were precarious (that is to say, at the beginning of his poetic activity and at the end of Domitian’s regime). The brief ‘cycle’ on this theme is then ascribable to the poet’s revision and, what is more, identifiable with Martial’s last attempt to save his position among the powerful. Finally, the representation of Spain. In his first books Martial never hid his foreign origins yet he never spoke at length about his homeland. In the second edition of Book 10 Spain becomes a dominant theme: the poet’s homeland is often depicted as an ideal space, opposed to the chaotic and – more importantly – ungrateful city of Rome.

 33 It has been optimally illustrated by Buongiovanni (2011) that the real addressee of 10.103 was not Bilbili’s potential new audience but Martial’s roman public: this epigram is in fact the poet’s last claim of dignity and literary authority, developed through a particular kind of sermo figuratus.

Antonio Iacoviello

Poetic Quotation in 4th Century BC Attic Oratory From the Court to the Written Text 1. Quotation of poetic texts within forensic speeches is a well-known practice in the Athenian 4th century BC oratory:1 however, cases of this expedient are not very common. First, a main point is the understanding of the dynamics which induced the orator to use poetic elements in his judicial argumentations. This technique fits in the constant presence of ‛equipment texts’, included in the speech in order to strenghten the thesis or the topic currently exposed; these texts are mainly documents2 of different natures (laws, decrees, oaths, epikleseis). Their use is essentially comparable to the quotation of poetic texts, because of their function in the oration and their similar textual history. Indeed, this history starts even before the ‛author manuscript’ creation: external constituents (i.e. documents) were in fact submitted in advance, within a previous step of attic trial procedure, the so–called ‛preliminary sessions’,3 as the litigants supplied the documents to be read during the real session. As surviving speeches attest, the quantity these texts could have been large. 4 This material – presumably written on sheets of papyrus rolls, or even on ostraka in the case of short testimonies – was then dropped off in an echinos, a vase which

 1 The first inspection which discusses the phenomenon analytically and with ambition of completeness – especially from the narratological point of view – is Perlmann (1964). 2 The presence of documents can be observed particularly in the demosthenic corpus. However, in Lycurgus’ Against Leocrates can be found martyriai (20, 23, 24), synthekai (24), proklesis (28), psephismata (36, 114, 118, 120, 122, 125, 146), the ephebic horkos (77), the oath pronounced before the battle of Plateia (81) and the nomos Lakeidamonion (129); nevertheless, we may already notice that none of these text, besides the Plateian oath, is quoted in extenso in Lycurgus’ manuscripts: there are just the titula, in only one branch of the tradition. 3 Canevaro (2013) 2; these ‘preliminary hearings’ were called anakriseis within public lawsuits (graphai), diaitai in private (dikai). 4 Theophr. Char. 6.8 describes the prototype of the man affected by aponoia, who reflects the typical behaviour of Athenian citizens always involved in trials, who goes to the preliminary sessions with a filled echinos. https://doi.org/10.1515/9783110632590-006

74  Antonio Iacoviello could reach huge dimensions5 and that was kept closed until the day of the trial. This iter concerned the documentary material but, in absence of accurate witnesses, it is likely that poetic texts (and their written support) had the same fate. It can be argued that both documents and literary texts were delivered during the trial by the grammateus, the clerk, an inferior rank Athenian magistrate6 who served as an ‛assembly secretary’. Since most of the poetic sections of surviving orations were pronounced by the clerk upon at the orator’s invitation,7 these texts as well were gathered up and then read at the right moment, namely the trial. Nevertheless, the correlation between poetry and documents does not invalidate the relevance poetry itself had in the 4th century, not only in Athenian society but especially in oratorical training and rhetorical practices. 8 The origins of quoting poetry are not clear; in addition, there is no overall modern study of 5th century practices.9 One of the most ancient cases is reported by Aristotle (Rh. 1.15 1375b) who includes the ancient poets among the martyres which may be employed in judicial argumentations. He quotes a chapter by a supposed Kleophon’s speech Against Kritias; we know nothing about such trial,10 except that Kleophon claimed to quote Solon by reporting a verse, which Aristotle considered as spurious. Moreover, the way Kleophon extrapolated and inserted it is uncertain: the word elegeion, indeed, pertains to poems of different dimensions.11 According to Harpocration (κ 15 Keaney s.v. καρκίνος), Lysias quoted verses of the tragic poet Charchinos in the speech Against Mnesimachos. It is

 5 For the echinos, see Boegehold (1995) 79–81, with archaeological and literary testimonia, among which Aristophanes’ Wasps have a particular role. For the procedure of documents depots, see Arist. Ath. Pol. 53. 2–3. 6 Alessandri (1982) 46; it is the grammateus tes boulès, different from the grammateus katà prytaneian, and from the epì tous nomous one, especially after 366–365 BC; cf. Arist. Ath. Pol. 54.3–5. 7 Sometimes clerk and orator speak to each other, for ensuring a correct reading of the document; cf. Dem. 20.84 where the orator invites the grammateus to hurry up finding a decree; the clerk was not able to find due to the great confusion inside the echinos. This is a singular case of passage describing procedural circumstances which entered the text. 8 North (1952); cf. Aristophanes Wasps 579–80 (where the poet refers to the delivering of a rhesis from Aeschylos’ Niobes) and Isoc. 2.48–49 (he invites the prose writers to appreciate the mythodes the ‛fabolous’ more than the ophelimon the ‛helpful’). 9 On Herodotus, Marincola (2006); for Thucydides, Jung (2006). 10 Doufour (1932) 137. 11 TGL s.v. ‘plurali de uno disticho’, or ‘de epigrammate quatuor versibus elegiacis comprehensa’; but the word means longer poems too, as ‘intellegitur autem ᾠδή vel ποίησις’; cf. Rowe (1972) 441.

Poetic Quotation in 4th Century BC Attic Oratory  75

interesting to note that, according to scholars,12 this speech was delivered within a public lawsuit, an element that sheds light on the use of poetry: indeed, the quoting of the great poets of the past would have been inadequate in a private trial.13 This habit appears clearer from the analysis of the speeches where poetry is found, both isolated and concentrated in a section of the speech: Homer Hesiod Tyrtaeus Solon Lycurgus



Aeschines – Demosthenes

Sophocles

  



Euripides

Epigrams

Adespota





–











These speeches (Aeschin. 1–2–3; Lycurg. Leoc.; Dem. 18–19) are comparable for several reasons: first, they are only forensic discourses, but on the other hand are characterized by a strong political mark: they were delivered in the period of Chaeronea, when social and collective destabilization was widespread. Furthermore, these trials (see esp. Aeschin. 1 and Lycurg.) distinguish themselves by showing a certain lack of serious probative elements which could support the accusation. For this reason, when there is no sufficient evidence that supports the orator – even in the form of documents – he calls on poetical authority to support the law and its interpretation:14 the poetic piéce is contextualized and explained as a document, and the orator steers the listeners to the right interpretation of the read or recited texts.15 The small number of speeches where we find poetry is not a coincidental phenomenon. These quotations, especially when they were extended, could damage the effectiveness of the speech: the Athenian public in the lawcourts (hoi periestekotes) was not quiet, rather they were easily distracted, compromis-

 12 Blass (1898) I 373; Carey (2007) 440. 13 In the extant production of Lysias, which not casually is composed by forensic private speeches, poetic quotations are completely absent; Scodel (2007) 134: ‘nobody quotes poetry in a private oration’; Carey (2015) 112 speaks of ‘overdramatizing’. 14 Dorjahn (1927) 91; Perlmann (1964) 167; Scodel (2007) 136. 15 Often by emphasizing the moral values perceivable from the text; Fisher (2001) 290.

76  Antonio Iacoviello ing the trial session;16 even though the delivery time of external texts (i.e. documents, poetic quotes)17 was defined by an hourglass (klepsydra), the risk of ‛boring’ the audience was high. Therefore, the orator had always to not lead his reasonings hexo tou pragmatos, ‛out of topic’.18 These observations, combined to the length of some quotations, caused some scholars to argue that these performances (or at least some of them) did not take place during the trial, as they were incompatible with practice and timing of forensic oratory.19 These quotations20 are likely to have been inserted afterwards, within an editorial context; it is an operation similar to that occurring with the documents, which are often omitted in the manuscript tradition.21 This mechanism is recognizable every time the quotation is introduced by the exhortation to read the text; this incitement means that it does not belong to the original speech draft. However, there are some cases when the poetic text is read by the orator himself, or it is doubtful whether the poetic quotation is delivered by the speaker or the clerk: Opening phrases

Passages

Reader

λέγε

Aeschin. . – Dem. .; 

Clerk

ἀναγίγνωσκε / ἀνάγνωθι

Aeschin. .; 

Clerk

ἄξιον τῶν ἰαμβείων ἀκοῦσαι

Lycurg. 

?

χρήσιμον ἀκοῦσαι τῶν ἐλεγείων

Lycurg. 

?

Ἔστι δὲ τὰ ἕπη, ἃ ἐγὼ νυνὶ μέλλω λέγειν

Lycurg. 

Orator

 16 With thorybos, a continuous and noisy buzz; the sessions could even be stopped because of people getting up and positioning themelves on the bema; [Dem.] 59.43; Lanni (1997); Carey (2015) 113. 17 We adopt the name ‘external texts’ since documents and poetic sections were not part of the orator’s draft of the speech. 18 This is the reason why the orator always has to justify himself for using poetry as a juridical instrument: Aeschin. 1.141; Lycurg. 102; but especially Dem. 19.241, 245, 255. 19 Especially the 16 verses of the quotation from Anthigones in Dem. 19; passing through the 39 vv. of solonian Eunomia in Demosthenes too; arriving to the monstrous 55 vv. of Lycurg. §100. 20 Durrbach (1932) 31; Dorjahn (1972) 62–63; Martin – De Budé (1952) 70; Pinto (2003) 7; Olding (2007) 168; there are obviously voices contra: Rowe (1972); Carey (2015), which yet are based on argumentative and literary reasoning, not dealing with the judicial practice and text history. 21 Drerup (1898); Canevaro (2013). In this case, poetic texts may be regarded as equivalent to all those which Canevaro calls ‘non stichometric documents’.

Poetic Quotation in 4th Century BC Attic Oratory  77

Opening phrases

Passages

Reader

None

Aeschin. .; Lycurg. ; ; 

Maybe the orator

This analysis concerns only those passages which were certainly read by the clerk; their status as external texts suggests that they share the same textual history of the documents. Moreover, whilst the latter are often replaced with spourious specimens22 in a later age, the insertion of a poetic section (e.g. Homeric) was a much easier thing. Nonetheless, these quotations are characterized by textual divergences from vulgatae, which are sometimes really marked. The aim of this study is the analysis of this peculiar kind of indirect tradition, the comprehension of the actual text read by the orator, the way the quotations were delivered and the vicissitudes of these insertions in the ancient text history. 2. Aeschines was, in his youth, a tragic actor;23 perhaps the fact that in every one of his three orations we find a poetic quotation is not accidental. The most interesting case is the Against Timarchus, speech pronounced in the early 346 BC24 against Timarchus, an active politician of that time. It was a preventive accusation, since Timarchus was supposed to join the graphe parapresbeias which Demosthenes was going to charge against Aeschines, after the second of the two embassies sent by Athens, which established the so-called ‘Peace of Philocrates’. Thus, the accusation had a clear political intention; and Aeschines had just rather specious reasonings for charging Timarchus. However, the legislative background on which he bases his accusation appears solid: first, he invokes three laws;25 the most relevant being the last perì hetaireseos, which prevented

 22 Cf. Aeschin. 1.12; 16; 21. 23 Even thought he was just a tritagonistès, the third actor; cf. [Plut.] X orat. 840b; Dem. 19.337; 18.15. It is not casual that, except the plays where the main character is a king, the third actor plays the role of the tyrant: ὁ Κρέων Αἰσχίνης (Dem. 19.247). 24 Schaefer (1885) II 336 dated the speech, with general consensus, to Archias’ archonship (346/365 BC); Harris (1985). 25 Aeschin. 1.12, on school rules; 16, perì hybreos (on violence against children); 21. It is noticeable that, even though all these laws appear in Aeschines’ manuscript tradition, today they are generally acknowledged as spurious, as written in a later age, trivializing what Aeschines said in the chapters immediately preceding these fake documents (in these nomoi lack notions which Aeschines indeed speak about in his introductions); cf. Drerup (1898); Martin – De Budé (1952) XXIV, 24.

78  Antonio Iacoviello men who prostituted from participating to public life; this law even implied a death warrant26 – this charge, however, was not common in the Athenian judicial pratice of the 4th century BC,27 and there was a lack of legal agreement about this matter.28 Moreover, Aeschines chooses as procedure of the lawsuit a dokimasia rhetoron (cf. 1.28): a college of judges had to decide whether Timarchus met the requirements for speaking at the tribune. However, the specific conditions on the basis of which Timarchus had to be judged were determined by Aeschines: he opted for accusations relating to issues of morality, which would have certainly struck Athenian sensibility:29 a) maltreating parents, b) military evasion or cowardice, c) having been a prostitute (pornòs) or having lived off men as an 'escort' (hetairesis), and d) having devoured one's inherited estate. Through usual dynamics in Athenian judiciary reality, Aeschines learned that Timarchus would have defendend himself with the most famous example of homoerotic philia, the one between Achilles and Patroclus.30 Aeschines’ primary objective is to anticipate the opponent’s reasonings and overturn them,31 quoting Homer frequently32 and giving the text his own interpretation, idealiz-

 26 In a final crescendo, Timarchus is accused of demou katalysis (191). The result of the trial was in favour of Aeschines: Timarchus was condemned to atimia and left public life (cf. Dem. 19.284–85). 27 Blass (1898) III/2 192; Dover (1978) 29–30; Usher (1999) 280. 28 Lape (2006) 145. In this speech, the orator acts as filling the role of the lawgiver; cf. Lycurg. 9 (on the crimes which πρόγονοι, the ‛ancestors’, did not speak about). 29 Aeschines chooses four main points for his dokimasia, but it is likely that his list is incomplete; furthermore, the themes resemble those of archons dokimasia, when the officers had to give an account on how they treated their parents; cf. Arist. Ath. Pol. 55.3; Fisher (2001) 160. 30 §132: οὐκ ἀφέξεται δέ, ὥς φασιν, οὐδὲ τῶν Ὁμήρου ποιημάτων. Aeschines conjectures that Timarchus could have used the love between Armodios and Aristogeiton as a renowned exemplum of his sexual relationship; Ober – Strauss (1990) 150; Ford (1999) 154. 31 Aeschines (who spoke first in the trial) is forced to anticipate possible accusations that Timarchus could have invoked about the ἐρωτικὰ ποιήματα which Aeschines wrote when he was young (§135–6); Michelakis (2002) 50. It is a puzzling information, which testifies the poligraphy of the orator. This might explain the strange sentence in the Ἀπολλωνὶου Αἰσχὶνου ῥήτωρος ἐξήγησις (Vita 2 Dilts): Λέγεται δὲ ἐρωτικός γεγονέναι. 32 The orator could have quoted another poem, closer in time and more patriotic: Aeschylus’ Myrmidones, a drama which dealt with the same mythical heritage of the Trojan cycle; Scodel (2007) 136. However, this a modern lucubration: the lacking quote may be explained through a statistical reason (Aeschylus is never quoted by orators; Fisher (2001) 293: ‘[Aeschylus] was the least relevant or easily intelligible’). Furthermore, the two surviving fragments of this drama which describe the relationship between Achilles and Patroclus (135–6 Radt), show licentious tones which may have legitimized Timarchus’ carnality. Aeschylus makes Achilles use the second person referring to the lover (as does Aeschines, modifying Il. 18.333 in §148) and there

Poetic Quotation in 4th Century BC Attic Oratory  79

ing33 the relationship between the heroes to distinguish it from the sexual ones of Timarchus.34 Aeschines’ accusation has been studied thoroughly, as evidence of Athenian perception of homosexuality35. But it must not be bypassed that this graphè hetairseos was a pretext for deeper political circumstances. The Homeric section of the speech is introduced by an important programmatic chapter (§144), whose first part is a captatio benevolentiae to the judges and public, called pepaideumenoi,36 and in the second part the speaker praises the eponymia tes philias autòn, ‘the proper name of their friendship’: the orator is inviting the audience not to perceive, in the quotation shortly thereafter delivered, allusions to the sexual relationship between the two heroes.37 The most peculiar feature of this section of the speech is that, in addition to being one of the most impressive case of poetic quotations in an oratorical context, it is an authentic Homeric philology laboratory, since all the quotes by Aeschines diverge from Homeric post–alexandrian vulgata. As well as other prose–writers’ quotations of the 4th century BC38 and ptolemaic pre-aristarchean papyri,39 Aeschines’ Homer is a key evidence of the great fluidity of Homeric text before Alexandrian age. First, it is important to understand whether textual divergences of these quotations are due to memory errors, common in indirect tradition.40 In at least two cases, it may be argued that variants are deliberate; Aeschines significantly  is even a mention of kisses: ὦ δυσχάριστε τῶν πυκνῶν φιλημάτων (TrGF 136 Radt); Michelakis (2002) 50–52. 33 For example, §145, which is a prose paraphrasis of the preceding six Iliadic verses; Fisher (2001) 290. He would have done the same thing with a decree. 34 Cf. §130, at the end of the section about the inexorability of pheme: ‘ποῖος Τίμαρχος; ὁ πόρνος’ (Aeschines widely adopted contemptuous nicknames; cf. 2.99 where Demosthenes is called Βάταλος). But Aeschines misreads the text, as the homosexual relationship between the heroes is, in Homer, just hinted and never proclaimed; thus, the litigants are likely to report the tradition (Hymn Aphrod. 202–6; Ibicus fr. 289 Page) which included the love between Achilles and Patroclus among the explicit sexual relationship between a god and a human; cf. Fisher (2001) 288. 35 Aeschines, deploring Timarchus’ lack of sophrosyne, aretè, eukosmia and enkrateia is here applying the so–called ‘public morality’ (Dem. 10.70). The inference shows a remarkable agreement with the contemporary ‘oligarchic’ theories of Plato (Timarchus is comparable to the ‘tyrant’ in Resp. 9; and Aeschines’ theory on real eros coinciding with aretè is the same reasoning of Symp. 179e); cf. Sissa (1999) 153; Fisher (2001) 350–51; Lape (2006). 36 Perlmann (1964) 156. 37 Ford (1999) 252; Sanz Morales (2001) 52. 38 Labarbe (1949). 39 West (1967); Haslam (2006); Bird (2010). 40 Van der Valk (1964) 272; Sanz Morales (2001) 55.

80  Antonio Iacoviello modified the text for his own cause,41 by conferring upon some verses an emotional and pathetic coloring, so that the relationship between Achilles and Patroclus seemed purer: in 1.148 (= 18.133) the vocative Πάτροκλε is replaced with a more affectionate ἑταῖρε,42 and in 1.150 Aeschines reworks the entire second hemistich43 of 18.99: Aeschines

Homer’s vulgata

.

ἀλλ'ἐπεὶ οὖν, φίλ'ἑταῖρε, σεῦ ὓστερος εἶμ' ὑπὸ γαῖαν

νῦν δ'ἐπεὶ οὖν, Πάτροκλε, σεῦ ὓστερος εἶμ' ὑπὸ γαῖαν

.

κτεινομένῳ ἐπαμῦναι, ὃ μοι πολὺ φίλτατος ἔσκεν.

κτεινομένῳ ἐπαμῦναι, ὁ μὲν μάλα τηλόθι πάτρης.

Furthemore this last verse is, in the Homeric poem, closely syntactically connected with the following, 18.100, which contains its predicate, ἔφθιτ(ο). This connection persuaded some interpreters that Aeschines was not aware of v. 100, or that, as he could not quote one without the other, he knew a lost verse 99a which softened this grammatical connection.44 The same issue, even more insidious, emerges in the most compelling (and lenghtiest) quotation from Aeschines: the delivery, by the clerk, of 23.77–91, by the speech of Patroclus’ ghost to Achilles on their common burial. The quotation shows a typical problem of Homeric tradition, namely the excess verses: there are three of them, and none left traces in the further tradition.45

 41 Nonetheless, the orator employs typical iuncturae of epic poetry. Aeschines was certainly aware of this production due to his education; however, all the pre-vulgata is composed of conventional sentences, pseudo-Homeric, of the style of Omerikà; cf. Olding (2007) 159. 42 Van der Valk (1964) 328; Fisher (2001) 292; Olding (2007) 157. Sanz Morales (2001) 57 argues that this modification is voluntary and worthless, noticing how the word had no particular affective meaning in 4th century. It does not appear a defensible opinion, because the variant modifies the text too much, in line with Aeschines’ reasoning, marking the difference between ἑταιρεία and ἑταίρησις; cf. Fisher (2001) 251. 43 In his edition of the Iliad, Allen (1931) refers to another Aeschinean variant, from ms. Par. gr. 3002 (l), φίλτατον, which changes the sense of the sentence, as it switches subject with object. Nevertheless, the lesson is ignored: neither Dilts (1997) nor Martin – De Budé (1953) insert it in their apparatus. The reason is maybe that the manuscript is not among the ‘potissimi ad recognoscendam Timarcheam’; Dilts (1997) IX. 44 Sanz Morales (2001) 57. 45 Haslam (1997) 75.

Poetic Quotation in 4th Century BC Attic Oratory  81

Aeschines

Homer’s Vulgata

οὐ γὰρ ἔτι ζωοί γε φίλων ἀπάνευ θεν ἑταίρων

οὐ μὲν γάρ ζωοί γε φίλων ἀπάνευ θεν ἑταίρων

…

…

a

μαρνάμενον δηίοις Ἑλένης ἕνεκ' ἡυκόμοιο

omittit



ἄλλο δέ τοι ἐρέω, σὺ δ'ἐνὶ φρεσὶ βάλλεο σῇσιν

ἄλλο δέ τοι ἐρέω καὶ ἐφήσομαι, αἴ κε πίθηαι



μὴ ἐμὰ σῶν ἀπάνευθε τιθήμεναι ὀστέ’, Ἀχιλλεῦ

μὴ ἐμὰ σῶν ἀπάνευθε τιθήμεναι ὀστέ’, Ἀχιλλεῦ

a

ἁλλ'ἵνα πέρ σε καὶ αὐτὸν ὀμοίη γαῖα κεκεύθῃ

omittit

b

χρυσέῳ ἐν ἀμφιφορεῖ τόν τοι πόρε πότνια μήτηρ

omittit



ὡς ὀμοῦ ἐτράφεμέν περ ἐν ὐμετέροισι δόμοισιν

ἀλλ'ὁμοῦ ὡς ἐτράφην περ ἐν ὐμετέροισι δόμοισιν

…

…



ὣς δὲ καὶ ὀστέα νῶιν ὁμή σορὸς ἀμφικαλύπτοι

ὣς δὲ καὶ ὀστέα νῶιν ὁμή σορὸς ἀμφικαλύπτοι



omittit

χρύσεος ἀμφιφορεὺς τόν τοι πόρε πότνια μήτηρ



A first variant occurs in v. 77; its reason is doubtful: memory mistake, variant of Aeschines’ Homeric edition, voluntary modification. The last case would not be surprising, as the two previous cases prove, but also because the insertion of ἕτι emphasizes the duration of Achilles and Patroclus relationship.46 This interpretation must be rejected,47 since the verse, as Aeschines quotes it, was more like 46 Olding (2007) 77. 47 Van der Valk (1964) 327 n. 330; Sanz Morales (2001) 58.

82  Antonio Iacoviello ly modifyed by the attraction to the previous v. 75: καί μοι δὸς τὴν χεῖρ’, ὀλοφύρομαι· οὐ γὰρ ἕτ' αὖτις.48 Furthermore, additional information appears in the scholia,49 where it is explicitly stated that this variant was attested in some politokà, namely Homer’s editions katà polin, ‛of the cities’. Hence, Aeschines is referring to an official Homeric version of some city;50 it is unclear to which one he refers, but this is plainly a noticeable clue of the textual history of the quotation. The surplus verses then follow, which, as said, are not uncommon in pre– Alexandrian indirect quotations.51 81a offers the topos of Helen’s guilt,52 vv. 83ab, which illustrate the urn where Achilles’ ashes will be placed, are more puzzling. The quote ends at v. 91: we may wonder whether Aeschines knew v. 92, systematically present in the vulgata, where the urn is mentioned (ἀμφιφορεύς). However, it was mentioned in the previous verse as well, at 91 (σορός). 18.92 is extremely similar to 83b, so the two unities 83ab and 92 cannot coexist.53 Ancient scholarship had already perceived the problem: scholia inform that Aristarchus athetized v. 92, as he thought it was a transposition from Od. 24.74.54 In this intricate piece of Homeric philology, the solution is not to invoke Aeschines’ fallacious memory.55 The main point is to understand whether he was aware of v. 92:56 Pasquali was convinced that Aeschines’ edition did  48 Although the Aeschinean variant has to be meant as pre–Alexandrian, it is interesting that all modern translators render the verse as if ἕτι is present, because it appears almost necessary for the context; Sanz Morales (2001) 58. 49 Erbse (1977) V.380. 50 As for other Iliadic scholia vetera, we only read a reducted version, epithomized, of the information and data of the great Iliadic scholiastic corpus in the ancient world; in the scholium from which this one derives, the exact polis to which the scholium refers was probably noted; Haslam (1997) 76. Another case of indirect tradition of 4th century BC which coincides with a katà polin edition is, according to scholia, Pl. Ion. 538d, where, quoting Il. 24.82, there is πῆμα instead of κῆρα; Labarbe (1949) 120. 51 To be limited to two significant examples, surplus verses are found in Arist. Eth. Nic. 3.8, 10 and Xen. Symp. 8.36 too. 52 An element which makes the verse ‘no torpe’: Sanz Morales (2001) 58. Patroclus’ pathetic tone justifies this insertion. 53 The different words here used by Homer to denote the funerary urn effectively indicate, according to modern lexicography, the same object: LSJ s.vv. λάρναξ ‘cinerary urn’; σορός ‘vessel for holding human remains; cinerary urn’; ἀμφιφορεύς ‘large jar of gold (or stone); used as cinerary urn in Iliad’ (the last word can also mean ‛food case’). 54 Erbse (1977) V 383. 55 Van der Valk (1964) 339. 56 An Heidelberg papyrus (P. Heid. Lit. 2 + P. Hib. 1.22 + P. Grenf. 2.4), signed Þ12 by Haslam (1997) and Bird (2010), of third century BC (not much later than Aeschines), although not easily

Poetic Quotation in 4th Century BC Attic Oratory  83

not contain it.57 Moreover, Aristarchus expunged it too, but not because it conflicted with 83ab, but instead with the previous 91: 23.92 was preserved in the tradition only because it had been athetized by an authority.58 Another hypothesis may be to consider verses 83 (including ‛a’ and ‛b’) – 92 as cohesive and independent unity: in this case, Homer’s passage shows a ringcomposition that justifies the redundancy of this tormented ‘golden amphora’.59 In this sequence of quotations, where pseudo Homeric texts appear as well,60 it is useful, in order to clarify the textual history, to analyze the ways in which these texts entered in the trial’s circumstances. It seems that Homeric quotations are based on the reading of an Iliad exemplar, and not on memory. The only certain reading case is §144 where Aeschines himself pronounces the first quotation.61 For the other three, he instead invites the clerk to read with the usual imperative verbs (λέγε; ἀνάγνωθι): this implies that poetic texts are taken from external sources, and not from the speech draft.

 readable, shows the presence of 23.92. The presence of 83ab is not verifiable due to a lacuna in correspondence to vv. 1–85. 57 Pasquali (1934) 221–2; followed by West (1967) 92. 58 It is not surprising that a verse athetized by Aristarchus entered current Iliad; on his 800 ca. attested interventions on Homer, c. eighty entered the tradition. 59 Di Luzio (1969); Dué (2001). 60 A desperate case is §128, where Aeschines declares that the will quote Homer, but instead delivers a dactilic emistich not present in our Iliad. It could belong to the pre-Ptolemaic surplus verses; cf. Martin – De Budé (1952) 63. However, the word φήμη never recurs in Iliad, but just in Odyssey (where is a synonym of φῆμις) and with a different meaning of the one meant by Aeschines, ‘rumor’ being a metaphor for ‘reputation’. This meaning in the Iliad belongs to the word ὄσσα, 2.93: is Aeschines referring to this last verse, modifying it? Cf. Van der Valk (1964) 331–32; Sanz Morales (2001) 84. But this word is an ἅπαξ in Homer, while Aeschines says: τὸν Ὅμηρον πολλάκις […] λέγοντα, ‛Homer […] often says’. The proposal of Marzullo (1953), to integrate ἐν τῇ Ἰλιάδι < μικρᾷ > (Dilts [1997] reports in his apparatus ἐν τῇ < μικρᾷ > Ἰλιάδι, creating an indifensible hiatus) is more understandable. This integration would assign the verse to the Κυκλικά, the cyclic poems, which Aeschines would have exchanged for real Homeric poems; Bultrighini (2014). Fisher (2001) wonders why Aeschines, speaking about φήμη, did not quote Ag. 938; but he is the same scholar who defined Aeschylus an author not beneficial in oratory argumentations. 61 Ἔστι δὲ τὰ ἕπη, ἃ ἐγὼ νυνὶ μέλλω λέγειν. Aeschines was known for his beautiful voice; cf. X orat. 840b where he is called εὔφωνος. In his youth he was not only tragic actor (Dem. 19.247), but even ὑπογραμματεύς; cf. X orat. 840b; Dem. 19.249. He was so quite skilled in the declamation of this kind of texts.

84  Antonio Iacoviello Thus, whithin Timarchus’ trial, the clerk did likely read a papyrus roll62 containing a current Iliad edition, perhaps marked with paragraphoi which showed to the clerk the text portions to deliver.63 Aeschines seems to have a distinct use of the clerk, based on a masterly turnover64 which created a second voice that caught the ears of the judges, as the speech gradually went on. As usual, the clerk’s interventions are concentrated in the first part of the speech, where there is a major necessity to submit the circumstantial and documentary evidence; this happens in the first part of the Timarchea as well, but after the initial submission of documents, the poetic section follows, strategically placed about in the three–quarters of the speech. The role of poetic quotations in Aeschines may, in a last analysis, help in the comprehension of the ancient history of this text. Aeschines’ manuscript tradition raises many issues:65 nonetheless, the medieval transmission is not the proper field where answers have to be found, because the extra–textual transpositions, as quotations, took place in a more ancient age, so the mobility of these elements is hardly perceivable. Some of the poetic quotations analyzed here were certainly delivered during the trial: 1.128, because Demosthenes will resume Aeschines’ argumentations about pheme in 19.243 with contemptuous irony; 1.144, as the orator explicitly declares he is delivering the quote (unless we should assume textual modifications out of the poetic section – but it would be unsuitable). An intriguing point is that reported by the scholium to Il. 23.77, which states that Aeschines is quot-

 62 Aeschines quotes Iliadic sections from Books 18 to 23; according to the standard measure of the papyrus roll in 4th century Athens– a length surely inferior to the Ptolemaic roll – presumably, more than one roll was employed and read. 63 Cf. the information in Isoc. 15.59, where he invites the clerk to read a section of the Panegiricus ἀρξὰμενος ἀπό τῆς παραγραφῆς, ‛starting from the paragraphoi’. Even though the Antidosis is a fictional speech, it reproduces real circumstances and practices of Athenian trials; Sanz Morales (2001) 52; Pinto (2003); Haslam (2007) 77. 64 Olding (2007) 160–4. 65 Aeschines’ tradition is based on 54 medieval manuscripts which contain the entire corpus, only or. 3 or only the Letters; Roncali (1969); Leone (1972); then, 48 papyri, which are characterized by textual eclecticism (they show errors, sometimes in common with one of the two families of medieval tradition, βD o f, sometimes are instead lectiones singulares); Monaco (2000). For example, an important papyrus for the Timarchea, P. Fuad. inv. 222, containing chapp. 145–7, seems oriented to family βD, with which shares four variants, against the two in common with f. It is a 5th century AD papyrus, fragment of a parchment codex, written in Biblical Uncial: it is evidence that Aeschines circulated on luxury copies.

Poetic Quotation in 4th Century BC Attic Oratory  85

ing a katà polin Homer.66 This city may have been Athens, or another place where Aeschines stayed enough for arranging an edition of his speeches: for instance, Rhodes. Indeed, many biographical sources tell about a rhetorical and teaching activity on the island, where he read his own speeches.67 Unsurprisingly, in the case of oratorical corpora, the first settings of conservation and textual safeguarding were indeed the political entourage, or schools.68 The quotations, except those that were voluntarily modified, could have been based on a pre– Alexandrian Homeric exemplar of this kind. Nevertheless, in this case, maybe‘the quotes seem to predate the text’s stichometry as well, which, again, the laws and decrees do not’.69 3. Lycurgus’ speech Against Leocrates, delivered in 331 BC, shows several peculiarities, which make it appear to the modern critics as a clumsy and awkward discourse.70 The ambiguity of this speech is due to the strange nature of the trial when it was pronounced. Formally, it was an eisaggelia,71 a changeable form of impeachment against a public personality. We know a considerable number of historical and judicial details about eisaggelia;72 the difficult point is to understand the circumstances when an eisaggelia could have been charged, thus the application criteria of the nomos eisaggeltikòs.73 Lycurgus’ speech raises this issue: in his reference to law (§9) Leocrates’ offences are not specifically included among the crimes which required an eisaggelia, thus the orator asks the court to adapt the procedure to the current trial.74 Scholars’ critical uncertainy is due to the fact that Leocratea, as Aeschines’ Timarchea, belongs to the group of speeches which sustained an accusation without irrefutable arguments75 – an element which appears connected with the use of poetic quotes within a trial. Leocrates – a private Athenian citizen – is  66 The scholium could be a precise reference to the Timarchea; Olding (2007) 169: ‘it is hard to imagine that this does not refer to Against Timarchos’. 67 Αἰσχὶνου ῥὴτορος βίος, 6 (Vita 1 Dilts); Περὶ Αἰσχίνου, 3–4 (Vita 3 Dilts); P. Oxy. 1800 (fr. 3 col. II), Vita Aeschinis; [Plut.] X orat. 840d–e; Phot. Biblioth. 61. 68 Canfora (1995) 164–6. 69 Olding (2007) 169 70 Allen (2000) 11. 71 Hansen (1975) 108; cf. Lycurg. 1, 5, 29, 30, 34, 55, 137. 72 One of the main sources is Harpocration’s lemma ε 7 Keaney, which divides these practices in three types, according to the proper court. 73 In this case, ancient sources are much less clear; just three ancient texts speak about nomos eisaggeltikòs: one from Hyperides’ Against Euxenippus; one from Theophrastus’ Perì nomon; and the ‘law of Timocrates’, Dem. 24.63). 74 It can be regarded as a case of ‘open texture of Athenian law’; cf. Harris (2014) 237. 75 Malcovati (1977) 816; Taddei (2012) 62–65.

86  Antonio Iacoviello accused by Lycurgus of lipotaxion,76 astrateia, asebeia e katalysis tou demou77 for having escaped Athens immediately after the report of Chaeronea defeat, and for having lived as a retailer in Rhodes, then Megaras, wasting his goods in Athens and then spreading the word of Piraeus’ siege. Leocrates then came back to Athens after years, convinced he could continue to live undisturbed. But Lycurgus’ ‘irremovability’78 did not spare him. The orator had to stretch the value of Leocrates’ treason in order to justify his remarkable charge.79 Within his argumentation, Lycurgus employed extra–textual material on great scale: mainly documents,80 and, even in this case, at the ¾ of the speech, an articulated poetic section takes over (§92–109; and a handful of isolated iambic trimeters in §132). It is essential to observe here that all of the Leocratea’s documents disappeared from the textual tradition; merely titula remained, and in a single part of the tradition only;81 the single entire quote is the oath pronounced by Greeks before Plataias’ battle (§81); nevertheless, this text appears in Diodorus82 and on an inscription of the middle 4th century BC as well.83 The comparison between these texts shows how Lycurgus changed the language, by ‛atticizing’ some words.84 These modifications are not likely to be voluntary: the orator quoted a version of the oath which has to be related to the Athenian tradition of that time. The first oath, the ‛ephebic oath’, is interesting as well, for it is absent in the textual tradition: it is inserted in modern editions using the text

 76 Lycurg. 77.147; it is the accusation against the one who escaped during a battle, while other soldiers were fighting. 77 Lycurg. 147. The charges get more serious as the speech goes on. 78 The definition of Petrie (1922) XXXIX. 79 Nonetheless, Lycurgus’ interpretation of εἰσαγγελία did not convince the jury, as Leocrates was acquitted the votes being equal. For Lycurgus’ unusual attitude towards the procedure, see Harris (2014) 234–41. 80 See n. 2. 81 The Medieval manuscript of Lycurgus, as for the others minor Attic orators (in particular Antiphon and Dinarchus), is based on two manuscripts, ms. Burney 95, the Crippsianus (A) which contains the Leocratea entirely, and ms. Bodl. Auct. T.II.08 (N) which contains chap. 1– 34 and 98–147, both of the early 14th century. Through an analysis of these manuscripts, it may be asserted the major trend of N in quoting titula: 5 psephismata on 7 have not titulum in A (36; 114; 122 – placed in N two lines before the start of the decree; 125; 146), as well as none of the two horkoi (77; 81) neither the ῥῆσις Εὐριπίδου (100). 82 11.9.2–3; his source was Ephoros; Olding (2007) 164–5. 83 Tod II, 306–7, n. 204. 84 He replaces ταχιολόχοι e ἐνωμοτάρχιαι in ll. 25–26 with a much more generic ἠγεμόνες; Allen (2000) 24; Olding (2007) 164.

Poetic Quotation in 4th Century BC Attic Oratory  87

of Stobaeus 4.1.48, Pollux 8.55–57 and an inscription in Acharnes discovered only in 1932.85 The poetic quotations are not less problematic; their eccentricity stays not in their textual issues, but in their role in the trial structure. The most significant quotes are: Euripides’ Erechteus (fr. 360 Kannicht), Il. 15.494–9 and Tirtaeus fr. 4 W. According to Hermogenes (Perì ideon 2 pp. 402–3 Rabe), Lycurgus was particularly acquainted with the use of poetry in his speeches: the rhetor probably referred to those which did not survive. Lycurgus’ Homer shows the same problems as those of the Aeschinean speech: there are three (or four) variants from the vulgata86 that need to be ascribed not to the orator’s memory,87 but to the already examined fluidity of the pre-Alexandrian Iliadic poem. However, the most puzzling quotation is the first: the Euripidean fragment immediatly stands out for its extraordinary length. There are 55 verses, the lengthiest fragment of this tragedy until the discovery of P. Sorb. 2328 (fr. 370 Kannicht), which contains the exodus of the drama.88 It is a rhesis where the female protagonist, Praxithea, wife of the Athenian king Erechteus, delivers an embellished and patriotic speech stating the necessity to sacrifice her daughter in order to avoid the conquest by the Thracian king Eumolpus, as suggested by the Delphic oracle.89 In this monologue, there are many themes, which interweave themselves both with lycurgean argumentations, and with the topoi of logoi epitaphioi, which had a great influence on the orator.90 Moreover, the choice of the drama, which is among the most patriotic of the tragic production91, is remarkable as well; nonetheless, an even more significant issue, totally neglected by scholars, is that Lycurgus, as belonging to the ancient and prestigious genos of Eteoboutadai,92 had control, with his family, over the priestly cults of the sanctuaries of Athena Polias and Poseidon Erechteus93 on the Acropolis.  85 Robert (1938) 302–3. 86 Sanz Morales (1991). 87 Durrbach (1932) 67. 88 Jouan–Van Looy (2000) 104. 89 Jouan–Van Looy (2000) 105–7; Sonnino (2010) 181–3. The myth chosen by Euripides is presented with a philoathenian version, but it existed a philobeothic variant as well, according to which the Beothics removed Thracians from Greece; cf. Di Benedetto (1971) 145. 90 Scodel (2007) 140. 91 Di Benedetto (1971) 146. 92 The familiar influence on the personality and political activity of Lycurgus is widely recognized; Durrbach (1889); id. (1932); Malcovati (1977) 798; Taddei (2012) 24–28. 93 Davies, APF 348; this is certain, despite Davies’ question about Eteoboutadai’s richness at the time of Lycurgus; his grandfather scarcely reached the census for being among pentakosiomedimnoi, and the silence of the Stratokles’ decree (a biographical document about

88  Antonio Iacoviello This biographic element may even shed light on the textual history of this excerpt. This quote has in fact many noteworthy features – first its length, not comparable to any other quotation in Attic speeches – which lead into several conclusions: a) The delivering of this quote in its full length is not admissible. Even if Lycurgus had paraphrased the content of the quotation in §98–99, it is likely that it was delivered, during the trial, in the form of excerptum; b) The delivering itself of the quotation has to be put into question. The presence of a paraphrasis of just two chapters does not justify the waste of time necessary for reciting such a long text. Furthermore, there is no invitation to the clerk to read; the quotation is introduced by a generic ἄξιον τῶν ἱαμβείων ἀκοῦσαι; c) The lack of titulum in ms A can be a proof that, in a certain period of the tradition, the quotation was absent – or that it was delivered by the orator himself. A lemma, indeed, is a clue of the clerk’s delivering of the external text. Nevertheless, the statement which introduces the quote (‘it is now the proper time to hear the iambus’) is unclear. d) The biographic information about Lycurgus’ relationship with the sanctuary of Poseidon Erechteus allows us to go further. It is obvious that, as he had control over the rites, the Eteobutadai took care of the traditions of Erechteus, the mythical Athenian king. We may argue that, in the second half of 4th century BC, a certain amount of texts about him circulated through aristocratic contexts, such as symposiums; the transmission of tragic rheseis in symposial catenae is, for instance, an extraordinary method of oral transmission of dramatic texts. Once this connection is assumed, it is easy to think that, within such an editorial context – in this case, evidently Athenian – hinging on Lycurgus’ figure, a fragment like that by Euripides could enter appear in speech with no particular trouble; e) Stilemas and verses depending on an Euripidean tradition were transmitted in different ways, not just through the indirect literary tradition; it is the case in CEG 594, a funerary marble inscription, in elegiac distich, dating to 340– 325 BC, the text of which, especially in the second of the two distichs, is clearly indebted to vv. 45–52 of fr. 360 Kannicht:94 in both the passages there is a reference to a στέφανος and to the necessity of exceed the father’s glory with the  Lycurgus, issued in honour of him by Stratokles after his death; IG II 2 457; [Plut] X orat. 852b) about liturgies the family could boast of. For the role of Eteobutadai in the administration of the sanctuary, see Blok – Lambert (2009) 109–11. 94 Tzagalis (2007).

Poetic Quotation in 4th Century BC Attic Oratory  89

help of virtue. The addressee of this inscription is not known; however, it is a noteworthy witness for the Euripidean text transmission out of literary grounds, and which points out how it could enter in disparate contexts. 4. In light of the examined examples, the textual–history mechanic which occurs may be regarded as very similar that of the documentary materials. The omission of documents in the oratory tradition is a fact; a reader of Attic oratory is not surprised in scanning the introduction of a decree, and then finding just its titulum. A similar fate is conceivable, at least for a phase of the history of the text, for poetic quotations as well. The most ambiguous point is that of their reintroduction: it was certainly simpler, in the contexts of the transmission – and modification – of the texts, than the reintroduction of documents: the latter were strictly connected to the Athenian judicial practice and to their specific conservations contexts, whereas texts such as Euripides or Homer were widely spread in written form since the end of the 4th century. This issue has to be explained by trying to clarify the ancient conservation settings, which constitute the basis of an ancient textual history.

Giulia Marolla

Jerome’s Two Libraries Sed et in duodecim prophetas uiginti quinque ᾽εξηγήσεων Origenis uolumina, manu eius exarata repperi, quae tanto amplector et seruo gaudio, ut Croesi opes habere me credam. (Jer. De vir. ill. 75)1

Jerome’s fondness for books in his possession is immediately perceptible from the passage quoted above; the twenty–five volumina he refers to belonged to Origen’s commentary on the Twelve Prophets, and had been copied by the presbyter Pamphilus, a friend of Eusebius of Caesarea. As Jerome points out in the same passage, Pamphilus was the copyist of most of the works of Origen preserved in the Library of Caesarea, and this was a considerable matter of pride, so that Jerome ‘thinks he has Croesus’ wealth’: the self–portrait of a proud bibliophile, indeed. A hint to the peculiar affection he has for his librarian possessions is also retrievable in Ep. 22.30, when, before describing the famous dream in which the Judge accused him of being a Ciceronianus instead of a Christianus, Jerome says: bybliotheca, quam mihi Romae summo studio ac labore confeceram, carere non poteram.2 He thus informs the reader that he has put a lot of effort in creating his personal library, a truthful assertion probably clouded by the overwhelming fame of the subsequent words of the same passage. It is common knowledge that, in the Holy Land, Jerome’s exegetical and philological activity on Scriptures required him to study and rigorously collate codices in Hebrew and Greek. As thoroughly explained by Brown (1992, 34) Jerome’s textual criticism was based on Origen’s use of critical sigla, and his Hebrew scrolls were items of his library particularly dear to him, as testified by the following passage. Accedit ad hanc dictandi difficultatem, quod caligantibus oculis senectute et aliquid sustinentibus beati Isaac, ad nocturnum lumen nequaquam ualemus Hebraeorum uolumina relegere, quae etiam ad solis dieique fulgorem litterarum nobis paruitate caecantur. Sed et Graecorum commentarios fratrum tantum uoce cognoscimus; nullique dubium quod alienis dentibus commoliti cibi uescentibus nauseam faciant.3

For Jerome, who despairs at his failing eyesight, it is impossible to read the Hebrew scrolls at night and even during the daytime, because of the smallness of  1 Ed. Ceresa–Gastaldo (1988) 182. 2 CSEL 54.189. 3 In Ezech. 7, Praef. (CCSL 75.277–8). https://doi.org/10.1515/9788110632590-007

92  Giulia Marolla the letters written; he also laments the impossibility of personally reading the Greek commentaries to the Scriptures, read out to him by his brothers, and explains the nauseating feeling he experiences, with the powerful and realistic metaphor of food, as if his ‘literary food’ had been chewed by someone else. Meghan Hale Williams ascribes the books in Jerome’s library in Bethlehem to three broad categories: Biblical texts, oeuvres of Christian authors and Hellenistic Jewish works,4 and lastly pagan literature.5 As to the second category, it is likely that Origen had a prominent place is Jerome’s library; according to Courcelle,6 the account of more than 800 items by Origen, deducible from letter 33 to Paula,7 is actually a list of books in Jerome’s possession, though Williams discards this assertion.8 There are passages that clearly testify that Jerome asked for copies of books which he did not own yet, as it was common practice at the time. For instance, in Ep. 10.3 there is a request to his correspondent Paul, to procure him some texts which presumably Paul already has at his disposal: the commentaries of Fortunatian, the History of Aurelius Victor, and the Letters of Novatian. Jerome explains his interest in the latter, adding that he wishes to read schismatic ideas only to ‘drink’ thereafter the antidote to this poison, that is Cyprian’s writings. Once again, as in the abovementioned passage, Jerome draws on a lexicon that pertains to physicality, and evokes realistic, and extremely effective, images. Interestingly, Jerome’s request is accompanied by a gift, he sends to the old Paul what he calls an ‘even older Paul’, and promises, should he appreciate the present, to send more texts that he has translated; they will venture across the sea to reach him, along with all sorts of eastern merchandise.9 The books whose presence in the library is more difficult to demonstrate are undoubtedly pagan oeuvres, since Jerome did not openly publicize having acquired codices of classical authors for rather obvious reasons. Therefore, even though it is possible to detect traces of the manuscripts of pagan authors that

 4 Especially Josephus, whom he considers a Greek Livy (Ep. 22.35) and a guide for his historical commentaries on Scriptures, see Courcelle (1969) 83. 5 Williams (2006) 147. 6 Courcelle (1969) 103–11. 7 Williams (2006) 154 ‘listed in that letter are 285 books of commentaries on the Bible, 468 homilies, excerpta on seven biblical books, and other materials totaling 38 books, including the ten books of the Miscellanies (in Greek, Stromateis) and four of On First Principles, for a total of just under 800 items of widely varying length’. 8 (ibid.) 155. 9 Si hoc munusculum placuerit, habemus etiam alia condita, quae cum plurimis orientalibus mercibus ad te, si spiritus sanctus adflauerit, nauigabunt (CSEL 54.38); See also Brown (1992) 22–23.

Jerome’s Two Libraries  93

belonged to Jerome, the evidence that can be gathered is destined to be fragmentary. Mostly it is possible to assume what books Jerome was interested in acquiring, but not to assert with certainty that those manuscripts were actually obtained and had become part of his personal library: as in Ep. 10, there is no way of knowing if Paul sent Jerome the copies he was interested in. Another testimony is provided by Rufinus, according to whom, Jerome was interested in having classical manuscripts copied for him from collections of other monasteries. Alioquin, si inficias eat, etiam testes quamplurimos fratrum habere possum, qui in meis cellulis manentes, in monte Oliueti, quamplurimos ei Ciceronis dialogos descripserunt, quorum ego et quaterniones, cum scriberent, frequenter in manibus tenui et relegi, et quod mercedes multo largiores, quam pro aliis scripturis solent, ab isto eis darentur agnoui. Mihi quoque ipsi aliquando, cum de Bethleem Ierosolymam venisset et codicem secum detulisset, in quo erat unus dialogus Ciceronis et idem ipse Graecus Platonis, quod dederit ipsum codicem et aliquandiu fuerit apud me, nullo genere negare potest.10

In this passage, Rufinus provides with a series of interesting details. First, he reports that Jerome used to pay the monks of Rufinus’ monastery on the Mount of the Olives to make numerous copies of almost all of Cicero’s dialogues.11 The accuser asserts that not only had he held in his hands the quaterniones – the quires themselves which constituted the codex – so that he could re–read the work copied for Jerome, but he also adds that his monks could testify that Jerome’s wages for the copies were even more conspicuous than usual. It would not seem so unlikely for Jerome to buy his own copy of Cicero’s oeuvres, paying more money to ensure himself an accurate copy; despite this, it is also true that this information is embedded in a long exchange of unpleasant accusations, to support Rufinus’ assertion that Jerome had perjured himself and violated his oath never to touch a non–Christian manuscript.12 Furthermore, Rufinus tells the reader that once Jerome had gone from Bethlehem to Jerusalem, bringing with him a manuscript containing a Ciceronian dialogue and the same work by Plato in Greek, presumably Cicero’s translation of Protagoras,13 a book which he had lent Rufinus for some time. Amongst similar allegations, there is the renowned accusation in  10 Rufin. Apol. Adu. Hier. 2.11 (CCSL 20.91–92). 11 Cf. Williams (2006) 162. 12 Ep. 22.30, domine, si umquam habuero codices saeculares, si legero, te negaui (CSEL 54.191). For diverging interpretations of the episode of the well-known dream of the Judge and Jerome’s promise never to re-read the classics see e.g. Hagendahl (1958) 91; 328 and also Adkin (1999) 162. 13 As for the identification of the Ciceronian and Platonic dialogue mentioned in the passage, see e.g. Simonetti (1957) 251 n.16; Grilli (2004) 389.

94  Giulia Marolla Apol. Adu. Hier. 2.7, a passage in which Rufinus accuses his long–lost friend of intertwining his oeuvres with more and longer quotations from classical authors than from Prophets and Apostles, even when writing to youngsters. Sed et in omnibus fere opusculis suis multo plura et prolixiora testimonia de his suis quam de prophetis nostris uel apostolis ponit. Puellis quoque uel mulierculis scribens, quae non utique nisi de nostris Scripturis aedificari et cupiunt et debent, exempla eis Flacci sui et Tullii uel Maronis intexit.14

It is peculiar that Rufinus stresses the difference between what he pretends to perceive as auctores nostri, and the auctores sui, drawing a line most firmly between the Christian tradition as a common patrimony and the classical tradition as distinctly Hieronymian. Actually, as Gamberale forcefully pointed out15, Rufinus’ translations are overflowing with classical reminiscences as well, and pagan classical authors were fundamental for the education of any cultivated person. Rufinus was not different from Jerome, but he, unlike Jerome, had not triumphantly promised to stop reading classical authors in one of his most famous pieces of writing. One might consider using a study of Jerome’s classical Latin quotations to infer the presence of the relevant authors in his library. However, this hypothesis inevitably collides with the fact that to learn by heart Latin classical texts was part of any youngster’s upbringing, especially Jerome’s. As he himself often recalls, Aelius Donatus was his praeceptor, and he had had the occasion of studying the best commentaries of the time.16 Moreover, in defending himself from Rufinus’ accusations (Adu. Rufin. 1.30), Jerome describes how, even though he rejected the idea of reading codices saeculares, it was impossible for him, as it would be for anyone, to forget the texts he had learnt during his years as a student of rhetoric. Allusions to Latin classics are therefore also ascribable to a ‘mental library’, one which probably Jerome refers to more often than it is necessary, to justify his quotations from detractors; but still, an enormous mental archive. It is peculiar, besides, that Jerome, while praising Nepotian for his thorough studies of the Scriptures and of Latin Fathers, says he had made his mind a ‘library of Christ’:

 14 CCSL 20.89. 15 Gamberale (2013) 166. 16 Adu. Rufin. 1.16, Puto quod puer legeris Aspri in Vergilium ac Sallustium commentarios, Vulcatii in orationes Ciceronis, Victorini in dialogos eius, et in Terentii comoedias praeceptoris mei Donati, aeque in Vergilium, et aliorum in alios, Plautum uidelicet, Lucretium, Flaccum, Persium atque Lucanum (CCSL 79). See Cavallera (1922) 7; Williams (2006) 16; Brugnoli (1965).

Jerome’s Two Libraries  95

lectione quoque adsidua et meditatione diuturna pectus suum bibliothecam fecerat Christi.17 It is almost impossible to assert with certainty whether Jerome’s quotations were taken from codices of classical authors at his disposal in Bethlehem, or were simple reminiscences of his juvenile studies18. The mental and physical library coexisted and interacted variously, and the famous vow in Ep. 22, which is filled with rhetoric and fiction, cannot be the main argument to assert there were no books of pagan authors, especially Cicero’s, in Jerome’s library at Bethlehem. Jerome’s purpose is to make use of the literary tradition of the past to enrich and strengthen Christian writings with an excellent style. An effective synthesis of Jerome’s thought is provided by Christa Gray, who states that Jerome’s ambivalent attitude towards the classics is part of a refusal of temptations, since the superior style of those texts has the same effect that sex or food would have, endangering Jerome’s ascetic resolution. At the same time, his need to obtain prestige and authority requires him to employ rhetorical strategies and a persuasive style; therefore, pagan texts are used as a mean to a higher purpose19. I also agree with Sabine MacCormack’s view20 on the relationship between Jerome and Ciceronian texts. MacCormack pointed out that Jerome’s attitude towards Cicero was very different from that of Arnobius, Minucius Felix and Lactantius, who were passionate in engaging with Cicero’s works,21 often expressing their agreement or disagreement. On the contrary, Jerome’s intention of putting the best language in service of Christian purposes is clear when he quotes Ciceronian expressions as an instrument, deploying, to use a Hieronymian expression, Tulliana puritas,22 but this assertion is most true also in relations to his approach to the Latin classics in general. In particular, the language of Jerome’s letters reveals an abundance of allusions and quotations concealed in the text, which could be easily detected and appreciated by the cultivated reader. Even though Jerome’s epistolary style is very different from Cicero’s, as Gasti asserts, a sort of Ciceronian model is perceptible from Jerome’s ability to easily change style and level with appropriateness and realism.23

 17 Ep. 60.10 (CSEL 54.561). 18 See e.g. Hagendahl (1958) 320–3. 19 Gray (2015) 23; for classics as pedagogical allies for Jerome, see also Giannarelli (2007) 296. 20 MacCormack (2013) 267; see also Gasti (2016) 29–31; 38–40. 21 For Lactantius’ complex relationship with Cicero, based on corrections, emulation and debate with the model, see Kendeffy (2015). 22 Ep. 58.8, Jerome employs this expression to compliment Paulinus of Nola. 23 Gasti (2016) 40 (n.37).

96  Giulia Marolla Jerome’s use of Cicero’s model in Ep. 123, is probably an effective example of his recurring to a mental library more than to a physical one, to embellish his Christian educational text with Ciceronian reminiscences. If Hagendahl’s assertion is indeed true that ‘if there were nothing left of Cicero’s literary oeuvre, we would in fact get a fairly good idea of its variety from the references to be found in Jerome’,24 it is also true that the lack of a linguistic analysis of a text inevitably condemns the reader to a limited understanding of its content. Therefore, the following are the results of a linguistic analysis conducted on Ep. 123, that, unsurprisingly, like most of Jerome’s epistles, would deserve a commentary but lacks one.25 The missive is a treaty on monogamy in the form of a letter to Geruchia,26 a young widow who was living in Gaul in 409, and offers a wide range of quotations and allusions to pagan authors. Both classical and biblical texts seem to be woven together, to quote Jerome’s definition of his technique of intertwining Scriptures with his own comments.27 Jerome’s intellectual loom allows him to connect words not as a literary ornament, and not even as a mere cento, but as an essential part of the fabric of his dissertation.28 Through Jerome’s weaving hands Dido is made the perfect example of a widow who repents for not honouring the memory of her late husband Sychaeus, regrets falling in love with Aeneas, and is therefore a Christian model for widows.29 Furthermore, Livy’s narration of the episode of Lucretia, undying exemplum of chastity, becomes in the letter the perfect example of how a Christian uniuira ought to behave (123.7). I should emphasize that, in this case, the choice of pagan examples is not peculiar, for it reflects Jerome’s model, which was the entirety of Tertullian’s oeuvres on monogamy.30 It is worth  24 Hagendahl (1958) 284. Before him, Luebeck (1872) 128 ff. detected more than one hundred Ciceronian echoes; see also Adkin (1992) 408. 25 All the quotations from letter 123 in this contribution are from CSEL 56/1.72–95. 26 See Fürst (2016) 198; PCBE IV, 72, s.v. Ageruchia; Rebenich (1992) 286. For the profile of correspondents in Gaul: Cain (2009) 178. 27 Ep. 123.1, Saepe ad uiduas scripsimus, et in exhortationem earum multa de scripturis sanctis exempla repetentes, uarios testimoniorum flores in unam pudicitiae coronam texuimus. Even though Jerome refers this expression only to Scriptures and not classical quotations. For a thorough analysis of the passage and of classical quotations in the text see Marolla (2017). 28 In Ep. 53.7 Jerome denigrates the centones as ‘puerilia’; cf. Curran (2012) 341. 29 For the re-use of Dido’s original myth in Christian authors and specifically in Jerome’s Ep. 123 see Nazzaro (1989) 209. 30 Especially Tert. Castit. 13.3 and Tert. De Monog. 17.2; see Micaelli (1979) 427. Ep. 123 is strictly connected with Jerome’s Adversus Iovinianum, for the different value of pagan exempla in Tertullian and in Jerome see Clausi (1995) 458 (n.3); 462. For the recurrence of Senecan themes, see also Torre (2000) 80.

Jerome’s Two Libraries  97

noticing though, that the language reveals the underlying Ciceronian model of many expressions, and that Cicero is the author to whom Jerome alludes more frequently in the letter, although never explicitly mentioned or quoted. In order to provide new information concerning this poorly studied letter, the Ciceronian iuncturae and syntagmata in Ep. 123 are analysed in what follows. While Jerome is going to explain the dangers of multiple marriages, he reports a grotesque story, immediately gaining the reader’s attention with a captivating beginning, and then lingering in the details of the tale. Ep. 123.9, rem dicturus incredibilem multorum testimoniis adprobabo. Ante annos plurimos, cum in chartis ecclesiasticis iuuarem Damasum, Romanae urbis episcopum, et orientis atque occidentis synodicis consultationibus responderem, uidi duo inter se paria uilissimorum e plebe hominum conparata, unum, qui uiginti sepelisset uxores, alteram, quae uicesimum secundum habuisset maritum, extremo sibi, ut ipsi putabant, matrimonio copulatos.

After reminding the reader, with a certain amount of pride, that he had once been in charge of Pope Damasus’ correspondence, Jerome tells he heard of a man who had been previously married twenty times who got married with a woman who had had twenty–two husbands. With contempt and disapproval, he also adds that after the wedding ceremony, a throng desired to know which of the two would outlive and bury the other spouse. Then, continuing the simple narrative of a moralizing tale, Jerome says that the wife had passed away, and the funeral procession, led by the husband, was turned into a sort of triumphal procession: the crowd, drawn to the event, praised the survivor, and crowned the man, who proceeded victoriously with a palm leaf in his hand. And yet, the whole story is introduced by the erudite incipit: rem dicturus incredibilem, which immediately reveals the irony of the author, who is trying to prove that a second marriage could easily lead to twenty more, and to such an outrageous outcome.31 The incipit marks the episode as incredible in a negative way, and enhances the stylistic level of what would otherwise seem a rather unsophisticated account. In fact, the expression is first attested in Cic. Inv. rhet. 1.16.23, ea, quae dicturi erimus, magna, noua, incredibilia esse. It also occurs in Verr. II.1.128 dicit enim rem incredibilem: C. Verrem insimulat auaritiae et audaciae, quae uitia uidentur in quemuis potius quam in istum conuenire.32 The locution

 31 This would also seem evocative of the Fabulam Graecanicam about to be told in Apul. Met. 1.1. 32 See also: Cic. Pis. 32, forsitan hoc quod dicturus sum, mirabile auditu esse uideatur, sed certe id dicam quod sentio.

98  Giulia Marolla seems to be attractive to Jerome, who also employs it in a Biblical commentary: in Hab. 2.3, rem incredibilem dicturus sum, sed tamen ueram.33 Peculiarly in Ep. 39, a consolation to Paula for the death of her daughter Blaesilla, Jerome introduces an account of the events concerning Melania the Elder with the same expression.34 In these cases, given the nature of the texts these expressions are gathered from, it would seem quite clearly that the Ciceronian echoes reflect traces of Jerome’s memories of his scholastic education, rather than books he was consulting at the time. While Jerome is warning his correspondent to avoid those who might conduct a dangerous way of living (123.14), he employs the expression e triuio, which literally means something that, coming from the crossroads, is in some way vulgar: 123.14 Fuge personas, in quibus potest malae conuersationis esse suspicio, nec paratum habeas illud e triuio: ‘sufficit mihi conscientia mea; non curo quid loquantur homines’.

It seems that, with this specific meaning, this expression is attested before Jerome only in Cicero’s pro Murena.35 Cic. Mur. 13, non debes, Marce, arripere maledictum ex triuio 36 aut ex scurrarum aliquo conuicio, neque temere consulem populi Romani saltatorem uocare.

Cicero is reproaching Cato for having called Murena a saltatorem, a dancer, which is not a vulgar expression itself, but is used as a term of abuse when referred to a Roman consul. The expression ‘e triuio’ has no other occurrence before Jerome, who uses it not only in letter 123, but also in Adu. Rufin. 3.42, nihilque super hoc amplius audies nisi illud e triuio: Cum dixeris quod uis, audies quod non uis37; and

 33 It is noticeable that the expression seems to be a calque of a Greek periphrastic form, as attested in Plut. Alex. Fort. 2.11. 341 ff., ἄτοπόν τι δόξω λέγειν, ἐρῶ δ᾽ ἀληθές. In Latin authors it also occurs in Sen. QNat. 4a, Praef. 4, incredibile est quod dicturus sum, sed tamen uerum. On this see Vottero (1989) 467. 34 Ep. 39.5 rem sum dicturus incredibilem, sed Christo teste non falsam (CSEL 54.305). The unbelievable story he is about to tell, in this case, concerns the abnegation of the woman, who was almost relieved by the death of her relatives, since she could dedicate herself with even more commitment to the ascetic life. 35 I find ‘e triuio’ also in Suet. Vesp. 5.4, Prandente eo quondam canis extrarius e triuio manum humanam intulit mensaeque subiecit. However, in this passage, the expression is not to be intended metaphorically but verbatim. 36 E triuio is translated ‘from the gutter’ in Macdonald (1995) 70. 37 CCSL 79.112.

Jerome’s Two Libraries  99

in other letters such as Ep. 38.5, si tunica non canduerit, statim illud e triuio: ‘inpostor et Graecus est’;38 Ep. 54.5, ubicumque uiderint Christianum, statim illud e triuio: ὁ Γραικός, ὁ ἐπιθέτης.39 The locution, therefore, seems to be born with Cicero and thereafter employed by Jerome in various informal statements, but it would seem possible that this expression simply happens not to survive elsewhere. Moreover, still in chapter 14, when Jerome argues that there is no one so physically weak not to think he might survive for at least another year, the author seems, once again, to be echoing Cicero. In fact, it has been already pointed out by Hagendahl that Ep. 123.14, nemo enim tam fractis uiribus et sic decrepitae senectutis est, ut non se putet unum adhuc annum esse uicturum, is an imitation of a similar Ciceronian sententia: Cato 24, Nemo est enim tam senex, qui se annum non putet posse uiuere.40 Was this a conscious reminiscence of his years as a student? Furthermore, it is worth mentioning that the Ciceronian reference is enhanced by the Virgilian iunctura ‘fractis uiribus’.41 Letter 123 is mostly known for the recollection of the barbaric invasions of Gaul in chapter 15, a passage that has granted the letter popularity, often quoted by scholars as one of the historical sources of the barbaric crossing of the Rhine frontier in 406. Apart from the content itself, though, it is worth mentioning that, while describing contemporary events, Jerome taps into his mental library enriching the description with panache and sophistication. Ep. 123.15, praesentium miseriarum pauca percurram. Quod rari hucusque residemus, non nostri meriti sed domini misericordiae est. Innumerabiles et ferocissimae nationes uniuersas Gallias occuparunt. Quicquid inter Alpes et Pyrenaeum est, quod Oceano Rhenoque concluditur, Quadus, Vandalus, Sarmata, Halani, Gypedes, Heruli, Saxones, Burgundiones, Alemanni et – o lugenda res publica!– hostes Pannonii uastauerunt. Etenim Assur uenit cum illis. Mogontiacus, nobilis quondam ciuitas, capta atque subuersa est et in ecclesia multa hominum milia trucidata, Vangiones longa obsidione finiti, Remorum urbs praepotens, Ambiani, Atrabatae, extremique hominum Morini, Tornacus, Nemetae, Argentoratus, translatae in Germaniam, Aquitaniae, Nouemque populorum, Lugdunensis, et Narbonensis prouinciae praeter paucas urbes cuncta populata sunt, quas et ipsas foris gladius, intus uastat fames.

 38 CSEL 54.293. 39 CSEL 54.470. 40 Hagendahl (1958) 253. 41 Verg. Aen. 2.170 (spes Danaum, fractae uires, auersa deae mens) where it concerns the Achaean forces, weakened and abandoned by Athena since Ulysses and Diomedes had stolen the Palladium. For the singular form of the iunctura see also Cic. Att. 4.3.4, contentio fratrum trium turpis, fracta uis, contemptus furor.

100  Giulia Marolla This section of the letter is introduced by the iunctura ‘praesentium miseriarum’, which expresses Jerome’s astonishment for the contemporary events and his lack of faith in a positive outcome. The very same sense of resignation is to be found in the first occurrence of the expression, when Cicero laments the sensu praesentis miseriae in Att. 3.15.2.42 Cicero, writing from Thessalonica in 58 BC, affirms that knowing the misery of present events worsens his melancholy.43 In addition to that, after recalling the name of all the invaders of Gaul, Jerome exclaims: O lugenda res publica! ‘Alas, pitiful State!’ While there is no previous record of the use of the expression with the verb lugere in the gerundive form, the locution lugere rem publicam is often used, exclusively by Cicero,44 and it would seem that Jerome is the first author to make use of this expression after him. In this case, Jerome’s echoes of Cicero would seem rather conscious; furthermore, the passage is overflowing with erudition, containing both Biblical (Ps. 83.9, Etenim Assur uenit cum illis) and Virgilian quotations (Verg. Aen. 8.727 extremique hominum Morini). Subsequently, Jerome recalls the ill–fated dies Alliensis: when in 390 BC Gauls led by Brennus, after defeating the Romans near the river Allia, looted the city, an episode Jerome refers to as an undying shame, a dedecus aeternum:45 123.16, aeterno quondam dedecore Romanum laborabat imperium, quod Gallis cuncta uastantibus fusoque apud Alliam exercitu Romam Brennus intrasset.

The iunctura ‘aeternum dedecus’ is previously attested only in Cicero’s Pro Fonteio and in the Ilias Latina. Cicero employs it while addressing his speech directly to the judges: he wants them to picture Marcus Fonteius in tears at the thought of leaving his mother and sister heartbroken and stained with eternal shame, Cic. Font. 48, cum acerbissimo luctu dedecus aeternum miseris atque ignominiam relinquat.46 Thereafter Jerome, in an act of self– pity, explains his fear of having spoken too freely because what he said is ‘dangerous’ (periculosa sunt), both for him and

 42 Nam ceteri dolores mitigantur uetustate, hic non potest non et sensu praesentis miseriae et recordatione praeteritae uitae cottidie augeri. 43 I find the locution also in Tac. Hist. 5.3.1. 44 E.g. Cic. Brut. 1.4, cum lugere facilius rem publicam posset, si uiueret, quam iuuare; Cic. Att. 12.28.2, Quod me ad meam consuetudinem reuocas, fuit meum quidem iam pridem rem publicam lugere, quod faciebam, sed mitius. 45 See the analysis in Rebenich (2009) 54–55. 46 Ilias Latina 3.257–8

Jerome’s Two Libraries  101

for his audience; therefore, he condemns the lack of freedom of speech. The topics he presented are so dangerous that the reader can’t freely express his pain, for paucity of willingness or because he does not dare crying over misfortunes: 123.16, et haec ipsa quae dixi periculosa sunt tam loquenti, quam audientibus, ut ne gemitus quidem liber sit nolentibus, immo nec audentibus nobis flere, quae patimur.

‘Gemitus liber’ is a Ciceronian iunctura, and it is first attested in the description of what happened when Caesar came back from Alexandria in Cic. Phil. 2.64, cum omnia metu tenerentur, gemitus tamen populi Romani liber fuit. The property that once belonged to Pompeius was sold in Rome, but despite the fear of Caesar’s men, people could freely express their grief: their gemitus was liber. Before Jerome, the iunctura is only used by Seneca the Elder (Controv. 7.2.5, proposito in rostris capite Ciceronis, quamuis omnia metu tenerentur, gemitus tamen populi liber fuit), who probably makes use of the Ciceronian expression gemitus liber on purpose to recall the death of Cicero himself, and people’s reaction at the sight of Cicero’s head on the Rostra; an homage to Cicero from Seneca the Elder, and, centuries later, from Jerome, one of his greatest admirers.47 Apart from Cicero, similar uses of classical reminiscences are detectable in the letter, for instance Jerome employs the language and images of comedy when he needs to speak effectively and firmly, such as in 123.12, quod si et nobis iure conceditur, adhinniamus ad omnes feminas, in which the author pictures men chasing women as if they were ‘horses in heat’. The image and the verb adhinnire seem to be borrowed from Plautus Cistellaria,48 when the senex tried to win the young Gymnasium’s attention. Jerome’s ability to change level and style, with realism and expressive vis, has already been discussed before in this contribution. When Jerome needs to enrich the exempla castitatis, he also resorts to historical models of chastity, such as in the following passage, where he describes the heroic mass suicide of three hundred Teutonic matronae after Marius’ victory at Aquae Sextiae.

 47 Jerome employs it not only in letter 123 and in other letters, but also in his exegetical works, such as in a Bible commentary: in Is. 5.21.2, seu quod tantis malis oppressa sit ipsa Babylon, ut ne gemitum quidem liberum habeat. (VL 27, 641). 48 Cist. 307, adhinnire equolam possum ego hanc. Ov. Rem. am. 634 has also been pointed out to me; for the verb adhinnire as a libidinis signum see Ambr. Hex. 6.3.10, si te edacitas equi intemperantiaque delectat et adhinnire ad feminas uoluptati est. See also TLL s.v. 650.80–81.

102  Giulia Marolla Ep. 123.7, Gens Teutonum ex ultimis oceani atque Germaniae profecta litoribus omnes Gallias inundauit saepiusque caesis Romanis exercitibus, apud Aquas Sextias Mario oppugnante, superata est. Quorum trecentae matronae, cum aliis se uiris captiuitatis condicione tradendas esse didicissent, primo consulem deprecatae sunt, ut templo Cereris ac Veneris in seruitium traderentur. Quod cum non inpetrarent, submouente eas lictore, caesis paruulis liberis mane mortuae sunt repertae suffocatis laqueo faucibus, et mutuis conplexibus se tenentes.

Similar extant versions of the episode are attested in Valerius Maximus (6.1. de ext. 3,49 – though his description is laconic and less detailed – and in Florus (1.38. passim)50 who, however, pictures these women as Cimbers (instead of Teutons), and describes their request to be priestesses, not slaves (as Valerius Maximus and Jerome write). It would not seem, therefore, that either Valerius Maximus or Florus are models for the passage, also because Jerome gives a number of the matronae, unlike the others, and specifies they wished to serve as slaves in temples dedicated to Ceres and Venus.51 On the other hand, the connection with Florus is undeniable when Jerome alludes to his programmatic praefatio of the Livian epitome. Florus explicated the difficulty of enclosing the vast historical narration in a succinct prose picturing himself as a cartographer, an image that is borrowed by Jerome in Ep. 123.13.52 Flor. Ep. it. 1, Praef. 3. faciam quod solent qui terrarum situs pingunt: in breui quasi tabella totam eius imaginem amplectar.

 49 Teutonorum uero coniuges Marium uictorem orarunt ut ab eo uirginibus Vestalibus dono mitterentur, adfirmantes aeque se atque illas uirilis concubitus expertes futuras, eaque re non impetrata laqueis sibi nocte proxima spiritum eripuerunt. Previously Plutarch (Mar. 27), but the passage is very different from Jerome’s; written a few years after Jerome’s letter 123 is Orosius’ description (Hist. 5.16). 50 Florus 1.38.17 Perinde speciosa mors earum fuit quam pugna. Nam cum missa ad Marium legatione libertatem ac sacerdotium non impetrassent (nec fas erat), suffocatis elisisque passim infantibus suis aut mutuis concidere uulneribus aut uinculo e crinibus suis facto ab arboribus iugisque plaustrorum pependerunt. 51 While Valerius Maximus mentions their request to be sent to vestal virgins as a gift. Useful comparisons in Salomone Gaggero (1981) 228–9 (n.24). 52 For the interpretation of Florus’ words see Facchini Tosi (1998) 91. Jerome actually employs the same image in various contexts, both exegetical and epistolary, such as Ep. 60.7, et sicut hi, qui in breui tabella terrarum situs pingunt, ita in paruo isto uolumine cernas adumbrata, non expressa signa uirtutum (CSEL 54.556); Ep. 73.5, Haec legi in Greaecorum uoluminibus et quasi latissimos terrarum situs in breui tabella uolui demonstrare non extendens spatia sensuum atque tractatuum, sed quibusdam punctis atque conpendiis infinita significans, ut in parua epistula multorum simul disceres uoluntates (CSEL 55.17–18).

Jerome’s Two Libraries  103

123.13 Quasi in breui tabella latissimos terrarum situs ostendere uolui.

The recognition of these allusions allows us to realise that Jerome’s borrowing of classical echoes, and in particular Ciceronian expressions, is absolutely a conscious choice. It does not seem fortuitous that, at the beginning of the treaty, the chapters containing his doctrinal dissertation on monogamy are overwhelming both in content and linguistically: his lexicon and his examples reflect Christian models, the Scriptures, Paul, and Tertullian, and there is no place for rhetorical fineries. However, starting from chapter 7, as soon as the author is not exclusively recalling the Christian tradition on monogamy and he backs up his arguments with contemporary facts, his style and language are richer in classical echoes and Ciceronian allusions. The language itself used by Jerome can be even more revealing than explicit quotations and might hint to what Jerome used to read when in Palestine, as well as to reminiscences of his juvenile years as a student. Therefore, we could never know which books were actually inside of Jerome’s library, but we do know that, even in the Holy Land, he continued to have manuscripts of pagan authors copied for him (as a malevolent Rufinus tells us), for instance of Cicero. It is also well known that despite his vow in Ep. 22, Jerome continued to allude to classical authors and possibly some of them had a place in his library, and letters dated from the years in Bethlehem hint to that possibility.53 It is certain that letter 123, like many other works of his that have not yet received a commentary, provides new evidence helpful in better understanding which authors did Jerome read, but also in glimpsing the interaction between Jerome’s two libraries: the physical library in Bethlehem, and the mental library, both filled with the best texts his predecessors had ever written.

 53 Pease (1919) 159 ‘How largely these reminiscences go back to his boyhood and to what extent they were refreshed by renewed reading we shall never know’, although dated, this analysis proves to be absolutely fitting. See also Adkin (1999) 164 who affirms that there is no evidence to support Hagendahl’s theory that Jerome had re–read classical texts in ‘Hagendahl’s second period’ and believes the classical quotations in his works are ascribable only to his memory.

Daniela Immacolata Cagnazzo

Some remarks on P. Lit. Lond. 63, a riddle epigram of an anthology? The word anthologion1 appears for the first time in Diogenanus around 150 AD. (s.v. Suidas)2 and seems to indicate a collection of extracts. The epigrams are grouped according to a mostly thematic criterion, forming the collection; the criterion is the same as the one employed in the sylloges and libelli, in which we find collections of material belonging to different authors, along with new epigrams, mostly aimed at publication. So, one may argue that for these kinds of works, the role of author and editor coincide: they are indeed in most of the cases the same person. During the Greco–Roman period, considered ‘il punto più alto della diffusione sociale della scrittura/lettura’,3 the editorial genre of epigrammatic anthology seems to have undergone a period of crisis in Egypt, especially in the 2nd century AD.4 In fact, although the imperial epigram shows its vitality, the contribution of the Greco–Egyptian culture in this field appears modest, which is evident by the small number of papyrus findings, which are otherwise conspicuous as regards the centuries prior to this period. Even if the themes changed, we can find texts of different genres that reflect ‘una realtà più provinciale nel gusto’,5 which can be seen in the increased breadth of social demand for literature, as well as the different perspective of social needs broadly speaking. Therefore, the great anthologies that were popular before this era ceased to be published, being replaced by a more monographic and private style and content, or by literary products related to a formal occasion or an academic purpose. Hence, there was an interesting phenomenon involving a rise in consumer goods and

 1 I would like to thank Prof. Tiziana Drago for her strong support and her willingness to help, as well as many thanks for her careful reading of this article. I am very grateful to Prof. Rosa Otranto, whose valuable suggestions contributed greatly to this paper. See Argentieri (1998). 2 See Cameron (1993) 84–90. 3 Cavallo (1986) 85. This idea is supported by the ‘trattatelli di lingua greca sulle modalità di scelta dei libri per l’allestimento di biblioteche private’ (Otranto [2000] xi). 4 It does not seem to be like this in other geographical areas, as witnessed by the presence of two leading exponents of epigrammatic Greek Imperial Age. The two being Straton of Sardis and Rufinus of Antiochia, although there is no certainty with regards to the chronological position of the two epigrammatists; in this regard see Floridi (2007). 5 Morelli (1994) 410. https://doi.org/10.1515/9783110632590-008

106  Daniela Immacolata Cagnazzo entertainment, namely horoscopes, γρίφοι, hexametric ethopoeia, erotic or symposium related epigrams, probably to be used for symposiums and banquets. This is the case of P. Lit. Lond. 63,6 dating back to the 2nd or 3rd century, and whose author is unknown. It is a wax tablet (22.5 x 12 cm), from Egypt, although the exact location of discovery is unknown. The text, which runs on two elegiac couplets, is written in eight stichoi and some verses are divided into two by a caesura. On the tablet, there are also visible traces of a previous composition. In the text, the errors definitely correspond to a way of writing that would be found in a school environment. This can be inferred from the repetition of the βρο– syllable in the first verse, by the confusion of the letters –η for –υ in the second verse, and by the over correction of –αι instead of –ε in the last verse. From a palaeographic analysis, we can determine that it is a ‘zero degree’ handwriting, in other words, it belongs to a schoolchild. The writing looks uncertain and shaky, the letters are big and the line spacing is wide.7 Subsequent corrections have been made to the text, some of which, according to Diels (1898), could correspond to another hand, more precisely that of a teacher. For such reasons, it has been conjectured that it is a textbook, although it is difficult to determine whether it can be considered a text copied from a model of a renowned author (at least from the 2nd century), or from a text specifically written by the teacher for the student. It is also questioned as to whether it was an already formulated school exercise or if it only became a calligraphy exercise during a subsequent phase.8 P. Lit. Lond. 63 concerns an object linked to writing, because it deals with the physical material on which the epigram is engraved: a wax tablet, as is clear from the text: Ἄψυχος γεγαυῖα βροτείαν | ἔνδοθεν αὐδὰν Βωστ[ρέω] | σὺν Μουσέων | φθεγγομένα στόματι. | Τὰν δ’εἰ δὴ τρωθεῖσα | [σι]δηρίῳ ἐντετύπωμαι | [ἀγ]γέλλω θνατοῖς κοὔπο|τε ἔχουσι πάρος. Even if I’m not alive, in my gut I echo the human voice with the mouth of the Muses. Engraved by the stylus, I print the characters of writing, even if they did not have it before, I send their voice to mortals.

 6 P. Lond. Lit. 63 = T. British Museum inv. 29527 = MP3 1765 = LDAB 5055. A photographic reproduction of the tablet is available in the editio princeps of Diels (1898) 858. 7 See Cribiore (1996) n. 202. 8 See for school papyrus Debut (1989) 251–78; Zalateo (1961) 160–234.

Some remarks on P. Lit. Lond. 63, a riddle epigram of an anthology?  107

The solution of the riddle is not perspicuous and there are many interpretive proposals. Milne (1927, ad loc.) suggested as a solution ‘the letter’, probably based on the analogy established by Diels with the Sappho of Antiphanes (fr. 194 K.– A.),9 supported by the presence of ἀγγέλλω.10 A different interpretation, which is rather convoluted, is the one theorized by Pordomingo,11 according to which the text is about a ‘mask’, which appears to come to life once it is fully formed (the stylus, however, would be the instrument used for engraving and carving the mask itself). Yet, the most plausible interpretation is that the solution is simply the ‘wax tablet’, as was previously interpreted by Beudel,12 as it better suits the ability to be engraved with the stylus, an instrument explicitly mentioned in the text.13 The material support of the text facilitates the assumption of a decreasing spread of the anthology–form, in favour of ‘consumer epigrams’, and also of the ones employed for schooling purposes, or to trigger intellectual reasonings. The latter is the purpose of the tablet previously mentioned, which although being a school exercise, forces the user to exploit their mental capabilities in order to solve a linguistic and metaphorical maze. This appears indeed to be the case of and epideictic epigram in the form of a riddle. In Greek, there are several names for this: γρῖφος, αἴνιγμα, πρόβλημα,14 depending on the resolution of the riddle that has been posed, which can be in elegiac couplets, hexameters, iambic trim-

 9 For the analogy between the riddle–epigram and the Sappho of Antiphanes I refer to Di Marco (2009) 86. 10 In the v. 4 Di Marco (2009) 85–86 reads ἀντέλλω. 11 See Pordomingo (2001) 724. 12 See Beudel (1911) 28. 13 This riddle resolution would not be, however, an unicum, as demonstrated by some cases in the Anthologia Graeca, e.g. Anth. Pal. 14.45, in which the contradictory statements, the consistency of the substance, as well as the ‘speaking without talking’ of the last verse, all indicate a wax tablet. On the other hand Anth. Pal. 14.60 is a riddle about a wooden tablet, according to the lemma of the manuscript that transmits it. This epigram has many similarities with the one we are addressing, not only because it contains contradicting ideas, but also because there is a reference to the Muses, the custodian of secrets, as well as the stylus that causes an inanimate object to come to life and be able to speak. The riddles were aimed to an almost exclusively oral use in the symposia, but experienced in writing a decisive moment because, even before being recited, they were written and then learned by heart. 14 The πρόβλημα, according to Plutarch (Mor. 150b), is the investigation about philosophical arguments, from which arose the dialectical confrontation, while the αἴνιγμα of which Plutarch informs us (154b) is a deliberately obscure expression, which contemplates only one answer to the riddle.

108  Daniela Immacolata Cagnazzo eters, or even prose. For our research, it is useful to consider the first type of riddle. What does γρῖφος mean? The term actually designates a ‘fishing net’, but metaphorically means ‘il tranello verbale che si cela dietro un indovinello’.15 This typology of epigrams can be found in the fourteenth book of the Anthologia Graeca, in which a large number of riddle epigrams belong; they are mostly anonymous, except for the numbers 63 of Mesomedes and 101 of Cleobulus. Indeed, the Anthologia was the next stage of a tradition that has its roots in the Hellenistic Age, when the riddle epigrammatic works of Philitas of Cos and Simias of Rhodes had started to be prepared. Even so, the riddle has even older origins. It was already present in the Indo–european culture,16 and some examples can even be found in the Homeric poems, and also in lyric poetry and comedy, as it seems to be confirmed by some fragments of Hesiod, Simonides and Eubulus, just to name a few. The riddle in the epigrammatic form would seem, however, much more recent. It is, in fact, to Philitas that we owe the transition from the riddle of symposium lyric poetry to the performance of the epigram.17 It is in the environment of the Hellenistic symposium that γρῖφος finds a greater use: according to Ath. 10.457e, there were three types (letter, syllable, name), and this gave rise to real linguistic skirmishes between symposiasts, with prizes such as meat slices, lascivious kisses, laurel wreaths, and punishments such as the obligation to drink pure wine or a cup of salt or sweet water. Therefore, the γρῖφος, which is easily comprehensible but at the same time highly ambiguous, started as a kind of game, and provided an opportunity for the duelists to ‘confrontarsi dialetticamente su questioni di poco conto, banali, le quali non prevedono una risposta giusta che ne escluda altre’.18 Due to the content of γρῖφος, the physical material used (wax tablet), and the poor quality of such material, it can be argued that P. Lit. Lond. 63 is part of a  15 Della Bona (2013) 172. 16 See West (2007) 367. 17 See, for example, P. Louvre inv. 7733 (= SH 983–4) – 2nd century BC –, known as the Elegy of the Oyster, which shows, on the verso, a fine riddle epigram with an extensive hypomnema, that probably belongs to a wider collection of riddles, because text and commentary were usually arranged in different books. The epigram is formed by six verses, and has a typical γρῖφος structure. It is interesting to note that the riddle solution is referred to even before it is actually announced. In fact, the first column reports Ο[ΣΤΡ]ΕΙΟΝ, which is a typical object of the symposium. We do not know if it was a common practice, because there are no similar findings, not even in the fourteenth book of the Anthologia Graeca. This would be a further confirmation of the hypothesis that it is part of a book of riddle epigrams, where the solutions were published before, then the text, and finally the commentary. On this papyrus, see Lasserre (1975) 173; Sbardella (2000) 182–4; Kwapsiz (2013), 164–5. 18 Della Bona (2013) 171.

Some remarks on P. Lit. Lond. 63, a riddle epigram of an anthology?  109

typology of school papyri that have come in a large number from the Egyptian environment.19 The P. Lit. Lond. 63 confirms the 2nd century crisis in Egypt of the scholarly output in favour of a production for a more private or scholastic use, although the books circulation and literacy were growing.20 To conclude, this research, which compares data from all the second and third centuries papyrus discovered in Egypt, including epigrammatic texts, seems to corroborate the picture about the crisis of the anthology form in the 2 nd century BC that emerges from previous studies. In fact, I have found only nine fragments of papyri, which suggest the collection of epigrams in anthology–form. Among these papyri, there is a wax tablet (P. Lit. Lond. 63), a wooden tablet (P. Ross. Georg. 1. 14 = MP3 1769)21 and an ostrakon (O. Bodl. 2. 2174 = MP3 1946 = LDAB 4894). However, only five of them contain more than just one epigram: P. Oxy. 54.3726 (= LDAB 2456);22 P. Oxy. 4.671 (= MP3 1614);23 P. Oxy. 1.15 (= MP3 1618)24; PSI 1. 17 (= MP3 1608);25 P. Stras. inv. 1016v (= MP3 1589 = LDAB 2449):26 P. Lips. inv. 1445v (= LDAB 9937).27 Therefore, there is only a small number of findings of quite a low quality and made in a style that is not particularly elegant, which unequivocally testifies to, during the Imperial Age, ‘un impoverimento qualitativo della produzione locale, un delicato interesse verso l’epigramma letterario e l’abbandono in ambito provinciale dei grandi autori del passato’.28 It would be necessary,

 19 See for the school papyri Zalateo (1961), Debut (1986), Cribiore (1996). 20 See Guidorizzi (1984) 313–4 and Morelli (1994) 392–3. 21 It is a particular example of a funerary epigram for a young man, Anubion, written on a wooden tablet to be placed over his mummy. 22 The papyrus fragment, which is in damaged conditions, is a calligraphic copy of the 2nd century and displays once again the kind of writing pieces that involved monuments attaining speaking abilities, typical of the funerary and votive epigrams. It is, probably, according to Nisbet (2003) 165, part of a series of epigrams, which was supposed to be at the beginning of a piece by Theocritus. 23 The papyrus is heavily damaged and contains hexametric verses influenced by style of Nonnus. 24 See Guidorizzi (1984) 314–5. 25 It is a rare case of an opistograph papyrus, containing six sepulchral epigrams, all dedicated to the same subject, the deceased Euprepius, for which his daughter had commissioned the verses to a poet. 26 From the 2nd century AD, it contains two epigrams in elegiac distich about women, along with some sententiae of Menander. 27 It is a papyrus dating back between 275 and 290 AD, that although later than the other, confirms the hypothesis of a decadence of the interest in drafting and copying anthology–form in Greco–Roma Egypt. See Luppe (2002) for a detailed analysis of the papyrus. 28 Guidorizzi (1984) 317.

110  Daniela Immacolata Cagnazzo at this point, to find the reason for the decline of this literary genre and anthology–form in a precise geographic–cultural environment, even though it was so widely disseminated elsewhere and even compiled in a large collection like the Anthologia Graeca. I therefore think that the work that still needs be done in this direction is remarkable, and that the study of the epigrammatic form should be always carried out in combination with the analysis of the book circulation mechanisms, the material development of the writing supports as well as with the social changes and the demand for such literature.

Leonor Hernandez Oñate

Textual Tradition and Reception in Theocritus The Case of αἰπολικὸν (Theoc. 1.56)

 Introduction The present study aims at analyzing the process through which the poetic corpus ascribed to Theocritus was formed. Hence, I will emphasize that the transmission1 of his work as a whole constitutes relevant evidence of its reception, since I consider that the presence, absence, ordering (among other procedures) of the poems in certain manuscripts or papyri are all elements that show us how his work was understood by critics, readers and editors throughout the time. This study, consequently, departs from the belief that textual criticism and transmission are profoundly associated to the history of reception,2 which is also a determining cause of the literary genre. Textual tradition, therefore, is generally speaking a key to understanding the evolution of the poetic phenomenon. Furthermore, this process will be exemplified by an examination of the lectio offered at v. 56 of Idyll 1 (in most modern editions), i.e. αἰπολικὸν, which is also unanimously present in the extant manuscripts3 in spite of the number of possibilities mentioned by the scholia, and of its problematic interpretation. I will argue that this can be explained as a bias caused by the horizon of expectations that the generic prejudices imposed upon scholars and the receiving audience of  1 According to my vision, transmission and tradition, although they are different concepts, are profoundly related and mutually dependent. Throughout this text, I will understand the former as the textual process of reading and copying that enables tradition, and the latter as the totality of testimonies that convey a literary work and its interpretation. 2 My concept of Reception is mainly based on the inquiries of Hans Robert Jauss, especially his famous Literaturgeschichte als Provokation der Literaturwissenschaft, to which I had access in its English version: Jauss (1970); and secondarily by more current approaches to this theory, like those displayed in Hardwick–Stray (2008), and Martindale–Thomas (2006). 3 The reading of all the manuscripts in the first hemistich of v. 56 is αἰπολικὸν τι θάημα (K, Va) or θαῦμα (La) but τι was deleted by Porson. Cf. Wilamowitz (1906) 36; Gow (19522) II.13: ‘The superfluous τι of the mss. is no doubt due to the quantity of the first syllable of θέαμα…’; Legrand (19463) 22: ‘Ante θάημα in codd. τι legitur contra metri leges [...]’. Neither this pronoun nor the variant θάημα/ θαῦμα / θέαμα (Hesych.) are under discussion in this article, since they present polemics and difficulties of their own. https://doi.org/10.1515/9783110632590-009

112  Leonor Hernandez Oñate the Theocritean text. In other words, one textual tradition4 chose to maintain this reading in all of its manuscripts, oblivious of the wide range of possibilities that must have circulated in antiquity, because of the strong influence of the pastoralization of the so–considered bucolic genre, and as a result the gradual consolidation of his text was strongly affected by the generic conception of it.

 The manuscript transmission of Theocritus and the conformation of the corpus The manuscript transmission of Theocritus includes both the poems present in the manuscripts and those that have been found thanks to some recently discovered papyri. First of all, one should recall that the extant work ascribed to Theocritus includes over thirty idylls, epigrams, fragments and a technopaegnion. Gow’s edition (19522) considers thirty–one idylls, because he includes a fragment of the Antinoe Papyri that he called παιδικά γ’, also taken by Gallavotti as Idyll 31, Fragmentum aeolicum. Nevertheless, Gallavotti’s collation of 1993 contains thirty–three of them, since he numbered the hexametric fragment of the Berenice as 325 and the spurious In mortuum Adonem as 33, along with 25 epigrams6 and the technopaegnion known as The Syrinx.7 The manuscript tradition of Theocritus is rather complex, considering that these documents also include the work of two other known Greek bucolic poets, namely Bion and Moschus, and of others that remain anonymous, all of which provide a remarkable heterogeneous character to the corpus bucolicum.8 Besides, the latter is composed by more than 180 codices that must date to the thirteenth and fourteenth centuries. Furthermore, as Gallavotti noted, this constitutes only

 4 Cf. Gallavotti (1951) 63–67. The Italian scholar informs us that all the extant manuscripts derive from one byzantine archetype and represent a medieval tradition. 5 This fragment was extracted from a quotation by Ath. 7.284, and does not pertain to the group transmission of the Corpus Theocriteum, since it does not appear in the manuscripts that compose it. Perhaps that is why Gow placed it as Fragment 3, among other two Theocritean fragmenta sine verbis transmitted by Eustatius (fr.1) and the Etymologicum Magnum (fr.2). 6 Gow (19522) includes 27. 7 The corpus and the Palatine Anthology preserve other technopaegnia, The Altar, ascribed to Dosiadas, The Wings, The Two–edged Axe and The Egg, allegedly by Simias, and another Altar by a character named Besantinus. Cf. Gow (19522). 8 Gallavotti (1951) 65.

Textual Tradition and Reception in Theocritus  113

one part of the tradition, the medieval tradition, even though it can be traced back to antiquity.9 The corpus is normally classified into three families: K. The Ambrosian family, considered the most reliable,10 although the recent papyrological discoveries prove that none of the medieval traditions offer an important approach to the ancient traditions.11 It contains Idylls 1–17 and 29, along with other sort of poems, in the following order: 1, 7, 3–6, 8–13, 2, 14, 15, 17, 16, 29, the epigrams, and two technopaegnia, the Altar and the Syrinx. La. The Laurentian family. These manuscripts contain Idylls 1, 5, 6, 4, 7, 3, 8– 13, 2, 14–16, 25, Moschus 4 (Megara), Idyll 17, Moschus 3 (Ep. Bion.), Idylls 22, 18, 20, 21, Moschus 1 (Amor fugitivus), Idyll 19, Bion 1 (Epitaphium Adonidis), Idylls 32 Gallavotti (In mortuum Adonem), 23, and Bion II (Epithalamium Achillei Deidamiaeque). Va. The Vatican family,12 which preserves Idylls 1 to 18 in the canonical order, for it was followed by the milanese editio princeps of 1480 and by the Aldine edition of 1495.13 As we can see, the manuscripts give a privileged position to the idylls considered bucolic, 1. 3–11, if compared with the other poems from the corpus, none of which are missing in the manuscripts, including Idylls 8 and 9, nowadays considered non authentic by most scholars.14 All of these constitute a sort of central unity, while the rest of the poems are relegated to a, say, peripheral position, and appear randomly in the manuscript branches. From this point of view, it would seem that the essential poems for this medieval tradition were those considered bucolic. For example, the codex G Laurentianus is divided into three parts: first we find (Va) 1, 5, 6, 2–4, 7.1–54, then we have (La) 7.55–13, and finally 15, 14 and Te. 2, the technopaegnion known as The Wings, ascribed to Simias.15 Besides, Idylls 24–28 are unknown to these three families and are found only in some humanistic codices (CDH); 30 was discovered in the 19th century from codex C, and 31 is a fragment contained in the Antinoe Papyrus and was published in 1930.

 9 Gallavotti (1951) 67–68. 10 Hunter (1999) 28. 11 Gutzwiller (1996); Cf. Gallavotti (1951) 71. 12 This family is considered the least reliable. Hunter (1999) 28. 13 Even though the common ordering of the Idylls in present days derives from the 1566 edition by Stephanus. 14 Nevertheless, Irigoin (1975) tried to prove the authenticity of these poems by means of numerical analysis. 15 Cf. Gow (19522) I.xxxvii–xxxix.

114  Leonor Hernandez Oñate At the same time, the extant papyri represent another important stage of the tradition, one set apart from the Byzantine manuscripts, since they must be related to a textual strand that does not descend from the hypothetical archetype of the medieval documents.16 Codices and papyri represent, thus, a different aspect of the tradition. Given that we are interested in the transmission of Theocritus’s poetry as a group, the papyri acquire special relevance as they too contain sequences of poems and provide unique information about the nature of the early collections and editions of the Theocritean oeuvre. The most important papyri for the study of Theocritus are P. Oxy. 17.2064, which can be dated back to the 2nd. century AD and is usually read with P. Oxy. 50.3548 (Po in Gallavotti), and the Antinoe Papyrus, from the 5th century AD (Pa). Both are the most cumulative documents and preserve the following poems: Po: 1, 6, 4, 5, 7, 3, 8–11. Pa: 1, 5, 7, 10, 14, 13, 12, 2, 18, 15, 26, 24, 17, 28–31, 22. Although Po, to a certain extent, resembles the ordering of the Laurentian family,17 the rest of the papyri portray the large variability of the way in which the poems were arranged in antiquity. We can take the 5th century Pergaminus Louvre 6678 et Rainer18 as an example of this, since is contains a few words of Idylls 1, 4, 5, 13, 15, 16, 22, 26; nevertheless, the only evidence about the order is that 4 was followed by 5 and that 26 is on the verso of 13, a sequence that is otherwise unattested.19 It seems evident, then, that in antiquity Theocritean poetry circulated in a different sequence from that which these three families inform us of. 20 The evidence pointed above, then, leads us to believe that our manuscripts represent a stage of the conformation of the corpus notoriously guided by thematic and generic criteria, in which the predominance of the poems labelled as bucolic over other literary forms is most clear. But when did this process take place? Can we trace it back to Theocritus, or to his earliest collections? Or is it a later phenomenon that cannot be attributed to the 3rd century conception of Theocritean literature? Current scholarship has stated that Hellenistic editions of poetry could contain works of a diverse and heterogeneous nature and that they were not guided by a unifying thematic principle, but rather by the concept of polyeideia, the multiplicity of forms (εἴδη),21 as Kathryn Gutzwiller has proven by  16 Gallavotti (19933) 13–20. 17 As noted by Gutzwiller (1996) 126. 18 Pe in Gallavotti’s sigla, P. 4 according to Gow’s edition. This pergaminus is constituted by eight leaves, seven of which were found in the Louvre and the eighth in Vienna. 19 Gow (19522) I. l–li. 20 Gutzwiller (1996) 120–5. 21 Also, a typical trace in Hellenistic poetics. Gow (19522) I.lxxii.

Textual Tradition and Reception in Theocritus  115

tracing the presence of all kinds of poems within the hypothetical 3rd century collection of Theocritus introduced by the famous ‘Chian epigram’ (Anth. Pal 9.434; Gow Ep. 27; Gall. Ep. 5).22 Another demonstration of this polyeideia and of its pre–eminence in Hellenistic poetics, not only at the moment of its creation but also of its editorial conformation, is the recently found Milan Papyrus, better known as the Posidippus Papyrus,23 a book roll containing a collection of about 112 epigrams, apparently all by the epigrammatist Posidippus of Pella; it probably dates back to the 3rd century BC.24 In this remarkable document, which allows us a better understanding of the nature of collections and anthologies in this era, we find poems of all kinds, ekphrastic, encomiastic, etc., in such a way that no poem must be read in isolation, since the act of reading the book as a whole constructs a specific literary experience. Furthermore, this recently discovered scroll, plausibly arranged by the author himself,25 shows a notorious conscience of the particularities of combining both editorial and authorial needs.26 Therefore, thematic uniformity does not seem to pertain to the Hellenistic conception of transmitting or creating poetry, and it certainly does not describe the variegated and refined techniques displayed by our poet. Nevertheless, the collection of bucolic poetry arranged by Artemidorus of Tarsus in the 1st century BC27 may capture an initial stage of the homogenization process through which the thematic criteria was expanded by readers and critics and imposed upon the

 22 Ἄλλος ὁ Χῖος, ἐγὼ δὲ Θεόκριτος ὃς τάδ' ἔγραψα / εἷς ἀπὸ τῶν πολλῶν εἰμὶ Συρακοσίων, / υἱὸς Πραξαγόραο περικλειτᾶς τε Φιλίννας·/ Μοῦσαν δ' ὀθνείαν οὔτιν' ἐφελκυσάμαν. ‘Another is the Chian, I am Theocritus, who wrote these things, one among the many Syracusans, son of Praxagoras and of the noble Philina. I dragged no stranger’s muse’. Cf. Hunter (1999) 133–5. Gutzwiller finds throughout this epigram allusions to Idylls 16 and 17, both encomiastic poems to Hiero of Syracuse and Ptolemy II respectively, and to epigram 14 Gall. (14 Gow, Anth. Pal. 9.435), all of which would have integrated this early collection of Theocritean poetry and which would have been sufficiently well known, even representative, among the Syracusan´s production, although they are not considered bucolic. Gutzwiller (1996) 129–32, 135–7. 23 P. Mil. Vogl. 8.309. First edition by Bastianini–Gallazzi (2001). 24 Gutzwiller (2005) 1–3, n. 3. Ferrari (2004) too believes all the poems belong to Posidippus, but Lloyd–Jones (2003) is opposed to this view. 25 Although that is a widely discussed matter; Cf. Gutzwiller (2005) 3. 26 Stephens (2006) 91. 27 Artemidorus is also known for his treatise on Doric and his comic lexicon. Accordingly, Van Sickle supposes that his interest in bucolic poetry must have been due to the Doric and the mimetic (as in Platonic terminology, Cf. Resp. 3.392d) characteristics of the compositions considered as such. Then, this linguistic criterion must have had a considerable position in his generic conception, as well as in his editorial choices. Van Sickle (1976) 30–31.

116  Leonor Hernandez Oñate preservation and reception of Theocritus’ oeuvre, as shown by the following epigram ascribed to him (Anth. Pal. 9.205):28 Βουκολικαὶ Μοῖσαι σποράδες ποκά, νῦν δ' ἅμα πᾶσαι ἐντὶ μιᾶς μάνδρας, ἐντὶ μιᾶς ἀγέλας. The Bucolic muses, once scattered, are all now in one sheepfold, in one flock.

We must notice that the former epigram already uses the word bucolic as a tagline, a literary concept that has a close connection with herding vocabulary and must be illustrated by these means.29 Also, this sylloge must have included a large amount of poets whose work was commonly recognized as bucolic. Reed maintains that non–herding poems of Bion and Moschus, the other two Greek bucolic poets, must have been part of this book; the term bucolic, as a consequence, was used very loosely and applied to several poems for apparently inconsistent reasons.30 Anyhow, it is difficult to trace which of the poems by our author were included in it, although this matter would be of the greatest importance for the study of the development of the genre, since this could have been the edition that Virgil had access to when composing his Eclogues.31

 The reading of αἰπολικὸν (Theoc. 1.56) This whole process, in which thematic concepts and expectations began to cause a selection of some poems over others because of their subjects by providing them with documentary, and thus transmission–related, preference, and to narrow the spectrum of what the Theocritean text must be and look like, can be illustrated in many ways; anyhow, I will focus on the case found in Idyll 1, verse 56, which, as I will argue, is a revealing example of how generic expectations can modify or

 28 This poem was also transmitted in the prolegomena of codices KAEGT with the indication of Artemidorus’ authorship, and we may reasonably suppose that it functioned as an opening epigram for his collection of bucolic poetry. 29 Gutzwiller (1996) 123–4. 30 Reed (2006) 211; Gutzwiller (1996) 123. 31 It is a known fact that Artemidorus’s edition was the one that circulated in the Augustan period. Besides, Virgil must have used it since he seems to follow the model of Idyll 8, spurious but added to the corpus probably already in this collection, in Eclogue 4. Cf. Hunter (1999) 27; Bernsdorff (2006) 196.

Textual Tradition and Reception in Theocritus  117

have an impact on the establishment of the text and, therefore, in the way it is transmitted and read. Idyll 1, one of the most popular poems of the Theocritean oeuvre, has been specially studied for its programmatic and metapoetical characteristics.32 The ivy cup, κισσύβιον, that a skilful goatherd promises to give the shepherd–poet Thyrsis in exchange for his song, and which is engraved33 with three scenes of everyday life described through a detailed ekphrasis (vv. 27–56),34 is believed to function as an emblem of Theocritus’ poetry that embodies his conceptual aspirations.35 In this manner, David Halperin claims that: ‘It is the elaborate carving embellishing the surface of the goatherd’s simple wooden bowl which […] allows it (κισσύβιον) to serve as an emblem of Theocritus’ distinctive contribution […]’.36 After describing the images contained in this beautifully crafted vessel, the goatherd mentions an apparently external decoration and immediately coronates the description with a concise apposition, a rather mysterious but short label to fit this literary object (vv. 55–56):

 32 By programmatic I mean a text employed to establish the poetic principles fundamental for an author’s work; poetry, then, is used to talk about poetry itself and to establish a literary programme that will be the base for further creations. Cairns (1984) 89, describes this concept with these words: ‘A poem or part of a poem can be described as programmatic if it is consciously composed in a way that it contains a very high concentration of material typical of its literary form… even more if it contains statements about poetry and the art of writing poetry, either implicit or explicit […]’. Van Sickle (1975) 58–61, (1976) 22, believes that Theocritus summarizes in Idyll 7 the generic outlines begun in 1. For Halperin (1983) 185–6, Idyll 1 is considered programmatic since antiquity and the best introduction to Theocritus, reason for which it is always found first on the manuscripts. Cf. Rossi (2000), Gutzwiller (2006). 33 The space disposition of the elements within the cup has been analyzed in detail by Gallavotti (1966). 34 About the ekphrasis, see Chapter 3, ‘Ekphrasis and the Reception of Works of Art’, in the book by Burton (1995), and chapter 2, ‘Ekphrastic Epigrams’ in the study of Rossi (2001) Cf. Payne, (2001). This ekphrasis clearly follows two main homeric models: the ekphrasis of Achilles’ shield (Il. 18.478–608) and the episode in which Odisseus makes the Cyclops drink from an ivy cup, κισσύβιον (Od. 9.346). Nevertheless, the term κισσύβιον is also found in a brief hesiodic fragment (frag. 393 Maerkelbach–West): ὀλίγωι δ' ἥδετο κισσυβίωι. 35 Cairns (1984) 92–93, 102–8; Edquist (1975) 105–6. 36 Halperin (1983) 173.

118  Leonor Hernandez Oñate παντᾷ δ' ἀμφὶ δέπας περιπέπταται ὑγρὸς ἄκανθος, αἰπολικὸν θάημα· τέρας κέ τυ θυμὸν ἀτύξαι.37 The undulating acanthus extends all over around the vessel, wonder of the goatherd’s, a marvel that could astonish your soul.

But the goatherd continues and now tells us about the origin of the ivy cup. According to Gow’s version,38 a Calydnian boatman gave it to him in return for a goat and a big piece of cheese (vv. 57–58): τῶ μὲν ἐγὼ πορθμῆι Καλυδνίῳ αἶγά τ' ἔδωκα ὦνον καὶ τυρόεντα μέγαν λευκοῖο γάλακτος· In return for this I gave a Calydnian boatman a goat and a large cheese of white milk.

We now must consider what does αἰπολικὸν θάημα means. Why is this a description of the wooden bowl, an adequate finale of this episode? How does it fit with its context? If this poem, or at least this section of it, is programmatic, what does it say about Theocritus’ poetry? This is the text printed in Gow’s, Dover’s and Hunter’s versions, which rely on what is attested by the manuscripts, since all the codices show αἰπολικὸν τι θάημα / θαῦμα; αἰπολικὸν is also the reading printed in the first editions of Theocritus (Milan [1480], Aldine [1495], and Kallierges [1516]). In this case, it should be read as marvel, wonder of a goatherd, meaning rustic and coming from the goatherd. Gow translates an object for a goatherd to marvel at.39 Hunter prefers a marvel of the goatherd’s world,40 while Dover states that it means a spectacle characteristic of goatherds’. Theocritus is making the point that his idealized pastoral world has much to teach us in the craft of carving, as in poetry.41 Previously,

 37 This is Gow’s text (19522). All translations are mine. 38 Calydna is the name some islands on the Aegean. This lectio is also problematic, since we find Καλυδωνίῳ on the manuscripts, but the scholia mention Καλυδνίῳ. Calidonia is, on the other hand, in Aetolia, close to the Corinthian gulf. 39 Gow (19522) ii.13. 40 Hunter (1999) 84. 41 Dover (1971) 82.

Textual Tradition and Reception in Theocritus  119

Bernhard Koehler, in his ‘Notae et emendationes in Theocritum’ (1767), had explained it as meaning ποιμενικòν.42 The word θάημα,43 the Doric form of θέαμα, does not surprise with its meaning wonder, a spectacle that gives pleasure, 44 since what has been just described, the ivy cup, looks astonishing and is a true work of art; the meaning of this noun is then rounded off by the sentence that follows, which, as if it were a lexicography lemma or dictionary entry, tells us that it is a truly stunning object whose beauty could strike one’s sight. The word that seems puzzling and problematic is the adjective αἰπολικὸν. Why should the cup be proper of the goatherd’s world? It did not belong to him, and he certainly had nothing to do with is fabrication. In fact, he was just a middleman who received it from the Calydnian boatman and passed it on to the poet Thyrsis, whose song about the mysterious death of Daphnis is the core of the idyll (vv. 64–145). Besides, none of the images of the ekphrasis have any relationship with him, neither are they primarily pastoral.45 We could take into account the latter proposal by Dover and think of this asseveration as a metaphor of pastoralization and rusticity. Nonetheless, I do not believe this explanation is completely convincing and I argue that this adjective interpreted as of the goatherds’ does not make complete sense in reference to the cup.46 In fact, this is not the only possibility. Even though the manuscripts do not preserve any notice of other ancient variants, the lexicographical tradition, such as Hesychius, and the scholia offer αἰολικόν and ποικίλον47 as alternatives, and several scholars throughout history have taken into account these valuable testimonies. Already Camerarius suggested αἰολικόν in the margin of his 1530 edition, although he printed αἰπολικόν; so did Stephanus in his 1579 version.48 Moreover, Winterton printed αἰολικόν in 1667, followed by the influential editions of  42 Koheler (1767) 11–12 prefers this lectio but he does not reject entirely the possibility offered by Hesychius, αἰολικὸν: ‘Mihi, quidem, fateor, magis placet vulgate lectio αἰπολικὸν, i. ποιμενικòν… Neque ego sprevero vulgatam in lexicographo […] scripturam’. 43 Printed by most modern editors, like Gow (19522), whose text I follow here. 44 Liddell–Scott, s.v. θέαμα. 45 The images depicted on the cup are: a woman courted by two young men is amused by their suffering (vv. 32–38) an old fisherman struggles to pull a fishnet (vv. 39–44) and two foxes try to steal food on a vineyard while the boy who was supposed to watch over it happily constructs a cage for crickets (vv. 45–54). 46 αἰπολικὸν, though, does have textual support and precedents in Anth. Pal. 9.101 and Auson. Epist. 14.33, which could be references to this passage. Cf. Gow (19522) II.13. 47 In fact, Wilamowitz thought that ποικίλον in the scholium 56a is an interpolation, but it is authentic in 56e: αἰολικόν. αἰόλον τι καὶ ποικίλον θέαμα. Cf. Wendel (1914) 51–52. 48 This is the year of the edition I had access to, not to his 1566 work. Nevertheless, just as Camerarius, he prints απἰολικόν and suggests αἰολικόν as one of his Emendationes in the prologue.

120  Leonor Hernandez Oñate Valckenaer (1773, 1781), Kiessling (1826), (Ziegler [1844]), Ahrens (1855 [editio maior], 18612) and Fritzsche (1868), who read αἰολίχον, among others; this trend continued in the early 20th century, as in Cholmeley’s and Edmonds’ work (both from 1919), despite the fact that the most important versions in this period, like Gow’s or Legrand’s (19463) decide to print αἰπολικòν. Further ahead, Carlo Gallavotti, preceded by Ahrens (‘αἰολίχoς deminutivum doricum adiectivi αἰόλος, ut 4, 20’),49 chose αἰολίχoν, like Fritzsche, for his edition (1993). If we take these suggestive variants into consideration, we need to determine their meaning and how they fit into the general semantics of the Theocritean poem. Hesychius states that the correct reading was: αἰολικὸν θέαμα. ἀντὶ τοῦ Αἰτωλικόν παρὰ Θεοκρίτῳ, ἡ γὰρ Καλυδὼν Αἰολὶς ἐκαλεῖτο.50 Wonder from Aeolis; instead of Aetolic in Theocritus, for Calydon was called Aeolis.

Therefore, this could mean wonder coming from Eolis, the district where Calydon was located, but we would have to read in v. 57 πορθμῆι Καλυδωνίῳ instead of Καλυδνίῳ.51 Another possibility is to follow Ahrens’ interpretation, where αἰολίχoν should be read as a Doric diminutive of the adjective αἰόλος,52 little variegated marvel.53 Given the difficulties of encompassing αἰολικὸν with the boatman’s provenance, which is yet another issue, we might try to find a different explanation for this variant, instead of dismissing it and deciding that αἰπολικὸν is a better lectio in the absence of a more suitable term, just as the whole manuscript medieval tradition has done. I believe, then, that it might designate a source, yet one that has no relationship with the boatman’s origin or with a specific place designated by Theocritus; rather, I think it outlines a literary affiliation, a statement in which the poet claims his debt to a certain poetic tradition, considering, also, that this

 49 Ahrens (1861) III. 50 Hesych. s.v. αἰολικὸν θέαμα. Cf. Wendel (1914) 52. Also, Edmonds (1919) 13, but he translates ‘a ferryman from Calymnus’, and then mentions that Calymnus is an island near Cos; Kiessling (1826) 12, also believes this adjective refers to the boatman’s origin. 51 This is Gow’s solution (19522) to the αἰολικὸν that the scholia mention. 52 Shifty, changeful. It would work here as a synonym of ποικίλος. About the variant αἰολιχὸν, cf. Wilamowitz (1906) 36–38. 53 This interpretation is supported by Cholmeley (19192) 192, and Fritzsche (1868) 33, although the latter maintains that the adjective αἰολίχoν does not refer to the whole cup, but only to the ὑγρὸς ἄκανθος in v. 55. Theocritus uses the adjective αἰόλος as a compound word in Theoc. 17.19: Πέρσαισι βαρὺς θεὸς αἰολομίτρας (Gow’s text).

Textual Tradition and Reception in Theocritus  121

text could be programmatic and that this ivy cup probably represents a literary metapoetical object. It is a possibility, hence, that αἰολικὸν, or αἰολίχον, refers to Aeolia, not to Aeolis, and in such a way it would be a sort of synonym of ποικιλίον, like Ahrens purposed, although it may not be related directly with αἰόλος, but to the Aeolian music and poetry, best represented by Sappho and Alcaeus, whose influence on Theocritus is clear.54 Interpreting ‘Aeolian’ would not be unsuspected or unprecedented within the history of Theocritean scholarship; Winterton, for instance, translated his reading αἰολικόν τι θάημα into Latin as aeolicum spectaculum.55 Moreover, the dictionary LSJ, s.v. Αἰολεύς, έως, ὁ, admits that the adjective αἰολικός, ή, όν may mean the aeolians, and uses as an example precisely these Theocritean verses. This makes plausible that the Syracusan was referring to Aeolia and consequently to its melic poets. The wooden vessel, the emblem of his poetry, would have come from Aeolia, and would have been fabricated and molded by their traditions and the prevailing poetic techniques of the region, eventually arriving at Thyrsis’ hands. The scholia that support these hypotheses establish that this αἰολικὸν is ἀπατητικόν, ‘treacherous’ or ‘fallacious’, since this is what ‘to eolisize’ means (ταῦτα γὰρ δύναται τὸ Αἰολιζιεν), because people from Aeolia usually lie a lot (oἱ γὰρ Αἰολεῖς ἐπὶ ἀπάτῃ διαβάλλονται).56 Or alternatively it means ‘variegated’ (αἰόλον τι καὶ ποικίλον θέαμα), from which the previous idea seems to derive. Therefore, these testimonies record that this term can be seen as an adjective of place referring to the Aeolians to some extent, although it is supposed to designate antonomastically the Aeolic capability to deceive, something very different form a literary debt. An important testimony that may provide evidence for my literary interpretation of this passage is a grammatical fragment from the poet and scholar Philitas of Cos (fr. 7 Dettori), a gloss transmitted by Athenaeus of Naucratis (11 98.12):

 54 The most evident examples are the poems known as the Aeolic Idylls, 28–30 (Gow); the parallels between Idyll 2 and Sappho are remarkable. Actually, the aeolic influence on Theocritus can be found largely throughout his oeuvre. Cf. Acosta Hughes (2010), Vox (1991), Fassino– Prauscello (2001). 55 Winterton (1667) 118. 56 Wendel (1914) 51.

122  Leonor Hernandez Oñate σκάλλιον‧ κυλίκιον μικρόν, ᾧ σπένδουσιν Αἰολεῖς, ὡς Φιλήτας φησὶν ἐν Ἀτάκτοις. Schallion: a small cup, from which Aeolians offer libations, as Philetas says in the Irregular words.

Athenaeus is talking about an obscure word, an object called σκάλλιον, which is a sort of kýlix, a small cup in which the Aeolians drink, according to the work by Philitas, Ἄτακτοι γλῶσσαι, Indeterminate or irregular words. Hence, Philitas, who is unanimously thought to have had an influence on Theocritus and may be related to him from a poetic point of view,57 would have written about a small Aeolian vessel; this term, furthermore, presents an Aeolic diminutive, just as ἰχὸν is a Doric form of diminutive in αἰολιχόν. As Dettori himself observes, this cup must be a rustic object and may be connected to our text: ‘[…] saremmo rimandati ad un oggetto rustico, come altri glossati da Filita, qualcosa di analogo al κισσύβιον teocriteo’.58 Athenaeus too identifies the κισσύβιον with Aeolia (11.53.10): Κλείταρχος δέ φησιν Αἰολεῖς τὸν σκύφον κισσύβιον καλεῖν· Cleitarchus says the aeolians call the rustic cup kissýbion.

Given the possible literary relationship between Theocritus and Philitas, it seems likely that Theocritus could have made an allusion to some object belonging to Philitean poetry or at least to a typical kind of Aeolian cup. It is possible, thus, that Theocritus in this passage admitted his debt to the melic oeuvre of the lesbian poets, emphasizing also the well–known and appreciated ποικίλία, the diversity of his poetry; maybe, as well, he could have been paying homage, as he usually does, to one of his most relevant models, Philitas. Nevertheless, these are only suggestions, since the profound difficulties posed by the text make it very complicated to assert one definite interpretation. But the conservation of variants for verse 56 in the lexicographical documents suggests that these were all valid versions that circulated in antiquity, and that this was a passage with large textual possibilities, even if they are absent it the manuscripts.

 57 Some sources claim that Theocritus was Philitas’s disciple, and that both met in Cos. (Schol. A a1–A a10), Wendel (1914) 1. This information must have originated from an interpretation of Idyll 7.40, where Simichidas says he cannot compare his poetry to Philitas’ or to Asclepiades’, and does not seem reliable. But the influence of Philitas is evident in many aspects; recently, much scholarly attention has been given to the presence of his work in Theocritus’s literary vision and in the construction of some of his most celebrated idylls, like 7. Cf. Montes Cala (2013), Sbardella (2000) 14, 34. 58 Dettori (2000) 83–84.

Textual Tradition and Reception in Theocritus  123

As Dover warns us: There was an ancient variant αἰολικὸν, Aeolian, but if Theocritus wrote that we do not know what he had in mind.59 To conclude, I do not intend to provide a solution for this matter or to establish a lectio for 1.56, although I do maintain that there are arguments to support αἰολιχόν, Ahren’s and Gallavotti’s choice. What is actually relevant about the former discussion is the demonstration of the variability of readings for this passage and that they may actually be feasible. There were no textual, linguistic, metrical, or any philological reasons to dismiss them, but they are nowhere to be found in the manuscript tradition, restricting this way much of the richness of the Theocritean text and many of the ways to approach it. This phenomenon also conditioned a certain posture on behalf of the new reader and gave the impression that the pastoral elements were preponderant in it. The generic expectations, then, did affect the way in which the codices preserved Idyll 1. One must consider, moreover, what impact the form of transmission of poetry in the Hellenistic age had on its conception and creation, since the pre–eminence of writing, the usage of collecting and editing both ancient and modern authors, and the eminent philological activities of the poetae docti, among other phenomena, affected the artistic production and its reception. The implications of gathering literary works, especially poetry, that were not originally made to circulate as a group, as was commonly made in the Hellenistic age, must be taken into account for the analysis of any poetic text. For example, when we collect in one book the Odes of Pindar, an editor creates a specific literary experience, as they are, to a certain extent, artificially making the book independent; and oral poems are read in continuity, and thus condition the reader to react according to the model constructed by the editor.60 This anthology habit clearly also had effects on the way Theocritus was received and read, despite him being a poet of literacy.61 Theocritus composed poems that have a remarkable internal coherence and are related to one another, they grow richer from their confrontation than from their segregation into thematic and generic categories. Anyhow, scholars, since antiquity, have re–arranged the text and disposed it according to their own interests. Thus, Theocritus’ oeuvre became public, so to speak, in the hands of his readers and critics, who poured in it their own aspirations, and adapted it to respond to these exigencies; this way it had a second life, almost independent from its author, under the passionate and creative understanding of his admirers.  59 Dover (1971) 82. 60 Hutchinson (2008) 2. 61 Based mostly on writing. Cf. Thomas (1992).

Federica Benuzzi

Eratosthenes’ studia Aristophanica The case of schol. Ar. Nub. 967aα–β, bα–β Holwerda Eratosthenes’ philological studies are preserved exclusively through indirect tradition, mostly in the form of quotations in scholiastic corpora to several authors and in the main lexicographical works by the imperial, Late–Antique and Byzantine era.1 Overshadowed by his achievements in geography, poetry and mathematics, Eratosthenes’ scholarly activity (and, especially, his work Περὶ τῆς ἀρχαίας κωμῳδίας)2 regained the interest of modern scholarship only at the end of the 20th century – long after Bernhardy’s (1822) and Strecker’s (1884) collections3 – thanks to three publications by Renzo Tosi (1994, 1998a, 1998b), which highlighted elements of continuity and innovation in the Eratosthenian scholarship, with regard to the Peripatetic tradition and the first phases of Alexandrian philology.4 The scholiastic material accompanying Ar. Nub. 967 preserves traces of an interpretative controversy that opposed Eratosthenes (fr. 101 Strecker) to the Peripatetic Chamaeleon.5 The debate stemmed from Aristophanes’ quotation, in the

 1 Papyrological evidence is limited to P. Oxy. 35.2737, which represents the most ancient extant witness of an extensive quotation from Eratosthenes’ scholarship (see Bagordo [1998] 133–4, Biles [1999], Trojahn [2002] 62–68 and Montana [2012]), PSI 2.144 (three excerpta of uncertain origin, see Wouters [1969] and Bagordo [1998] 134–5) and P. Turner 39, a list of book titles including an otherwise unknown Eratosthenian work on the Iliad (see Otranto [2000] 72–77; Geus [2002] 291 and 302–3). P. Oxy. 13.1611 should most probably be included in this group as well, although Eratosthenes’ name is only conjecturally restored. 2 The full title is preserved by Diog. Laert. 7.5, Gal. Gloss. 19 p. 65 Kühn (see also Perilli [2017] 144–5), Harp. δ 13 and μ 25 Κ (= 86.1–7 and 203.8–11 Dindorf). 3 In the absence of a modern critical edition (which is currently being prepared by Maria Broggiato, Università degli Studi di Roma ‘La Sapienza’), the numbering of Eratosthenes’ philological fragments continues to be the one given by Strecker (1884). Di Tullio (1915) does not provide any significant improvement compared to Strecker’s work. 4 These aspects are absent in Pfeiffer’s still authoritative account of Eratosthenes’ intellectual persona ([1968] 152–70). 5 For an overview of his exegetical activity, see Podlecki (1969) 120–4; Bagordo (1998) 26–28; Schorn (2012). Arrighetti (1987) 141–59 provides a detailed analysis of the so–called Chamaileonsmethode. Observations on specific features of his exegesis can also be found passim in Bouchard (2016). https://doi.org/10.1515/9783110632590-010

126  Federica Benuzzi Clouds, of the incipit of two hexametric hymns,6 mentioned by Strong Argument as evidence of the superiority of ancient παίδευσις over contemporary education:7 εἶτ’ αὖ προμαθεῖν ᾆσμ’ ἐδίδασκεν, τὼ μηρὼ μὴ ξυνέχοντας, ἢ ‘Παλλάδα περσέπολιν δεινάν’ ἢ ‘τηλέπορόν τι βόαμα’, ἐντειναμένους τὴν ἁρμονίαν, ἣν οἱ πατέρες παρέδωκαν. (Nub. 966–8) Then again he (scil. the music–master) would teach them to learn a song by heart, without keeping their thighs together. Either ‘Pallas, the terrible destroyer of cities’ or ‘a far–reaching cry’, pitching the tune their fathers handed down.

The present discussion will be based on two premises: (1) The Aristophanic context clearly excludes the possibility of any interest of the playwright in the authorship of the two hymns,8 and rather relies on their authority as renowned traditional songs linked to the education of Athenian youth. Evidently, the simple allusion to the first words of these texts was enough for Aristophanes to trigger the stereotype of ancient aristocratic παίδευσις in his audience. Nevertheless, (2) the identification of the two texts became a highly debated ζήτημα already among the Peripatetics – as the involvement of Chamaeleon shows – and the querelle grew more and more confused with a gradual overlap of subsequent phases of exegesis of the Aristophanic line through the centuries. As a consequence of (1) and (2), the present paper will focus exclusively on the different stages of the ancient philological interpretation of Nub. 967 and must not be regarded as an attempt at devising any sort of alleged hypotext in the Aristophanic text. This is the text of the scholia as preserved by EMNp and RV:9 ἢ EM Παλλάδα EMNp περσέπτολιν ἢ τηλέπορόν τι βόαμα E: †ἀμφοτέρων†. τὸ μὲν οὖν πρότερον Λαμπροκλέους (PMG 735) εἶναί φασι τοῦ ἀθλητοῦ, Μίδωνος υἱοῦ· ἔχει δὲ οὕτως ‘Παλλάδα περσέπτολιν κλῄζω EM πολεμοδόκον ἁγνάν, παῖδα Διὸς μεγάλου δαμάσιππον’. τὸ

 6 As far as the use of dactylic hexameters in comedy is concerned, see White (1912) 149–54; Dale (1968) 28; West (1982) 98; Parker (1997) 48–55; Quaglia (2007); Bellocchi (2009); Bakola (2010), 163 n. 138: ‘in Nub. 967 and V. 652, Aristophanes accommodates dactyls in lyric anapaests, and in Eq. 529 and Lys. 520 dactyls in recitative anapaests’. 7 See Anderson (1995) 72. 8 See Pòrtulas (2012) 231. 9 The printed text is that of the manuscripts and not the one proposed by Holwerda. Punctuation follows Arrighetti’s autoptic examination of E (see Arrighetti [1987] 231: ‘l’interpunzione del codice E [f. 71v] non appare riprodotta fedelmente dall’Holwerda, né nell’ed. definitiva né nel lavoro preparatorio pubblicato in Mnemos. 1952’).

Eratosthenes’ studia Aristophanica  127

δὲ ‘τηλέπορόν τι βόαμα’ μὴ εὑρίσκεσθαι, ὅτου ποτ’ ἐστίν· ἐν γὰρ ἀποσπάσματι εὑρεῖν Ἀριστοφάνη ἐν τῇ βιβλιοθήκῃ. τινὲς δέ φασιν αὐτὸ Κυδίδου Ἑρμιονέως (= Κυδίας PMG 714– 5?) ENp ‘τηλέπορόν τι βόαμα λύρας’. ἄλλως. E οὕτως Ἐρατοσθένης· Φρύνιχος (fr. 78 K.– A.)10 δὲ αὐτοῦ τούτου τοῦ ᾄσματος μέμνηται ὡς Λαμπροκλέους ὄντος τοῦ Μίδωνος υἱοῦ ἢ μαθητοῦ. ἔχει δὲ οὕτως ‘Παλλάδα περσέπτολιν δεινὰν θεὸν ἐγρεκύδοιμον’. Χαμαιλέων (fr. 28 Steffen, 29a–b Wehrli, 28a–b Giordano, 31B Martano) δ’ ἀπορεῖ. πότερον· ‘κλῄζω πολεμοδόκον ἁγνάν, παῖδα Διὸς μεγάλου δαμάσιππον’ ENp καὶ κατὰ Λαμπροκλέα ὑποτίθησι κατὰ λέξιν. E τὸ δὲ ‘τηλέπορόν τι βόαμα’ Κυδίδου τοῦ Ἑρμιονέως κιθαρῳδοῦ ἀπό τινος τῶν ᾀσμάτων ‘τηλέπορόν τι βόαμα λύρας’. ENp Either ‘Pallas, destroyer of cities’ or ‘a far–reaching cry’: †of both†. They say that the first is by Lamprocles the aulete,11 son of Midon. It goes like this: ‘Pallas, the destroyer of cities I invoke, war–sustaining, venerable, daughter of great Zeus, horse–taming’. It is not known by whom the hymn ‘a far–reaching cry’ is. Indeed, Aristophanes (scil. of Byzantium) found it in a fragment in the library (scil. of Alexandria). Some say it is by Cydidas of Hermion: ‘a far–reaching cry of a lyre’. Otherwise.12 Thus, says Eratosthenes: Phrynichus quotes this very hymn, ascribing it to Lamprocles, son or pupil of Midon. It goes like this: ‘Pallas, the terrible destroyer of cities, strife–stirring goddess’. Chamaeleon is at a loss. Whether: ‘I invoke, war–sustaining, venerable, daughter of great Zeus, horse–taming’ and reports the version of Lamprocles word for word. ‘A far–reaching cry’ is (a quotation) from one of the hymns by Cydidas of Hermion, the citharode, ‘a far–reaching cry of a lyre’. ἢ Παλλάδα· ἀρχὴ ᾄσματος Φρυνίχου (fort. Λαμπροκλέους), ὡς Ἐρατοσθένης φησίν. Φρύνιχος δὲ αὐτοῦ τούτου τοῦ ᾄσματος μνημονεύει ὡς Λαμπροκλέους ὄντος· ‘Παλλάδα περσέπτολιν κληΐζω πολεμαδόκον ἁγνάν, παῖδα Διὸς μεγάλου’. τὸ δὲ τηλέπορόν τι βόαμα· καὶ τοῦτο μέλους ἀρχή. φασὶ δὲ μὴ εὑρίσκεσθαι, ὅτου ποτέ ἐστιν· ἐν γὰρ ἀποσπάσματι ἐν τῇ βιβλιοθήκῃ εὑρεῖν Ἀριστοφάνη. τινὲς δέ φασι Κυδίδου τινὸς Ἑρμιονέως ‘τηλέπορόν τι βόαμα λύρας’ RV Either ‘Pallas’: beginning of a hymn by Phrynichus, as Eratosthenes says. Phrynichus quotes this very hymn ascribing it to Lamprocles: ‘Pallas, the destroyer of cities I invoke, war–sustaining, venerable, daughter of great Zeus’. ‘A far–reaching cry’ is also the beginning of a song. They say that it is not known by whom it is. Indeed, Aristophanes (scil. of Byzantium) found it in a passage in the library (scil. of Alexandria). Some say it is by Cydidas of Hermion: ‘a far–reaching cry of a lyre’.

 10 No relevant contribution to the present discussion is provided by Stama (2014) 349–54. 11 Wallace (2003) 74–75 proposes the emendation of τοῦ ἀθλητοῦ (also in schol. Tzetz. Ar. Nub. 966a οἷον τὸ Λαμπροκλέους τοῦ ἀθλητοῦ, υἱοῦ Μίδωνος ‘Παλλάδα περσέπολιν κλῄζω κτλ) into a more probable τοῦ αὐλητοῦ, in the light of the several occurrences that link the name of the poet to the aulos. 12 The translation of the adverb ἄλλως fails to convey its real function as linguistic mark of the compiling of different Late Antique commentaries, from which the scholia originated, see Wilson (1967); Zuntz (19752) 117–9; Maehler (2000); Dickey (2007) 108–9; Montana (2011).

128  Federica Benuzzi The distribution of the exegetic material is quite clear. One can identify three annotations (one in RV,13 two in ENp,14 separated by ἄλλως in E) that can be reduced to two main redactions. Indeed, the first annotation in ENp is clearly a slightly abridged form of the text of RV (the main trace of epitomization being the absence of Eratosthenes’ name), while the second is a different version. The content of the two redactions can be summarized as follows (from now on α will stand for the second annotation of ENp [i.e. schol. Ar. Ra. 967bα Holwerda] and β for the one in RV and for its abridged version in ENp [i.e. schol. Ar. Ra. 967aβ, bβ and aα Holwerda]).

1.

2.

Redaction α: mention of Eratosthenes’ position on the interpretative issue of the Παλλάδα περσέπολιν (he ascribed it to Lamprocles,15 Midon’s son or pupil,16 on the ground of the evidence provided by Phrynichus), quotation of the beginning of the hymn, Παλλάδα περσέπτολιν17 δεινὰν θεὸν ἐγρεκύδοιμον,

 13 The exegetic material of RV is split in two parts (scholl. Ar. Ra. 967bβ and aβ) in Holwerda’s edition. 14 Corresponding respectively, in Holwerda’s edition, to scholl. Ar. Ra. 967aα and bα. 15 See PMG 735–6. Not much is known about Lamprocles, but what can be inferred by the textual evidence is that he was a relevant figure in the Athenian music scene in the early 5th century. Indeed, Ath. 11.491c refers to him as Λαμπροκλῆς ὁ διθυραμβοποιός, while [Plut.] De mus. 1136d– e mentions his investigation of the Mixolydian harmonia (ἐν δὲ τοῖς ἱστορικοῖς † τοῖς ἁρμονικοῖς Πυθοκλείδην φησὶ [fort. ἐν δὲ τοῖς ἱστορικοῖς οἱ ἁρμονικοί Πυθοκλείδην φασὶ, see Einarson–Lacy 1967, 386] τὸν αὐλητὴν εὑρετὴν αὐτῆς γεγονέναι, αὖθις δὲ Λαμπροκλέα τὸν Ἀθηναῖον, συνιδόντα ὅτι οὐκ ἐνταῦθα ἔχει τὴν διάζευξιν ὅπου σχεδὸν ἅπαντες ᾤοντο, ἀλλ’ ἐπὶ τὸ ὀξύ, τοιοῦτον αὐτῆς ἀπεργάσασθαι τὸ σχῆμα οἷον τὸ ἀπὸ παραμέσης ἐπὶ ὑπάτην ὑπατῶν). Moreover, a scholium to [Pl.] Alc. I.118c.6 states that he was Damon’s teacher (Πυθοκλείδης μουσικὸς ἦν, τῆς σεμνῆς μουσικῆς διδάσκαλος, καὶ Πυθαγόρειος, οὗ μαθητὴς Ἀγαθοκλῆς, οὗ Λαμπροκλῆς, οὗ Δάμων). With regard to the possible confusion – in Ath. 1.20e, on the identity of Sophocles’ teacher – between Lamprus ὁ μουσικός (also mentioned by Phrynichus [fr. 74 K.–A.]) and Lamprocles, see Abert (1924); Campbell (1991) 315; Robbins (1999a); Wallace (2003) 74 n. 6. 16 The only comparable occurrence of such a doubt is schol. Aeschin. 3.18 Εὐμολπίδας: Εὐμόλπου ἀπογόνους. ἐγένετο δὲ Εὔμολπος Μουσαίου υἱὸς ἢ μαθητής, τοῦ τὰ μυστήρια εὑρηκότος. Eratosthenes’ source for the relationship between Lamprocles and Midon is unknown. However, it is not unlikely that he derived the piece of information directly from Phrynichus’ context. 17 The manuscript tradition of all sources that quote or discuss the hymn to Pallas shows an alternation between περσέπολιν and περσέπτολιν. The latter (for which see Aesch. Pers. 65 and Callim. 5. 43) is non-metrical in the context of the incipit of the ode to Pallas, and its presence is probably influenced by the pervasive use of πτόλις–compounds and πτόλις–derivatives in

Eratosthenes’ studia Aristophanica  129

3. obscure reference to an aporia of Chamaeleon (Χαμαιλέων δ’ ἀπορεῖ), 4. second quotation (κλῄζω πολεμαδόκον ἁγνάν, παῖδα Διὸς μεγάλου δαμάσιππον), introduced by a πότερον that finds no corresponding ἤ in the following text, ‘he quotes word by word in accordance with Lamprocles’, without a clear subject (the nearest is Chamaeleon, but the sequence ἀπορεῖ– καί is most probably corrupt), 5. ascription of the τηλέπορόν τι βόαμα to Cydidas of Hermion (maybe identifiable with the Κυδίας mentioned by Plato and Plutarch [= PMG 714–5], see Robbins 1999b), with a quotation of the beginning of the hymn. Redaction β: 1. initial ascription of the ode Παλλάδα περσέπολιν to Phrynichus by Eratosthenes, 2. attribution to Lamprocles by Phrynichus (Φρύνιχος δὲ αὐτοῦ τούτου τοῦ ᾄσματος μνημονεύει ὡς Λαμπροκλέους ὄντος), 3. quotation of the hymn in a ‘hybrid’ form between the two quotations preserved in α (Παλλάδα περσέπτολιν κληΐζω πολεμαδόκον ἁγνάν, παῖδα Διὸς μεγάλου), 4. information about the finding of a text beginning with τηλέπορόν τι βόαμα by Aristophanes of Byzantium in the library of Alexandria, 5. mention of the ascription of the hymn τηλέπορόν τι βόαμα to Cydidas of Hermion by some commentators (τινὲς δέ φασι), with a quotation of the beginning of the hymn (τηλέπορόν τι βόαμα λύρας). The two redactions differ in several respects. Indeed, α seems to give quite a different text of the hymn to Pallas in comparison to the one given by β. What is more, β ignores Chamaeleon’s stance in the debate on the Παλλάδα περσέπολιν, a stance which α at least alludes to. On the other hand, β is the only witness of Aristophanes’ impasse with regard to the τηλεπορόν τι βόαμα, a text that he found ἐν ἀποσπάσματι18 in the library and apparently would not (or could not)  Homer (see e.g. Il. 1.164 πτολίεθρον, 2.278 πτολίπορθος, 6.305 ἐρυσίπτολιν), in other texts using epic language (see e.g. Hes. Th. 936 πτολιπόρθῳ, h.Hom. 11.2 ἐρυσίπτολιν, Pind. Ol. 8.35 πτολιπόρθοις, Ap. Rhod. 1.186 πτολίεθρον), and in tragedy (see e.g. Eur. IA 1476 ἑλέπτολιν). With regard to the orthography of this adjective, the text here printed follows for every occurrence the respective editor’s choice. 18 The expression usually indicates a passage preserved in a longer text or in a set of texts (see Diog. Laert. 9.113–4 εἰκῆ τε αὐτῷ ἔκειτο τὰ ποιήματα, ἐνίοτε ἡμίβρωτα· ὥστε καὶ Ζωπύρῳ τῷ ῥήτορι ἀναγινώσκοντά τι ἐπιτυλίττειν καὶ κατὰ τὸ ἐπελθὸν διεξιέναι· ἐλθόντα τ’ ἐφ’ ἡμισείας, οὕτως εὑρεῖν τὸ ἀπόσπασμα τέως ἀγνοοῦντα and Cornutus ND 26,18 ἔοικε γὰρ ὁ ποιητὴς μυθοῦ

130  Federica Benuzzi link to Cydidas (indeed, the ascription to the citharode of Hermion is given as a possibility, while α presents it as a fact).19 As the aim of this paper is to analyse Eratosthenes’ approach to the Aristophanic text, the focus will primarily be on the first hymn quoted in Nub. 967, the Παλλάδα περσέπολιν. Indeed, while Eratosthenes’ involvement in the authorship debate on the hymn to Pallas is certain, there is no clear textual evidence that links him to the ascription of the τηλέπορόν τι βόαμα, although it is very unlikely that he completely overlooked the issue. In particular, if the section on Aristophanes of Byzantium logically derives from a later source, the bit attributing the song to Cydidas/Cydias might theoretically derive from Eratosthenes, or even from an older source. As far as the hymn to Pallas is concerned, any attempt to reconstruct the authorship controversy (and Eratosthenes’ position in it) requires a more detailed analysis of the scholiastic Wortlaut, especially because of the obscurity of Chamaeleon’s stance in the debate. Indeed, the inconsistent wording of α generated the assumption20 that there were actually two hymns beginning with Παλλάδα περσέπολιν (one continuing with δεινὰν θεὸν ἐγρεκύδοιμον, the other with κλῄζω πολεμαδόκον) and that Chamaeleon’s aporia regarded the identification of the text quoted by Aristophanes. The first and principal supporter of this approach is Holwerda, whose conjectural interventions on α and β result from a number of premises: 21

 παλαιοῦ παραφέρειν τοῦτο ἀπόσπασμα, καθ’ ὃν ὁ Ζεὺς ἐμυθεύετο κτλ). The description given by the scholium could refer to a roll containing ἀντιλεγόμενα, texts of doubtful origin and/or debated authenticity (see e.g. Vit.Eur. 3.15–17 Schwartz τὰ πάντα δὲ ἦν αὐτοῦ [scil. Εὐριπίδου] δράματα ϟβ, σῴζεται δὲ αὐτοῦ δράματα ξζ καὶ γ πρὸς τούτοις τὰ ἀντιλεγόμενα, σατυρικὰ δὲ η, ἀντιλέγεται δὲ καὶ τούτων τὸ α). On the identification of bookrolls in the Alexandrian library see e.g. Blum (1977) 15–17; Canfora (1993) 16–28; Montana (2015) 84, 90–91. 19 Although it could naturally result from epitomization in β, the opposition between the clear statement of authorship of α and the doubtful tone of β might reflect subsequent interpretative stages: an older one relying on an extant authorial text by Cydidas/Cydias and a more recent one, represented by the librarian Aristophanes and his (by that time anonymous) ἀπόσπασμα. Other texts dealing with the τηλέπορόν τι βόαμα are schol. anon. rec. Ar. Nub. 967b, schol. Tzetz. Ar. Nub. 967b–c–e. 20 Wilamowitz (1900) 84–85; Grenfell–Hunt (1919) 146; Dover (1968) 215; Wehrli (1969) 81; Arrighetti (1987) 231; Cingano (1990) 211; Giordano (19902) 156–8. 21 The text of the scholia printed by Holwerda (=schol. Ar. Nub. 967b) reads as follows: α) οὕτως Ἐρατοσθένης. Φρύνιχος δὲ αὐτοῦ τούτου τοῦ ᾄσματος μέμνηται ὡς Λαμπροκλέους ὄντος τοῦ Μίδωνος υἱοῦ ἢ μαθητοῦ. ἔχει δὲ οὕτως Παλλάδα περσέπτολιν δεινὰν θεὸν ἐγρεκύδοιμον. Χαμαιλέων δ’ ἀπορεῖ, πότερον κλῄζω πολεμοδόκον ἁγνάν, παῖδα Διὸς μεγάλου δαμάσιππον ἢ κατὰ Λαμπροκλέα ὑποτίθησι κατὰ λέξιν. β) ἢ Παλλάδα· ἀρχὴ ᾄσματος Στησιχόρου, ὡς

Eratosthenes’ studia Aristophanica  131

1.

2.

3.

firstly, Chamaeleon’s doubt was about the existence of two independent hymns to Pallas (Παλλάδα περσέπολιν δεινὰν θεὸν ἐγρεκύδοιμον by Lamprocles, Παλλάδα περσέπολιν κλῄζω πολεμαδόκον κτλ by Stesichorus)22, hence the alleged lacunae after πότερον and after κατὰ Λαμπροκλέα, and the emendation of καί into ἤ in α; secondly, despite what α states, Eratosthenes ascribed the hymn to Stesichorus (both the supposed lacuna before οὕτως Ἐρατοσθένης in α and the replacement of Φρυνίχου with Στησιχόρου in β – already proposed by van Leeuwen [1898], 154 – result from this hypothesis); finally, the Phrynichus quoted by the scholia is not the comic poet, but a grammarian from the Imperial age,23 and it was the latter who ascribed the Παλλάδα περσέπολιν to Lamprocles.

In fact, two ancient witnesses state that the authorship dispute on the hymn to Pallas involved both Lamprocles and Stesichorus24 as possible authors. The first text is schol. Ael. Ar. Or. 46.162.7–8,25 that relies on the authoritative source of

 Ἐρατοσθένης φησίν. Φρύνιχος δὲ αὐτοῦ τούτου τοῦ ᾄσματος μνημονεύει ὡς Λαμπροκλέους ὄντος· Παλλάδα περσέπτολιν κληΐζω πολεμαδόκον ἁγνάν, παῖδα Διὸς μεγάλου. 22 This clearly contradicts all the witnesses that ascribe Παλλάδα περσέπολιν κλῄζω πολεμαδόκον κτλ to Lamprocles, see Sgobbi (2006) 290. 23 Although initially identifying him with the comic poet (Holwerda [1952] 229), while editing the scholia to the Clouds Holwerda eventually opted for the grammarian ‘cum hic inter Eratosthenem et Chamaeleontem locum obtineat’ (id. [1977] 186). The grammarian Phrynichus surely worked on Aristophanic comedy, but he directed his attention towards the linguistic aspects (due to his atticist interests, see e.g. Strout–French [1941] 921), rather than towards content–related ones (reviewer’s observation). What is more, Holwerda’s choice is untenable as it disregards the incontrovertible wording of P. Oxy. 13.1611 where the main argument against Chamaeleon’s interpretation is that he should have been aware of Phrynichus quoting Lamprocles, see Sgobbi (2006) 291. The identification with the comic playwright can be considered sure, for the idea of an explicit and extensive quotation from a lyric poet in tragedy seems very hard to justify, see Sgobbi (2006) 292 n. 40. 24 See Davies–Finglass (2014) 595–6. 25 In this portion of the speech, the author quotes several times Strong Argument’s description of the ancient παίδευσις (Nub. 961–5, 967, 968–9, 971–2, 984–5), while imagining Miltiades’ possible answers to some hypothetical questions regarding the Athenian education. On the quotations of comic poets in Aelius Aristides, see Vix (2016) 375–92. Another reuse of the hymn to Pallas in rhetorical context is Dio Chrys. 13.19 (καὶ νῦν ἐπὶ ταύταις ταῖς ἐλπίσιν οἰκεῖτε τὴν πόλιν καὶ τοὺς υἱέας παρασκευάζετε ὡς δυνατοὺς ἐσομένους χρῆσθαι τοῖς τε αὑτῶν καὶ τοῖς δημοσίοις πράγμασιν, οἳ ἂν ἱκανῶς κιθαρίσωσι ‘Παλλάδα περσέπολιν δεινὰν’ ἢ τῷ ποδὶ βῶσι πρὸς τὴν λύραν). Unfortunately, Arethas’ scholia to Dio (see Sonny [1896] 83–130) do not deal with this

132  Federica Benuzzi Rufus26 and Dionysius,27 but shows some clear signs of confusion too, as it seems to consider Phrynichus a possible author of the hymn: ἐξηγεῖται δὲ αὐτός, λέγων τὸ ‘μὴ πρὸς ἡδονὴν, ἀλλὰ ἢ Παλλάδα περσέπτολιν’. εἶδος δὲ τοῦτο ᾄσματος καὶ ἀρχή. τὸν δὲ ποιητὴν αὐτοῦ Ῥοῦφος (see FGrHist 826). καὶ Διονύσιος (scil. ὁ Μουσικός)28 ἱστοροῦσιν ἐν τῇ Μουσικῇ Φρύνιχόν τινα, ἄλλοι δέ φασι Λαμπροκλέα, ἢ Στησίχορον. τὸ δὲ δεινὰν Παλλάδα γελοίως ἀντίκειται. τὸ γὰρ ᾆσμα οὕτως ἔχει· Παλλάδα περσέπτολιν κλεισοπολεμαδόκον, ἁγνὰν παῖδα Διὸς μεγάλου, δαμνοπῶλον, ἄϊστον παρθένον. He explains, saying ‘not (the kind of music) for pleasure, but either ‘Pallas destroyer of cities’. This is a kind of song29 and an incipit. Rufus and Dionysius, in the Mousike, say that the author of this (scil. hymn) was a certain Phrynichus, others say Lamprocles, or Stesichorus. The (expression) ‘terrible Pallas’ corresponds parodistically (scil. to the original text). Indeed, the hymn goes like this: ‘Pallas, the destroyer of cities, I invoke, war–sustaining, venerable, daughter of great Zeus, horse–taming, unseen virgin’.

The second is P. Oxy. 13.1611.30 Indeed, the lacunous papyrus – which was probably a collection of exegetic excerpta on various topics (maybe drawn from a hypomnema by Didymus on a lost play of Aristophanes)31 – contains a section (frr. 5+43+40+6+?)32 on the Παλλάδα περσέπολιν:

 specific passage. However, the variant Ἀθηνᾶν pro δεινὰν in the α family of Dio’s manuscript tradition might not be due to a simple majuscule error (as suggested by Verrengia [1999] 154). 26 See Bagordo (1998) 72, 162. 27 The former was probably quoting the latter (see Wilamowitz [1900] 85). It is impossible to determine how independent this scholium is from the tradition of the scholia to Ar. Nub. 967. 28 See Montanari (1997); Rocconi (2010). 29 The expression is unclear, as εἶδος usually refers to actual types of songs (see e.g. Poll. 4.82 ἔνιοι δὲ καὶ ἐμβατηρίους αὐλοὺς ὠνόμασαν τοὺς ἐπὶ τοῖς προσοδίοις, καὶ δακτυλικοὺς τοὺς ἐπὶ τοῖς ὑπορχήμασιν· οἱ δὲ ταῦτα οὐκ αὐλῶν ἀλλὰ μελῶν εἶναι εἴδη λέγουσιν, schol. Ar. Eq. 1279d ὄρθιον νόμον] εἶδος μέλους, schol. Eur. Or. 1390 εἶδος θρήνου ὁ ἰάλεμος, schol. Hom. Il. 18.570b λίνον] εἶδος ὕμνου … ὡς παιὰν καὶ διθύραμβος). 30 MP3 2290; LDAB 5430; CPP 0050. 31 Several works scrutinized and partially corrected Grenfell and Hunt’s (1908) hypotheses on the content of the papyrus (see the general introduction to P. Oxy. 13.1611 and the notes to ll. 158–73). See, in particular, Arrighetti (1968) 76–98; (1977) 63–65; (1987) 215–28, and Montanari (1989) 414–8. See also Perrone (2009) 115; Montana ap. Bastianini et al. (20122) 238–9. As for the relationship between ancient scholarship and the scholia to Aristophanes, see also Montana (2006). 32 The identification of the papyrus fragments was carried out by Montanari (1989) 415, who also established the text that is printed here (ibidem, 414–5). The only alterations to Montanari’s text are the insertion of Allen’s conjecture [φησὶν Ἐρατοσθέ]|νης (l. 158) and the exclusion of

Eratosthenes’ studia Aristophanica  133

σιν̣ [ ] αδ|εσ[ ] α|φορο̣[ κα] θ|απερ[ φησιν Ερατοσθε] νης ε|.[ ] ται̣ς Φ|ρυν[ιχο]|υ [ ] ἀφηγούμεν̣|ọς τ.[ ] Πα[λ]λά[δ]α π̣ερ̣|σ̣έπ[ολιν] [κλῄζ]ω π̣[ολ]εμ̣[αδό] [κο]ν̣ ἁγ̣νὰ̣ν π̣[αῖδα Δι] [ὸς] μ̣εγάλου δ[αμάσιπ] πον οὕτω π̣αρα[φέρει.] – διαπoρ̣ουσ̣ι γ̣αρ ο̣ὐ[κ ὀ] λ̣ίγ̣οι π̣[ε]ρὶ τ̣[ουτ]ω̣ν, κα [θ]άπερ Χαμαιλέων πό [τ]ερόν ποτε Στη[σι]χόρου ἐστὶν ἢ Λαμπροκλ[έ] [ο]υς κ[αίπ]ε̣ρ τοῦ Φρυν̣[ί] [χου Λαμ]προκλεῖ μα[ [ ] προσ̣νέμο̣ν [τος. καὶ Ἀ]ριστο̣φάνης [δὲ παραπ]οιεῖ λέγων· [Παλλάδα] π[ε]ρ̣σ̣έ[π]ο[λιν δε ινάν

… … … according (to what Eratosthenes says?) … … of Phrynichus … reporting … ‘Pallas, destroyer of cities I invoke, war–sustaining, venerable, daughter of great Zeus, horse–taming’, so (he quotes?). Many do not know, with regard to these things,





a











a



 Körte’s tentative proposal μά[λα ἀκριβῶς in l. 172 (see id. 1924, 242), where the restoration doubtfully suggested by the editores principes μα[θη Μίδωνος], see Grenfell–Hunt (1919) 147 is at least consistent with the information of the scholia. Holwerda’s attempt ἐ[ν δὲ τοῖς Κωμασ]|ταῖς (ll. 159–60. See id. 1952, 230) is fascinating but lacks evidence.

134  Federica Benuzzi like Chamaeleon, whether this is by Stesichorus or by Lamprocles, even though Phrynichus to Lamprocles (pupil?) … ascribes. And Aristophanes (makes a parody?) saying: ‘Pallas, terrible destroyer of cities…’





The best preserved part of the papyrus can be divided into three sections: firstly, the quotation (ll. 162–5) of the hymn to Pallas in the β version; secondly, the mention (ll. 166–74) of the hesitation of several commentators (Chamaeleon in particular) between Lamprocles and Stesichorus, with regard to the ode’s authorship, in spite of the explicit attribution of the hymn to Lamprocles by the comic poet Phrynichus;33 thirdly, the reference (ll. 174–6) to an underlying parodic aim in Aristophanes’ quotation of the first line of the same hymn (the papyrus’ verb is probably παραποιεῖ, whereas the scholia recentiora have εἰρωνεύεται/διασύρει)34. The same reference, differently formulated, is also in the abovementioned scholium to Aelius Aristides (τὸ δὲ δεινὰν Παλλάδα γελοίως ἀντίκειται).

 33 Serena Perrone (2010) analysed the papyrological evidence for the use of comic intertextual comparisons in ancient scholarship on comedy: the hypomnemata from the Imperial age abound with lexical comparisons but content–related ones are also present. As underlined by Perrone (ibidem, 100), most of the comparisons are preserved in extremely epitomised form (often reduced to the bare mention of the name of the author used as term of comparison). Though quoting P. Oxy. 13.1611 (ibidem, 99–100), Perrone does not include the papyrus’ section about the hymn to Pallas, probably because the text does not represent a standard example of exegetical intertextuality, but rather a specific case of solution to a problem of poetic intertextuality. Still, the papyrus provides much more information than many other examples taken into account. Indeed, besides preserving the name of the comic poet used as comparison and quoting extensively from the reference text, the papyrus also mentions the opinions of several commentators and probably included the title of the play from which the parallel was drawn (see the lacuna in l. 159, as restored by Holwerda [1952]). 34 See schol. anon. rec. Ar. Nub. 967a ᾆσμα Λαμπροκλέους· διασύρει δὲ αὐτόν· ‘Παλλάδα περσέπτολιν κληΐζω πολεμοδόκων ἁγνὸν παῖδα Διὸς μεγάλην δαμάσιππον’, b Παλάδα περσέπτολιν· Κύδοιπός τις ποιητὴς εἰσῆξέ τινα λέγοντα ἐν τῷ ἑαυτοῦ ποιήματι τὸ παρὸν ᾆσμα μεθ’ ὕμνου· ὁ δὲ εἰρωνεύεται αὐτόν.

Eratosthenes’ studia Aristophanica  135

Given the undeniable relationship between the papyrus’ text and the scholia, Allen’s restoration [φησὶν Ἐρατοσθέ]|νης (l. 158)35 seems more than plausible. Moreover, the supplement is especially corroborated by the occurrence of Phrynichus’ name (clearly visible in l. 160) near Eratosthenes’ in both versions of schol. Nub. 967. In actual fact, the two versions seem to preserve opposite information with regard to Eratosthenes’ stance in the controversy on the authorship of the Παλλάδα περσέπολιν, as α claims that Eratosthenes ascribed the hymn to Lamprocles (οὕτως Ἐρατοσθένης· Φρύνιχος δὲ αὐτοῦ τούτου τοῦ ᾄσματος μέμνηται ὡς Λαμπροκλέους ὄντος τοῦ Μίδωνος υἱοῦ ἢ μαθητοῦ), while β states that he attributed it to Phrynichus (ἀρχὴ ᾄσματος Φρυνίχου, ὡς Ἐρατοσθένης φησίν). As already observed above, the contradiction was solved by Holwerda through the replacement of Φρυνίχου with Στησιχόρου in β.36 However (provided that Allen’s proposal is to be considered correct), P. Oxy. 13.1611 constitutes significant evidence in favour of the opposite information preserved by α, i.e. that Eratosthenes ascribed the Παλλάδα περσέπολιν (that he found alluded to in the verse of the Clouds)37 to Lamprocles, because the comic poet Phrynichus quoted the very same text in an extended form, explicitly saying that it was by him38 (see P. Oxy. 13.1611 ll. 157–61 and 171–4). Perhaps a different intervention on the text of β is required. Indeed, emending the same genitive Φρυνίχου in Λαμπροκλέους could make β consistent with both α and P. Oxy. 13.1611. A confusion, in the textual transmission, between Lamprocles’ and Phrynichus’ name is in actual fact not unlikely, because of the close distance at which the two words occur in the

 35 See Grenfell–Hunt (1919) 146. The restoration was not equally accepted by scholars. According to the papyrus’ editors (ibidem), Eratosthenes’ name ‘can be restored’, whereas for Sgobbi (2006) 292 it is only ‘prospettabile – ma certamente tutt’altro che sicura’. 36 See also the expunction of Φρυνίχου by Scorza (1932) 10. 37 One could hardly agree with Sgobbi (2006) 292, when he suggests: ‘Eratostene […] avrebbe potuto parlare dell’inno trattando di Frinico’. 38 Nothing can be inferred by the textual evidence with regard to the context of the citation of Lamprocles by Phrynichus. Nevertheless, quotations of lyric poetry in comedy are widely attested (see Cingano [1990] 207 n. 66 for a complete list of citations, adaptations and allusions to such poets in Aristophanes), although the authors’ names are rarely mentioned (see Nub. 1356). One of the distinctive traits of Eratosthenes’ philology lies in the frequent comparisons between different texts, in order to solve the literary ζητήματα (see frr. 9, 52, 66, 81, 101 Strecker). The importance of this ‘methodological’ use of intertextuality – especially when compared to the exegetic methods of Eratosthenes’ predecessors (such as Lycophron and the Γλωσσογράφοι) – was first highlighted by Tosi (1998b) 346. For the papyrological evidence of the use of intertextual comic comparisons in the exegesis to comedy, see Perrone (2010).

136  Federica Benuzzi text.39 With this correction, Eratosthenes’ argument in β becomes clear, as does the emphatic phrasing δὲ αὐτοῦ τούτου,40 which highlights the authority of the intertextual comparison (‘beginning of a hymn by Lamprocles, according to Eratosthenes: Phrynichus quotes this very hymn ascribing it to Lamprocles’). Allen’s restoration also affects the general interpretation of the whole text of ll. 158–76, as the presence of Eratosthenes’ name at the beginning of the argumentation seems to entail that the scholar of Cyrene – in his interpretamentum to Nub. 967 – explicitly quoted and refuted Chamaeleon’s declaration of aporia,41 thus constituting the source from which the commentary drew the information concerning the Peripatetic’s stance on the topic. Indeed, the analysis of his predecessors’ opinions42 (whose explanations were often invalidated thanks to proving intertextual comparisons) was an essential step of Eratosthenes’ interpretative approach (see Tosi [1998b]). Besides representing an essential element in order to reconstruct Eratosthenes’ position in the authorship debate on the Παλλάδα περσέπολιν, P. Oxy. 13.1611 preserves further valuable information. Indeed, the statement concerning the intentional alteration of the beginning of the ode to Pallas by Aristophanes (ll. 174–5 καὶ Ἀ]ριστoφάνης [δὲ παραπ]οιεῖ λέγων) led Sgobbi (2006) to consider the theory of the existence of two independent hymns untenable. Before considering Sgobbi’s perspective, it is useful to sum up the different extant wordings of the hymn to Pallas. P. Oxy. 13.1611 and the redaction β of the scholia to Nub. 967 present one hexameter, followed by a dactylic segment:

 39 It cannot be determined whether the corruption of Λαμπροκλέους in Φρυνίχου originated at the time (and because) of the hasty epitomization of the original annotation (which is better preserved by α and P. Oxy. 13.1611) or, on the contrary, whether it happened at a later stage, when the two indipendent redactions were already clearly defined. 40 Sgobbi (2006) 292 n. 42 does not touch the text of the manuscripts and justifies the phrasing as ‘traccia di un discorso articolato di Eratostene sull’inno’. See also Martano (2012) 238. 41 Chamaeleon wrote a Περὶ κωμῳδίας and a Περὶ Στησιχόρου (frr. 43–54 and 27–29 Giordano, 46–7 and 30–32 Martano). Both works are likely to have hosted a disquisition on the problematic ascription of the ode to Stesichorus, see Grenfell–Hunt (1919) 147; Montanari (1989) 418, but the commonly accepted position is that the mention of Chamaeleon’s aporia derives from the book On Stesichorus, see Giordano ad. frr. 28a–c; Arrighetti (1987) 218. On Eratosthenes possibly being the source preserving the information on Chamaeleon’s aporia, see Arrighetti (1987) 218; Montanari (1989) 416–7; Sgobbi (2006) 293. 42 See e.g. the confrontation with Callimachus in schol. Ar. Nub. 553 (fr. 97 Strecker) and the frequent conflicts with Lycophron (frr. 7, 25 and 74 Strecker).

Eratosthenes’ studia Aristophanica  137

Παλλάδα περσέπολιν κληΐζω πολεμαδόκον ἁγνάν παῖδα Διὸς μεγάλου δαμάσιππον.

Two full hexameters are instead preserved by schol. Tzetz. Ar. Nub. 967a UCs: Παλλάδα περσέπολιν κλῄζω πολεμαδόκον ἁγνάν παῖδα Διὸς μεγάλου δαμάσιππον ἄϊστον Ἀθήνην.

The name of the goddess is replaced by παρθένον contra metrum in another scholium by Tzetzes43 (966a UAmb) Παλλάδα περσέπολιν κλῄζω πολεμαδόκον ἁγνάν παῖδα Διὸς μεγάλου δαμάσιππον ἄϊστον παρθένον

and in the Chiliades (I.25.685–7) by the same author, with δάμοπλον (probably δαμνοπῶλον, schol. Ael. Ar. Or. 46,162,7–8) instead of δαμάσιππον: Παλλάδα περσέπτολιν κλῄζω πολεμαδόκον ἁγνάν παῖδα Διὸς μεγάλου δάμοπλον ἄϊστον παρθένον.

The same version, with a clear misunderstanding of κλῄζω πολεμαδόκον is to be found in schol. Ael. Ar. Or. 46.162.7–8: Παλλάδα περσέπτολιν κλεισοπολεμαδόκον, ἁγνὰν παῖδα Διὸς μεγάλου, δαμνοπῶλον, ἄϊστον παρθένον.

Finally, the redaction α of the scholia to Nub. 967 is the only witness of a completely different version, of only one hexameter: Παλλάδα περσέπτολιν δεινὰν θεὸν ἐγρεκύδοιμον.

Despite the alternative variants Ἀθήνην–παρθένον and δαμάσιππον– δαμνοπῶλον,44 the three occurrences by Tzetzes45 and the scholium to Aelius Aristides constitute one consistent group, the internal relationships of which are  43 For the main causes of discrepancies within Tzetzes’ scholia to Aristophanes see Massa Positano (1960) xxv. 44 While δαμάσιππος has at least one parallel in B. 3.23, δαμνοπῶλος is a hapax possibly influenced by widespread compounds such as πωλοδάμνης (see e.g. Xen. Oec. 3.10) and πωλοδαμνέω (see e.g. Soph. Aj. 549; Eur. Rh. 187, 624). 45 Tzetzes contradicts himself (see Sgobbi [2006] 289 and n. 23–24; Ercoles [2013] 514 n. 858) with regard to the authorship of the hymn, which he ascribes to Lamprocles in the scholia (clearly on the basis of the vetera) and to Stesichorus in the Chiliades. The inconsistency (i.e. the

138  Federica Benuzzi bound to remain unclear. Moreover, nothing can be inferred about the tradition from which these testimonies derived ἄϊστον Ἀθήνην.46 With regard to the version of the hymn preserved by α, Holwerda – and many before and after him – believed that this was the beginning of a different hymn to Pallas. Nevertheless, the hypothesis of two independent hymns makes Chamaeleon’s aporia hardly understandable. Indeed, if the Peripatetic knew two such hymns as the ones postulated by Holwerda (Παλλάδα περσέπολιν δεινὰν θεὸν ἐγρεκύδοιμον on the one hand, Παλλάδα περσέπολιν κλῄζω πολεμαδόκον κτλ on the other), then he would have had no reason to be in doubt, for the δεινάν of Nub. 967 would have logically implied the quotation of the former and excluded an identification with the latter. While opposing Holwerda’s perspective and textual arrangement, Sgobbi (2006, 295–6) suggested an alternative explanation, based on the evidence regarding Aristophanes’ alleged parodic rewriting of the incipit of the ode in Nub. 967. According to him, there was only one hymn to Pallas, and the text of α results from subsequent phases of corruption. In particular, the form of the hymn Παλλάδα περσέπτολιν δεινὰν θεὸν ἐγρεκύδοιμον is due to the hesitation of a scholiast, who noticed the difference between the Aristophanic verse and the wording of the hymn as quoted in the scholium on his antigraph,47 but was unaware of the alleged parodic alteration by Aristophanes. He consequently (1) stopped copying the Wortlaut Παλλάδα περσέπολιν κλῄζω πολεμοδόκον κτλ, (2) added ἔχει δὲ οὕτως (‘[the hymn] goes like this’), (3) inserted the Aristophanic δεινάν, expanding the incipit of the hymn (or allowing it to be expanded at a later stage of the tradition) with the ‘Hesiodic’ reminiscence θεὸν ἐγρεκύδοιμον (see Hes. Th. 924–5 αὐτὸς δ’ ἐκ κεφαλῆς γλαυκώπιδα γείνατ’ Ἀθήνην, / δεινὴν ἐγρεκύδοιμον ἀγέστρατον ἀτρυτώνην. See also Sgobbi 2006, 291 and 294), (4) finished copying the hymn’s text as given in the scholium. It must be pointed out that Sgobbi’s perspective presumes that the scholium originally

 loss of every trace of Chamaeleon’s aporia with regard to the authorship in the Chiliades) could be due to Tzetzes’ faulty memory of the controversy. Worth mentioning is also the citation (Chil. I 25.688–94, right after the hymn to Pallas) of the allegedly Archilochean hymn τήνελλα καλλίνικε, which is juxtaposed by Tzetzes to the Παλλάδα περσέπολιν at the end of the section On Stesichorus without a clear logical link. Like the hymn to Pallas, the τήνελλα καλλίνικε was alluded to by Aristophanes (in Ach. 1227–34 and Av. 1765) and was at the centre of an exegetical debate involving Eratosthenes (fr. 136 Strecker), Callimachus (fr. 384.39 Pfeiffer) and Aristarchus (apud scholl. Pind. Ol. 9.3; Nem. 3.1), and preserved by several scholia (see e.g. scholl. Pind. ll. cc. and schol. Ar. Av. 1764). Tzetzes’ mention of this other unrelated aristophanic ζήτημα after the Παλλάδα περσέπολιν might reinforce the idea that the scholar was relying on his memory. 46 See Sgobbi (2006) 289 n. 22. 47 See also Scorza (1932) 10.

Eratosthenes’ studia Aristophanica  139

contained only the section regarding Eratosthenes’ position in the controversy, followed by the citation of the incipit of the Παλλάδα περσέπολιν. According to the scholar (but see also Arrighetti 1987, 218), the wording of α results from a later insertion of the section concerning Chamaeleon’s aporia with regard to the authorship.48 In Sgobbi’s eyes, this section did not integrate properly in the text, ended up ‘splitting’ the quotation of the beginning of the hymn between ἐγρεκύδοιμον and κλῄζω, and lost its last part, thus becoming the Χαμαιλέων δ’ ἀπορεῖ, πότερον now visible in α. If, on the one hand, the idea of the Hesiodic epithet intruding into the text can be accepted, the hypothesis of a later insertion of Chamaeleon’s section is, on the other, not equally convincing. Before giving an alternative explanation for the problematic wording of α it is useful to underline how eliminating – both on logical and on textual grounds – the option of two independent hymns49 affects our understanding of Chamaeleon’s doubt. Indeed, even if a hesitation between two existing texts is excluded, the details of the Peripatetic’s aporia are all but clear. At least two scenarios can be considered: 1. Chamaeleon ignored Phrynichus’ testimony of the wording and of the authorship of the hymn (as the central role of the playwright in Eratosthenes’ answer to Chamaeleon may suggest). If this is the case, he must have had some other element linking Lamprocles to the ode to Pallas, making him a possible candidate for the ascription. However, the evidence connecting Lamprocles to the hymn was apparently not convincing enough to overthrow and exclude a traditional ascription to Stesichorus.50  48 Originally, the note might have sounded like Χαμαιλέων δ’ ἀπορεῖ, πότερον , see Sgobbi (2006) 295. 49 It must be pointed out that refusing the idea of two concurrent versions of the ode to Pallas reading respectively Παλλάδα περσέπολιν κλῄζω πολεμοδόκον κτλ and Παλλάδα περσέπολιν δεινὰν θεὸν ἐγρεκύδοιμον does not imply presuming that only one hymn to the goddess (Παλλάδα περσέπολιν κλῄζω πολεμοδόκον κτλ) circulated in the 5th century Athens and was quoted extensively by Phrynichus and parodically by Aristophanes. On the contrary, it is unlikely that Nub. 967 actually referred to the same hymn quoted by Phrynichus, as maintained by Eratosthenes. However, the problematic text of α shows that the wording Παλλάδα περσέπολιν δεινὰν θεὸν ἐγρεκύδοιμον is not ancient and originated from a corruption. As already underlined above, the level of the cultural and literary context of composition of Nub. 967 must be kept separated from the level of the Peripatetic and Alexandrian interpretation of the same passage. 50 See Pòrtulas (2012) 232: ‘Stesichorus is a ‘zero degree’, so to speak, as far as the attribution of this sort of choral composition is concerned’. For an inquiry on the possible reasons linking Stesichorus to the hymn to Pallas, see Sgobbi (2006) 297–8. On authorship, authority and double attributions in archaic poetry, see Condello (2007). Cingano (1990) offers a detailed analysis of all cases of authorship debate involving Stesichorus and Ibycus.

140  Federica Benuzzi 2.

Chamaeleon knew Phrynichus’ passage but was not sure whether Aristophanes was really quoting Lamprocles. In this case, the main reason for the hesitation must have been the discrepancy between Lamprocles’ and Aristophanes’ text (i.e. the adjective δεινάν). This discrepancy might have led the Peripatetic into postulating another hymn beginning with Παλλάδα περσέπολιν δεινάν, which he tentatively ascribed to Stesichorus. In these conditions, the core of Eratosthenes’ objection must be the mention of the parodic alteration of the hymn by the comic playwright (but this finds less support in the textual evidence).

As far as Sgobbi’s explanation of the Wortlaut of α is concerned, the compresence of Eratosthenes’ and Chamaeleon’s section in one cohesive exegetical cluster in P. Oxy. 13.1611 makes the hypothesis of a dismemberment of the ancient annotation somewhat unlikely. Indeed, it is hard to believe that the two parts arrived independently (and at different times) in the text of the scholium, originating with their overlap the text of α. A comparison between the latter and the papyrus clearly shows that the two essentially preserve the same ancient content, and that the problematic wording of α must be due to errors within the textual transmission of this exegetical cluster. An alternative – still highly tentative – explanation might be that the scholiast (after stopping while copying the hymn’s text, after adding ἔχει δὲ οὕτως51 and after inserting the Aristophanic δεινάν followed by the ‘Hesiodic’ θεὸν ἐγρεκύδοιμον) did not finish transcribing the ode’s incipit as quoted by the scholium in the exemplar, and that – at a later stage of the transmission – the comparison with another annotation, which preserved the text in its entirety, caused the addition of the omitted words κλῄζω πολεμοδόκον ἁγνάν κτλ. This segment did not integrate properly in the text, overlapping and partially obliterating the section concerning Chamaeleon (which was always transmitted in conjunction with the Eratosthenian one). In this perspective, the text that can be reconstructed as the original redaction of the scholium coincides almost exactly with P. Oxy. 13.1611,52 reading as follows:53  51 The insertion of the phrase (which is the standard way, in the scholia, to introduce any literary quotation) shows the misinterpretation of the original wording. Indeed, the expression breaks the syntax, which implies that Phrynichus is the subject of ὑποτίθησι see Sgobbi (2006) 296. 52 As already suggested by Sgobbi (2006) 296, according to whom the scholium ‘verrebbe ad essere il più preciso, tra le nostre fonti, nel riportare le parole di Eratostene’. 53 The reconstruction must be regarded exclusively as a working hypothesis aimed at explaining the form of the text transmitted by the manuscripts, and not as an emendation proposal.

Eratosthenes’ studia Aristophanica  141

οὕτως Ἐρατοσθένης· Φρύνιχος δὲ αὐτοῦ τούτου τοῦ ᾄσματος μέμνηται ὡς Λαμπροκλέους ὄντος τοῦ Μίδωνος υἱοῦ ἢ μαθητοῦ. ‘Παλλάδα περσέπτολιν κλῄζω πολεμοδόκον ἁγνάν, παῖδα Διὸς μεγάλου δαμάσιππον’ καὶ κατὰ Λαμπροκλέα ὑποτίθησι κατὰ λέξιν.54 Χαμαιλέων δ’ ἀπορεῖ, πότερον . Thus says Eratosthenes: Phrynichus quotes this very hymn, ascribing it to Lamprocles, son or pupil of Midon, ‘Pallas, destroyer of cities, I invoke, war–sustaining, venerable, daughter of great Zeus, horse–taming’ and reports the version of Lamprocles word by word. Chamaeleon is uncertain, whether .

To sum up, the form of α can be explained without resorting to a hypothetical splitting of the ancient annotation (which is already unitary in P. Oxy. 13.1611) into two parts, one concerning Eratosthenes and the other concerning Chamaeleon. As far as β is concerned, the relatively economic intervention (Φρυνίχου > Λαμπροκλέους) re–establishes the correct logic sequence of the Eratosthenian interpretamentum also in β and confirms that, in actual fact, α and β represent two redactions of the same ancient annotation.55

 Indeed, emending a text such as the one of ENp – the wording of which reflects several subsequent phases of reception both of the comic and of the scholiastic content – would be unthinkable from an ecdotic perspective. The reconstructed wording could only be printed in the apparatus criticus with a clear explanation of its merely theoretical value. 54 The sentence, the subject of which is the comic poet, corresponds to l. 165 in P. Oxy. 13.1611 (where one should probably read oὕτω παρα[φέρει]), and states that Phrynichus quoted the hymn in its original version, the one by Lamprocles, whereas Aristophanes parodically distorted it. 55 As far as the dispute on the Παλλάδα περσέπολιν is concerned, α and β are respectively the plenior and brevior version of the same ancient annotation. However, β preserves more information than α with regard to the τηλέπορόν τι βόαμα.

Sara Panteri

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics The fragment in Theo Sm. 81.17–82.5 Hiller Despite the fame of Eratosthenes’ scientific activity, the ancient and medieval tradition has often preserved only scanty fragments of his written works. The same situation affects the contribution of Eratosthenes as philosopher: very little remains of the man nicknamed by the Ancients δεύτερος ἢ νέος Πλάτων.1 This nickname is even more remarkable if we consider that a work entitled Πλατωνικός is attributed to Eratosthenes by Theon of Smyrna2 in his Expositio rerum mathematicarum ad legendum Platonem utilium. The fog which surrounds Eratosthenes’ lost work is thick and gloomy. The title is reported only twice3 and the fragments which can be led back to it are few.4 In this paper I will deal with Theo Sm. 81.17– 82.5 Hiller, one of the two fragments in which the title Πλατωνικός is specified. In the first part I will provide the Greek text, the critical apparatus and the translation of the fragment. In the second part I will contextualize the fragment within the work of Theon, in particular, and within the exegesis on Plato’s Timaeus, in general. In the third part I will deal with the interpretation of an interesting proportion which is established as an example in the fragment. The four terms of this geometrical proportion are the four mental states (νοῦς, ἐπιστήμη, δόξα, αἴσθησις). Using this kind of material, Eratosthenes, on the one hand, carries on his exegetical point with technical acumen. On the other hand, he converses with Plato’s philosophy and, specifically, with the Republic and the Timaeus. Finally, in the fourth part, I will summarize the central arguments. The main aim of my  1 Suda ε 2898 A. (= FGrHist 241 T 1). 2 Suda θ 204.1 A. Θέων, Σμυρναῖος, φιλόσοφος. The information about Theon’s life is meagre, but he probably lived during the first half of the 2nd century AD and was a professor of Platonic Philosophy; his only extant work is the Expositio. Cf. Petrucci (2012) 10ff. 3 Interestingly enough, Eratosthenes’ Πλατωνικός is the first source mentioned by Theon at the very beginning of the Expositio (2.3ff. Hiller). The second mention is Theo Sm. 81.17 Hiller. 4 A complete edition of Eratosthenes’ fragments does not exist. As for the fragments which have a mathematical or philosophical content, the following contributions have to be mentioned: Bernhardy (1822) (collection of fragments); Hiller (1870); Solmsen (1942); Wolfer (1954); Dörrie (1987) 116–30, 352–77; Geus (2002). The editor of Theon, whose apparatus is reported here (with modifications based on the view of codex A) is E. Hiller, Expositio rerum mathematicarum ad legendum Platonem utilium, Lipsiae 1878. For the fragment here studied, see Geus (1995). https://doi.org/10.1515/9783110632590-011

144  Sara Panteri paper is to provide an interesting case of study where ‒ starting from a terminological distinction – particular attention and interest is paid to a concept, the proportion, which has both philosophical and mathematical implications.

 Text, Apparatus and Translation Theo Sm. 81.6–82.5 Hiller διαφέρει δὲ διάστημα καὶ λόγος, ἐπειδὴ διάστημα μέν ἐστι τὸ μεταξὺ τῶν ὁμογενῶν τε καὶ ἀνίσων ὅρων, λόγος δὲ ἁπλῶς ἡ τῶν ὁμογενῶν ὅρων πρὸς ἀλλήλους σχέσις. διὸ καὶ τῶν ἴσων ὅρων διάστημα μὲν οὐδέν ἐστι μεταξύ, λόγος δὲ πρὸς ἀλλήλους εἷς καὶ ὁ αὐτὸς ὁ τῆς ἰσότητος· τῶν δὲ ἀνίσων διάστημα μὲν ἓν καὶ τὸ αὐτὸ ἀφ’ ἑκατέρου ἑκάτερον, λόγος δὲ ἕτερος καὶ ἐναντίος ἑκατέρου πρὸς ἑκάτερον· οἷον ἀπὸ τῶν βʹ πρὸς τὸ ἓν καὶ ἀπὸ τοῦ ἑνὸς πρὸς τὰ βʹ διάστημα ἓν καὶ τὸ αὐτό, λόγος δὲ ἕτερος, τῶν μὲν δύο πρὸς τὸ ἓν διπλάσιος, τοῦ δὲ ἑνὸς πρὸς τὰ βʹ ἥμισυς.

Ἐρατοσθένης δὲ ἐν τῷ Πλατωνικῷ φησι, μὴ ταὐτὸν εἶναι διάστημα καὶ λόγον, ἐπειδὴ λόγος μέν ἐστι δύο μεγεθῶν ἡ πρὸς ἄλληλα ποιὰ σχέσις· γίνεται δ’ αὕτη καὶ ἐν διαφόροις ˂καὶ ἐν ἀδιαφόροις>. οἷον ἐν ᾧ λόγῳ ἐστὶ τὸ αἰσθητὸν πρὸς τὸ νοητόν, ἐν τούτῳ δόξα πρὸς ἐπιστήμην, καὶ διαφέρει καὶ τὸ νοητὸν τοῦ ἐπιστητοῦ ᾧ καὶ ἡ δόξα τοῦ αἰσθητοῦ. διάστημα δὲ ἐν διαφέρουσι μόνον, ἢ κατὰ τὸ μέγεθος ἢ κατὰ ποιότητα ἢ κατὰ θέσιν ἢ ἄλλως ὁπωσοῦν. δῆλον δὲ καὶ ἐντεῦθεν, ὅτι λόγος διαστήματος ἕτερον· τὸ γὰρ ἥμισυ πρὸς τὸ διπλάσιον ˂καὶ τὸ διπλάσιον πρὸς τὸ ἥμισυ˃ λόγον μὲν οὐ τὸν αὐτὸν ἔχει, διάστημα δὲ τὸ αὐτό.

 Cod.: A (= Venetus Marcianus Gr. 307, XI/XII saec.) ║2 ἐπειδή Α : ἐπεί Bernhardy ║μεγεθῶν A1 quod rec. Bernhardy : μεγέθων Α │ ἡ add. A1 quod rec. Bernhardy│ πρὸς ἄλληλα Bernhardy quod rec. Hiller : προσάλληλα Α │δ’ A : δέ Bernhardy │ αὕτη Hiller : αὐτή A quod rec. Bernhardy ║3 καὶ ἐν ἀδιαφόροις add. Hiller ║ τὸ αἰσθητὸν πρὸς τὸ νοητόν ex apogr. Bernhardy et Hiller : τὸ νοητὸν πρὸς τὸ αἰσθητόν A ║4 [καὶ] διαφέρει exp. Bernhardy │[καὶ] ἡ δόξα exp. Bernhardy ║5 ἐν Hiller : ἕν A : ἐνδιαφέρουσι Bernhardy ║8 καὶ τὸ διπλάσιον πρὸς τὸ ἥμισυ add. Hiller

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  145

. Translation Interval and ratio differ, because interval is between terms of the same kind and unequal, while ratio is simply the relation between two terms of the same kind to one another. Therefore there is no interval between equal terms, while the ratio to one another is one and the same, the equality; and the interval between unequal terms is one and the same of one to the other, while ratio is other and contrary of one to the other. For example, from two to one and from one to two, the interval is one and the same, the ratio is other, on the one hand from two to one (scil. the ratio) is the double, on the other hand, from one to two (scil. the ratio) is the half.

Eratosthenes in Platonicus says that interval and ratio are not the same thing, because ratio is a kind of relation between two magnitudes to one another; this exists both between different (scil. magnitudes) ˂and not–different ones˃. For example, in the ratio in which the perceptible is to the intelligible, in this ratio is opinion to knowledge, and the intelligible is different from the knowable in the same way opinion is different from the perceptible. On the contrary, an interval exists only between terms that are different in magnitude, in quality, in position or in another sense. It is evident also from this, that ratio is other from interval: indeed, the half to the double and the double to the half does not have the same ratio, but the interval is the same.

 The context Theon of Smyrna’s Expositio rerum mathematicarum ad legendum Platonem utilium can be divided into three main parts. The mention of Eratosthenes appears in the second part, which is concerned with music (46.20–119.21 Hiller). As for the sections on arithmetic (17.25–46.19 Hiller) and on astronomy (120.1–205.6 Hiller), the perspective of Theon when he deals with music is first and foremost didactical.5

 5 Theon expresses the didactical aim of the entire work at the beginning of the Expositio (1.10– 2.2 Hiller). On the one hand, the Expositio should cover the five subjects at the basis of education listed in Resp. 7 (as a consequence, the fact that the sections on solid and plane geometry do not appear in Expositio is taken by some scholars as evidence of the incompleteness of Theon’s work; cf. Petrucci (2012) 37–40. On the other hand, the notions provided by Theon throw light on the obscure passages of Plato’s dialogues, considered difficult to comprehend because of their technical content.

146  Sara Panteri In the second section, indeed, Theon provides the fundamental notions of harmonic theory.6 According to Petrucci (2012),7 the entire section on music should be read in the light of Plato’s Timaeus and, more specifically, of the generation of the World Soul (Tim. 34b–36d). This passage is considered a key text within Platonic Philosophy; it is one of the most famous and widely studied.8 In the ψυχογονία (‘generation of the soul’), described by Timaeus in a long monologue, the Demiurge cuts and matches the matter of the soul. The first division (35b–c) follows a double geometrical progression, that of the double and that of the triple.9 Then (36a), the Demiurge fills each interval (ἐν ἐκάστῳ διαστήματι) cutting– off parts from the matter and putting them between the intervals previously created so that two means (μεσότητες) are found: one harmonic,10 the other arithmetic.11 If the ratios between the terms obtained by means of these subsequent operations are considered, they all are harmonic ratios (called διαστάσεις in 36a).12 Then, each harmonic interval (διάστημα) of 4/3 is filled with intervals of 9/8 so that its remaining part will be filled with the interval (διάστασις) of leimma, namely 256/243.13 The description of the divisio animae ends at this point (36b).

 6 In order to understand the harmony of the cosmos, it is first necessary to understand music in numbers. This kind of music belongs to the realm of intelligible; its sensible counterpart is music in instruments. Only after grasping these two kinds of music, one can approach the harmony of the cosmos (Theo Sm. 46.20–47.17 Hiller); cf. Barker (1989) II.209. As for the content, Theon refers both to the Pythagorean and Aristoxenian harmonic theories and often compares the two. For the general structure and the content of this section, see Mathiesen (1999) 412–28 (especially 416ff.) and Petrucci (2012) 26–30 (especially 26). 7 Cf. Petrucci (2012) 26–30. 8 The list of authors who worked on Plato’s Timaeus, from antiquity to the imperial era, is long but some of them are: Xenocrates (4th BC); Crantor (4th/3rd BC); Clearchus (4th/3rd BC); Posidonius (2nd/1st BC); Adrastus of Aphrodisias (1st/2nd AD); Plutarch (1st/2nd AD); Porphyry (3rd AD); Calcidius (4th AD); Proclus (5th AD). For the commentary tradition on Plato’s Timaeus, cf. Ferrari (2000). 9 The two progressions are 1, 2, 4, 8 and 1, 3, 9, 27. 10 The two geometric progressions with the insertion of the harmonics terms are 1, 4/3, 2, 8/3, 4, 16/3, 8 and 1, 3/2, 3, 9/2, 9, 27/2, 27 (insertions in bold). 11 The two geometric progressions with harmonics terms and with the insertion of the arithmetic terms are 1, 4/3, 3/2, 2, 8/3, 3, 4, 16/3, 6, 8 and 1, 3/2, 2, 3, 9/2, 6, 9, 27/2, 18, 27. The entire sequence becomes 1, 4/3, 3/2, 2, 8/3, 3, 4, 9/2, 16/3, 6, 8, 9, 27/2, 18, 27. 12 The ratios involved are 3/2 (the fifth), 4/3 (the fourth), 9/8 (the tone). Obviously, no all the ratios are harmonic, but the ratios found here are recognised as being ἐν συμφωνίᾳ (Theo Sm. 74.15–75.25 Hiller). 13 Indeed 4/3 can be divided in order to create the three ratios of 9/8, 9/8, 256/243. For the exact division, see Brisson (1992) 286ff.; Brisson (1998) 316ff.

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  147

However, it is clear that, by means of this last division, the structure of a tetrachord takes shape.14 And, again, it is worth noticing that the sequence tetrachord (9/8, 9/8, 256/243) ~ 9/8 ~ tetrachord (9/8, 9/8, 256/243) corresponds to the ratio 2/1, the octave represented in its disjunctive form.15 Now, between 1 and 27, we will find the diatonic octave repeated four times (at the beginning, from 1 to 16), but at the end (from 16 to 27) the sequence is closed with a fifth (3/2) and a tone. Interestingly enough, this wide extension is only potential in music, where it seems that the maximal sequence we can find is that of the two octaves and a fifth.16 Nonetheless, the reason why this particular division is accomplished is that, as Adrastus17 says (Theo Sm. 65.1–3 Hiller), Plato made it πρὸς τὴν φύσιν ὁρῶν, ἐπειδὴ τὴν ψυχὴν ἀνάγκη συνισταμένην καθ’ ἁρμονίαν μέχρι τῶν στερεῶν προάγειν (‘looking at the nature, because the soul structured according to harmony is necessarily extended until solid numbers’). Evidently, as for the creation of the world (Tim. 32a–b),18 we are dealing with a reality, the soul, which is solid in nature and, therefore, when represented by numbers, has to be extended until solid numbers (i.e. 8 and 27) in the two geometrical progressions. Now, going back to the Expositio, Theon (72.21–87.3 Hiller) pays particular attention to the fundamental concept of λόγος (‘ratio’). First of all, he clarifies the different meanings that a so common and widely used word can have. Then, he states that he will focus on the technical sense of ratio of the proportion (λόγος τῆς ἀναλογίας in 73.14ff. Hiller). After a definition and explanation of λόγος

 14 Literally, it is an instrument made up of four strings. In harmonic theory, it is a series of four notes; it spans a fourth, that is, the ratio between the two boundary notes is 4/3. These two boundary notes are fixed while the other two notes in between occupy different positions (according to certain ratios) depending on the melodic genus of the tetrachord (see, Barker [1989] II.11ff.). 15 The octave divided in this way is made according to the ditonic diatonic genus. Cf. Cornford (1937) 66–72; Barker (1989) II, 58–61; Brisson (1992) 284–7; Brisson (1998) 314–8; Barker (2007) 318–23. 16 Cf. Aristox. Harm. 23.22–32. 17 Adrastus of Aphrodisias was a Peripatetic philosopher lived in the 2nd century AD. He wrote a commentary on the Timaeus and he is probably the main source for the second part of Theon’s Expositio. See Goulet (1989) and Petrucci (2012) 514–21. 18 In this passage it is said that, since the world is a solid reality, a mean proportional between fire and earth is not sufficient. At least two means have to be found. This is the reason why air and water are introduced as means proportional between fire and earth. Some scholars, for instance Brisson (1998) 386, think that behind this statement there is a reference (more or less explicit) to the problem of doubling the cube (to which Eratosthenes gave its own solution; see Eutoc. Sph. cyl. 88.3–96.15.27 Heiberg). If this is the case, the same idea would be in the background here, in the procedure of the divisio animae. I will come back to this point below.

148  Sara Panteri (73.16–74.14 Hiller), different kinds of λόγοι are listed (74.15–81.5 Hiller). At the end of the list, we find the section dealing with the difference between διάστημα (‘interval’ or ‘distance’) and λόγος.19 The section immediately after treats proportions (82.6–85.8 Hiller). Likely, the reason why Theon in his Expositio is interested in the difference between διάστημα and λόγος lies in the fact that this distinction was important for understanding the structuring of the World Soul in Plato’s Timaeus. In this passage, as seen above, the word used to refer to the sections in which the soul is divided is διάστημα/διάστασις which can be ἡμιόλιος, ἐπίτριτος, ἐπόγδοος or διάστασις ἀριθμοῦ πρὸς ἀριθμόν. The three adjectives can be translated as ‘containing one and a half’, ‘containing one and one third’, ‘containing one and one eighth’ respectively; these are, more specifically, the ratios 3/2, 4/3, 9/8, that is, the fifth, the fourth, and the tone, the fundamental ratios in music. Probably, the fact that Plato uses the word διάστημα/διάστασις while he actually refers to ratios could have produced doubts and ambiguities in Antiquity. The issue is even more striking if we consider that the difference between διάστημα and λόγος became a distinctive point of departure between the two main Greek schools of Harmonics. This distinction was emphasized after Aristoxenus by his followers and, later, by Harmonic theorists like Ptolemy and Porphyry.20 On the one hand, the Pythagorean School considered the relation between two different pitched notes as represented by means of numerical ratios.21 On the other hand, the followers of Aristoxenus regarded this relation as a linear

 19 The inscription at the beginning of the section in codex A is: τίνι διαφέρει διάστημα καὶ λόγος, ‘in what interval and ratio differs’. 20 Raffa (2013) 250 ‘In fact, a sharp opposition between διάστημα and λόγος could hardly be seen before Aristoxenus, who paved the way for it, although with no explicit theorisation, with his rejection of λόγοι as irrelevant to the understanding of the diastematic movement of melody. A further step in that direction was arguably taken by his followers, as is also discernible in Porphyry’s text, where Aristoxenus is mentioned only once (93.26), most of his references being to οἱ Ἀριστοξένειοι instead’. It should also be remembered that ‘neither school is monolithic ‒ there are important internal distinctions to be drawn ‒ and [...] the work of each did not flow onwards quite independently of the other. There were occasional attempts to bring the two approaches together in a coherent synthesis, as well as more frequent polemical interactions across the doctrinal divide’ (Barker [1989] II.4). 21 See, e.g., Theo Sm. 56.9–61.17 Hiller. Here Theon explains how some Pythagoreans (Pythagoras, Lasus of Hermione, Hippasus of Metapontum, Phytagoreans in general, Archytas and Eudoxus) have found correspondences between consonances and numerical values. See Barker (2014).

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  149

distance.22 Probably, at certain point in the exegetical tradition on Plato, a terminological clarification was felt as compelling. As a consequence, the topic, the ζήτημα, of the difference between διάστημα and λόγος became traditional.23 In this regard, Eratosthenes’ Πλατωνικός seems to be one of the most ancient sources.24

 Commentary In order to understand properly the problems connected with the mention of Eratosthenes in Expositio 81.17–82.5 Hiller, it is useful to consider the entire section of the text dealing with the difference between διάστημα and λόγος. If we consider the argumentative structure of 81.6–82.5 Hiller, it is clear that Theon repeats the same point twice, that is, that there is a difference between the two terms. If we look at the text, it is possible to divide the entire section into two parts: 81.6–

 22 See, e.g., Aristox. Harm. 8.12–10.11. As Barker (1989) II.5 states, this idea of intervals as linear distances seems to be one of the features of Aristoxenus’ approach which persists among his followers. 23 Indeed, Petrucci (2012) 47 considers the question of the difference between διάστημα and λόγος as a ζήτημα, a particular topic which has become traditional in the context of exegesis. For the exegesis κατὰ ζητήματα, see Ferrari (2000). Theon is not the only source for this Eratosthenic fragment. Porphyry, in his Commentary to Ptolemy’s Harmonica, mentions Eratosthenes for the same topic: the difference between διάστημα and λόγος. The fragment in Porphyry (In Pt. Harm. 91.4–10 Düring) is less rich in content if compared with the fragment in Theon. At least we find the same example (that of the double and the half) to explain the same point (to one interval two ratios correspond). Yet, Porphyry criticizes Eratosthenes because, according to him, the scholar of Cyrene not only would not have made evident the distinction between the two words (91.9ff. Düring), but also he would have used διάστημα instead of λόγος (92.24ff. Düring). It is Porphyry’s opinion (supported by the sources he mentions) that, in music, the two terms are interchangeable. Indeed, even though they belong to two different approaches to the harmonic theory (the Pythagorean school and the Aristoxenian school), these two terms can be used, at least in music, in reference to the same thing (that is a harmonic interval). To testify this terminological interchangeability Porphyry mentions many ancient authors who used διάστημα instead of λόγος (92.27–93.4 Düring) such as Plato in the Timaeus, above all, but also Dionysius of Halicarnassus, Archytas in De Musica and Euclid himself in the Sectio Canonis. 24 Doubts exist on the literary genre of this work: according to some scholars (Wolfer [1954] 4– 19; Hirzel [1963] 405–7; Fraser [1970] 180ff.), it was probably a dialogue in which Plato was one of the characters, maybe the main one. According to Geus (2002) 192ff. Πλατωνικός was a treatise.

150  Sara Panteri 16 and 81.17–82.5 Hiller. The dividing point is exactly the mention of Eratosthenes. The two parts have the same structure, yet there are significant differences. Both similarities and differences can be appreciated if we compare the two: Theo Sm. .- = A

Theo Sm. .-. = B

statement of a difference

διαφέρει δὲ διάστημα καὶ λόγος

Ἐρατοσθένης δὲ ἐν τῷ Πλατωνικῷ φησι μὴ ταὐτὸν εἶναι διάστημα καὶ λόγον

definitions

διάστημα: τὸ μεταξὺ τῶν ὁμογενῶν τε καὶ ἀνίσων ὅρων λόγος: ἁπλῶς ἡ τῶν ὁμογενῶν ὅρων πρὸς ἀλλήλους σχέσις

διάστημα: not defined λόγος: ἐστι δύο μεγεθῶν ἡ πρὸς ἄλληλα ποιὰ σχέσις

different conditions of existence: when διάστημα cannot exist

διάστημα: does not exist between λόγος: exists both ἐν διαφόροις ˂καὶ ἴσοι ὅροι ἐν ἀδιαφόροις˃ λόγος: between ἴσοι ὅροι is λόγος τῆς ἰσότητος example not given

proportion with four terms

when a διάστημα exists

not clarified but implicit

διάστημα: exists only ἐν ἀδιαφόροις ἢ κατὰ τὸ μέγεθος ἢ κατὰ ποιότητα ἢ κατὰ θέσιν ἢ ἄλλως ὁπωσοῦν.

different perspectives

διάστημα: ἓν καὶ τὸ αὐτὸ ἀφ’ ἑκατέρου ἑκάτερον λόγος: ἕτερος καὶ ἐναντίος ἑκατέρου πρὸς ἑκάτερον.

not clarified but implicit

example  and  as example: the interval is the same (); the ratio is different (/ is the half, / is the double: they are two different λόγοι)

the half and the double as example: the interval is the same; the ratio is different (the double to the half and the half to the double are two different λόγοι)

As can be seen, the structure of the two parts (here called A and B) shows several common points. First of all, the existence of a difference between διάστημα and λόγος is stated. Then, definitions are given. While in part A both διάστημα and λόγος are defined, in part B only λόγος has its own definition. It is clear from A that by defining both διάστημα and λόγος Theon thinks that the difference between these two concepts immediately arises as evident since the two definitions

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  151

differ.25 As a consequence of the definitions given (διό), διάστημα exists only between unequal terms, while λόγος, when the terms are equal, does exist and is one and the same – it is λόγος τῆς ἰσότητος. No example of this statement is given. On the other hand, in B, λόγος is defined and its existence between both different and not–different terms is stated. This statement needs the integration proposed by Hiller which seems to be necessary to the sense and to the structure of the passage. Clearly, without this integration, the opposition with διάστημα expressed immediately below would not make sense. Then, still in B, an example, introduced by οἷον, is provided. After this example, it is stated that a διάστημα can exist only between different terms and it is specified that this difference can be of three kinds plus one (κατὰ τὸ μέγεθος ἢ κατὰ ποιότητα ἢ κατὰ θέσιν ἢ ἄλλως ὁπωσοῦν). Finally, the last sentence of both A and B affirms the difference between διάστημα and λόγος from another point of view: to one διάστημα, two λόγοι correspond. While in A this difference is clearly explained and then an example is provided, in B the evidence has to be deduced from the example. It is worth noticing that both examples are constructed on the opposition between 1 and 2 or the half and the double.26 Once analysed the structure of the entire section, we could try to come back to the question of why Theon repeats the same argument twice. Since, as stated, the distinction between διάστημα and λόγος had become a major traditional topic, Theon could have intended to report a general distinction between the two at the beginning of the section; for this reason, he could have quoted Eratosthenes’ Πλατωνικός as an eminent confirmation.27 The fact that in the first part Theon gives both the definition of διάστημα and of λόγος deserves attention. When dealing with this opposition, Theon provides a definition of διάστημα which has an Aristoxenian flavour,28 yet implies the same terminology in the definition of  25 This is the sense of the causal clause introduced by ἐπειδή and of the opposition established by μέν... δέ… 26 For the possible Aristotelian flavour of this example, see Uguaglia (2011). 27 This would be also an argument in favour of the fact that the entire section 81.17–82.5 is Eratosthenic material. It is true, however, that while we have the certainty both that Eratosthenes did stress the difference between διάστημα and λόγος, and that he used the example of the double and the half, it might be that the proportion used as example is Theon’s insertion. However, it makes more sense that the entire section comes from Eratosthenes: Theon, indeed, had already clarified the difference between the two terms and had already said what he wanted to say about the topic. Why should he repeat the same ideas if he was not quoting entirely his authoritative source, that is Eratosthenes? 28 The formal definition of διάστημα in Aristoxenus (Harm. 20.20–21.1) is: διάστημα δ’ ἐστὶ τὸ ὑπὸ δύο φθόγγων ὡρισμένον μὴ τὴν αὐτὴν τάσιν ἐχόντων (‘interval is that which is bounded by two notes which do not have the same pitch’, tr. Barker [1989] II.136).

152  Sara Panteri λόγος. In turn, the definition of λόγος is halfway between the Euclidean definition (El. V def. 3)29 and the definition of διάστημα. And it is true that, even if the two terms are different, they can easily overlap, if only two conditions are set: (1) the terms (ὅροι) have to be different; (2) the direction of the ratio between the two terms has to be stated. In this case, διάστημα can be used instead of λόγος in the field of Harmonics without creating problems. Thus, the use Plato makes of διάστημα in the Timaeus, from the point of view of a harmonic theorist, is completely understandable.30 Still, the two words do not perfectly overlap. The definition of λόγος provided in the section behind the authority of Eratosthenes and reproducing the Euclidean definition (El. V def. 3) deserves attention. This nearly perfect overlap is made possible because the word used while referring to the terms of the proportion is μεγεθῶν. The choice of this word acquires even more relevance if we consider that in the other two places in the Expositio31 where λόγος is defined, it is said that it is a relation (σχέσις) between two ὅροι ὁμογενεῖς (‘homogeneous terms’). In both definitions, λόγος is seen as a relation between two things that are quantifiable. Saying that something is quantifiable in Greek as well as in modern thought implies the condition of measurability. With this idea in mind the term λόγος can be used in a technical sense meaning ratio.32 If we speak about a ratio between μεγέθη, as in the Euclidean definition, we have in mind geometrical quantities.33 It is remarkable that Eratosthenes specifies that, contrary to διάστημα, λόγος can exist also between ‘non– different (scil. magnitudes)’. As Theon has previously said, a λόγος exists even when the two terms of a ratio are equal, but in this case, it has a particular form and name, that of ἰσότης. This idea should be connected with other two fragments attributed to Eratosthenes and reported by Theon (82.22–83.334 and 111.10–  29 Eucl. El. V def. 3 λόγος ἐστὶ δύο μεγεθῶν ὁμογενῶν ἡ κατὰ πηλικότητά ποια σχέσις. (‘a ratio is a sort of relation in respect of size between two magnitudes of the same kind’, tr. Heath [1956] II.114). 30 It is worth remembering that for Plato the use of the term διάστημα was not problematic at all. As Raffa (2013) 250 states, ‘at that stage in the development of harmonics the word διάστημα had not yet acquired any problematic overtones’. 31 I.e. 73.16ff. Hiller λόγος δέ ἐστιν ὁ κατ’ ἀνάλογον δυοῖν ὅρων ὁμογενῶν ἡ πρὸς ἀλλήλους [αὐτῶν] ποιὰ σχέσις (‘λόγος in reference to proportion (i.e. ratio) is a kind of relation between two homogeneous terms between each other’). 32 As Barker (1994) 115 notes ‘by the early fourth century [...] pitch itself is a quantitative variable [...] We need only note the crucial consequence that pitches themselves, and not merely the dimensions of sound–producing objects, can be described as standing to one another in numerical ratios’. 33 I.e. (points), lines, surfaces or solids. 34 This fragment is extremely problematic because of its textual corruption.

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  153

13 Hiller). These fragments create a hierarchy among τὸ ἴσον/ἡ ἰσότης, λόγος and ἀναλογία. τὸ ἴσον is there described as the principle of λόγος since, Theon explains, from τὸ ἴσον/ἡ ἰσότης it is possible to find out sequences of ratios which create geometrical progressions (ἀναλογίαι) because the ratio between each number with the following one is the same. Nevertheless, Theon says that the demonstration of this ‘rule’ does not appear in Eratosthenes and for this reason Theon borrows it from Adrastus. Now, let us return to Theo Sm. 81.17–82.5 Hiller. After clarifying the point of the existence of λόγος between non–different terms, an interesting statement is set and two particular ratios are considered. Even though the term ἀναλογία does not appear, the expression ἐν ᾧ λόγῳ [...] ἐν τούτῳ [...], which links the two ratios, evidently describes a proportion and, in particular, a geometrical proportion. The first ratio is between τὸ αἰσθητόν and τὸ νοητόν, the second one is between δόξα and ἐπιστήμη. Therefore, these two ratios are equal. Immediately after, it is specified that τὸ νοητόν differs (διαφέρει) from τὸ ἐπιστητόν35 in the same way ἡ δόξα differs from τὸ αἰσθητόν. Since this proportion appears in a work entitled Πλατωνικός, the first idea is to consider the Platonic corpus in order to find a possible model. It has been noticed by scholars, especially by Dörrie (1981 and 1987)36 and Geus (2002),37that the proportion established in our fragment recalls the similitude of the line in Plato’s Resp 6.509d–511e. There Socrates uses this image in order to explain to Glaucon his ontology and consequent epistemology. A line (γραμμή) is divided into two unequal parts; one (probably the shorter) is associated to τὸ ὁρώμενον (the domain of the visible, that is the objects of perception), while the other part (probably the longer) is associated to τὸ νοητόν (the intelligible objects). Then, each part is divided again according to the same ratio (ἀνὰ τὸν αὐτὸν λόγον, 509d) in which the line was divided at the beginning. As a result, the three following par-

 35 In my opinion, this is a mistake of the copyist. The word that should appear there is ἑπιστήμη. 36 Cf. Dörrie (1981) 34–37 and (1987) 370. 37 Cf. Geus (2002) 145 n. 23 and n. 25.

154  Sara Panteri titions demarcate segments which are related to each other according to a geometrical proportion.38 Firstly, each segment is associated with the objects of different mental states, secondly, in 7.533c–534a, each section is connected with the mental states. A possible representation is: ὁρώμενον : νοητόν = εἰκόνες : ζῷα etc. = lower νοητά : higher νοητά39

for the objects of mental states, and δόξα : νοήσις = εἰκασία : πίστις = διάνοια : ἐπιστήμη

for the mental states. As can be seen, the proportion established by Eratosthenes and the similitude of the line share at least five features: 1) the used terminology (τὸ νοητόν, δόξα, ἐπιστήμη);40 2) the ontological and epistemological idea of a division between the sphere of the intelligible and that of the sensible; 3) the idea to associate the mental states to the objects of these mental states; 4) the idea to treat these terms as quantities (Plato speaks about a concrete line, which is cut);41 5) the subsequent idea to create ratios and geometrical proportions between these terms.

 38 It should be noticed that this kind of proportion, made out of three λόγοι, implies that the two central segments of the line have necessarily the same length: A:B=d:e=e:f. 39 Here the terminology of Adam (1963) II. 64ff. is adopted. In particular, as for the νοητόν’s section, Plato distinguishes between the case in which the soul carries his research from hypotheses in order to reach conclusions (as it happens in mathematical subjects), and the case in which the soul starts from hypotheses but with the aim to reach a principle which is ἀνυπόθετος, ‘not hypothetical’ (Resp. 6.510b). 40 As for the terminology used and, in particular, for the words ἐπιστήμη and νόησις; see Cornford (1987). 41 Notargiacomo (2009) 201 stresses this point.

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  155

Picture 1: Visual scheme

It has been demonstrated in a sufficient way that the model for the proportion most likely comes from Plato’s Republic. Nevertheless, it is worth noting that the context in which this proportion is used is completely different from the Platonic one. In Plato, the proportion is used to make Glaucon understand Plato’s own epistemology; therefore the main reason why a proportion is established is to make clear the relation between the states of mind and the associated objects. In

156  Sara Panteri Eratosthenes’ fragment the proportion is within a technical context. As said before, the fact that the proportion is introduced by the neuter adverbial οἷον means that it is supposed to be considered as an example. And, since it follows the statement that a λόγος can exist both between different and not–different terms, it should explain somehow this particular point. This, again, reinforces the idea that the proportion created between mental states and their respective objects has to be necessarily a geometrical proportion: only the ratios within a geometrical proportion can be reduced to ἰσότης, by exchanging the means with each other or the extremes with each other.42 The privileged position occupied by the ratio of equality is perfectly explained by the so–called rule of Adrastus (107.15– 111.9 Hiller), by means of which it is possible to find out series of geometrical proportion starting from 1,1,1 and to reduce each geometrical proportion to 1,1,1. As said above, Eratosthenes himself sustained that the principle of ἀναλογία is λόγος and the principle of λόγος is τὸ ἴσον. A philosophical implication could be found behind this discourse. Usually some of the problems which appear when the scholars43 work on the similitude of the line are: (1) how to divide it? That is, which part is longer and which one is shorter (the intelligible or the sensible)? (2) Which are the consequences of the fact that the central segments are equal? These questions are striking since they  42 In other words, if we consider the proportion a:b=c:d and if, for instance, we interchange the means, each other, the proportion acquires the structure a:c=b:d (which is, again, a geometric proportion). This is possible since the terms a and c, on the one hand, and b and d, on the other, are equivalent. We could say that, since the terms are equivalent, between them there is no διάστημα but only a λόγος. Now, a particular case is a:b=a:b , which expresses an equality between two ratios which are, quantitatively speaking, exactly the same. In this case, if we interchange the means, we get an equality between two ratios whose terms are the same: a:a=b:b. Quantitatively speaking, between the terms of each ratio there cannot exist a διάστημα but only a λόγος. I think that, having in mind the diagram of the similitude of the line, the proportion built in Eratosthenes’ fragment could be represented as a:b =A:B, where the difference between a and A is not quantitative but qualitative (one is the mental state, the other represents the objects). The conclusion is that, if we interchange the means, we can still have the ratio a:A, but between the terms there is no διάστημα (from the quantitative point of view they correspond to the same section of the line). 43 The bibliography on the similitude of the line is huge. Among various problems the following ones can be listed: (1) the division (using a commensurable or an incommensurable ratio); (2) the orientation (vertical, horizontal, diagonal); (3) the equality of the central segment (was Plato aware of it or not?); (4) the attribution of the longer part to the intelligible and of the smaller part to the sensible; (5) the problem of the relationship original–copy; (6) the objects of διάνοια (and of mathematics); (7) whether the sensible is knowable or not?; (8) the comparison between the similitude of the line with other similitudes (such as the cave and the sun). See, LaFrance (1986); Dreher (1990); Balashov (1994); Smith (1996).

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  157

imply the understanding of Plato’s ontology and epistemology. However, it seems like in the Eratosthenic proportion, these issues are not particularly problematic: since the proportion is geometrical, it can always be reduced to the equality. This point is stressed by the way Eratosthenes uses the proportion. If there is a difference, the difference is between mental states and objects. It looks more like a difference in quality, rather than in quantity. Now I would like to make some remarks on the terminology employed. It is evident that the concepts of νοητόν, αἰσθητόν, ἐπιστήμη and δόξα occupy a fundamental position within Plato’s epistemology and appear in many dialogues.44 However, at a certain point in the Platonic (or, better, Platonic Pythagorean) tradition these four concepts have been crystallized in the form of the τετρακτύς. Theon, in particular, lists eleven τετρακτύες: the eighth (97.25–98.7 Hiller) is the τετρακτύς of the states of mind (νοῦς, ἐπιστήμη, δόξα, αἴσθησις). Theon says explicitly (l.c.) that they can be associated to the numbers of the Pythagorean τετρακτύς, the first listed by Theon (93.17–94.11 Hiller), which is in the form of the arithmetic progression 1, 2, 3, 4. Therefore: νοῦς=1, ἐπιστήμη=2, δόξα=3, αἴσθησις=4. All the τετρακτύες from the fourth to the eleventh seem to be able to be reduced to the first numerical one. The only other numerical τετρακτύς ‒ the second one listed by Theon (94.11–96.8 Hiller) ‒ is called Platonic. This can appear in two ways, namely the two geometrical progressions (which can be seen as geometrical proportions) used by Plato to describe the division of the matter of the World Soul in the Timaeus: 1, 2 , 4, 8, and 1, 3, 9, 27. The third τετρακτύς (point–line–plane–solid) is associated to these two progressions (96.12–97.3 Hiller). As it can be seen, the main and fundamental difference between the Pythagorean and the Platonic τετρακτύς lies in the fact that, while the former is in the form of an arithmetic proportion,45 the latter is a geometrical proportion. As Petrucci suggests (2012),46 this evidence is extremely remarkable since in the Timaeus the generation of the World Soul depends exactly on the numbers of the

 44 See, in particular, Smp. 210c; Phlb. 66b; Resp. 6.511d, 533e, 9.585b; Tht. 151e; Prm. 155d; Tim. 28b, 37b, 51d. 45 Here a terminological clarification is necessary. The only properly called proportion (in Greek ἀναλογία) is the geometric one, while the arithmetic and the harmonic proportions are more properly only μεσότητες (‘means’). Still, the Greek terminology is often not coherent since there are many cases where ἀναλογία is used in reference to arithmetic or harmonic μεσότητες (see, e.g., one of the most striking examples: Archytas VS 47 B2; see Huffman [2005] 180). But the awareness of the terminological difference is well attested (for instance, by Theon himself, who reports what has already been noticed by Adrastus; see Theo Sm. 106.12–20 Hiller). In this paper I will use the term proportion for the geometric, the arithmetic and harmonic proportion. 46 Cf. Petrucci (2012) 412, 416ff.

158  Sara Panteri second τετρακτύς. Its numbers, Theon affirms (95.14–16 Hiller), provide a more complete set of consonant ratios, that is all the ratios identified by the first τετρακτύς47 plus the tone (9/8). Now we should consider again the proportion in the Eratosthenic fragment. As we said, the terms involved recall that of the eight τετρακτύς, but with an interesting difference: instead of two mental states (νοῦς and αἴσθησις), we have the two substantivized verbal adjectives τὸ νοητόν and τὸ αἰσθητόν which refer to the objects of those mental states. Still, the terminological overlap is plain and this is the reason why it is likely to say that Eratosthenes, even having in mind the passage from the Republic, has been affected by the Platonic tradition of the τετρακτύς of the four mental states. In Plato’s dialogues, we do not find a τετρακτύς of the four mental states expressed as such, with a fixed structure, nor do we find any association between these forms and numbers. Yet, as Aristotle witnesses, this idea was elaborated in the Academy after Plato.48 And, interestingly, this material also appears in the mysterious so called  47 I.e. 2/1 (the octave), 3/2 (the fifth), 4/3 (the fourth), 4/1 (the double octave). 48 See Brisson (1998) 327. But we have to take into consideration the fact that we are dealing with a topic that implies the problematic and ongoing question which involves Plato, his theory of first principles and his ‘unwritten doctrines’. For my purposes, I restrict myself to some considerations and I will take Burkert (1972) as starting point. Also Gaiser (1994) should be taken into consideration. First of all the testimony of Aristotle, who is supposed to be our main source for ‘Plato’s unwritten doctrines’, deserves to be mentioned. In a famous passage of De Anima (404b 16–27), the Stagirite deals exactly with the τετρακτύς of the mental states and with its association to 1, 2, 3, 4 in a context where the Timaeus is mentioned. τὸν αὐτὸν δὲ τρόπον καὶ Πλάτων ἐν τῷ Τιμαίῳ τὴν ψυχὴν ἐκ τῶν στοιχείων ποιεῖ· γινώσκεσθαι γὰρ τῷ ὁμοίῳ τὸ ὅμοιον, τὰ δὲ πράγματα ἐκ τῶν ἀρχῶν εἶναι. ὁμοίως δὲ καὶ ἐν τοῖς περὶ φιλοσοφίας λεγομένοις διωρίσθη, αὐτὸ μὲν τὸ ζῷον ἐξ αὐτῆς τῆς τοῦ ἑνὸς ἰδέας καὶ τοῦ πρώτου μήκους καὶ πλάτους καὶ βάθους, τὰ δ’ ἄλλα ὁμοιοτρόπως· ἔτι δὲ καὶ ἄλλως, νοῦν μὲν τὸ ἕν, ἐπιστήμην δὲ τὰ δύο (μοναχῶς γὰρ ἐφ’ ἕν), τὸν δὲ τοῦ ἐπιπέδου ἀριθμὸν δόξαν, αἴσθησιν δὲ τὸν τοῦ στερεοῦ. Similarly, Plato in the Timaeus derives soul from the elements, saying that like is known by like, and that things are derived from the first principles. Similarly, in our De Philosophia we showed that in his view the Animal itself is derived from the Idea of the One and the primary length, breadth and depth and all other things are correspondingly. He also puts it otherwise, saying that reason is the one, knowledge the two (since it moves only in one direction), opinion the number of the plane, sense–perception that of the solid (tr. Ross 1961, 172). While talking about the elements of the soul, Aristotle mentions Plato’s Timaeus. Then he refers to a work titled De Philosophia in which it was showed that αὐτὸ τὸ ζῷον derives from the one, the idea of length, breadth and depth. Ross (1961, 177ff.) thinks that here Aristotle is referring to

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  159

Pseudo–Pythagorean texts, which might go back to the Hellenistic period49. In any case, as said, Eratosthenes inserts these terms within a geometrical proportion. Even if the association of the terms of the Eratosthenic proportion with exact  his own (lost) work titled Περὶ φιλοσοφίας and that the doctrine here mentioned should be attributed to Plato. Aristotle reports the idea that the series 1,2,3,4 was associated both to the four dimensions and to the four mental states. Since, actually, no mention of the fact that Plato held these ideas is explicitly given, this passage leaves interpretative doubts. Burkert (1972) 26, aware of the debate on this issue (25 n. 48 and 49), states: ‘the identification of the One with the mind (νοῦς) is attested for Xenocrates, as well as the series knowledge–opinion–sensation, and he knows the derivation–series line–plane–solid [...]. But then Xenocrates’ definition of soul is introduced as something new, a conclusion of ‘some’; therefore what precedes can not refer to Xenocrates, and it is a tempting conclusion that what follows the citation of the Timaeus and its interpretation in the work On Philosophy is a reference to the ‘unwritten doctrines’’. It is natural to look at the Platonic dialogues and see whether the topic appears somewhere. As Burkert notes (l.c.), the only locus which could be related to this issue is Leg. 894a: γίγνεται δὴ πάντων γένεσις... ὁπόταν ἀρχὴ λαβοῦσα αὔξην εἰς τὴν δευτέραν ἔλθῃ μετάβασιν καὶ ἀπὸ ταύτης εἰς τὴν πλησίον, καὶ μέχρι τριῶν ἐλθοῦσα αἴσθησιν σχῇ τοῖς αἰσθανομένοις. The origin of each thing takes place... when a first principle, taking on increment [the line], passes into its second transformation [the plane] and from this to its neighbour [the solid], and having made three transformations makes perception possible to those who perceive it (tr. Burkert [1972] 26). The translation of the passage given above includes the commonly accepted interpretation. According to this interpretation, behind this passage, there would be a ‘geometrical allegory’ and, accordingly, perception could be applied only to the solid reality. However, as Burkert (l.c.) notes, no association with numbers is given. Still, in Speusippus (see, in particular, fr. 28 Tarán, which goes back to a work of Speusippus titled Περὶ Πυθαγορικῶν ἀριθμῶν) and Xenocrates (see, in particular, frr. 83, 118, 120 Isnardi–Parente) we can find references to the topic of the sequences of the four dimensions/four geometrical objects point–line–surface–solid, of the numbers of the τετρακτύς and of the four mental states. Likely, these ideas, rooted in Plato’s own philosophy, had to have been developed within the Academic environment. 49 In particular, a fragment of a work titled Περὶ νοῦ καὶ αἰσθήσεως, attributed to Pseudo–Archytas and transmitted by Iamblichus in De communi mathematica scientia (35.27–36.18), presents an interesting combination between the τετρακτύς of the mental states and the similitude of the line ἒτι δὲ σαφέστερον Ἀρχύτας ἐν τῷ Περὶ νοῦ καὶ αἰσθήσεως διακρίνει τὰ κριτήρια τῶν ὄντων, καὶ τὸ τῶν μαθηματικῶν οἰκειότατον κριτήριον παρίστησι διὰ τούτων· ‘ἐν ἁμῖν’ γὰρ ‘αὐτοῖς’, φησί, ‘κατὰ ψυχὰν γνώσιές εἰσι τέσσαρες, νόος ἐπιστάμα δόξα αἴσθησις, ὧν αἱ μὲν δύο τοῦ λόγου ἀρχαί ἐντι, οἷον νόος αἴσθασις, τὰ δὲ δύο τέλη, οἷον ἐπιστάμα καὶ δόξα· τὸ δ’ ὅμοιον ἀεὶ τοῦ ὁμοίου γνωστικόν. φανερὸν ὦν ὅτι ὁ μὲν νόος ἐν ἁμῖν τῶν νοατῶν γνωστικόν, ἁ δὲ ἐπιστήμη τῶν ἐπιστατῶν, ἁ δὲ δόξα τῶν δοξαστῶν, ἁ δὲ αἴσθασις τῶν αἰσθατῶν· διόπερ ὦν

160  Sara Panteri numbers is not a crucial issue, it is interesting to deal with the evidence that Eratosthenes does not conceive these four terms within an arithmetic proportion (which might mean an association to the Pythagorean τετρακτύς).50 As said before, the Platonic passage where the problem of the difference between διάστημα and λόγος appears is the Timaeus. This is testified by Porphyry’s text, which is the other testimony for Eratosthenes’ fragment. As Barker (2003) notices, the entire section (90.24–93.4 Düring) gives us an interesting picture: many ancient scholars dealt with this terminological distinction and many of them raised the problem within Commentaries or studies on the Timaeus (at least this is the case for Aelianus, Demetrius and Panaetius). Among the authors mentioned by Porphyry, Eratosthenes seems to be the oldest source to oppose διάστημα and λόγος seems to be Eratosthenes. Given that, it is reasonable to focus  δεῖ μεταβαίνεν ἀπὸ μὲν τῶν αἰσθατῶν ἐπὶ τὰ δοξαστὰ τὰν διάνοιαν, ἀπὸ δὲ τῶν δοξαστῶν ἐπὶ τὰ ἐπιστατά, καὶ ἀπὸ τούτων ἐπὶ τὰ νοατά· ταῦτα δὲ σύμφωνα ποιητά, θεωρούμενα δι’ αὐτῶν ἀλάθεα. διωρισμένων δὲ τούτων τὰ μετὰ ταῦτα δεῖ νοῆσαι. καθάπερ γὰρ γραμμὰν δίχα τετμαμένην καὶ ἴσα πάλιν ἑκατέρων τμήματα τετμαμένα ἀνὰ τὸν αὐτὸν λόγον, καὶ οὕτω διῃρήσθω καὶ τὸ νοατὸν ποττὸ ὁρατόν Even more clearly Archytas in his On Knowing and Sense Perception distinguishes the criteria of beings, and sets that one criterion, the most common to mathematics by means of these words: ‘by us’, he says indeed ‘as regards the soul there are four means of knowing, mind, knowledge, opinion and sense perception, two of which are the principles of reasoning, mind and perception, the other two are the ends, that is knowledge and opinion; the similar knows always the similar. It is evident that our mind allows us to know the intelligible, our knowledge the knowable, our opinion the opinable, the perception the perceptible; this is the reason why our thought passes necessarily from the perceptible to the opinable, from the opinable to the knowable, from the knowable to the intelligible. When these things are in harmony, by means of these it is possible to see the truth. Once these divisions are made, it is necessary to consider the things after these. Like a line divided into two sections and these sections divided into two again according to the same ratio, the intelligible has to be divided in reference to the visible. Even if there are many questions around the pseudo–Pythagorean material, still it is worth noting that immediately after having explained the mental states, the unknown author of this mysterious work deals with the similitude of the line, which appears in Plato’s Republic. The pseudo– Pythagorean texts probably go back to the Hellenistic period. They are usually distinguished into two categories: the ones attributed to the relatives of Pythagoras and the ones attributed to other Pythagoreans (often to the more well–known names, like Archytas). Both the date and the place are uncertain. Thesleff (1961) 50–75 criticizes the scholars who propose a late date (after 1st BC) for these texts. According to Thesleff, there are good reasons to date back these texts to the 2nd or even the 3rd century BC. Cf. also Centrone (2014). 50 Even if we should keep in mind that the numbers 1, 2, 3, 4 carry not only an arithmetical value but they also have philosophical implications.

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  161

also on the Timaeus and how the four mental states are treated in it. Two sections deserve attention: the very beginning of Timaeus’ monologue (between 27d and 33b) and the passage a little before the description of the four regular solids (51d– 52a). In the former section, Timaeus is making a distinction between Being and Becoming. While the first one is περιληπτόν by νόησις with λόγος, the second one is δοξαστόν by δόξα with ἄλογος αἴσθησις. While the first one does not have generation, the second one has it. In particular, the world belongs to Becoming because it γέγονεν· ὁρατὸς γὰρ ἁπτός τέ ἐστιν καὶ σῶμα ἔχων, πάντα δὲ τὰ τοιαῦτα αἰσθητά, τὰ δ’ αἰσθητά, δόξῃ περιληπτὰ μετ’ αἰσθήσεως, γιγνόμενα καὶ γεννητὰ ἐφάνη (28b–c).51 But it has been generated as an image, εἰκών of the eternal world, which is its example, παράδειγμα (29b). The same opposition, expressed with the terms ὁρατόν and νοητόν, appears thereafter, when Timaeus speaks about the reason why the Demiurge has created the world (30b–31a). As for the second section mentioned above (51d–52a), the opposition is even more interesting: a distinction, on the one hand, between νοητόν and αἰσθητόν, on the other hand, between νόησις and δόξα is clearly established. In conclusion, in the Timaeus we do not find a striking and comprehensive image like the similitude of the line in the Republic, but the stress on the relation and the tension between sensible/thinkable and respective objects is always there. Furthermore, we should consider also the role of proportion in the Timaeus. As stated, in the Republic, Plato chooses a proportion, a geometrical proportion, to explain in the clearest way his epistemology. But it is precisely in the Timaeus that proportion becomes a fundamental topic. Indeed, in that dialogue the geometrical proportion is used to structure both the world and the World Soul. As for the World Soul, we have already mentioned how all the three kinds of proportion are involved in the divisio animae. As for the generation of the world, the passage of interest is the following (32a–b): εἰ μὲν οὖν ἐπίπεδον μέν, βάθος δὲ μηδὲν ἔχον ἔδει γίγνεσθαι τὸ τοῦ παντὸς σῶμα, μία μεσότης ἂν ἐξήρκει τά τε μεθ’ αὑτῆς συνδεῖν καὶ ἑαυτήν, νῦν δὲ στερεοειδῆ γὰρ αὐτὸν προσῆκεν εἶναι, τὰ δὲ στερεὰ μία μὲν οὐδέποτε, δύο δὲ ἀεὶ μεσότητες συναρμόττουσιν· οὕτω δὴ πυρός τε καὶ γῆς ὕδωρ ἀέρα τε ὁ θεὸς ἐν μέσῳ θείς, καὶ πρὸς ἄλληλα καθ’ ὅσον ἦν δυνατὸν ἀνὰ τὸν αὐτὸν λόγον ἀπεργασάμενος, ὅτιπερ πῦρ πρὸς ἀέρα, τοῦτο ἀέρα πρὸς ὕδωρ, καὶ ὅτι ἀὴρ πρὸς ὕδωρ, ὕδωρ πρὸς γῆν, συνέδησεν καὶ συνεστήσατο οὐρανὸν ὁρατὸν καὶ ἁπτόν.

 51 ‘It has come to be; for it can be seen and touched and it has a body, and all such things are sensible; and, as we saw, sensible things, that are to be apprehended by belief together with sensation, are things that become and can be generated’ (tr. Cornford [1937] 22).

162  Sara Panteri Now if it had been required that the body of the universe should be a plane surface with no depth, a single mean would have been enough to connect its companions and itself; but in fact the world was to be solid in form, and solids are always conjoined, not by one mean, but by two. Accordingly the god set water and air between fire and earth, and made them, so far as possible, proportional to one another, so that as fire is to air, so is air to water, and as air is to water, so is water to earth, and thus he bound together the frame of the world visible and tangible (tr. Cornford [1937], 44).

The point here stressed is the fact that, since the world is a solid reality, one mean proportional between fire and earth is not sufficient but, on the contrary, four terms are necessary. Notably, the four elements are ἀνὰ τὸν αὐτὸν λόγον, that is they are structured according to a geometrical proportion. This is the reason why by means of this proportion of elements, the world is said to be ὁρατὸν καὶ ἁπτόν, that is, perceptible. In other words, in order to have αἴσθησις of something, this something has to be solid. Interestingly enough, the world is a solid reality just as the four elements, which are its components, are σώματα. As a consequence, τὸ δὲ τοῦ σώματος εἶδος πᾶν καὶ βάθος ἔχει. τὸ δὲ βάθος αὖ πᾶσα ἀνάγκη τὴν ἐπίπεδον περιειληφέναι φύσιν (53c).52 Here there is the idea that from a bi–dimensional reality, we have to reach a tri–dimensional reality. The same happens in the difficult passage which describes the generation of the four regular solids from triangles. These solids are then associated with the four elements (56a). The necessity to find two means proportional while generating the world because this is a solid reality is consistent with the necessity to extend the divisio animae up to the solid numbers 8 and 27. In my opinion, the reference to a plane (ἐπίπεδον) or a solid (στερεοειδῆ) nature of the world should remind us of the sequence σημεῖον, γραμμή, ἐπίπεδον, στερεόν, that is, the third τετρακτύς listed by Theon. Interestingly enough, Theon associates the third τετρακτύς both with the Pythagorean one (1, 2, 3, 4) and with the Platonic one (1, 2, 4, 8; 1, 3, 9, 27).53 From a numerological point of view, this second association is better, since geometrical  52 ‘[...] and all body has depth. Depth, moreover, must be bounded by surface; and every surface that is rectilinear is composed by triangles’ (tr. Cornford [1937] 212). Cornford (1937) 212ff. n. 4 explains in this way why in the Timaeus (contrary to Leg. 894a), after mentioning the triangles (that is, surfaces), there is no mention of lines and points: ‘The Timaeus is a myth of the physical world, and therefore has no need to go further back than the surface, the stage where in descending from the ἀρχή the third dimension becomes possible; for without the third dimension there is no sensation’. 53 The third τετρακτύς, which is point–line–surface–solid, is explicitly associated with the numbers of the Platonic τετρακτύς (96.12–97.3). But then the sixth τετρακτύς, the one of the growing (seed–things which grow in length–things which grow in breadth–things which grow in depth), shows again the series point–line–surface–solid but this time associated to the Pythagorean 1,2,3,4 (97.17–20 Hiller).

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  163

solids find their equivalents in solid numbers, like 8 and 27. In conclusion, what seems to appear from this reasoning is the fact that in Plato there are the grounds to establish an association between perception, solid nature, depth/solid and solid number. And these elements have to be considered fourth terms within series of elements which can be consistently and properly described as geometrical proportions with four terms.

 Conclusion Eratosthenes’ interest for Plato’s philosophy should have been significant if he got the nickname of second or new Plato. Nevertheless, the text probably more devoted to the topic, the so–titled Πλατωνικός, is extant only in a few fragments, reported by Theon, a late source. In one of those fragmenta, Eratosthenes deals with a problem which is both terminological and substantial. The scholar of Cyrene, who was also an eminent philologist, stresses a terminological problem: διάστημα has not the same meaning as λόγος. As another important testimony for the same Eratosthenic fragment, that is Porphyry, confirms, we have good reasons to think that the Platonic passage which made this clarification interesting for a scholar of Platonic Philosophy was Plato’s Timaeus (34b–36d).54 Definitely, in the field of harmonic theory these two words occupy a primary position and can be referred to the same concept. Yet, for instance, they are also used in the field of geometry and, in that case, the overlap is not correct.55 One of the main differences between the two terms is that λόγος, unlike διάστημα, can exist between two not–different terms. In order to clarify this point, Eratosthenes establishes a geometrical proportion between the four mental states and their objects, where the equality of ratios allows seeing also equality between the two terms of each ratio. Reciprocal relations and proportional structures among mental states

 54 According to Barker (2003) 76ff., the case–sample of the difference between διάστημα and λόγος and its importance among the sources mentioned by Porphyry, testifies that during the Hellenist period there was an attempt to combine the two theoretical perspectives, the Pythagorean one and the Aristoxenian one. This attempt was carried on not only within the Peripatetic school but also by other authors who, in particular, worked on Plato’s Timaeus. And, among the authors who worked on Platonic Philosophy during the Hellenistic period, we have to place also Eratosthenes. 55 In Euclid (see, e.g., El. I dem. 3) διάστημα is the distance between a point and the centre of a circle, that is, the radius, used in order to construct the circle.

164  Sara Panteri and sensible and thinkable objects are established in Plato’s dialogues and, noticeably, in the Timaeus. But the idea to create a proportion between these philosophical concepts seems to lie on the similitude of the line in Plato’s Republic. Since very little is extant of Eratosthenes’ Πλατωνικός, caution is mandatory. However, if we consider what we can reconstruct from the early commentaries on Plato’s Timaeus, the commentators examined and tried to explain those passages considered obscure, especially for their technical and mathematical content. Among those, the creation of the World Soul in the Timaeus occupied one of the first positions.56 In the commentary tradition on Plato’s dialogues it is custom to quote passages from other dialogues while commenting on a particular passage of a certain dialogue. This exegetical method perfectly fits the expression Πλάτωνα ἐκ Πλάτωνος σαφηνίζειν, a meaningful coined phrase from the best Alexandrian philological tradition.57 Without speculating too much on the Eratosthenic fragment, it is nonetheless true that the Platonic terminological problem comes from the Timaeus, but, to explain it, an example taken from the Republic which involves the similitude is used. Interestingly, the shape that the Eratosthenic proportion takes, especially from the terminological point of view, reminds us of the τετρακτύς of the mental states which, developed within the Academic environment, can be referred back to the terminology used by Plato in different dialogues. What is interesting is that, in order to work as an example, this proportion has to have a geometrical structure; it has to be a proper ἀναλογία. So, in a context which is very technical and which ultimately belongs to the field of harmonic theory, Eratosthenes did not choose to use numbers or geometrical quantities, but he imported traditional Platonic epistemological material and used it as example. The importance and the value that Plato attributes to the geometric proportion are quite evident, especially in the Timaeus. There the four elements which create the world are arranged according to a geometrical proportion (in order for the world to be perceptible, as it is explicitly said) and, similarly, the World Soul is extended until the solid numbers 8 and 27. In the Timaeus we can also find reference to the sequence plane–surfaces–solids. Therefore the idea of ordering the four mental states in a geometrical proportion is something which is consistent with what we find in Plato. Already from this short fragment, it is  56 As Isnardi Parente states (1998) 221 ‘l’esegesi del passo del Timeo relativo all’anima del mondo appare essere cominciata subito nell’Accademia’. Indeed, we can mention Xenocrates (mentioned by Plutarch, De an. proc. 1012b–f) and Crantor (defined by Proclus, In Tim. 1.75.30– 76.2 Diehl ὁ πρῶτος τοῦ Πλάτωνος ἐξηγητής). 57 This expression appears for the first time in Procl. Theol. Plat. 1.10.1–4, but this does not mean that the method was not used before (see, Ferrari [2000] and Petrucci [2012] 44ff.). As for the Aristarchean διόρθωσις, see e.g. Montanari (1998).

Eratosthenes’ Πλατωνικός between Philosophy and Mathematics  165

evident that Eratosthenes’ interest for Plato’s philosophy was particularly focused on some philosophical issues which were in a way connected to mathematical (broadly speaking) concepts. In this light, Eratosthenes seems to well deserve the prominent position Theon of Smyrna gave him in his Expositio.

Nicola Reggiani

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus Between Ancient Text Production and Modern Digital Representation Fragments are much more important than books because they testify to texts potentially at the apex of an ancient period and now, for whatever reason, no longer extant. The claim of a wisdom from the ancient period extant only in fragments represents a secret apologetic weapon directed against the supremacy of the majority. […] Fragments of ancient wisdom constitute that literary past which the libraries want to canonize in their attempt to establish their supremacy over other cultural moving forces of ancient society.1

Ancient medical prescriptions (usually also known as ‘recipes’) are a very peculiar text typology. They are at the same time therapeutical / pharmaceutical indications delivered by practicing physicians to the users (whether patients, colleagues, or pharmacists)2 and ways of handing down an articulated set of knowledge.3 Based as they are on both scientific experience and practical know– how, they essentially aim at preserving and transmitting diagnostic–therapeutic records for future replication. It is apparent that such reference tools are subject to modifications, adaptations, and updates according to actual clinical cases and individual expertise.4 Medicine is not an exact, established science, but an evolving set of strategies that imply theoretical reflection and practical action, both of

 1 Veltri (2006) 91. The present contribution stems from the project ‘Online Humanities Scholarship: A Digital Medical Library Based on Ancient Texts’ (ERC–2013–AdG DIGMEDTEXT, Grant Agreement no. 339828, Principal Investigator Prof. Isabella Andorlini) held at the University of Parma on funding of the European Research Council (see http://www.papirologia.unipr.it/ERC). 2 Cf. Nutton (1995). 3 The most complete structure of a recipe usually comprises, after the title (prographe) with the name of the remedy (usually after its shape, format, or original author), the therapeutical indication (epangelia), a pharmacological section describing its composition in details (synthesis or symmetria), and practical directions for preparation and administration (skeuasia): cf. Marganne (2004) 78–79 and Andorlini (2007) 26. They can be more or less abridged according to their final destination – the plain lists of ingredients, with very quick reference to composition and therapy, are the closest to our concept of ‘recipe’. 4 Cf. Andorlini (2017) 26. https://doi.org/10.1515/9783110632590-012

168  Nicola Reggiani which must be adjusted and fine–tuned on the ground of experience.5 It is, in a sense, a ‘reperformance’;6 and even when this complex of ‘liquid’ knowledge is entrusted to writing, this happens in fragmented textual forms that resemble very much the oral discourse. ‘Fragment’ is, indeed, the baseline keyword to understand the textual transmission of medical recipes in the papyri: prescriptions, under whichever format they may come, are always fragments of an oral discourse, subject to adaptations and updates, and they keep this distinct appearance even in their written form. This will be the general bottom line of the present contribution, which will be focused on the transmission of medical prescriptive texts as attested by the papyrological sources from Graeco–Roman Egypt,7 as a key to understand also the earlier phases of the process, as well as its contemporary outcome in terms of digital representation. An early stage of oral transmission of medical prescriptions can be ascertained from the survival of characteristic expressive modules in the later extant texts, repeated according to precise and constant schemes (formulaic structures, technical language) – a feature that occurs also in magical–religious rituals and juridical regulations, which partake in the same global set of knowledge that can be defined as ‘performative wisdom’.8 When, at some point, they are entrusted to writing on material supports, it is mainly for practical purposes,9 which influence the mechanics of their transmission. Hieratic Papyrus Ebers, dated to as early as the last quarter of the 16th century BC, is a huge repertory (108 columns) of recipes

 5 Cf. Von Staden (1997) 79–81; Andorlini (2006) 144–5; Reggiani (2015) 133–4; Reggiani (2016) 130–2. The incipit of Hippocrates’ De Victu is devoted to the medical science as a progressive knowledge, to which each generation adds its own contribution: cf. Perilli (2007) 60. On the relations between prescription and practice see also Andorlini (2007) 25–30. 6 I borrow this term from the studies about the interplay between text transmission and representation in classical drama; see recently Finglass (2015). 7 On the contribution of the Graeco–Roman papyri to the history of ancient medicine see, for an overall overview, Marganne (1981) and Andorlini (1993). On some fundamental characters of ancient Egyptian medical transmission see now Pommerening (2018) for Hieroglyphic/Hieratic texts and Jacob (2018) for Demotic prescriptions. On medical prescriptions in Greek papyri see Andorlini (1981); Andorlini (1993) passim; Andorlini (2006); Andorlini (2007); Andorlini (2018b); Reggiani (2018b). 8 Cf. Andorlini (2006) 143 n. 3, referring to Goltz (1974) 303ff. In general, on prescriptive formularity see Andorlini (2006) passim. On the formularity of Hippocratic prescriptions see Totelin (2009) 21–66. 9 On the origin of the earliest technical manuals composed of rules, prescriptions, and practical recommendations, see Cambiano (1992). At any rate, ‘the fact that there was a tradition of pharmacological knowledge is much more important than whether this tradition took a written or oral form’: Totelin (2009) 17.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  169

– loosely classified by diseases – and clinical cases, likely collected along with anatomical sections in a sort of reference manual for practicing physicians.10 Even in written form, however, medical prescriptions maintain their oral aspect, not only in the textual clues that I mentioned few lines above, but also in a set of graphic and layout features, the purpose of which is patently to preserve the shape of ‘fragment’ that each prescription acquires as a basic unit of discourse. Thus, in the Ebers papyrus itself, we can see each textual unit distinguished by a heading coloured red, and some other display devices, like variations in the column widths and displacements of text portions to distinguish different contents (e.g. lists of ingredients),11 which must be kept in mind for subsequent observations (see Picture 1).

Picture 1: Papyrus Ebers compared with PSI 10.1180.

Another striking feature, which illustrates the still active role played by actual practice in the transmission of medical prescriptive knowledge even in written  10 On Papyrus Ebers see recently Scholl (2002), with earlier bibliography. 11 Cf. Andorlini (2006) 148 n. 13.

170  Nicola Reggiani form, is represented by the annotations that we sometimes find added to the margins of the main text, recording the physicians’ individual experience (‘really effective’).12 At the incipit of Papyrus Ebers, we read that ‘Thoth [the well–known Egyptian god of wisdom and writing] is the guide, he who causes the script speak. He is who elaborates the receptarium (lit. ‘completions of writings’), who imparts force to the scholars and to the physicians, his followers’.13 Thus, the earliest transcriptions of medical writings are sacred in scope, and take place in temple environments. This is somehow implicit in the well–known sacred status of writing in ancient Egypt14, but also corresponds to the idea that ancient medicine initially developed – mostly or totally – around sanctuaries. This argument has been claimed by Lorenzo Perilli with particular reference to ancient Greece,15 but the same observations can easily be applied to the Egyptian context as well.16 In the ancient Greek world, we do find indeed early attestations of medical writings in temple settings. Strabo, writing under Augustus, says apropos of Epidaurus: καὶ αὕτη δ’ οὐκ ἄϲημοϲ ἡ πόλιϲ καὶ μάλιϲτα διὰ τὴν ἐπιφάνειαν τοῦ Ἀϲκληπιοῦ θεραπεύειν νόϲουϲ παντοδαπὰϲ πεπιϲτευμένου, καὶ τὸ ἱερὸν πλῆρεϲ ἔχοντοϲ ἀεὶ τῶν τε καμνόντων καὶ τῶν ἀνακειμένων πινάκων, ἐν οἷϲ ἀναγεγραμμέναι τυγχάνουϲιν αἱ θεραπεῖαι, καθάπερ ἐν Κῷ τε καὶ Τρίκκῃ. (Strab. 8.6.15.5–10) The city was renowned, in particular because of Asclepius’ reputation: it was believed that he healed any type of illness, and his temple was always crowded with sick people and full of votive tablets attached to the walls, on which the medical treatments used to be transcribed, just as in Cos and Tricca.

Strabo attests to a very traditional and long–standing practice, connected to the sanctuaries of the healing gods,17 which we know was already stabilized at least

 12 On the marginal additions attesting the individual experimentations cf. Ebers (1873) 43, Andorlini (2006) 143 n. 5, and now Pommerening (2018). 13 I take this translation from Tanja Pommerening’s lecture delivered at the International Conference ‘Parlare la medicina: tra lingue e culture, nello spazio e nel tempo’ held in Parma on September 5–7, 2016 in the framework of the DIGMEDTEXT project, cf. Pommerening (2018). The wording is slightly adapted in accordance to Andorlini (2006) 143. 14 Cf. Marganne (2004) 17–18. 15 Cf. Perilli (2005); Perilli (2006) 38–40; Perilli (2007) 55–58. 16 Cf. Perilli (2006) 40–44 and now Pommerening (2018). 17 Cf. Perilli (2007) 66–67.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  171

in Hippocrates’ times (around 5th century BC). The famous father of Western medicine is indeed in the centre of a controversial episode involving temple medical writings, as recounted (among others) by Pliny the Elder in his Natural History: is, cum fuisset mos, liberatos morbis scribere in templo eius dei quid auxiliatum esset, ut postea similitudo proficeret, exscripsisse ea traditur atque, ut Varro apud nos credit, templo cremato ⟨i⟩is instituisse medicinam hanc, quae clinice vocatur. (Plin. HN. 29.4.4–8) It has been passed down that since there [sc. in Cos] the people who had been freed from the illnesses used to write in his [sc. Asclepius’] temple what had been of help to them [sc. the remedies that healed them] so that it could be useful for future circumstances, he [sc. Hippocrates] transcribed them and, as Varro believes among us, templo cremato, with them established the medicine, which is called ‘clinical’.

Depending on the correct translation of the ablative absolute templo cremato, interpretation is twofold: some believe that Hippocrates himself set fire to the temple, in order to remain solely depositary of that science; others have a bias towards an accident.18 At any rate, whatever it was, the bottom line of the story is the foundation of Hippocratic medicine on a well–established tradition of depositing and safekeeping medical writings in sanctuaries (mostly Asklepieia),19 which contained detailed descriptions of therapeutical procedures, as professional records of clinical cases to be kept reserved in order to transmit relevant knowledge to whom would face similar circumstances in the future, and eventually re–elaborated in votive dedications to the healing god:20 repertori, cartelle cliniche ante litteram, nelle quali registrare la ricca casistica che si offriva alla considerazione dei medici, e che è alla base delle testimonianze superstiti, epigrafiche e non. Schede di questo genere dovettero costituire il nucleo di repositori, dai quali si originarono più tardi le biblioteche dei santuari di Asclepio, costituite verosimilmente, soprattutto nella fase più antica, da materiali specialistici per uso interno, sia professionale che didattico.21

 18 The tradition about Hippocrates’ alleged arson is complex: it is not registered by Pliny only, and other sources give partially different details; see Von Staden (1999) 149–57 for a full picture of the matter. 19 Cf. Perilli (2007) 57 and ff. 20 Cf. Perilli (2007) 67 for some reference to extant votive inscriptions, which he argues to have derived from professional medical records, likely deprived of the most technical medical details for celebrative purposes. 21 Perilli (2006) 39; cf. Perilli (2007) 55–59.

172  Nicola Reggiani This can be connected to the background issues related to the earlier transmission of recipes as investigated by Laurence Totelin (2009), who in the first three chapters points out how some Hippocratic treatises containing prescriptions, whether embedded in the text or listed in homogeneous sections, could stem from ‘smaller catalogues’ integrated with oral sources. It is in fact possible to verify Totelin’s picture on the ground of the epigraphical and papyrological sources at our disposal. As attested by Strabo, the earliest medical writings appeared in the form of pinakes, i.e. inscribed tablets,22 and indeed we find some similar references in fourth–century temple inventories recorded on inscriptions from the Asklepieion of Athens. IG II.2.1533 (dated to ca. 329/8 BC) reports a list of medical objects and surgical instruments kept at that temple: at lines 116–7, a [ἰα]τρικὸν | γραμματε⟨ῖ⟩ον (‘medical tablet’) is mentioned. Whether the listed items were ex voto dedications or actual equipment belonging to the operating temple doctors,23 this epigraph attests to the existence of medical tablets in sacred environments. A similar inventory, IG II.2.47 (dated to the early 4th century BC as well), recording πινάκια (‘tablets’, l. 18) among different objects labelled as ϲιδηρᾶ (‘iron things’), would speak in favour of dedications, since dedications of metal tablets are attested elsewhere.24 Again, details are perhaps irrelevant for our sake: suffice it to stress that the first stage of written transmission of Greek medical prescriptions, as snapshots of an oral knowledge in progress, took place within the temples and in a tablet format. The collocation in temples is a fundamental means for assuring the sacredness of written objects in a cultural environment where writing is ‘laic’ and desacralized.25 Moreover, the tablet format corresponds to the types of supporting media that were specifically devoted to record transient discourses with immediacy, for ease of reference and use. Antiquity indeed seems to articulate its phenomenology of writing into two big categories, which we can label ‘canonized

 22 Cf. Degni (1998) 11ff. 23 Cf. Perilli (2007) 62. On physicians operating in the temples of Asclepius see e.g. Perilli (2006) 31–38. 24 For example, Pausanias (9.31.4) was showed by the Boeotians a lead tablet, which contained Hesiod’s Works and Days and was dedicated near a sacred spring. However, Perilli (2007) 62–64 does not completely exclude that such iron tools could have pertained to everyday medical practice. A distinction is probably to be made between grammateia and pinakia, also on the ground of IG I.2.91.11 (Callias’ financial decrees, dated to the early Thirties of the 5th century BC) referring to τά τε πινάκια καὶ τὰ γραμματεῖα, seemingly as two different entities, but this discourse must be undertaken elsewhere. 25 For the secularity of Greek script in relation to medical writings cf. Marganne (2004) 18; for the enshrinement of Greek writings in temples see Perilli (2007) passim.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  173

writings’ and ‘offhand writings’.26 The former are accomplished written expressions aimed at lifelong preservation and textual canonization (mainly papyrus rolls); the latter are extemporaneous and temporary written expressions intended to be quickly discarded, updated, or modified somehow, and mostly for internal use (mainly tablet–like supports).27 Interactions between both typologies are somehow well attested, and are particularly striking as to the dynamics of textual transmission. The most appropriate to be mentioned here is probably the testimony provided by Diogenes Laertius (3.1.37), who reports that according to some sources Philip of Opus ‘transcribed’ (μετέγραψεν, certainly on papyrus rolls) Plato’s Laws, which had been left on waxed tablets (ἐν κηρῷ). The passage clearly deals with transmission stages of an oral discourse: for practical purposes (preliminary versions of works to be corrected and refined)28, the utterance is first transcribed on tablet–like supports, which for technical reasons look like the preferential media for ephemeral writings, provisional sketches, and then flow into papyrus rolls, for authoritative canonization and preservation.29 Thus medical prescriptions, expressions of a ‘fluid’ science and always subject to adaptations and updates, are preferentially transmitted through the former means: originally, wooden waxed tablets or the like. This fact, in a sense, offsets the strong concreteness of the usual Greek term indicating a medical recipe: γραφή, that is ‘writing’ tout court.30 ‘Hippocrates’, in whatsoever manner, broke a long–standing tradition by taking medical written transmission out of the temple.31 In doing so, however, he did not dare to change the transmission medium. Indeed, it seems that he retained the pinakia as the preferential writing media for medical knowledge. A subsection of the sixth book of the Epidemics exhibits the heading τὰ ἐκ τοῦ

 26 For more detailed discussion see Reggiani (2018b); see already Reggiani (2010) passim. 27 A similar statement has been made by Calame (2011) on the difference between Orphic texts on tablets, intended for an ‘internal’, mystic use, and Orphic texts on rolls like the Derveni papyrus, intended for an ‘official’, religious use. 28 We know from other sources (i.e. Quint. Inst. 1.8.64; Dion. Hal. Comp. 25) that Plato did use to take notes on tablets, as also other authors did: cf. Dorandi (2007) 13–28ff. 29 For transcribing on scrolls as a means of authoritative canonization, see the well–known case of the Homeric poems in Peisistratid Athens: cf. Reggiani (2010) 109–12. 30 See for instance Gal. Comp. med. gen. 13.676.2–4 K.: ἡ τοῦ φαρμάκου γραφή, ‘the prescription of the medicine’. 31 For the possible religious implications of some Hippocratic recipes see Totelin (2009) 111–39. Of course, here I am using ‘Hippocrates’ in a mere metaphorical sense, being well aware of the complex authorial issues of the so–called Corpus Hippocraticum.

174  Nicola Reggiani ϲμικροῦ πινακιδίου (Hippoc. Epid. 6.8.7), i.e. ‘from the small tablet’, which is glossed by Galen as follows:32 τιϲὶ δὲ οἵπερ καὶ ἀκριβέϲτερά μοι δοκοῦϲι καταμαθεῖν τῶν βιβλίων τὴν δύναμιν, ὑπὸ μὲν τοῦ Θεϲϲαλοῦ γεγράφθαι δοκεῖ τὰ εʹ, δύο δ’ εἶναι τοῦ μεγάλου Ἱπποκράτουϲ, καὶ ἐπιγεγράφθαι γέ που διὰ τοῦτο τὰ ἐκ τοῦ μικροῦ πινακιδίου (Gal. Diff. resp. 7.855.1–5 K.) Some people, whom by the way I believe have better understood the power of those books [sc. the Epidemics], think that five have been written by Thessalus [i.e. Hippocrates’ son], while two are of the great Hippocrates himself, and that for this reason they were titled ‘from the small tablet’.

The provenance from a small tablet seems to act as a stylistic brand of Hippocrates’ writing authority,33 failed to have been re–elaborated into more ‘literary’ an appearance.34 In fact, the clinical cases described in the section of Epidemics introduced by the said caption appear as brief notes, almost lacking syntactic structure, in a juxtaposition of data that has been described as quasi–stenographic.35 The lack of any literary refinement immediately gives the impression of being a quick jotting down of oral considerations and practical experiences, or at least the transcription of ephemeral annotations. Apparently, the first written transmission of medical prescriptions – which we may identify with Laurence Totelin’s ‘smaller catalogues’ – takes the shape of schematic clinical repertoires aimed at safekeeping and future reference. This must have been the appearance of the medical pinakia kept in the temples – whether votive accounts or actual reference files – and preserved in Hippocrates’ annotations. The oral and fragmentary aspect of prescriptive texts, as we perceive it through the disrupted yet formulaic syntax of Hippocratic ‘annotations’,36 corresponds with what is found in later times: symbols, abbreviations, and a rather  32 The same observation is contained in Galen’s commentary to Hippocrates’ Epidemics (Gal. In Hp. Epid. 7.441.18–20 Pfaff = CMG 5.10.2.2), which is preserved in Arabic translation only: cf. Marganne (2007) 324 n. 43; Perilli (2007) 65. 33 Cf. Manetti–Roselli (1982) 166–9; Marganne (2007) 323–6; Perilli (2007) 65. 34 ‘Edition’ of the Hippocratic works may well have followed the same tradition ‘from tablet to papyrus roll’ that we see at work with Plato’s Laws. On the ‘library’ form (i.e. papyrus rolls) of Hippocratic treatises cf. Marganne (2007) 320–2 and Totelin (2009) 225–58, with some striking observations on their fruition. On the derivation of some Hippocratic treatises from the re–elaboration of clinical records cf. Perilli (2007) 57–58 and 64–67. 35 Perilli (2007) 65–66; cf. Manetti–Roselli (1982) 167ff. 36 On the formulaic structure of early Greek medical treatises – namely those belonging to the Corpus Hippocraticum – see Andorlini (2006) 148–9 and Totelin (2009) 21ff.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  175

tachygraphical handwriting constantly characterize the ‘graphic and expressive jargon’37 of medical recipes as preserved by Graeco–Roman papyri from Egypt (an example among many possible others: P. Oxy. 8.1088, from the 1st century AD;38 Picture 2) and as attested by coeval literary sources. Galen (Comp. med. sec. loc. 12.423.13ff. K.) writes of a friend and colleague of his, Claudianus, who found some recipes belonging to a third colleague, then deceased, transcribed ‘in symbols’ (ϲυμβολικῶϲ δὲ γεγραμμένον: ibidem 424.1–2) almost incomprehensible, which forpictuced him to countless single–step attempts in order to understand and decode them (ibidem 424.2–4).

Picture 2: P. Oxy. 8.1088

Galen’s testimony depicts a further stage in the transmission of medical recipes. Now they are passed on, whether intentionally or not, among practicing physicians, in written form yet even more contributing to the ‘fluid’ appearance of the matter, because each user is compelled to customize the texts according to his own personal experiences or needs.39 The strong presence of technical writing

 37 Andorlini (2006). 38 Cf. Marganne (2004) 79 and Andorlini (2006) 164–5. A palaeographical study of this interesting papyrus is provided by Lundon (2004). 39 ‘Some degree of information sharing must have taken place in regard to initially unfamiliar drugs, though subsequently adapted by individual physicians to particular needs and experi-

176  Nicola Reggiani devices such as symbols, abbreviations, special terms, tachygraphy, is itself indication of a circulation among specialized circles, as much as ‘Hippocratic’ or temple pinakes were reserved to single closed ‘schools’.40 L’osservazione di tali fenomeni, e del loro riproporsi costantemente nella tradizione dei testi medici greci su papiro, permette di riconoscere diverse fasi e livelli in cui il sapere tecnico contenuto nella ricetta medica veniva materialmente veicolato al lettore/consumatore attraverso moduli espressivi e dispositivi tecnici, visivi, fisici, che formano una sorta di koiné, un tutt’uno tra lingua tecnica e scrittura speciale dei testi. Di qui la suggestione di rintracciare una specie di ‘gergo’ nei connotati di quel particolare linguaggio criptico, grafico ed espressivo, che comunica all’interno di una determinata categoria professionale: il medico, gli altri medici (i colleghi), il farmacista, il commerciante di farmaci, il paziente. Si tratta di modi speciali di usare parole e segni attraverso i quali le competenze medico–terapeutiche tendono a specializzarsi all’interno di una corporazione di addetti alla professione medica.41

It must not be underestimated that it is seemingly from the first centuries of the Roman Empire onwards that we find annotated evaluations of efficacy resembling those of the Papyrus Ebers,42 while the practice of indicating precise quantities for ingredients seems to date back to just few centuries before, to second– century BC Alexandria43. Both of them may be regarded as new features, operational for a more widespread circulation of written medical prescriptions.44 This

 ence’: Lang (2013) 181. ‘However communicated, empirical constraints and variation in the experiential use of medical recipes produced a considerable degree of adaptation, alteration and personalization among professionals, folk healers, and self–medicating users’: Lang (2013) 180. A nice case of self–medicating learned layman might be the Psenpaapis of O. Claud. 2.220 (ca. AD 137–145), who asks his brother Gemellus to go to the doctor to get some saffron and to send it to him, because he did not receive the medicinal kollyria; it seems apparent that he is willing to reproduce the eye–salve by himself: cf. Hanson (2010) 192 and 199, where she evokes laymen interested in self–medication as possible holders of single prescriptions. 40 On the expressive devices of the medical prescriptions on papyrus in relation with circulation among specialists see Andorlini (2006) 153ff.; on earlier medical knowledge reserved to specialized centres see Perilli (2007) 69–70. 41 Andorlini (2006) 153. 42 Cf. Lang (2013) 180 n. 164: ‘Similar claims in the Greek medical corpus before the early 2nd century CE are rare. Diseases of women 1.78 is a possibly unique example: ‘You will not find anything better’’. 43 Cf. Lang (2013) 181–2. At any rate, the issue of quantifying measures is complex, as is shown in Reggiani (2015) and (2016) passim. 44 Totelin (2009) 17 notes that Hippocratic compilers never name the sources of their recipes. This is in countertrend with what we know from the Greek papyri, and perhaps this change can be ascribed to the new transmission stage.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  177

new transmission stage involves some further relevant issues, pinpointed by Galen himself – who, as a matter of common knowledge, carried on an important work of retrieval and collection of medical prescriptions – as follows:45 ἐπεὶ δ’ἔνιαι κακῶϲ εἰϲι γεγραμμέναι, τινῶν μὲν ἐν τῷ τοῖϲ αἰτήϲαϲι διδόναι τὰϲ γραφὰϲ ἑκοντὶ ψευδομένων, ἐνίων δὲ καὶ διαϲτρεφόντων ἃ παρά τινων ἔλαβον ἀντίγραφα. τὰ δὲ δὴ βιβλία τὰ κατὰ τὰϲ βιβλιοθήκαϲ ἀποκείμενα, τὰ τῶν ἀριθμῶν ἔχοντα ϲημεῖα, ῥᾳδίωϲ διαϲτρέφεται (Gal. Antid. 1.5 = 14.31.10–15 K.) Some prescriptions are transcribed wrongly, because some people alter them intentionally when they give them to whom requested them, while others distance themselves from the copies received; and indeed the books kept at the libraries, those containing the symbols for the dose figures, are easily forged.

Forgery of medical prescriptions is a by–consequence of this new stage of transmission: circulation among physicians could harm the ‘intellectual copyright’ of the original developer of a remedy, so that he altered intentionally his written annotations (or made them unintelligible, as in the case of Claudianus’ colleague) in order to prevent unauthorized uses and to remain solely depositary of his own inventions (a concern that, as we saw, was attributed – in ‘flaming’ terms – to the same Hippocrates). At any rate, even intentional circulation of recipes was prone to corruptions and misunderstandings, as the following case – a letter sent by a doctor, Chairas, to a colleague of his, Dionysios, in AD 59 – clearly illustrates:46 ἀντιρ̣ράφια (l. ἀντιγράφια) δέ μοι δύο ἔ̣πεμψαϲ, | τὸ μὲν τῆϲ Ἀρχαγαθήου (l. –είου) τὸ δὲ τῆϲ | ἑλκωτικῆϲ. ἡ μὲν Ἀρχαγάθ⟨ε⟩ι̣⟨ο⟩ϲ̣ ὑγιῶϲ | περιέχει, ἡ δὲ ἑλκωτικὴ ῥητ{ε}ίνηϲ | ϲυνϲταθμίαν οὐ περιέχει. ἐρωτῶ | δὲ ϲε περὶ ἑ̣λ̣κωτικῆϲ γενναίαϲ | δυναμένηϲ ἀκινδύνωϲ πέλματ(α) | ἑλκῶϲα̣ι̣· ̣ ̣ ̣ γ̣[ὰ]ρ̣ [κ]ατʼ ἀνάγκην | ἐπείγομαι (P. Mert. 1.12.13–21)

 45 Cf. Marganne (2004) 79. See also Gal. Comp. med. gen. 4.7 = 13.726 K., where Galen again stresses some issues in deciphering symbols and conventional indications in medical prescriptions, praising the possibility to understand the differences between remedies and ingredients as one of the benefits of medical knowledge. On tachygraphical manuals comprising medical vocabulary see Marganne (2004) 85–86. 46 Cf. Hanson (2010) 191 and Andorlini (2016) 153–8, with a deep analysis of the technical devices deployed by the writer; cf. also Andorlini (2017) 15–16.

178  Nicola Reggiani You sent me two copies [i.e. of recipes], one of Archagathus’ plaster, the other one of the cicatrizant. Archagathus’ one is correctly composed, while the cicatrizant lacks the resin dosage. Please let me know of a strong cicatrizant that can scar [the sores in] the foot planum, because I urgently need it.

This stage of transmission does not take place on wooden tablets any more. The recipes that belonged to the late colleague of Galen and Claudianus, in the abovementioned passage, are said to be written ἐν π⟨τ⟩υκτίδι διφθέρᾳ, ‘on a folded parchment’ (Gal. Comp. med. sec. loc. 12.423.13–15 K.). This mention is paralleled by other passages by the same Galen, who in the recently discovered treatise De indolentia describes the loss of his own books in the AD 192 fire in Rome47 and, along with them, of a considerable collection of pharmacological recipes, which had been carefully kept κατὰ δύο διφθέραϲ π⟨τ⟩υτκτάϲ, ‘in two folded parchments’ (Gal. Indol. 33). The reference is overtly to those ‘notebooks’ made up of parchment sheets folded and bound together, the diffusion of which in the first centuries of the Roman Empire is now a fact.48 Such parchment notebooks, seemingly a Roman innovation (after the papyrus shortage occurred in 170–168 BC due to the invasion of Egypt by Antiochus V – so runs the vulgate),49 were meant to bear the very same function of the codices of bound wax tablets that are attested in the ancient Mediterranean since earlier times. Terminology is striking: as early as Homer, ptuktis is used to define such a tablet codex (ἐν πίνακι πτυκτῷ, Hom. Il. 6.168–9: Bellerophon’s ‘folded’ tablet).50 This makes clear that (a) the singular form could conceal a binding of more than one tablet (this can perhaps explain how Hippocrates’ clinical records could have fit ‘one’ tablet), and (b) parchment

 47 Cf. Boudon–Millot (2008); Boudon–Millot–Jouanna (2010). Further bibliography on the treatise in Garofalo–Lami (2012) xxxi–xli. 48 Boudon–Millot–Jouanna (2010) 103–5 and 108 argue that what Galen is describing in the two passages are parchment codices of large format, not just notebooks. This is a matter of interpretation, which deserves a longer and deeper discussion to be undertaken elsewhere – see Reggiani (2018b). The editors also discuss the meaning of the term ptuktis, and identify one of these two ‘codices’ with Claudianus’ one. 49 Cf. e.g. Roberts–Skeat (1987) 15–23. We should not underestimate the fact that Galen, as well as the other people involved in his exchange of parchment dossiers, came from Pergamum, which is the alleged motherland of parchment itself. 50 For discussion cf. Degni (1998) 11–13.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  179

notebooks were regarded as belonging to the same bibliological category as tablet codices,51 and therefore used for the same purposes – ephemeral transcriptions of discourses subject to adaptations, modifications, updates.52 At this stage, single medical prescriptions seemingly circulated in the form of small papyrus slips, purposely cut from larger sheets or rolls: they are what Isabella Bonati calls ‘etichette–ricetta’ (recipe labels) because they closely resemble, in shape and purpose, the ‘labels’ that were used to tag containers of medicaments.53 What is striking of such items is that when they flow into personal collections, for the sake of safekeeping, reference, and further transmission (as the examples cited above show), they are transcribed on those parchment notebooks,54 some examples of which came down to us – namely P. Ryl. 1.29 (mid 3rd century AD) and PSI 6.718 = SB 26.16458 (4th century AD: Picture 3) – and maintain their original aspect of unitary fragments, which is indeed unavoidable to ensure a correct, yet specialis circulation, as well as the preservation of scholarly authority, as the frequent mention of the inventors’ names apparently indicates.55 Thus the writers deploy an entire set of graphic and layout devices with the sole aim of preserving the unity of each single recipe: separating lines (paragraphoi,56 forked paragraphoi), indented/extended heading lines (eisthesis/ekthesis), blank spaces, line fillers, as well as formulaic verbal expedients. This fact is comprehensible on one hand, because it was important to keep each prescription well separated to avoid dangerous confusions in a writing system that did not envisage word breaking as a rule (scriptio continua); but is remarkable on the other hand, since it is a requirement of such fragmentary textual categories that were closer to the parameters of ‘liquid’ transmission.

 51 Striking are Martial’s references to parchment notebooks as pugillares membranei (‘tablets of parchment’, 14.7; cf. 14.184, and see also his famous hos eme, quos artat breuibus membrana tabellis, ‘buy those, which parchment wraps in small tablets’, 1.2). For discussion of these attestations, which stand at the very birth of the parchment codex, see e.g. Roberts–Skeat (1987) 24– 29 and Degni (1998) 55–59. 52 For Roman parchment booklets to be used for drafting see e.g. Hor. Sat. 2.3.1–2; Id. Ars P. 386–90. 53 Bonati (2016) part. 52–57. Ostraka are other common written media for the circulation of single recipes; they were likely intended for a very quick use and soon discarded. 54 Cf. Nicholls (2010) 383. 55 Cf. Hanson (1997) 303. 56 On the sign paragraphos cf. Giangrande (1978); Barbis Lupi (1994).

180  Nicola Reggiani

Picture 3: PSI 6.718 = SB 26.16458

Consequently, recipes acquired the status of ‘textual fragments’ to be quoted or re–used by medical authors or compilers of more articulated reference works, the catalogues also known as receptaria, which were transcribed in more ‘library’ bookforms such as papyrus rolls, though maintaining their fragmentary status. Among the numerous possible examples, P. Berl. Möller 13 is a stunning instance. This papyrus is a comparatively large portion of a roll coming from Hermoupolis Magna and palaeographically dated between the late third and the early 4th century AD.57 It is written on the recto along the fibres, which makes it clear that it was purposely produced as a collection of medical prescriptions, of which only two columns survive: the first one, fully preserved, is numbered ‘11’ on the top, indicating that it was preceded by ten other columns; the second one, the lines of which are almost completely lost in the right–hand break, should have been numbered ‘12’ accordingly. The first column contains a single prescription, on  57 The image is available online in the databank Berliner Papyrusdatenbank (http://ww2.smb.museum/berlpap). I follow text and translation as republished by Corazza (2016).

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  181

nine lines, preceded by a two–line heading slightly indented, which corresponds to the epangelia section, i.e. the therapeutic indication: π̣ρ̣ὸ̣ς̣ τ̣ὸ̣ μ̣ὴ̣ ἀπορρ̣ε̣ῖν̣ τ̣ὰς̣ ἐν τῇ κε|φα̣λ̣ῇ̣ τ̣ρ̣ί̣χαϲ, ‘to prevent hair loss on the head’. The recipe – which does not provide the synopsis of the ingredients but move directly to the skeuasia, the practical instructions – has been identified by Marganne (1980) as a prescription that Galen ascribes to Heras of Cappadocia, a pharmacologist active between 20 BC and AD 20. The text on the papyrus parallels Gal. Comp. med. sec. loc. 12.430.8–15 K. verbatim,58 while other variant versions of the same remedy are recorded by Galen himself (ibidem 12.435–6 K.) as antecedents of Heras’ one.59 The rest of the column is blank. Subsequently, Corazza (2016) discovered that also some remnants of the second column can be identified with other recipes by Heras, this time against headache, mentioned by Galen as well, with some wording variants.60 Some features of the papyrus are remarkable as they illustrate some basic characteristics of the transmission of this text typology. (a) The recipes of the second column are separated by a horizontal line (paragraphos) from one another, thus preserving the usual appearance of single textual units. The indented heading of the first column should have been replicated at the top of the second column, and since all the three identified recipes deal with headache, we may think that the they were titled with a collective caption explaining their common application;61 this use might be related to the fact that single prescriptions were often transcribed without their introductory sections.62 (b) Two of the  58 In fact there are some interesting variants, which as usual show how papyri can contribute to the history of the texts: in particular, at line 10 (καλοῦϲι pap. : καλοῦϲι καί Gal.) the papyrus offers a superior reading, since the conjunction is syntactically unfit; further discussion in Corazza (2016) ad locc. 59 Cf. Marganne (1980) 182–3. 60 In particular, the first prescription of the second column (ll. 1–3) parallels Gal. Comp. med. sec. loc. 12.593.14 K. verbatim, while the following two (ll. 4–8 and 9–15) show partial overlaps with ibidem 12.594.1–4 (= Aet. 6.50.75–9) and 12.594.7ff. K. All these recipes are ascribed to Heras. The remaining traces of fifteen lines, articulated in four more recipes, could not be identified with any known text. 61 Corazza (2016) does not mention this feature. Unfortunately, nothing certain can be said about the remaining four prescriptions of the second column. Some possible integrations, as already noted by Corazza (2016) passim ad locc. may be compatible with other headache remedies; on the other hand, it could be possible that the somehow larger interlinear space after the paragraphoi between ll. 15–6 and 22–3 conceal further headings of different thematic sections (ibidem ad locc.), though they appear narrower than the space at the top of the column. 62 The graphic and paratextual devices deployed by such collections of recipes to keep the original fragmentary unity of the texts is even more evident in PSI 10.1180. A papyrus roll from 1st / 2nd century AD Tebtunis. The prescriptions, here, are separated by paragraphoi as in P. Berl. Möller 13; moreover, their first line is always extended in ekthesis, and the ingredients are listed

182  Nicola Reggiani recipes patently parallel Galen, but the papyrus is by no means a copy of On the composition of medicaments by places: they do not follow the canonical order in which they are cited in Galen’s treatise, clearly attesting a work of selection, extraction, and thematic re–arrangement, in which each recipe is treated as a unit to be managed on its own. Moreover, the other two identified prescriptions look like variants of Heras’ texts as reported by Galen, thus attesting a ‘fluid’ stage of transmission, in which recipes are modified and adapted according to the users (Galen himself, as we saw, attests some earlier versions of Heras’ recipe against hair loss). (c) Combining both of the preceding points, the explanation of the huge blank space in the first column becomes apparent. As already suggested by Corazza (2016) 41, this was likely ‘intended to contain further prescriptions against hair loss […], possibly taken from various sources’ or from the same Heras, whether through Galen or not.63 This case therefore perfectly illustrates how collections of prescriptions were ‘living texts’, as defined by Bonati (2016) 66, ‘fluid’ and open repositories of an ever–changing fragmentary knowledge rather than fixed containers of canonized texts, as more formalized treatises may have been even when they are transcribed on writing media of library tradition.64 The operation of collecting, comparing, and transcribing prescriptions – a fundamental step in the history of their transmission – is well attested in the everyday practice of ancient physicians, in the

 in columns with the quantities well detached to the right – a mise en page that recalls very closely the layout of Papyrus Ebers (see Picture 1). 63 ‘This remedy against hair loss is likely to have been followed in the Narthex [sc. Heras’ pharmacological work] by another one (or more than one) for the same disease apparently not preserved in the papyrus, cf. Gal. Comp. [med.] sec. loc. [12.]430[.15–6] K.: ταῦτα περὶ τῆϲ προτέραϲ ϲυνθέϲεωϲ εἰπὼν ὁ Ἥραϲ ἐφεξῆϲ γράφει ταῦτα’: Corazza (2016) 44. Corazza (2016) 41 mentions the possibility that the blank may have served as a separation between two different thematic sections of the collection, as attested elsewhere, and namely refers to P. Lit. Lond. 132. However, this is a very well–shaped collection of Hyperides’ orations, and the blank separates the orations from each other (cf. Caroli [2007] 263), which is something somehow different than our case. The blanks in the Hyperides roll are indeed subsequently treated as initial agrapha, where a second scribe adds the title of the successive oration (on the use of placing titles in the opening agrapha of bookrolls see e.g. Caroli [2007] 53–57), while in our receptarium the ‘titles’ are certainly placed at the top of the column. 64 I suspect that when such technical reference books show up in the format of papyrus roll, they come from a somehow ‘higher’ temple tradition. It is a hypothesis to be verified, but, for instance, PSI 10.1180, the receptarium roll mentioned above, also written on the recto, certainly comes from the temple of Soknebtunis at Tebtunis: cf. Andorlini (2004) 82–83. P. Oxy. 8.1088, cited above as well, comes from a papyrus roll cut off and reused on purpose, so it represents a slightly different case.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  183

circulating antigraphia of the abovementioned letter by Chairas65 as well as by Galen himself, who after describing how he got in possession of the parchment dossier recounts: […] πολίτην τε καὶ ϲυμφοιτηὴν ἐμὸν ὀνόματι Τεύθραντα διατρίβοντα κατὰ τὴν πόλιν εὗρον ὃϲ διεδέδεκτο τὰϲ Εὐμενοῦϲ τοῦ ἰατροῦ διφθέραϲ, ὄντοϲ μὲν καὶ αὐτοῦ Περγαμηνοῦ, φιλοφαρμάκου δέ καὶ πολυφαρμάκου πάντων τῶν ἰατρῶν μάλιϲτα. καὶ αὖται δέ αἱ διφθέραι ἐν ἐνὶ ϲχεδὸν ἐξ ὅληϲ τῆϲ οἰκουμένηϲ ἦϲαν ϲυνηθροιϲμέναι διὰ τὰϲ ἐπιγινομέναϲ ἀποδημίαϲ αὐτῷ [...]. ἐκ τούτων οὖν τῶν παραϲκευῶν, εἴ τιϲ τι τῶν θαυμαϲίων εἶχε φαρμάκων, ἐλάμβανον οὐ χαλεπῶϲ, ἀντιδιδοὺϲ δύο καὶ τρία τῶν ὁμοίων. (Gal. Indol. 34–5) I found a fellow citizen and pupil of mine, Teuthras, who was living in the city, and who had received the parchments of Eumenes the physician, who was from Pergamum as well, who loved medicines and was rich in medicines more than all physicians. Such parchments were collected in one [notebook?] – so to say – from all around the world, due to his travels, following one another […]. From such sketches, if someone possessed an extraordinary medicine, I got it without difficulty, giving two or three similar recipes in exchange.

Such a circulation of collected, even swapped prescriptions takes place in the form of parchment sheets or notebooks, which we should imagine to have been gradually filled in as the holder found new interesting items.66 As a new stage of transmission, prescriptions are gathered in more structured ‘catalogues’, where the texts are usually organized and arranged thematically, though nonetheless maintain a certain openness to ‘modular’ additions, as we previously observed, and as is revealed by the wide variety of prescription versions and variants that we find in the receptaria: although individual recipes in a collection on papyrus often resemble items in the known authors, each extensive collection on papyrus has thusfar proved to be a unique assemblage.67

 65 Cf. Andorlini (2003) 15–16: ‘il nostro medico di provincia ci pare impegnato nel tenere in mano più versioni di uno stesso testo da collazionare’; see also Andorlini (1993) 462–4. 66 It is particularly remarkable the abovementioned PSI 6.718, where the prescriptions follow one another without any apparent thematic coherence: a recipe against enuresis, one against hoarseness, and a third one ‘for making old wine good’. No attention is usually paid to this last prescription by the editors, but note that aromatic wine could have had also a medical use (see e.g. Fournet–Magdelaine [2001] and Maravela [2010]; in general on wine in ancient medicine see Jouanna [2012] 173–93) and it seems likely that this recipe advised to spice the wine in some unknown way. This may account for its inclusion in such a collection of medical prescriptions. 67 Hanson (2010) 199.

184  Nicola Reggiani The most suitable book format for such a ‘liquid’ typology of text seems to have been the codex, for its flexibility to accommodate in–text or marginal additions.68 The most intriguing example is the so–called ‘Michigan Medical Codex’ (P. Mich. 17.758),69 a receptarium on a small–format papyrus codex, dated to the 4th century AD, of which thirteen leaves survive to an amount of twenty–six pages, in which numerous recipes are collected – seemingly – according to type of medication (pills and lozenges, then wet and dry plasters, at least in the extant pieces). As usual, recipes start with an indented heading, announcing the type of remedy, and are separated from each other with lines and small blank spaces; they typically contain the list of ingredients, followed by directions for composition and use. Many prescriptions are ascribed to famous doctors, and ‘the presence of plasters from a variety of different physicians suggests that the basic text of the codex was combining and taking its shape over considerable time’.70 They likely came from those personal notebooks, which were collected over the time, and then passed on. But an even striking feature is the intervention by the owner of the codex, certainly a practicing physician himself: not only did he intervene in the text by correcting a few of the scribe’s obvious errors of copying, but he also squeezed into the copious bottom margins some twenty additional recipes on related topics, doubling the number of recipes in the preserved sections. Because empty space was limited, he emphasized separation between recipes through lines and marginal markers’,71

contributing to build up an articulated network, in which the baseline concern was to keep the original fragmentary unit (and authoritativeness) of each prescription by the deployment of the usual paratextual devices. The activity of the anonymous owner of the Michigan Medical Codex is a true ‘philological’ behaviour, the same that underlies Galen’s efforts to collect hundreds of prescriptions, and to collate library manuscripts, carefully considering all possible marginal annotations and textual variants, in order to compile his huge treatises,72 among which we must certainly situate the pharmacological compilations. These were intended to be circulating publications of canonized  68 Literature about the codex and its technical feature is immense; see at least Turner (1977), beside the bibliography collected by Reggiani (2010). 69 Originally published by Youtie (1996). For the following observations I refer to Hanson (1996), Hanson (1997) 302–4, and Andorlini (2003) 26–28. 70 Hanson (1997) 303. 71 Hanson (1997) 303. 72 Cf. Andorlini (2003) 15–16 and Bonati (2016) 63–65 for discussion of Galen’s philological work.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  185

texts, and are therefore entrusted to more standardized book forms such as papyrus rolls. Indeed, in describing the AD 192 disaster, he clearly distinguishes three types of writing media, which patently exemplify the three different stages of transmission of medical prescriptions: οὐ μόνον οὖν ἀπώλοντο κατὰ τὴν πυρκαιὰν αἱ διφτέραι πάϲαι […] ἀλλὰ πραγματεία μοι γεγονυῖα μετὰ πολλῆϲ ἀκριβείαϲ, ἡ περὶ ϲυνθέϲεωϲ φαρμάκων, ἐν ᾗ πῶϲ ἄν τιϲ αὖθιϲ ϲυνθείη[ν] φαρμάκων ὀλίγων γραφαὶ φθάνουϲαι δίδοϲθαι τοῖϲ ἑταίροιϲ (Gal. Indol. 37) Not only did perish in the blaze all the parchments […], but also a treatise composed by me with great care, the one On the composition of medicaments, in which I showed how the most celebrated remedies can be recomposed, and are preserved only the recipes of few medicines, which had already been given to my fellows.

The single recipes (graphai), likely on papyrus slips, and the parchment notebooks (diphtheriai)73 are clearly contrasted with the copies of the first two books of his De compositione medicamentorum per genera, which elsewhere he states to have been burnt in that occasion, referring to them as bibloi (Gal. Comp. med. gen. 13.362.1–5 K.).74 Working copies, most likely in the codex format just like the Michigan ‘book’, are not mentioned, though ‘l’uso dotto delle copie di lavoro doveva essere una modalità già praticata per la predisposizione e la fruizione del libro nella forma libraria del rotolo’.75 Such an editorial practice, necessarily in continuous development, unavoidably open to endless revisions, annotations, modifications, goes on until the very latest examples among Greek papyri. P. Ant. 3.186 is a very fragmentary large–format papyrus codex from Antinoupolis, dated to the 6th century AD on palaeographical grounds, and contains sections of Galen’s De compositione medicamentorum per genera. Some scanty marginal annotations survive, and typical layout devices like blank spaces and overlines are

 73 It seems to me that for Galen diphtherai means the single parchment sheets, which could be folded (ptuktis) and then gathered in one or more bounded notebooks (see above, ἐν ἐνὶ and κατὰ δύο διφθέραϲ π⟨τ⟩υτκτάϲ). 74 Cf. Boudon–Millot (2008) 12–13. 75 Andorlini (2003) 16.

186  Nicola Reggiani deployed to isolate the headings ἄλλο introducing new prescriptions in a homogeneous series (see e.g. fr. 8).76 This shows how even canonized texts were transformed into ‘living’ texts over and over again, not without remarkable effects on the medieval tradition of the ancient authors.77 The following words by Ann Hanson best summarize the bottom line of textual transmission of medical prescriptions: The appearance of the ancient catalogues underscores the potential frangibility of recipe collections, for their writers and scribes took specific care to articulate not only the rubrics under which a series of prescriptions were grouped, but also to articulate individual prescriptions, in order to maintain the recipe's textual integrity in a scribal culture in which the majority of prose texts were written continuously, without word division or breaks at the end of sentences (scriptio continua). In the case of recipes it was, of course, crucial to show where one stopped and the new one began, to demonstrate which ingredients and which instructions must be understood together. The full range of articulating devices then in use are brought into play to highlight the beginning of a new recipe: ekthesis, or extension into the margin, and eisthesis, or indentation; paragraphoi, or other separating lines; on–line spaces; enlargement of the first word’s initial letter. Verbal separators, such as beginning a recipe with ‘If someone suffers from this, do that’, or, for the second and following recipes in a series, beginning with ‘another’, are both ubiquitous dividers, shared not only with earlier non–Greek medical traditions of the eastern Mediterranean, but passed on to the Greek manuscripts of the Byzantine tradition.78

All of these considerations are basic to the evaluation of modern representation of such medical sources in electronic format. Papyrology, which admittedly has a long–standing familiarity with digital resources, in the last few years has built powerful textual databanks that have proven fundamental for enhancing scholarly research in the field.79 Digitisation is particularly important for textual identifications and reconstructions,80 but the traditional papyrological databases (namely, the Duke Databank of Documentary Papyri at http://papyri.info) have not been conceived to encode the so–called ‘literary’ and ‘paraliterary’ text categories, to which medical papyri (and prescriptions among them) belong. New  76 Andorlini (1992) 14–20 illustrates other similar cases from a different late papyrus codex from Antinoupolis (abbreviations, symbols, text displacements, etc.). On the context of the medical collection from Antinoupolis cf. Marganne (1984). 77 On the relationship between papyrus collections of recipes and the manuscript tradition cf. Andorlini (1992) and Hanson (2010) 198. On the ‘afterlife’ of Hippocratic recipes see Totelin (2009) 257–96. 78 Hanson (1997) 302. 79 See now Reggiani (2017a) passim, in particular chapters 1 and 8. 80 Cf. Hanson (1997) 300–4 and (2002) 196 for some examples of an effective use of literary databases (namely the Thesaurus Linguae Graecae) in the study of medical papyri.

Transmission of Recipes and Receptaria in Greek Medical Writings on Papyrus  187

platforms are currently being developed to fill in this gap – above all, the Digital Corpus of Literary Papyri (DCLP at http://litpap.info) – but are based on the already extant tools, which were designed to support documentary texts only. This raised the necessity to develop new scientific concepts and new technical schemes in order to face the peculiar features of such complex texts as the ones I presented in this contribution. Not by chance the medical papyri project held in Parma81 has been one of the first partners of DCLP. As I underlined above, medical texts – and prescriptions above all – distinguish themselves for a strong fluidity, which is rather unfitted by the traditional way of representing ancient texts as deployed by the current databases, which still owe very much to the traditional critical edition as reconstruction of a fixed archetype, overcoming all possible variants as deviations from a canonical text, and which do not support a precise encoding of paratextual and layout features. On the contrary, as we saw, textual variants and individual idiosyncrasies get to be intimately part of the transmission of medical prescriptions, as the full set of paratextual devices deployed to keep the fragmentary unit of recipes do. The inadequacy of the traditional philological/stemmatological model to represent in full the textual features of these complex and fluid technical writings has already been pointed out by Hanson (2010) and Totelin (2009) part. 21–66. The ‘accretive model of composition’, advanced by the former to provide a suitable description of the phenomenon, which corresponds to the latter’s anthropological standpoint, is perfectly pertinent to a global rethinking of the matter, in search of new standards of representation of such ‘dynamic’ texts. I will not deal with merely technical information here82, but with the theoretical concept that lies behind them. Essentially, when conceiving a digital edition, we ought not to focus on the sole text but must keep the complete dimension of the documentary object. Each ancient textual feature, which contributes intimately to its meaning and its cultural significance, deserves careful preservation. Therefore, we ought to re–establish the original concept of text fragmentation by developing and enhancing the current databanks, towards the ideal of a multitextual and comprehensive

 81 The DIGMEDTEXT Project is aimed at creating an online database of the Greek medical papyri: see at http://www.papirologia.unipr.it/ERC. 82 I refer to Reggiani (2018c), (2018b), and (2019b) for a deeper discussion of the technical issues.

188  Nicola Reggiani digital critical edition83. In other words, in the new multi–dimensional electronic environment a new stage in the history of textual transmission is conceivable.

 83 Multitext is the innovative infrastructure model developed by the Center for Hellenic Studies for the digital edition of Homeric poems (including Homeric papyri). It is based on the consideration that Homeric textual evidence, fluid in nature, does not comply with the traditional philological/stemmatological view of variants stemming from a single archetype, since an original Homeric text never existed. The multitextual model is also suitable for the digital encoding of fragmentary sources: cf. Reggiani (2017a) chapter 9 and (2019a).

Rosa Lorito

Latin Epigraphy and Literary Texts in 4th Century AD Rome The Case of Vettius Agorius Praetextatus Epigraphy could be an important means of transmission for literary texts. 1 This is clear for instance from the funerary inscription of Vettius Agorius Praetextatus and Aconia Fabia Paulina. 2 The text, engraved on a marble ara, not only has the typical features of a funerary monument, but also the complexity of a lyric text, as we can see from the use of iambic senarii. This inscription is also a testimony of epigraphy as an instrument to convey celebrative and propagandistic texts, but also to keep safe and preserve certain values, in the wider context of the 4th century AD, characterised by the clashes between paganism and Christianity. In the text, we can also see how in Praetextatus’ age these clashes were not so harsh: on the contrary, there was a permeation between these two worlds. 3 Vettius Agorius Praetextatus,4 vir clarissimus, praefectus urbi in 367–368 and praefectus praetorio in 384, is a well–known character of Macrobius’ Saturnalia and played an important role in the culture and society of 4th century Rome. As we can see from epigraphic,5 historical6 and legal sources,7 he had a typical senatorial cursus honorum: he was quaestor candidatus, praetor urbanus, corrector Tusciae et Umbriae, consularis Lusitaniae, proconsul Achaiae, praefectus urbi, legatus a senatu missus V,8 praefectus praetorio II (sic) Italiae et Illyrici9 and consul  1 On versified Latin epigraphy, cf. Sanders (1981) 707–20, while, on literary features of carmina latina epigraphica, cf., in general, Cugusi (1985). On engraved poetry and parallelism between epigraphy and literature, cf. Massaro (2009) 225–51; Massaro (2013a) 365–413; Massaro (2013b) 253–74; Massaro (2014) 65–102. 2 CIL VI.1779 = Dessau, ILS 1259 (Roma). Cf. Ensoli–La Rocca (2000) 506–7, too. 3 About pagan and Christian world and the debate between them, cf. Kahlos (2008) 63–9; Kahlos (2009a) 283–302; Kahlos (2009b) 79–95; Kahlos (2010a) 167–79; Kahlos (2011) 187–92; Trout (2015b) 213–31. On the laws against Pagans, cf. Lizzi Testa (2009) 385–409. 4 PLRE I, Praetextatus 1; cf. in general Kahlos (2010b). 5 CIL VI.1777 = Dessau, ILS 1258; CIL VI.1778; CIL VI.1779 = Dessau, ILS 1259 (Roma). 6 Amm. Marc. 22.7.6; Symm. Relat. 12.4; cf. Relat. 10–11, Jer. Ep. 23.23. 7 CTh 9.16.7; CTh 8.14.1; CTh 9.40.10; CTh 14.4.4; CTh 6.35.7; CTh 13.3.8; CTh 1.6.6; CTh 6.5.2; Cod. Iust. 1.54.5. 8 Legatus amplissimi ordinis septies et ad impetrandum rebus arduis semper oppositus in CIL VI.1777 = Dessau, ILS 1258 (Roma). 9 Praef. praetorii Illyrici Italiae et Africae in CIL VI.1777 = Dessau, ILS 1258 (Roma). https://doi.org/10.1515/9783110632590-013

190  Rosa Lorito ordinarius designatus.10 In Maijastina Kahlos’ opinion, his cursus honorum was noteworthy because he was both corrector Tusciae and consularis Lusitaniae: 11 senators usually had only one consular governorate, but he had two. Apart from Praetextatus, only one other man, Volusius Venustus, had two consular governorates: he was corrector Apuliae et Calabriae and consularis Siciliae.12 But Kahlos’ statement is not true, because in Italy between Diocletian’s and Julian’s reign there are many cases of senators that had more than one consular governorate, so Praetextatus’ cursus is quite ordinary.13 As a member of the senatorial aristocracy of 4th century Rome, it seems that he had not had political offices for 15–16 years after his urban prefecture: probably, he spent these years in otium litteratum, devoting himself to philosophy or literature, as it was typical for a nobleman in ancient Rome. This is evident from Symmachus’ letters addressed to him and from the example of Manlius Theodorus, a senator that retired from public life to study philosophy.14 Finally, he played a very important role in the religious sphere: he had both religious offices of the traditional Roman cult, but also practiced mystery cults. 15 In Macrobius’ Saturnalia, he is the leading character: the typical perfect Roman aristocrat with all the virtutes to which the 5th century AD aristocracy aspired. In this work, the idealised generation to which he belonged is called saeculum Praetextati.16 His funerary marble ara is a compendium of all the aspects of Praetextatus’ life. It comes from Rome, but the place of finding is not known. It is now displayed in the Musei Capitolini of Rome. The marble ara has a height of 127 cm, a width of  10 Cons. designatus in CIL VI.1777 = Dessau, ILS 1258 (Roma). 11 CIL VI.1777 = Dessau, ILS 1258; CIL VI.1778; CIL VI.1779 = Dessau, ILS 1259 (Roma). 12 CIL IX.329 = Dessau, ILS 5557a = AE 1999.511 (Canusium); Symm. Ep. 4.71. 13 For Kahlos’ theory, cf. Kahlos (2010b) 30–31. Against this, goes the cursus honorum of senators such as L. Crepereius Madalianus, corrector Flaminiae et Piceni and consularis Ponti et Bithyniae, or Fabius Titianus, corrector Flaminiae et Piceni and consularis Siciliae. For a list of senators who covered senatorial charges, see Cecconi (1994) 210–24. 14 Kahlos (2010b) 43–44; for his cultural interests, cf. Kahlos (2010b) 124–50. 15 We can see it in CIL VI.1779 = Dessau, ILS 1259 (Roma) above all; on his activity of restoration and protection of pagan temples, cf. Kahlos (1995) 39–47, while on his role in the clashes between the bishops in Rome, cf. Kahlos (1997) 41–54. For the Roman religious panorama of the 4th century AD, the cults and Praetextatus in this context, cf. Kahlos (2010b) 50–123; Jones (2011) 132–72; cf. Warrior (2006) 42–54, 80–92, and Machado (2011) 493–516, too. On the epigraphic testimonies about the last pagan priests of Rome, cf. Orlandi (2011) 425–66. 16 On Macrobius’ work on Praetextatus as one of the main characters, cf. Kahlos (2010b) 180– 200; Jones (2011) 231–72.

Latin Epigraphy and Literary Texts in 4th Century AD Rome  191

73,5 cm and a depth of 53 cm. The epigraphic field has a height of 46 cm and a width of 46 cm, while the letters size is of 3–1,5 cm. The ara is rebuilt by two merging fragments and it has not the upper right angle, while there is a lacuna in the end of the inscription on the right side. On the upper molding, an inscription engraved in 1750 can be found. It was restored in 1750 and in 2000 and it can be dated after 384.17 The inscription is engraved on all the four sides of the ara and commemorates this man, his wife and the concordia of their marriage.18 On the front side, we find Praetextatus’ cursus honorum, with both the civil offices (LL. 9–18) and the religious offices (LL. 1–8). Then, there is the presentation of Aconia Fabia Paulina,19 clarissima femina, with her religious offices and the mysteries in which she was initiated (LL. 19–22). At the end, we can read that they lived together for forty years (L. 23). The inscription on the right side is dedicated to Paulina: it lists all the virtues of the woman and her great devotion towards her husband and Rome (LL. 3–8). On the left side, we find Praetextatus’ praise to Paulina. He exalts the virtues of his wife highlighting the strong bond that tied them (LL. 3–15). Finally, the part engraved on the back side concerns Paulina’s celebration of Vettius Agorius Praetextatus: she praises her husband both from a civil (LL. 3–7) and a religious point of view (LL. 8–29). The latter point is more extensive and lists Praetextatus’ functions as a pagan priest and his role as a witness during the initiation of his wife. Moreover, Paulina attributes to Praetextatus her prestige among Roman matrons (LL. 30–37). Praetextatus’ epitaph is peculiar because of its structure. While the text engraved on the front side is a classic funerary commemorative inscription for the two dedicatees, the other parts are a proper funerary poem engraved on a marble ara. The text of these other three parts, indeed, is in iambic senarii and, just for a few lines, in elegiac couplets.20 CIL VI.1779, cf. pp. 3174, 3814, 4757–4759 = Dessau, ILS 1259 (Roma) = CLE 111 = SupplIt Imagines – Roma 1, 36; Ensoli, La Rocca (2000) 507–8, nr. 130; Kolb, Fugmann (2008) 66–70, nr. 14, LL. 1–18 = HD017684 (A. Niquet) = EDR121930 = La Rocca, Parisi Presicce (2010) 344–7, nr. 2 with photo (A. Danti) = http://www.musei

 17 Ensoli–La Rocca (2000) 507, nr. 130. 18 On this inscription and the role of Paulina in relation to her husband, cf. Kahlos (1994) 13– 25. 19 PLRE I, Paulina 4. 20 L. 41 back side; LL. 3–15 left side; LL. 3–5 right side, following the order in http://www.mqdq. it/public/ce/testo/CLE/111.

192  Rosa Lorito capitolini.org/it/percorsi/percorsi_per_sale/palazzo_nuovo/stanzette_terrene/a ra_funeraria_di_vettius_agorius_praetextatus_e_aconia_fabia_paulina = http:// www.mqdq.it/public/ce/testo/CLE/111 = CCCA 3, 246 = RICIS 2, 501/180. Front side D(is) M(anibus) Vettius Agorius Praetextatus augur, p[o]ntifex Vestae, pontifex Sol[is], quindecemvir, curialis Herc[u]lis, sacratus Libero et Eleusi[ni]s, hierophanta, neocorus, tauroboliatus, pater patrum, in [r]e publica ver[o] quaestor candidatus, pretor urbanus, corrector Tusciae et Umbriae, consularis Lusitaniae, proconsule Achaiae, praefectus urbi, legatus a senatu missus V, praefectus praetorio II Italiae et Illyrici, consul ordinarius designatus, et Aconia Fabia Paulina c(larissima) f(emina), sacrata Cereri et Eleusiniis, sacrata apud Eginam Hecatae, tauroboliata, hierophantria. Hi coniuncti simul vixerunt ann(os) XL.









Right side Vettius Agorius Praetextatus Paulinae coniugi Paulina, veri et castitatis conscia, dicata templis atq(ue) amica numinum, sibi maritum praeferens, Romam viro, pudens, fidelis, pura mente et corpore, benigna cunctis, utilis Penatibus caldas [– – –] us [– – –] n.

Left side Vettius Agorius Praetextatus Paulinae coniugi. Paulina, nostri pectoris consortio



Latin Epigraphy and Literary Texts in 4th Century AD Rome  193

fomes pudoris, castitatis vinculum amorque purus et fides caelo sata, arcana mentis cui reclusa credidi, munus deorum, qui maritalem torum nectunt amicis et pudicis nexibus, pietate matris, coniugali gratia, nexu sororis, filiae modestia, et quanta amicis iungimur fiducia, aetatis usu, consecrandi foedere, iugi fideli simplici concordia iuvans maritum, diligens, ornans, colens.







Back side [Sple]ndor parentum nil mihi maius dedit [quam] quod marito digna iam tum visa sum, [se]d lumen omne vel decus nomen viri, Agori, superbo qui creatus germine patriam, senatum coniugemq(ue) inluminas probitate mentis, moribus, studiis simul, virtutis apicem quis supremum nanctus es. Tu namque quidquid lingua utraq(ue) est proditum cura soforum, porta quis caeli patet, vel quae periti condidere carmina, vel quae solutis vocibus sunt edita, meliora reddis quam legendo sumpseras. Sed ista parva: tu pius m[y]stes sacris teletis reperta mentis arcano premis, divumque numen multiplex doctus colis, sociam benigne coniuge nectens sacris, hominum deumque consciam ac fidam tibi. Quid nunc honores aut potestates loquar hominumque votis adpetita gaudia, quae tu caduca ac parva semper autumans divum sacerdos infulis celsus clues? Tu me, marite, disciplinarum bono puram ac pudicam sorte mortis eximens, in templa ducis ac famulam divis dicas; te teste cunctis imbuor mysteriis; tu Dindymenes Atteosqu[e] antistitem teletis honoras taureis consors pius; Hecates ministram trina secreta edoces Cererisque Graiae tu sacris dignam paras. Te propter omnis me beatam, me piam celebrant, quod ipse bonam disseminas,













194  Rosa Lorito totum per orbem: ignota noscor omnibus. Nam te marito cur placere non queam? exemplum de me Romulae matres petunt subolemque pulchram, si tuae similis, putant. Optant probantque nunc viri, nunc feminae, quae tu magister indidisti insignia. His nunc ademptis maesta coniunx maceror, felix, maritum si superstitem mihi divi dedissent, sed tamen felix, tua quia sum fuique postque mortem mox ero.





Front side To the gods Mani. Vettius Agorius Praetextatus, augur, pontifex Vestae, pontifex Solis, quindecemvir, curialis, Herculis sacratus, Libero et Eleusiniis hierophanta, neocorus, tauroboliatus, pater patrum in re publica vero, quaestor candidatus, praetor urbanus, corrector Tusciae et Umbriae, consularis Lusitaniae, proconsul Achaiae, praefectus urbi, legatus a senatu missus V, praefectus praetorio II Italiae et Illyrici, consul ordinarius designatus, and Aconia Fabia Paulina, very illustrious woman, sacrata Cereri et Eleusiniis, sacrata apud Aeginam Hecatae, tauroboliata, hierophantria, these lived united together for forty years.









Right side Vettius Agorius Praetextatus to the wife Paulina. Paulina, partner of the truth and of the chastity, devoted to temples and friend of the gods, who preferred her husband to herself, Rome to the husband, demure, faithful, pure in mind and body,



Latin Epigraphy and Literary Texts in 4th Century AD Rome  195

benevolent towards everyone, useful to the Penates, [– – –]

Left side Vettius Agorius Praetextatus to his wife Paulina. Paulina, partnership of my heart, hearth of modesty, bond of chastity and pure love and fidelity sown in the sky, secrets of the mind in which I believed open, gift of the gods, who tie the bridal bed with propitious and modest ties, with the mercy of a mother, with bridal grace, with the bond of a sister, with the modesty of a daughter, and with how big trust we are united with the friends, with the custom of the age of consecrating the marital yoke by a pact, with faithful simple concord, helping the husband, loving him, celebrating him, honoring him.





Back side The glory of my parents gave me nothing bigger, than seeming since then worthy of the husband. But all the light or the glory is in the name of the husband, Agorius, that, born from a superb seed enligten the homeland, the senate, the wife with the rectitude of the soul, with the good customs, together with the dedication, with these things you reached the supreme apex of the virtue. Indeed, you have unveiled everything with the one and the other one language cure of the wise men, to whom the door of the heaven is open, or the verses that experts had composed or that have been produced with a free voice, you recite better things than the ones you have learned by reading. But these are little things: you, pious mystes, hide the things discovered by the sacred initiations with the secret of the mind and, learned, venerate the divine deity, binding beningly the partner as a colleague with sacred things conscious of the men and of the gods and faithful to you. What could I tell now about the honors and the powers and the joys desired by the vows of the men? That you, always considering these things frail and small, have a reputation for high priest of the gods with the sacred bandages? You, husband, freeing me from the fate of death pure and demure with the good of the teachings lead me to the temples and call me servant of the gods.









196  Rosa Lorito Being you witness, I am initiated into all the misteries, you honor the priest of Dindymenes and Attis with taurine initiations, as a pious partner. Educate me as a priestess of Hecate to the three mysteries and prepare me worthy of the sacred things of the Greek Ceres. Because of you everybody celebrate me as blessed, pious, because you propagate me as good, unknown worldwide, I am recognized by everybody. Why could I not be liked with you as a husband? Roman mothers search me as an example and think that their offspring is handsome, if it is similar to you. Both the men and the women choose and approve the extraordinary things that you as a master imposed to me. Being these things stolen, I fret myself as a sad spouse, happy, if the gods had given the survivor husband to me, but however happy, because I am yours and I was and I will be then after death.









First of all, we shall analyse the text from a metrical point of view, focusing on the back part of the epitaph as an example. It is in a sort of dramatic dialogue form. In general, the use of iambic senarii and, less, of elegiac couplets, is quite peculiar, because it applies features which are typical of a literary text to an epigraphic text, transforming it in poetry. The iambic senarii, in particular, were the meter mainly used in Plautus’ and Terentius’ works and it derives from the Greek iambic trimeter, so close to the manner in which people talked and therefore used for the episodes in Greek comedy and tragedy. In this case, I think that the use of iambic senarii wants to convey the idea of a ‘dialogue’ between the two spouses, creating a parallelism with theatre texts.21 From textual hints, we can also infer that it was written as an epitaph from the beginning and then fit into being engraved surface of the marble ara. It is not clear, however, whether it was made after Praetextatus’ or Paulina’s death or after they both passed away. Likewise, it is not known who wrote it. In Kahlos’ opinion, only one author is to be identified, because of the text’s uniformity.22 Cameron, by analyzing the structure of the text, suggests that the three poems were not originally composed to be inscribed on a funerary monument, but rather  21 Another funerary inscription in iambic senarii is ILLRP 819 = CIL I.2 3146; cf. Buonopane (2009) 206, fig. 8.22. An example of epitaph in verses is also given by the one of Petronius Probus, cf. Trout (2001) 157–76, or the eight elegiac couplets epitaph engraved on the sarcophagus of Iunius Bassus signo Theotecnius, CIL VI.41341a, cf. PLRE I, Bassus 15, Porena (2005) 223. 22 On the theories about the composition of the poem, cf. Kahlos (2010b) 220.

Latin Epigraphy and Literary Texts in 4th Century AD Rome  197

adapted to this function by adding few lines, such as LL. 38–41. Then, observing the spacing on the front side, he also supposes that Praetextatus’ poems on Paulina were added later, after her death.23 I believe that both opinions are right: the poem could have been written in a first moment, while both were alive, to celebrate their union, and then become a funerary text thanks to the addition of a few lines. The funerary inscription of Vettius Agorius Praetextatus is also noteworthy because of the predominance of religious over civil offices. We can see this in two parts of this text. First, in the front side of the ara, where the religious offices are placed before the civil ones in the cursus honorum. Secondly, on the back side, Paulina emphasises his merits and religious duties, remembering how her husband considers civil offices temporary and trivial, and prefers to be regarded as a priest (LL. 18–21).24 Because of the preference given to his ‘religious’ functions, Praetextatus is seen as a ‘champion’ of the pagan world. Nonetheless, this was typical of the epitaphs belonging to the aristocratic men, was a way to express the values of the senatorial aristocracy.25 The text on the back side is also rich in references to pagan gods and cults: Dindymene (Cybele) (L. 26), Attis (L. 26), Hecate (L. 28), Ceres (L. 29). The list of the pagan gods is relevant in a period in which Christianity was becoming increasingly powerful. As such, this text seems to be an attempt at resisting Christianity in 4th century AD Rome, a city witnessing clashes between these two powers. However, during Praetextatus’ age there was still a climate of reasonable tolerance and co–existence.26 If we compare epigraphic Christian testimonies and Praetextatus’ poems, we can see a lot of common features that underline how pagan and Christian culture lived together. For example, in both pagan and in Christian funerary inscriptions, the elogium insisted on the social and familiar condition of the deceased. Noteworthy is also the use of terms such coniunx, typical of Christian epitaphs,27 or the emphasis on values such as pudor (L. 6 on the right side, LL. 4, 8 on the left side), fidelitas (L. 6 on the right side, LL. 5, 13 on the left side), pietas (L. 9 on the left side), modestia (L. 10 on the left side), benignitas (L. 7 on the right side), puritas (L. 6 on the right side), or the idea of the life in heaven after death.

 23 Cameron (2011) 301–3. 24 Kahlos (2010b) 143–4. 25 Kahlos (2010b) 212. 26 Kahlos (2010b) 50–61; Jones (2011) 4–5. 27 Carletti (1986) 18 and inscriptions nr. 23, 24, 25, 26, 27, 29, 31, 49, 52, 91, 92, 103, 115, 124.

198  Rosa Lorito One example of a Christian funerary epitaph similar to Praetextatus’ text is ICURns V.13196:28 ICURns V.13196 = Carletti (2008) 183 nr. 66 with photo = EDB 2645 Vona fidelis casta dulcissima prudens femina Patricia innox, / q(uae) vixit ann(os) XXI m(enses) VI d(ies) XVIII, q(uae) fecit cum Cleopatro virg(inio) suo / orfano nunc relincto cum omne religione et castitate ann(os) IIII m(ensem) I d(ies) XIIII. / Depos(i)t(a) est d(ie) XI kal(endas) mart(ias). Benemerente in pace. L(ocus) b(isomus). Good, faithful, chaste, very sweet, Patricia, innocent woman, who lived for 21 years, 6 months, 18 days, who spent with his virginal Cleopater, now left as an orphan, with every religiosity and chastity 4 years, 1 month, 14 days. She was laid on the eleventh day before the calends of March. Being praiseworthy in peace. Tomb made for two people. The sepulchral inscription is engraved on a marble table and comes from the underground cemetery at the Catacombs on the via Appia. It can be dated between 300 and 399. Here we can see the use of adjectives such as fidelis, casta, dulcissima, prudens, innox (L. 1), that we can find in Paulina’s presentation. Around the middle of the 4th century AD, we can see also how Christian funerary epigraphy started to employ more and more verses, using schemes and themes typical of the late–Roman funerary elogium, with influences of Virgil’s poetry. This is another point in common between pagan and Christian cultures and it points towards the spread of Christianity in higher social ranks.29 The funerary inscription of Vettius Agorius Praetextatus and Aconia Fabia Paulina is an important testimony on many different levels. First, it is an example of a literary text in an epigraphic form, an epitaph that unites the typical features of a funerary inscription and those of an epigram. At the same time, it unites features of pagan and Christian texts. Finally, it may also have been a propagandistic text because, through the commemoration of this very important pagan man, his wife and the concordia of their marriage, pagan aristocratic values and  28 ICURns V.13196, cf. Carletti (1986) 55–56 nr. 33 and Carletti (2008) 183 nr. 66. Another example of a Christian epitaph similar to Praetextatus’ is ICURns III.8470, cf. Carletti (2008) 185 nr. 69. 29 Carletti (1986) 22–23. For the ‘pagan features’ in Christian epigrams, cf. Carletti (2008) 72–74. An example of Christian funerary memory in epigrams could be the ‘Episcopal epigraphy’ of Damasus, cf. Carletti (2008) 62–63; 78–90. On this topic, cf. also Trout (2015) 12–20, in particular 19–20 on Damasus’ verses. Moreover, the inscription installed in the first Roman basilica dedicated to martyr Agnes founded by the empress Constantina is an example of fourteen–exameter inscription with references to Virgil and Ovid, ICUR II.1, 44–45, cf. Ihm (1895), nr. 84, cf. Trout (2014) 263–82, part. 264–5. See also Trout (2013) 215–34.

Latin Epigraphy and Literary Texts in 4th Century AD Rome  199

religion were celebrated and defended from the attacks of Christianity, which was becoming increasingly powerful. In this context, the use of epigraphy as an instrument to make the memory of these values eternal was certainly more relevant.

Alan Taylor Farnes

The Scribal Habits of Codex Sangermanensis in Greek and Latin in Light of its Exemplar James R. Royse, in his 2008 Scribal Habits in Early Greek New Testament Papyri, calls upon New Testament text critics to re–examine NT witnesses according to what he calls the ‘Colwell method’. This method analyzes singular readings in each manuscript—readings which only exist in one manuscript. It is then assumed that a singular reading is the invention of the copyist of the manuscript.1 Royse calls upon text critics to assist in this work, saying Ideally, of course, all the major witnesses to the text of the New Testament — that is, the continuous–text Greek manuscripts, the lectionaries, the versional manuscripts, and the Fathers — would be studied in detail in order to provide this same kind of information concerning scribal habits, translational tendencies, and so on. One’s assertions could then be based on empirical evidence about the witnesses.2

Royse entreats text critics to systematically and methodologically reanalyze all of the witnesses of the New Testament with respect to singular readings in order to ascertain each scribe’s tendencies.3 Royse invites his fellow text critics to perform the same type of analysis of the witnesses as the study he performed on the six

 1 See Royse (2008) 39: ‘[Colwell’s] view is that the singular readings of a manuscript are the textual creations of the scribe, and thus than analysis of the patterns found within these singular readings will reveal the habits of the scribe’. See also Colwell (1969) 108: ‘This study is restricted to singular readings (readings without other manuscript support) on the assumption that these readings are the creation of the scribe. The restriction of this study to singular readings can be made with confidence in view of the wealth of manuscript attestation for the Greek New Testament. A singular reading has been defined as a reading which has no Greek support in the critical apparatus of Tischendorf’s 8th edition’. 2 Royse (2008) 13. See also his earlier work: Royse (1995) 239–52 = Royse (20132) 461–78. 3 The need for these studies are emphasized: ‘The general habits serve, then, as the basis of our knowledge of transcriptional probability (and improbability): what sorts of alterations scribes are likely (or unlikely) to have made in the text’, Royse (2008) 4. Quoting Ernest C. Colwell, Royse urges text critics to ‘begin at the beginning’ and ‘commence with the oldest witnesses and work down the stream of tradition’, Royse (2008) 14. Here he quotes Colwell (1969) 159. Again, Royse reemphasizes, ‘The scribal (and translational) peculiarities of all the major sources for the New Testament should be catalogued’, Royse (2008) 738. https://doi.org/10.1515/9783110632590-014

202  Alan Taylor Farnes extensive early Greek papyri. To Royse, such an undertaking is imperative in order to determine the very foundational principles of textual criticism.4 Royse’s massive work disproved, at least with respect to the six early papyri which he analyzed, the long held text critical maxim lectio brevior potior.5 In its place Royse concludes that, ceteris paribus, these early New Testament scribes tended to omit rather than to add.6

 4 ‘We wish to find a way to characterize the habits of scribes that will avoid, as far as is possible, both any question–begging assumptions about scribal behavior and any controversial presuppositions about the history of the text. Some presuppositions are, as we shall see, necessary; but they should be as limited as possible in order that the results attained can be used with confidence in deciding textual issues’, Royse (2008) 31. 5 This text critical maxim states: ‘The shorter reading (unless it lacks entirely the authority of the ancient and weighty witnesses) is to be preferred to the more verbose, for scribes were much more prone to add than to omit. They scarcely ever deliberately omitted anything, but they added many things; certainly they omitted some things by accident, but likewise not a few things have been added to the text by scribes through errors of the eye, ear, memory, imagination, and judgement. Particularly the shorter reading is to be preferred, even though according to the authority of the witnesses it may appear to be inferior to the other,— a. if at the same time it is more difficult, more obscure, ambiguous, elliptical, hebraizing, or solecistic; b. if the same thing is expressed with different phrases in various manuscripts; c. if the order of words varies; d. if at the beginning of pericopes; e. if the longer reading savours of a gloss or interpretation, or agrees with the wording of parallel passages, or seems to have come from lectionaries. But on the other hand the longer is to be preferred to the shorter (unless the latter appears in many good witnesses),—a. if the occasion of the omission can be attributed to homoeoteleuton; b. if that which was omitted could have seemed to the scribe to be obscure, harsh, superfluous, unusual, paradoxical, offensive to pious ears, erroneous, or in opposition to parallel passages; c. if that which is lacking could be lacking without harming the sense or the structure of the sentence, as for example incidental, brief prepositions, and other matter the absence of which would be scarcely noticed by the scribe when re–reading what he had written; d. if the shorter reading is less in accord with the character, style, or scope of the author; e. if the shorter reading utterly lacks sense; f. if it is probable that the shorter reading has crept in from parallel passages or from lectionaries’, Metzger–Ehrman (20054) 166–7. Lectio Brevior Potior is still widely accepted by many scholars today: Zuntz refers to Griesbach’s canons as ‘a series of carefully worded rules which gave the essense of his vast experience’, Zuntz (1953) 6; ‘The canon states that “the shorter reading […] is preferable to the more verbose;” this, says Griesbach — quite correctly, is based on the principle that scribes are for more prone to add to their texts than to omit’, Epp (1976) 225–6; ‘The venerable maxim lectio brevior lectio potior (“the shorter reading is the more probable reading”) is certainly right in many instances’, Aland–Aland (19872) 281; ‘In general the shorter reading is to be preferred’, Metzger (19942), 13* (see also 10*–4*); ‘Griesbach showed great skill and tact in evaluating the evidence of variant readings […]. The importance of Griesbach for New Testament textual criticism can scarcely be overestimated’, Metzger–Ehrman (20054) 167. 6 See Royse (2008) 719–20.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  203

Royse’s method, however, is not without its shortcomings mostly due to the fact that the method necessitates the postulation of an unknown Vorlage. David C. Parker voices two questions with the singular readings method: first that, due to incomplete collations, some supposed singular readings may not in fact be singular.7 Second, singular readings are only singular readings until another manuscript is found which contains the same reading. In such a case, the reading is no longer singular.8 Parker accepts however that overall such difficulties may not be fatal and that ‘it has to be acknowledged that such tendencies do emerge, and with them evidence about the way in which scribes went about their work’.9 Parker questions Royse’s conclusion that lectio brevior be reversed in favor of the longer reading on a grand scale but accepts Royse’s findings with respect to the early papyri. Parker wants to be sure that Royse’s findings are not expanded outside of their proper time frame unless further research is conducted. Holger Strutwolf finds flaws with the singular readings method noting some of the same problems that have been noticed by other scholars such as Parker; namely that because ‘it is far from sure that the singular readings a manuscript contains really are the individual readings [i.e. scribal creations] of the scribe who produced the manuscript’ therefore ‘the singular readings of a certain manuscript might have been invented by its scribe, but could also already have existed in the ancestor or even been passed down through three or more generations of ancestors now lost’.10 I agree with Strutwolf’s critique and it is the main reason why I endeavor to find a different method for determining scribal habits. Royse’s method employing singular readings necessitates the use of hypothetical and reconstructed Vorlagen. Royse concedes that the method is not perfect and that not all scribally created readings will be found among the singular readings.11 Additionally, there is no way to know if a singular reading is indeed scribally created or if the scribe inherited the reading from a now lost exemplar. Royse’s method is indeed one of the best and most fruitful methods available but it is not without its flaws. This essay, therefore, presents another method of determining scribal habits by examining manuscripts with extant exemplars.   Parker (2009) 256.  Parker (2009) 256. Such is the case in my analysis of P127. Many readings which were previously singular readings in Codex Bezae (05) are no longer singular because of their support by P127.  Parker (2009) 256–7. 0 Strutwolf (2012) 142. Malik agrees saying: ‘It needs to be noted that, since singulars do not constitute all of the errors in the manuscript and since some of them may even have been inherited from the exemplar, what follows must be regarded as an approximation of the scribe’s copying behaviour rather than an absolute account thereof’, Malik (2017) 167, emphasis in original. 1 See Royse (2008) 42.

204  Alan Taylor Farnes Where possible (which is admittedly rare), an analysis of scribal habits of manuscripts with a known Vorlage provides a clearer insight into actual scribal habits. Royse says as much himself: There has been (it seems) a failure to explore the problem of scribal habits for the text of the New Testament in the best possible case, namely, where the Vorlage of an extant manuscript is also known to be extant. In such a situation we can virtually look over the scribe’s shoulder and compare the text he is copying with his result.12

He continues: ‘The earliest example among New Testament manuscripts of an exemplar and its copy is D (06), Codex Claromontanus, and Dabs1 [Codex Sangermanensis]’.13 This essay attempts to ascertain the scribal habits of Codex Sangermanensis in light of its exemplar Codex Claromontanus. Codex Claromontanus (06) is a fifth–century majuscule bilingual parchment codex of the Pauline epistles housed at the Bibliothèque nationale de France in Paris.14 Its exact copying location is unknown but was ‘manifestly in the West and possibly in South Italy’15 and ‘possibly in Sardinia’.16 Jean Irigoin preferred Sicily as their place of origin rather than southern Italy.17 Alexander Souter argued that the Vorlage of the Latin text of Claromontanus was the same Vorlage used by Lu-

 12 Royse (2008) 34, emphasis added. 13 Royse (2008) 35. 14 Aland (1994) gives a date of 6th century but Parker, Lowe, Frede, Gryson, Eymann and Fröhlich all date it to the 5th century. See Parker (2008) 259; Lowe (Oxford 1950) V.3; Frede (1973) I.76; Frede (1975) I.23–24; Frede (1962) I.11*–12*; Gryson (1999) I.119. Eymann (1996) I.19; Fröhlich (1995) I.26. See Aland (1994) 19. See also Tischendorf (1832) ix–xiv for further description of the manuscript. 15 Lowe (1950) V.3. See also Lowe (1925) 205 where Lowe concludes that Claromontanus may have been written in Italy due to the fact that it at times employs a large letter at the beginning of each page on the Latin side of the text. Kenyon also assigned Claromontanus to Southern Italy: Kenyon (1900) 297. Of its location in light of the use of the bd uncial Frede comments: ‘Die bd– Unziale ist ein toter Nebenzweig am Baum der Entwicklung der lateinischen Schrift, und ihre Verwendung in unserer Handschrift läßt von vornherein darauf schließen, daß sie nicht in den großen lateinischen Schreibzentren Italiens entstanden sein kann, sondern nur in einem abgelegenen, provinziellen Scriptorium, das der Entwicklung nicht folgte’, Frede (1964) 19. Eymann, Fröhlich and Gryson also say southern Italy: Eymann (1996) I.19; Fröhlich (1995) I.27; Gryson (1999) I.119. Perhaps Sardinia would qualify as not a part of the great Latin writing centres of Italy and as a remote provincial scriptorium. In 1973, however, Frede states that Claromontanus was copied in Southern Italy: Frede (1973) I.76. 16 Parker (2008) 259. 17 See Irigoin (1996) 11.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  205

cifer of Cagliari and that that bilingual manuscript originally belonged to Sardinia.18 Souter described the historical situation that necessitated a Greco–Latin copy of the bible detailing that in 533 ‘Sardinia came into the possession of the Eastern Byzantine empire, the language of which was Greek. Hence the necessity for a Greek version of the Bible in the island. The inhabitants spoke Latin, the invaders Greek. A bilingual bible was a necessity for Church services. Such a codex I believe Claromontanus to have been’.19 Claromontanus is a descendent of a fourth–century bilingual edition of Paul.20 The whole of 06 was originally copied by the same scribe21 except for a few supplemental folia which were added in the 6th century. Frede described it in 1964 as having 533 sheets of very fine parchment22 and Scrivener says that the sheets are ‘the thinnest and finest vellum’.23 Codex Claromontanus is lacunose in Rom 1.1–7 and is supplemented at Rom 1.27– 30 and 1 Cor 14.13–22. In the second half of the 8th century Claromontanus was brought to Corbie, France.24 The codex is notable for having a total of nine correctors with varying amounts of corrections among them. Most of the corrections were made by a scribe who Tischendorf called D*** and who worked in the 9th century.25 This same scribe also added the accents and breathings. Sometime after D*** finished correcting, Codices Sangermanensis (0319/VL76) and Waldeccensis (0320/VL83) were copied from Codex Claromontanus. Between 1565 and 1582 Theodore Beza found the manuscript in the abbey of Clermont, France (from whence it received the name Claromontanus).26 It then somehow soon thereafter came into the possession of Claude Dupuy27 and was passed down through the Dupuy family until it was donated to the Bibliothèque nationale de France in Paris in 1657 28 where it remains to this day.

 18 See Souter (1905) 241. 19 Souter (1905) 242. 20 See Gryson (1999) I.119; Frede (1975) I.23. 21 See Lowe (1925) 204. 22 See Frede (1964) 16: ‘Die Handschrift besitzt heute insgesamt noch 533 Blätter aus sehr feinem Pergament’. 23 Scrivener (1883) 163. 24 Thus Gryson: ‘In der zweiten Hälfte des 8. Jahrhundert gelangte der Claromontanus nach Corbie, wo ein ausgeprägtes Interesse an alten Texten bestand [...] und wurde dort oder in der Nähe mehrfach kopiert’, Gryson (1999) I.119. See also Frede (1975) I.23. 25 Tischendorf (1832) xxv. 26 See Gryson (1999) I.120; Kenyon (19122) 80; Scrivener (1883) 164. 27 See Frede (1975) I.23. 28 See Salgado (2002) 40. See also Gregory (1900) 107–8.

206  Alan Taylor Farnes

 Codex Claromontanus and its Relatives Parker then gives a stemma of this group of manuscripts:29

Fig. 1: Stemma explaining the relationships of this group of manuscripts.

0319 is our earliest example of a known duplicate New Testament manuscript. Additionally, Codex Claromontanus is the only extant New Testament manuscript to have two known duplicates in 0319 and 0320. 0319 and 0320 are also the only known duplicate majuscule manuscripts. Lastly, all of these manuscripts are diglots containing both Greek and Latin. According to Parker, there are only twenty–four bilingual Greco–Latin New Testament manuscripts of which we will be discussing three in this study and the above 010 and 012 are two others.30 In studying this group of duplicate manuscripts I was fortunate that Tischendorf and others have already studied Codex Claromontanus in depth and have provided a transcription.31 I relied heavily on Tischendorf’s transcription to double–check my work as a resource especially where the manuscript is difficult to read. His appendix concerning corrections was particularly useful. The Bibliothèque nationale de France provides high–resolution full colour images of this codex.32

 29 Parker (2008) 260. This stemma is found in Parker’s Introduction but the explanation of the stemma originates with Corssen (1887). See also Frede (1964); Royse (2012) 179. 30 Parker (2008) 69. 31 Tischendorf (1832). Swanson also includes 06 in his parallel line transcription in the volumes which I consulted which were Romans and 2 Corinthians. See Swanson (2001); Swanson (2005). 32 Images can be accessed through the library’s website (www.bnf.fr) or found more directly here: http://gallica.bnf.fr/ark:/12148/btv1b84683111 and here: http://gallica.bnf.fr/ark:/12148 /btv1b10515443k.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  207

. Codex Claromontanus (06) and Codex Sangermanensis (0319) Codex Sangermanensis is a ninth–century majuscule bilingual parchment codex of the Pauline epistles housed at the National Library of Russia in St. Petersburg.33 The codex contains 177 folia with 31 lines per page. It is 36 cm tall and 27.5 cm wide. The Greek text is a ninth–century biblical majuscule34 described as a ‘large, coarse hand’.35 The Latin text is a ninth–century uncial.36 The exact copying location is difficult to ascertain but many commentators conclude that 0319 was copied near Corbie, France where Claromontanus was located.37 Sangermanensis found its way from Corbie to Saint–Germain–des–Prés (from whence it received the name Sangermanensis) in Paris (140 kilometers south of Corbie) at an unknown time.38 During the French Revolution, Peter Dubrowski brought it to Saint Petersburg where it remains today.39 In two places in 0319 the words ‘Ex museo Petri Dubrowsky’ can be found (1r, 176v). The page is written in two columns with Greek on the left and Latin on the right. Being a copy of Codex Claromontanus, it

 33 See Aland (1994) 19; Treu (1996) 38. 34 See Barbour (1981) xvii. Note that Barbour uses the word uncial but accepts the difficulty with the term: ‘The word “uncial” is not very aptly used, especially of the later styles that developed from biblical uncial; but, since the precise meaning of the word has never been established nor any satisfactory alternative suggested, it remains the most convenient term to describe writing that is not miniscule and yet not entirely capitals’, Barbour (1981) xvi. I prefer to refer to the hand as a majuscule hand but have here used the term that Barbour supplies. See Turner (19872) 1: ‘Let us call handwriting in which each letter “stands independent and for itself” writing in capitals (not “uncials”, a word which has a precise meaning in Latin writing, but only a derived and imprecise one in Greek)’. 35 Kenyon (19122) 102. Scrivener: ‘Its uncials are coarse, large, and thick’, Scrivener (1883) 166. 36 See Houghton (2016) 243; Frede (1964) 35–36. 37 See Frede (1973) I.76; Frede (1964) 42–45; Frede (1975) I.25; Gryson (1999) I.122; Bredehorn, (1999) 470–1; Treu (1996) 38. Fröhlich shows that one reason why Corbie is thought to be the location of origin for Sangermanensis is because there is a section in 1 Cor 14 where a different form of the Latin text appears. Rather than the D–text, the I–text appears in 1 Cor 14. This I–text is peculiar to Corbie: ‘Es liegt nahe, sich den Sangermanensis im Bereich von Corbie entstanden zu denken, da sich der I–text, den 76 in 1 Cor 14 für einige Zeilen als Lückenbüßer verwendet, über die Ambrosiasterüberlieferung nach Corbie lokalisieren läßt’, Fröhlich (1995) I.28. Houghton says: ‘possibly in France’, Houghton (2016) 243 and Eymann simply says northwest France, Eymann (1996) I.20. 38 See Frede (1975) I.25; Fröhlich (1995) I.28; Gryson (1999) I.122; Gregory (1900) 110; Kenyon (19122) 84; Treu (1996) 38. 39 See Frede (1975) I.25; Fröhlich (1995) I.28; Gryson (1999) I.122; Gregory (1900) 110; Treu (1996) 38; Nestle (1901) 77.

208  Alan Taylor Farnes was formerly referred to as Dabs1. Like 06, it contains all of the canonical Pauline epistles, including Hebrews, in canonical order but with Colossians and Philippians transposed. One notable difference between 0319 and 06 is that 0319 copies both Greek and Latin on the same page whereas 06 (and 0320) copies Greek on the verso and Latin on the recto. 0319 for the most part follows the same line division as 06 but since 0319 uses 31 lines while 06 only uses 21 the page divisions of course do not line up. David Parker shows that 0319 actually employed two different scribes: the first for folia 1–68 and the second for the remainder.40 Eymann agrees that 0319 was written by two different scribes41 and Gryson in 1999 agrees that the scribes change at folium 68.42 Scribe A copied folia 1r–68r and Scribe B copied 68v–176v. The scribes changed places at the beginning of a page which is at the end of 1 Cor 16.14 at a sense break. Because there are two scribes in 0319 I will therefore divide the analysis of scribal habits between Scribe A (0319A) and Scribe B (0319B). I have used test passages to gain insight into the copying habits of the scribe of 0319. The test passages for 0319A are Rom 1.7–32. For 0319B the test passages are 2 Cor 12.7–20; Eph 1.3–9, 13–19; 2.11–18; and Titus 1.9–3.11. In total, the test passage for 0319A comprises 58 N–A lines which is just over two N–A pages and is 468 words. The test passages for 0319B comprise 140 N–A lines which is five and a half N–A pages for 1,150 words.43

. Codex Sangermanensis as a copy of Codex Claromontanus As I will show more fully below, there can simply be no doubt that 0319 is a direct copy of 06. Text critics have long accepted this theory. Griesbach was convinced that 0319 copied 06: ‘There is no doubt at all that our codex [0319] is a copy of Claromontanus’.44 Hort commented at length on the relationship between these two manuscripts: ‘The St Germain MS E3 [0319], apparently written in Cent. X or later in Cent. IX, has long been recognised as a copy of the Clermont MS D2 [06], […] all possible doubt as to the direct derivation of the one from the other is taken  0 See Parker (1992) 65. 1 See Eymann (1996) I.20. 2 See Gryson (1999) I.122. 43 See Royse (2008) 898–99 for the definition of and how to calculate N–A lines. The method is an attempt to estimate approximately how many variant readings there would be on a typical page of Nestle–Aland with a typical page being twenty–five lines. 44 ‘Dubitari prorsus nequit, nostrum codicem Claromontani esse apographum’, Griesbach (1793) 77.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  209

away by the senseless readings which the scribe of [0319] has constructed out of a combination of what was written by its correctors’.45 Most recently, Houghton has added his voice in agreement that Sangermanensis is indeed a copy of Claromontanus.46 I am in full agreement that 0319 is indeed a copy of 06 but I will reserve my comment on the quality of copying until later. There are a few places in 0319 where there can be no doubt that the scribes were looking directly at 06. Additionally, it will become clear that 0319 was copied by letter and not by dictation.47 Both Scribe A and Scribe B make obvious errors that reveals that they are clearly copying directly from 06 rather than from an intermediary source or through dictation. At 2 Cor 12.15 NA28 contains the word ἧσσον. The TR has the Atticized ηττον.48 The first hand of 06 reads ησσον but 06*** corrected this to ηττον by writing tiny tau inside the sigma with the result that the final text looked something like a majuscule epsilon or an uppercase Roman G. 0319B apparently was confused and so copied the text as graphically correct as possible by writing sigma with small tau inside. In this instance we can be certain that 0319B was looking directly at 06 while copying this verse. Because they49 correctly copied 06 this reading is not considered an error and is not included in the statistics.  45 Hort (1882) II.254. 46 Houghton (2016) 80, 243–4. 47 Parker states that copying by letter was generally the norm in the Byzantine period: ‘That Byzantine scribes copied by eye is suggested by the typical portrait of an evangelist, sitting in a scribal posture with the exemplar on a lectern in front of him’, Parker (2008) 155. Parker then refers to an image of a scribe from GA 713 F2r. 48 See Gignac (1976) I.145–6: ‘The fluctuation between –σσ– and –ττ– likewise reflects the dialectal heritage of the Koine. Non–Attic words appear only in –σσ–, e.g., feminine nouns in –ισσα. The multiplicative numerals δισσός, τρισσός, etc., are spelled only with –σσ–, as is βυσσός, but other words are found in both spellings. Forms in –ττ– predominate in Byzantine times. Words which appear only in –ττ– are words which are specifically Attic or later formations or loanwords, e.g., πιττάκιον, ὁλοκόττινος’. See especially Gignac’s examples of this particular word on I.147. James Hope Moulton comments: ‘Κοινή has σσ almost exclusively where Attic had ττ, which was hardly used outside the contiguous districts of Attica and Boeotia’, Moulton (1906) II.107. Moulton later discusses the word ἧττων specifically on this same page. 49 Throughout this study I will employ the gender–neutral singular ‘they’ pronoun when referring to a scribe whose preferred personal gender pronoun is unknown (on the use of the singular ‘they’ see Warenda [1993] 99: ‘Observers have long pointed out the ambiguity of the use of the pronoun HE in generic contexts and the advantages of having a true generic singular pronoun, which would be sex–neutral. In the absence of such a sex–neutral pronoun, speakers of English have been expected to utter sentences such as “Everybody should bring his book tomorrow”, where the everybody referred to includes forty women and just one man. For centuries, speakers and writers of English have been happily getting around this obstacle by using THEY in such

210  Alan Taylor Farnes Another place where it is obvious that 0319B is copying from 06 is Eph 1.6. At the end of this verse the first hand writes υιω αυτου which is not present in NA28 or the TR. This addition is found in 06*, 010, 012, 0319, and 0320. This addition is then corrected by 06*** by placing a small slash through the first and last letters to be omitted. When 0319B arrived at these words they are again confused and, striving for perfect graphical accuracy and without understanding that the corrector intended these words to be excised, they copied these words with the slashes. Because, however, they correctly copied the Vorlage this reading is not counted as an error. The Latin text here does contain filio suo without any marks indicating that they should be excised. This is because the Latin side of 06 also contains filio suo without any deletion marks. To echo Griesbach, there can be no doubt that 0319 copied 06.50

 Notable Attributes of 0319 As mentioned above, 0319 was copied by two different scribes. Scribe B began copying on folio 68v and maintained the thirty–one lines per page format started

 situations, yielding sentences such as “Everybody should bring their book tomorrow”’. See also Foertsch–Gernsbacher [1997] 106: ‘With increasing frequency, writers and speakers are ignoring grammatical proscription and using the plural pronoun they to refer to singular antecedents. […] singular they is a cognitively efficient substitute for generic he or she’. For more recent treatments see https://www.grammarly.com/blog/use–the–singular–they/ and http://www.arrantpedantry.com/2016/10/31/stupidity–on–singular–they/). Ancient scribes were not in all cases male, on female scribes see Haines–Eitzen (2000) 41–52; Haines–Eitzen (1998) 629–46. David C. Parker also mentions that the 13th century Theodoros Hagiopetrites had a daughter named Eirene who was a scribe. See Parker (2012) 70. The British Library published the following blog post on the occasion of International Women’s Day 2017 detailing some of the manuscripts in the British Library which were written by women: http://blogs.bl.uk/digitisedmanuscripts/2017/03/female–scribes–in–early–manuscripts–.html. But the fact that it is a possibility that a woman was a scribe should be tempered by the fact that most scribes were indeed men. See, for example, Georgi Parpulov’s statement: ‘All but a few scribes were men, yet Hagiopetrites had a daughter who inherited the profession’, Parpulov (2012) 313. Nonetheless, this project will use gender– neutral language. I, however, cannot avoid instances when I cite a previous scholar who refers to a scribe as a male. In all other cases when the gender of a scribe is unknown I will use the gender–neutral pronoun ‘they’ or will repeat the noun itself to avoid using a pronoun. This will at times sound awkward but I will not speculate nor assume the gender of the scribes in this study. In short, this is an attempt to use inclusive language. 50 See Royse (2008) 75–76 for more examples of obvious copying of 06 by 0319 especially as it relates to difficulty interpreting corrections in the Vorlage.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  211

by Scribe A. 0319B completed the last remaining verses of 1 Corinthians and then began 2 Corinthians. When they began writing 2 Corinthians still on their first page of copying, however, 0319B greatly reduced the size of their letters in both Greek and Latin. It seems that 0319B understood that if they did not write smaller then they would run out of parchment or would use too much parchment. The codex in the end amounts to 177 folia and 0319B still had to write from 2 Corinthians to the end of the Pauline corpus in just the remaining 109 folia. One notable reading in 0319 is found in Rom 1.7. The crux of the issue is whether the epistle is meant specifically to those in Rome or to a general audience and therefore whether the words ‘in Rome’ are original or are later interpolations. Parker says that these words might be an addition ‘in the collected edition in order to identify the letters more clearly’ from one another.51 Connected to this issue is the mention of Rome also in Rom 1.15. Although 06 is deficient at Rom 1.7, 0319 indeed reads εν Ρωμη at this verse. Additionally, both VL75 and VL76 read Romae (VL76: Romę). We can therefore be confident that 06 originally contained εν Ρωμη even though it is lacunose at this point. Additionally, in Rom 1.15 both 06 and 0319 read εν Ρωμη and VL75 and VL76 have ‘in Rome’ (in Roma). We can therefore be confident that both 06 and 0319 (and their Latin renditions) are firm in the tradition that this epistle was intended for Rome itself and was not a general epistle. 0319/VL76 may even be intensifying the importance of the location in Rome by omitting αγαπητοις θεου in Greek and in caritate Dei (VL75)/dilectis Dei (Vulgate) in Latin. If such omission was intentional then to 0319/VL76 this letter is not addressed to those who are beloved of God in Rome but to all of Rome itself. That the omission is committed in both Greek and Latin perhaps shows intention by the scribe. The omission of ‘beloved of God’ is a singular reading by 0319/VL76 in both Greek and Latin (but is not counted in the statistics in Greek because the Vorlage is not extant. It would be counted in the statistics in Latin since the Latin Vorlage is here extant except that in Latin it is not a singular reading because there is a Greek precedent in 0319).

. Scribal Habits of Codex Sangermanensis (Greek) Scribe A This section will concern the scribal habits of the Greek side of Codex Sangermanensis Scribe A. A section below discusses the scribal habits of 0319B and of the scribes of the Latin side of the codex.

 51 See Parker (2008) 270.

212  Alan Taylor Farnes .. Insignificant Variant Readings Insignificant variant readings include variant readings that have been corrected to match the base text, errors due to corrections in the Vorlage, errors due to graphical confusion, orthographic variants and nonsense variants.52 ... Corrections There are two corrections in the test passages of 0319A. One of these corrections was by the original scribe and a later hand made the other. One correction in 0319 is found at Rom 1.8 τω κοσμω. There appears to be a very faint ν after τω with the result that the scribe originally wrote των κοσμω. There is, in any case, if not ν, some letter erased here since there is a large space after τω. Since the correction is made by erasure rather than with an expunging dot or a slash it is very difficult to tell which scribe made the correction. Turner outlines four main methods of correction in Greek manuscripts: 1) ‘a sponge may be used to delete a whole word or line (or more)’; 2) ‘deletions may be indicated by enclosing a passage in round brackets’; 3) ‘by cancelling a letter or letters by means of a stroke drawn horizontally or obliquely through them’ as is frequently seen in Claromontanus; and 4) ‘by placing a dot (“expunging dot” […]) or a line above, or above and below, or to either side; or a combination of these methods’.53 0319 is clearly not in the practice of marking deletions with slashes through the letter as 06 does. If 0319 had used this method to delete letters then 0319 would not have misunderstood the many places where 06 slashes a letter to mark that it should be deleted. There remain three other popular methods for deleting unwanted text: 1) place an expunging dot over or under the letter, 2) sponge out the letter itself, or 3) enclose the letter in round brackets. We will see later that 0319 employs expunging dots to delete unwanted text but never round brackets. The offending letter was sponged out but it appears that some time has passed which has allowed the ink to set in. 0319 generally uses the expunging dot method. Therefore, I believe that this correction was made by a later scribe and not by 0319. The Latin side of the page reads in uniuerso mundo in the ablative case. The original variant of των κοσμω will feature below under nonsense variants. Another correction is found in Rom 1.29. This correction is made by the original hand to match the Vorlage and will therefore not be included in the statistics. 0319 originally wrote πεπλερωμενους which 0319 then changed to

 52 See Royse (2008) 74–91 and Shipley (1904) 63ff. for discussions of these categories. 53 Turner (19872) 16. See also Haines–Eitzen (2000) 109.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  213

πεπληρωμενους by placing an expunging dot under the offending letter and then writing η above ε. It is important to point out that 0319 is clearly in the practice of using expunging dots to delete unwanted letters. ... Errors due to Corrections in 06 It has been difficult to ascertain whether 0319 ‘correctly’ copied their Vorlage at the countless corrections in 06. My method has, therefore, been to consider ‘correct’ a transcription that matches any of the previous readings. That is to say, whether 0319 followed the original reading written by 06 or the corrected reading by 06*** is irrelevant. If 0319 correctly followed any previous reading, then the transcription is correct. On the whole, both scribes of 0319 followed whatever reading stood in the main text itself. If a correction was made within the main text then they were likely to pick it up. If, however, the correction was made in the margins, then they were likely to ignore it. This is mostly because both scribes of 0319 were not fluent in the meaning of symbols which refer to something written in the margins. Speaking of P66 Parker comments: ‘How easy or difficult the readers of this manuscript found [the system of symbols] to use we do not know’.54 0319’s confusion over the corrections may have been because 0319 may have been used to a different system of signs than the system used by 06. Turner says, ‘As with the signs used for punctuation, it seems that critical signs did not always have the same meaning or the meaning assigned to them by our authorities. And other signs are employed, the meaning of which is not always easily explicable: for instance, ⸓ or ÷, which sometimes is used as a caret to indicate an omission to be made good […], sometimes acts as a reference mark for a marginal note […], and sometimes may have a quite separate signification’.55 It seems that both scribes of 0319 found the system of symbols in 06 very difficult to use. For example, at Rom 1.13 (0319A) where NA28 reads ἵνα τινὰ καρπὸν 06 writes ινα τι καρπον. 06*** corrects this reading to match the NA28 reading but does so by writing a tiny symbol just after τι in the text and then writing να in the left margin on the very edge of the page. I believe the tiny symbol to be a sigma, an abbreviation for συν56 with the intent that τι be written ‘with’ να. 0319A ignores whatever correction is occurring and simply writes what is clear in the main body text. This is how 0319A treats corrections in the majority

 54 Parker (2008) 191. 55 Turner (19872) 14. 56 See Oikonomides (1974) 129.

214  Alan Taylor Farnes of instances: they follow the clearest reading in the main text. Because they correctly transcribed one of the existing readings in 06, this reading is not counted as an error and does not factor into the error rate. In total, 0319A contains two errors caused by corrections in 06 in the test passages accounting for 14.3% of the variant readings. All of these errors result in a singular reading. At Rom 1.19a 0319A makes an error that results in a singular reading. This error can only be explained as a confusion concerning the correction. The text as written by 06* matches that of NA28: ὁ θεὸς γὰρ. One can see that the corrector 06*** placed a dotted obelos (⸓) to the top left of the nomen sacrum θς̅. 06*** then wrote another symbol above the gamma of γὰρ. This was to indicate to move γὰρ before θεὸς so that the reader would read ὁ γὰρ θεὸς. The correction was probably made in an attempt to place the post–positive γὰρ in the proper location. 0319A seemingly confused the correction to mean to transpose ο and θς. The resulting text in 0319 reads θς̅ ο γαρ. This variant is a singular reading which was scribally created. When viewed without the benefit of its Vorlage this variant is ruled as a singular transposition. In Rom 1.30 06* wrote θεοστυγις which is corrected by 06** to θεοστυγεις matching the TR and NA28. 06** corrected the word by inserting a small epsilon under the roof of the gamma. 0319A wrote θεοστυπις. Mistaking the insertion of ε by 06** for the right hasta of π, they substitute a π in the place of γε. This error is a classic example of 0319A misunderstanding the corrections in 06. Of such corrections, Treu noted: ‘Korrekturen der Vorlage oft mißverstanden, so daß sinnlose Fehler entstanden’.57 This variant is a singular reading which was scribally created. When viewed without the benefit of its Vorlage this variant is ruled as a singular nonsense variant. Errors due to Correction Rom 1.19a θεος ο γαρ Rom 1.30 θεοστυπις

... Errors due to Graphical Confusion 0319A made one error due to graphical confusion which constitutes 7.14% of the variant readings in our test passages. This variant was caused when 0319A had difficulty reading the text of 06 for some other reason than a difficult correction in 06.  57 Treu (1996) 38.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  215

At Rom 1.27 06 writes κατεργαζομενοι agreeing with NA28 and the TR. 06 wrote this word, however, very near the end of the line with the end of the word bleeding into the margin. The final two letters have become so faint that they are almost invisible. Tischendorf transcribes this word in 06 as κατεργαζομενοι and does not include that there are any corrections to this word. One can very faintly see two characters after –εν. Perhaps this text was in a similar condition when 0319A attempted to copy this word. Perhaps 0319A saw that there were two characters there but they were so illegible that 0319A had to hazard a guess as to what they were. 0319A’s concluding reading turned out to be nonsense and a singular reading. 0319A wrote κατεργοζομενεν which is not a real word. When viewed without the benefit of its Vorlage this variant is ruled as a singular nonsense variant which was scribally created. Errors due to Graphical Confusion Rom 1.27 κατεργαζομενε ... Orthographic Variant Readings 0319A has seven orthographic variant readings accounting for 46.67% of all variant readings in 0319A’s test passages.58 Unsurprisingly, all of these variants are also singular readings. In Rom 1.9 06***, along with the standard text, reads μαρτυς but 0319A writes μαρτις. Although 06*** corrects the error by 06* of μαρτυρ, this correction in no way obscures or problematizes the reading of the upsilon in this word. That is to say, no one could claim that the ypsilon could be graphically mistaken for an iota. The cause of the error here is probably aural confusion from the tongue to the ear to the hand. The scribe reads μαρτυς aloud and pronounces it as μαρτις due to

 58 Here I am also following Royse’s definition of an orthographic variant. See Royse (2008) 79– 81 for what qualifies as an orthographic variant. In short, in order to ‘reduce the material involved in the present study to a more manageable level’ Royse ignores common orthographic variations such as the ‘interchanges of ει/ι, αι/ε, and οι/υ, presence or absence of movable ν, as well breathings, accents, punctuation, iota adscript or subscript, and other clear forms of abbreviations or writing conventions’, Royse (2008) 81.

216  Alan Taylor Farnes the Historical Greek Pronunciation (HGP)59 and therefore writes it as pronounced.60 It is possible that 0319A’s Greek pronunciation is somewhat influenced by the Latin word corresponding to μαρτυς: testis. 0319A may have written μαρτις because it sounds similar to testis. This explanation would suggest that the scribe of 0319A knows enough Greek to at least pronounce Greek words but not enough Greek to recognize a nonsensical word. 0319A often writes ι in place of υ as also seen in Rom 1.13 and Rom 1.20c. These variants are caused in the same way as the variant in Rom 1.9. When viewed without the benefit of the Vorlage these variants are ruled as singular orthographic variants. Interestingly, 0319A makes the opposite interchange by writing υ in place of ι at Rom 1.19b and Rom 1.21. In Rom 1.19b 06 reads αυτοις but 0319A writes αυτους. Similarly, in Rom 1.21 06 reads διαλογισμοις but 0319 writes διαλογισμους. Although both of these words (αυτους and διαλογισμους) are indeed valid words and could be substitutions rather than orthographic variants, 61 I believe they are better understood as orthographic variants. Especially in the case of διαλογισμους which is preceded by τοις it seems that a simple orthographic υ for ι interchange is occurring here rather than the scribe intentionally changing the word from one word to another. So we see that 0319 interchanges υ and ι occasionally. When viewed without the benefit of the Vorlage these variants are ruled as singular substitutions. At Rom 1.20b NA28, TR and 06 read καθοραται but 0319A writes κατθοραται. Here 0319A has doubled the theta by prefixing the corresponding voiceless stop tau. Gignac explains that this orthographic phenomenon occurs occasionally:  59 Caragounis clarifies that the term ‘itacism’ refers only to a certain type of pronunciation and that this term is inappropriately broadened by text critics: ‘So–called itacism explains only the confusion of the i–sound vowels and diphthongs, although strictly, it should be used only of the η being pronounced the same way as the ι. The HGP, however, which is responsible for this state of affairs, exhibits many other confusions, which under no circumstances can be subsumed under the designation itacism: e.g. αι with ε and vice versa, ο with ω and vice versa, the double pronunciation of the diphthongs αυ, ευ, and ηυ (as αν, and af etc.) and the pronunciation of consonants such as β, γ and δ as well as the absence of aspiration. […] It would be desirable to dispense altogether with the inaccurate and misleading term itacism, and to speak, instead, of the HGP’, Caragounis (2004) 500–1. 60 Gignac notes that υ x ι interchanges ‘occur occasionally throughout the Roman and Byzantine periods’, Gignac (1976) I.267. See also Gignac (1976) I.267–71; Caragounis (2004) 537; Moulton (1906) II.79: ‘Changes between υ and ι are practically always due to assimilation or dissimilation’. 61 It is important here to remember Royse’s caution here that ‘The distinction between an orthographic variation and a more substantive one is, of course, sometimes extremely subtle’, Royse (2008) 79.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  217

‘Aspirates […] are occasionally strengthened by the insertion of the corresponding voiceless stop’.62 When viewed without the benefit of its Vorlage this variant is ruled as a singular orthographic variant. At Rom 1.25 0319A reads τιν, a singular reading, in place of την as in 06. I have classified this variant as orthographic but there is, however, I believe, another alternative albeit perhaps not as likely as the first. I’ll begin by discussing this reading as an orthographic variant as I have classified it. 0319A interchanges η for ι here. Even following Caragounis’ strict definition of itacism63 this example can confidently be called itacism which is a part of the larger HGP. Gignac states that the η for ι interchange ‘occurs very frequently in all phonetic conditions throughout the Roman and Byzantine periods’.64 We probably have here a simple orthographic variant as a result of itacism. A more interesting, albeit not quite as likely, possibility is that this variant arose due to graphical confusion. This example, whether decidedly an orthographic variant or a graphical confusion variant, highlights the benefit and importance of visually consulting the Vorlage of a witness. Only by visually consulting the base text manuscript can we fully understand a scribe’s habits. Of course, most of the time the Vorlage is not available but when it is it should be consulted  62 Gignac (1976) I.100. He also states ‘Doubling of an aspirate by prefixing the corresponding voiceless stop, as well as the doubling of the aspirate itself, is attested elsewhere in the Koine. Mayser explained it as a graphic representation of the syllable boundary within the aspirate, but most examples from later papyri show the doubling in a position in which the syllable boundary does not lie within the aspirate. Schwyzer thinks that the prefixed voiceless stop (or aspirate) represents the plosive, while the aspirate represents the fricative, of an affricate [kkh, tth, pph]; but the papyri generally show this doubling only in a position in which the aspiration could be lost, e.g., before another aspirate, or before a nasal. This partial doubling of an aspirate appears parallel to the ordinary doubling of the simple consonants which is so common’, Gignac (1976) I.101. ‘Single consonants are very frequently doubled in writing and contiguous identical consonants very frequently represented by a single letter through the Roman and Byzantine periods, as elsewhere in the Koine’, Gignac (1976) I.154. See also Moulton (1906) II.100–1, 110. 63 See Caragounis (2004) 500–1. 64 Gignac (1976) 235. Additionally, Gignac adds: ‘The process of itacism, which resulted in the eventual identification of the sounds originally represented by ι, ει, η, ηι, οι, υ, and υι in /i/, was well advanced in Egypt by the beginning of the Roman period. ει and ι are alternate representations of /i/; η and ηι are identified; οι, υ, and υι all represent /y/. Moreover, there is a very frequent interchange of η with ε and ει, indicating that η also represented /i/ at least in the speech of many writers. On the other hand, there is a frequent interchange of η with ε (and sometimes with its phonetic equivalent αι) throughout the Roman and Byzantine periods, in similar documents and sometimes in identical phonetic conditions and even in the same words in which an interchange of η with ι or ει is found’, Gignac (1976) 235. See also Caragounis (2004) 533–7 and Moulton (1906) II.72.

218  Alan Taylor Farnes visually. That is to say, when the Vorlage is available, working only with transcriptions is not sufficient and care must be given to consult the base text visually. While this example of possible graphical confusion may not be convincing, a more compelling argument for this same type of graphical confusion can be seen in the variant reading in Rom 1.24 in καν which I have classified as a substitution. The base text here in 06 reads την but 0319A wrote τιν. It is easy to see, however, how a scribe could miscount the correct number of hasta and drop out a hasta or insert an additional hasta. My thought here is that 0319A glanced at the word and mistook the second hasta of the eta as the first hasta of the ny. This would result in the word τιν. While this example may be weak, I think it is an example of a larger phenomenon that can be seen elsewhere. When viewed without the benefit of its Vorlage this variant is ruled as a singular orthographic variant. Orthographic Variant Readings Rom 1.9 μαρτις υ→ι Rom 1.13 εκωλιθην υ→ι Rom 1.19 αυτους ι→υ Rom 1.20b κατθοραται θ→τθ Rom 1.20c διναμις υ→ι Rom 1.21 διαλογισμους ι→υ Rom 1.25 τιν η→ι ... Nonsense Variant Readings 0319A contains two nonsense readings in our test passages accounting for 13.33% of the total variant readings. Both of these nonsense readings are also singular readings. At Rom 1.8 0319A originally writes των κοσμω which is changed by a later scribe to τω κοσμω. When viewed without the benefit of its Vorlage this variant is ruled as a singular nonsense variant. In Rom 1.28 06 reads παρεδωκεν but 0319A wrote the nonsensical τιαρεδωκεν. 0319 is here graphically mistaking the π for τι (Π > ΤΙ). Such an example also illustrates the subtle line between a graphical confusion and a nonsense reading. In this case however, since both graphical confusion and nonsense reading fall within the category of insignificant variants, it does not much matter tremendously in which category they are placed. Had the writing been more difficult to read or illegible then I would have placed it in the graphical confusion category but the reading here is pretty clear. The problem seems to be then with 0319A’s grasp of the Greek language. It is nonsensical variants such as this and other variants caused by graphical confusion and errors due to corrections

The Scribal Habits of Codex Sangermanensis in Greek and Latin  219

in the Vorlage that leads one to think that 0319A did not know Greek. As stated above, I believe 0319A knew enough Greek to pronounce the words but I do not think 0319A knew the Greek language. When viewed without the benefit of its Vorlage this variant is ruled as a singular nonsense variant. Nonsense Variant Readings Rom 1.8 των κοσμω Rom 1.28 τιαρεδωκεν

.. Accuracy and Copying Technique The following constitute significant singulars and include additions, omissions, transpositions, and substitutions.65 ... Addition There are no additions in our test passages. ... Omission 0319 did not omit any text in our test passages. ... Transposition There is one transposition of three words in the test passages of 0319A at Rom 1.19a. This variant, however, is classed as an error due to correction in the Vorlage and does not therefore count as a transposition. But when considered without the benefit of the Vorlage it is ruled as a singular transposition. ... Substitution 0319A features two substitutions which consist of 13.33% of all variants in our test passages. At Rom 1.20a the standard text reads αορατα but 0319A writes ορατα. 012 also writes this variant. If the scribe of 0319A intentionally wrote this variant without the alpha privative then 0319 is substantially altering — even reversing — the meaning of this verse. Instead of Paul speaking of invisible attributes of the creator he is now speaking of visible attributes of the creator. Such would be a substantive change indeed if it is more than a simple error.  65 See Royse (2008) 91–93 for his discussion of these categories.

220  Alan Taylor Farnes Nonetheless, in light of all of 0319A’s other scribal habits, I am inclined to think that this is an error. I do not think that 0319A knows enough Greek to change the text for dogmatic purposes in this way. This variant is probably simple haplography by dropping out one of the many alpha in this verse. The text here reads τα γαρ αορατα αυτου (ΤΑΓΑΡΑΟΡΑΤΑΑΥΤΟΥ). It is easy to see how one of the many repetitive alpha could be dropped from this verse. I think this is the most likely explanation for this variant although I categorize it as a substitution because the resulting word is a real word that is different from the word in the base text. The Latin facing text further supports this conclusion. VL76 reads inuisibilia. Had 0319A intended a doctrinal change here then perhaps they would have also made the same change on the Latin side of the page. I believe that this variant was scribally created even though 012 also has this reading. At Rom 1.24 0319A writes καν in place of και as in 06. This is classified as a substitution because the resulting text is a real word. While I categorize this variant as a substitution, it may better be viewed as a graphical error. If it is indeed a substitution then 0319A intensifies the subject resulting in ‘Therefore even God […]’. This is one possibility. But if it is a graphical error then the substitution is of no significance. This variant reading is a better example of a graphical confusion as a result of inserting an additional hasta similar to what we saw in Rom 1.25 with τιν. It is possible that 0319A misread the hasta and thought that the first hasta of the pi of παρεδωκεν was actually the final hasta of a ny. There seems to be a slight smudge or trace of ink on the bottom left hasta of the pi that could lead one’s eye to read a ny here. This is not bleed–through since the reverse is blank here. The first hasta of the pi of παρεδωκεν is slightly elevated above the iota of και. The result is that the alleged ny looks very much like 06’s other ny. So I think the most plausible cause of error in this instance is a graphical confusion due to the first hasta of the pi of παρεδωκεν. 0319A then reused the offending hasta when writing παρεδωκεν. On the Latin side of the page VL76 simply wrote et without any strengthening which supports my claim that this change is unintentional. When viewed without the benefit of its Vorlage this variant is ruled as a singular substitution. In the end, 0319A features two substitutions both of which have other possible explanations that would categorize them as insignificant variants. To restate, the only two significant variants in 0319A are somewhat weak and may better be explained as insignificant variants. Substitutions Rom 1.20a αορατα > ορατα Rom 1.24 και > καν

The Scribal Habits of Codex Sangermanensis in Greek and Latin  221

. Scribal Habits of Codex Sangermanensis (Greek) Scribe B .. Insignificant Variant Readings ... Corrections There is only one correction in the test passages of 0319B. At Eph 2.15 0319B originally writes καταργισας (an orthographic singular reading), which a later scribe corrected to καταργησας (the standard reading). This confusion was influenced by 06’s original reading καταρτισας (a singular reading) which is corrected by 06*** to καταργησας. This variant will feature below under errors due to corrections in 06. ... Errors due to Corrections in 06 0319B has three errors due to a correction in the Vorlage accounting for 37.5% of 0319B’s total variants. The correction in 06 at 2 Cor 12.20 caused an error in 0319B which resulted in a singular reading which was scribally created. Here NA28 and TR read φοβοῦμαι but 06* wrote φοβουμε. 06** corrected this reading to match the standard reading but the correction was made in an interesting and confusing way. Rather than cross out the letters to be corrected or writing the correct letters above, 06** simply attached a small loop to the lunate stroke of the original epsilon and inserted an iota after the word. When viewed without the benefit of its Vorlage this variant is ruled as a singular orthographic variant. It is easy to see how this could be confusing to a scribe or to any person trying to make sense out of this word. 0319B was very confused by the correction and chose to write φοβουμει which is a singular reading. Because 0319B created a singular reading which was not intended in 06 this reading counts as a variant reading contributing toward the error rate. Another error caused by a correction in 06 is found at Eph 2.15. 06 originally wrote καταρτισας which was later corrected to καταργησας and matches the standard reading. 0319B copied καταργισας which is a scribally created singular reading. When viewed without the benefit of its Vorlage this variant is ruled as a singular orthographic variant. The correction in 06 at Titus 2.15 caused an error in 0319B. Here NA28 and the TR read ἔλεγχε but 06* wrote ελενχε. 06*’s text exhibits a common scribal orthographic error by writing the word as it is pronounced instead of as it is spelled. 66 06*** corrects this reading to ελεγχε to match the standard reading but does so

 66 See Gignac (1976) I.165–77; Caragounis (2004) 377–80.

222  Alan Taylor Farnes in a confusing manner. 06*** attempted to erase the diagonal stroke and final hasta of 06*’s ν but did so only partially. 06*** then added a horizontal stroke to the first hasta of ν turning the letter into a gamma. One can easily see how anyone reading the correction would be confused as to the intended final reading. 0319B was thoroughly confused and wrote ελειχε — a singular reading which was scribally created. When viewed without the benefit of its Vorlage this variant is ruled as a singular orthographic variant. Errors due to Correction 2 Cor 12.20 φοβουμει Eph 2.15 καταργισας Titus 2.15 ελειχε ... Errors due to Graphical Confusion There are two errors due to graphical confusion in the test passages of 0319B accounting for 25% of 0319B’s total variants. At Titus 2.12 0319B reads αιων in place of αἰῶνι as in NA28 and TR. 06 reads αιωνι but this word is written at the end of the line. This means that the text on the other side of the page starts, and is therefore most dense, directly underneath (on the other side of the page) where this word is written. Because the vellum is so thin and there is so much writing at that place on the other side of the page, it is difficult to discern what text belongs on the facing page and what text belongs on the other side of the page. Such is the case in our example in Titus 2.12. There is an iota after αιων but text from the other side of the page bleeds through. Combine the bleed through with fading ink and it is easy to see how 0319B would have difficulty accurately reading this word and creates a scribally created singular reading. When viewed without the benefit of its Vorlage this variant is ruled as a singular orthographic variant. An even better example of graphical confusion, especially due to bleeding text, is found at Titus 3.7. 0319B has written the nonsensical reading κληρθνομοι (a scribally created singular reading) where NA28 and TR read κληρονόμοι. Such a nonsense reading, however, is much better understood after consulting 0319B’s Vorlage. The text on the other side of the page has bled through to make the text here almost illegible. The Latin text on the other side of the page from Titus 3.3 reads errantes. The medial horizontal stroke of the final e of errantes (shaped very similarly to a majuscule Greek lunate epsilon) crosses perfectly through the first omicron in κληρονομοι on the other side of the page. This bleed through leads 0319B to read theta in place of omicron. When viewed without the benefit of its Vorlage this variant is ruled as a singular nonsense variant. Errors due to Graphical Confusion

The Scribal Habits of Codex Sangermanensis in Greek and Latin  223

Titus 2.12 αιων Titus 3.7 κληρθνομοι ... Orthographic Variant Readings Only one orthographic variant is found in the test passages of 0319B accounting for 12.5% of 0319B’s total variants. It is found in 2 Cor 12.13. 06** and the TR read ηττηθητε while 0319 reads ητηθητε. NA28 and 06* read ησσωθητε. This same variant was made (probably independently) by 1646. This reading can be explained as a simple orthographic variant by simplifying a contiguous identical consonant.67 Orthographic Variant Readings 2 Cor 12.13 ητηθητε ττ → τ ... Nonsense Variant Reading 0319B only made one nonsense variant in our test passages which is found at Titus 3.3. The standard reading here is διαγοντες στυγητοι. Below is the text as found in Claromontanus. 0319B copied τυγητοι. Perhaps 0319B sees a small dot beneath the sigma of στυγητοι and therefore omits it. Or 0319B could have accidentally dropped one of the sigmas that are side by side. Either way, the resulting text is nonsense. To be fair to 0319B this variant is very close to being classified as an error due to correction in the Vorlage but since 0319 does not usually interpret expunging dots correctly and because I think that is a stray dot and not a proper expunging dot, this variant is classified as a nonsense variant. Again, since both nonsense variants and errors due to correction in the Vorlage are classified as insignificant variants, the exact classification does not much matter. When viewed without the benefit of its Vorlage this variant is ruled as a singular nonsense variant. Nonsense Variant Readings Titus 3.3 τυγητοι

.. Accuracy and Copying Technique ... Addition There are no additions in our test passages.

 67 See Gignac (1976) I.154–5, 161; Moulton (1906) II.101–2.

224  Alan Taylor Farnes ... Omission 0319B did not omit any text in our test passages. ... Transposition There are no transpositions in our test passages. ... Substitution 0319 made one substitution in our test passages. This substitution aligns the singular reading of 06 to the more common standard reading. At 2 Cor 12.15 06 writes the nonsensical and singular reading εκδαπανησησομαι.68 0319B corrects this reading to match the standard text of εκδαπανηθησομαι. Substitutions 2 Cor 12.15 εκδαπανησησομαι > εκδαπανηθησομαι

. Conclusions concerning the Greek Text Much of what I have said about 0319A can be applied to 0319B. While 0319B has a much lower error rate than 0319A at about twelve variants per thousand words compared to 0319A’s almost thirty variants per thousand words, 0319B makes the same type of mistakes as does 0319A just not as often. So I think 0319B was a Latin scribe whose first language was not Greek and who had a very basic understanding of the Greek language if any. 0319B likewise had a difficult time interpreting correction marks in the Vorlage. The main difference between 0319A and 0319B was 0319A’s creation of orthographical variants. 0319A created seven orthographic variants compared to 0319B’s one orthographic variant even though 0319A’s test passage was half the amount of 0319B’s test passages. Aside from this striking difference in scribal practice the rest of their scribal habits were somewhat comparable proportionally. Tab. 1: Variants by Test Passage Words Scribe A

Rom



Sub

Nons 

Orth 

ErrCorr 

GraphErr 

Total 



 68 Tischendorf erroneously reads εκδαπανησομαι here, dropping out a set of ησ. See Tischendorf (1832) 249.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  225

Words Scribe B

Sub

Nons

Orth

ErrCorr

GraphErr

Total

 Cor















Eph















Titus















Total

,













One of the most shocking and significant conclusions from the study of the test passages of 0319 is that neither scribe added nor omitted a single word. This is not to say that they were perfect scribes — far from it — but with respect to whether they added or omitted, the scribes of 0319 did neither. In fact, over 1,618 words or eight N–A pages, they made only three significant variants all of which were substitutions and all of which could possibly be explained away. One of the substitutions was correcting 06’s singular nonsense reading and the other two could be graphical errors. In the end, the scribes of 0319 made hardly any significant variants at all. They did make plenty of insignificant variants — orthographic changes, misreading corrections in the text or even misreading the source text entirely — but they made very few significant variants. James R. Royse argued against lectio brevior potior in favor of lectio longior potior arguing that the scribes who he studied tended to omit more words than add words.69 After our analysis of these test passages of 0319, we cannot make a conclusion either for or against lectio brevior. Previous scholars who studied Codex Sangermanensis concluded that 0319 was a pretty poor copy of Claromontanus and useless for textual criticism. It is clear that the scribes of 0319 made obvious blunders that anyone who knows Greek would not have made. They created a good number of nonsense readings. But when viewed in light of 0319’s difficult Vorlage with nine correctors and parchment so thin that text bleeds through from the other side, perhaps we can be a little more kind to the scribes of 0319. They faithfully copied the text for the vast majority of the passages but when they made a mistake it was a glaring mistake — especially in the case of 0319A. But we should not let these glaring mistakes, however obvious, blind us from seeing how dutifully 0319 copied the majority of the text. The conclusion is that the scribes of 0319 surely did not know Greek very well at all. They knew Greek well enough to pronounce it but did not to actually know the language. This ignorance of the Greek language is responsible for most of their blunders. Concerning the scribes of 0319’s knowledge of Greek, Houghton  69 Royse (2008) 719–20, 734.

226  Alan Taylor Farnes comments: ‘Nonsense readings in the Greek confirm that the copyist’s principal language was Latin’.70 As we will see below, their Latin knowledge is far superior to their Greek knowledge. The fact that the scribes of 0319 did not know Greek very well brings up another set of questions concerning the copying skill of non–native Greek speakers versus native Greek speakers. A similar question is how well does a scribe who intimately knows the text copy compared to someone who is unfamiliar with the text? Again, the scribes of 0319 copied the Vorlage fantastically but when they made an error it was an egregious one. The opposite is likely true for Greek speaking scribes: they may be slightly more prone to make changes but the changes they make would be less egregious and without so many nonsense variants. More research concerning non–native scribes is needed in order to understand if this theory holds true. We have seen here the benefit of having the Vorlage of a manuscript at our disposal. As can be seen in the tables in the appendix, a study of 0319’s scribal habits in light of the Vorlage greatly enhances our ability to understand a scribe’s habits. If we were to study this manuscript as if we did not have access to its Vorlage then our results would look slightly different. Instead of three significant variants we would have four and instead of twenty–two total variants, we would only have nineteen. The reason for this is because even when it is clear according to the Vorlage that a reading is scribally created, it is possible that another scribe has independently made the same error and that this reading is therefore not a singular reading — even though we know that it is scribally created. We can conclude that, when available (which is admittedly very rare) a study of the scribal habits in light of a manuscript’s Vorlage is superior to a study of scribal habits by means of singular readings or any other method.

 70 Houghton (2016) 80. Griesbach agrees that the scribe’s primary language was Latin saying, ‘Graecarum litterarum ductus non rudiores quidem sunt, quam latinarum. At satis tamen certum est, scribam fuisse latinum hominem. Graeca enim, quae scripsit, plane non intellexit, atque hinc inaudita verborum portenta innumera confinxit, ac graecas adeo litteras cum latinis, v. c. Graecorum Ρ cum Latinorum P, turpiter confudit, et alicubi, notante Wetstenio, loco ΩΡΑΙΟΙ scripsit ΩΠΑΙΟΙ, in quem errorem Graecus, quantumvis indoctus et negligens, incidere non potuisset’, Griesbach (1793) 77. See also Montfaucon (1708) 218–20 and Wettstein (1751) II.7–8.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  227

 Codex Claromontanus (VL75) and Codex Sangermanensis (VL76) Just as the vast majority of scholars agree that the Greek text of 0319 is a copy of codex Claromontanus, the majority opinion is that the Latin text of VL76 is also a copy of the same codex.71 Because the Latin text of 06 (VL75) does not contain nearly as many corrections (which are easily misinterpreted) and because the scribe of VL76 knew Latin well, there are few obvious indicators that VL76 copied VL75. The text of VL76 does match VL75 very closely. Where VL76 diverges from VL75 it is to vulgatize VL75’s Old Latin text. It follows that if 0319 copied 06 then VL76 most likely copied VL75. The scribes of the Greek side of Sangermanensis, I believe, are the same as the scribes of the Latin side of the codex, respectively. As mentioned above, the Greek hand is a ninth–century biblical uncial and the Latin hand is a ninth–century uncial. The contemporary hands share many of the same features. An analysis of ductus and paleography of similar letters in both Greek and Latin shows that the Greek letters are formed in a similar manner to the Latin letters. Since I believe that the scribe of the Greek text and the Latin text is the same person, to get the complete picture of the scribal habits of 0319 it is necessary to also analyze the scribal habits of the Latin text. The Latin text can yield a more complete picture of this scribe. Of course this means that since there were two scribes on the Greek side of the page then there are also two scribes writing the Latin. They will be referred to as VL76A and VL76B.

. Scribal Habits of VL76A The test passages for the Latin of Sangermanensis are the whole of Rom 1 (VL76A) and 2 Cor 11.33–2 Cor 12.14; Eph 1.5–13; 2.3–11; and Titus 1.1–3.3 (VL76B). In total the test passages contain 1,547 words.72

 71 See Houghton (2016) 244; Frede (1964) 36–39. 72 I have not used Royse’s method of converting number of words into N–A pages for the Latin text. Instead I provide statistics for variants per 1,000 words.

228  Alan Taylor Farnes .. Insignificant Variant Readings ... Corrections There are no corrections in the test passages of VL76A. ... Errors due to Corrections in VL75 There are no errors in VL76A or VL76B that can be conclusively attributed to corrections in VL75. It is interesting, however, that both of the omissions in VL76A and the only addition in VL76A occur where there is a correction in VL75. ... Errors due to Graphical Confusion There are no errors due to graphical confusion in the test passages of VL76A. ... Orthographic Variant Readings There are three orthographic variants accounting for 50% of the total variants in VL76A. Two of these represent a later stage in Latin orthography and match the Vulgate while the remaining orthographic variant (Rom 1.24) shares the reading with at least eleven other Latin manuscripts. Orthographic Variants Rom 1.23 serpentum > serpentium Rom 1.24 adficiant > afficiant Rom 1.30 adfectione > affectione ... Nonsense Variant Readings There are no nonsense readings in our test passages.

.. Accuracy and Copying Technique ... Addition VL76A created no additions. ... Omissions VL76A includes two omissions of four words total accounting for 33.3% of the total variants. The first omission is at Rom 1.7 where VL76A omits in caritate Dei. VL76 is the only Latin manuscript with this reading but 0319 here omits ἀγαπητοῖς θεοῦ so VL76 does not here have a singular reading.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  229

Omissions Rom 1.7 in caritate Dei Rom 1.32 et ... Transpositions There are no transpositions in our test passages. ... Substitutions VL76A does not make any substitutions when copying VL75.

. Scribal Habits of VL76B .. Insignificant Variant Readings ... Corrections There is one correction in the test passages of VL76B found in Titus 1.9 made by a later hand. Since this correction is made by a later hand it has no relevance to our study of this scribe. ... Errors due to Corrections in VL75 There are no errors due to corrections in the Vorlage. ... Errors due to Graphical Confusion There are no errors due to graphical confusion in the test passages of VL76B. ... Orthographic Variant Readings There are two orthographic variants accounting for 16.67% of the total variants. Both of these represent a later form of Latin orthography and match the Vulgate. Orthographic Variants Titus 1.7a uinulentum > uinolentum Titus 1.7b percossorem > percussorem ... Nonsense Variant Readings There are no nonsense readings in our test passages.

230  Alan Taylor Farnes .. Accuracy and Copying Technique ... Omission VL76B omitted one word in the test passages at Titus 3.3 accounting for 10% of the significant variants in VL76B. Omissions Titus 3.3 a ... Transposition There are no transpositions in our test passages. ... Substitution VL76B creates nine substitutions for a total of nine words which account for 90% of significant variants and 75% of the total variants. Seven of these substitutions diverge from VL75 and match the Vulgate (2 Cor 11.33; 2 Cor 12.11; Titus 1.5, 12a, 12b, 16; 3.1). Four of the substitutions change a reading which is a singular reading in VL75 (2 Cor 11.33; Eph 1.9; Titus 1.5, 12a). One substitution is a singular reading (Titus 1.13). Substitutions 2 Cor 11.33 ut > et 2 Cor 12.11 commendare > commendari Eph 1.9 ostenderet > ostenderent Titus 1.5 grati > gratia Titus 1.12a propetat > propheta Titus 1.12b cretensis > cretenses Titus 1.13 quem > qua Titus 1.16 ab > ad Titus 3.1 ab > ad

. Conclusions concerning the Latin Text VL76B’s overwhelming characteristic is a strong habit of vulgatizing the text. The large majority of VL76B’s variants represent a later stage in the development of the Latin language and happen to match the Vulgate. VL76A made no such substitutions. VL76A created proportionally more orthographic variants that did VL76B and they both made omissions and VL76A made an addition. While their overall error rate was almost identical it is important to remember that VL76A made most of the orthographic changes and VL76B made all of the substitutions which happen to match the Vulgate. It is important, however, to remember

The Scribal Habits of Codex Sangermanensis in Greek and Latin  231

Royse’s comment that the difference between substitutions and orthographic variants is often very subjective and extremely subtle.73 In VL76 we see two scribes who are familiar with the language that is being copied, which is unlike the situation in 0319A and 0319B. We therefore are provided an insight into the scribal habits of scribes when they are familiar with one language and unfamiliar with another. VL76A and VL76B omit more than they add when collating the text on the Vorlage. This confirms Royse’s theory about lectio brevior with respect to these scribes. Again, since the scribes of VL76 know Latin, their scribal habits with respect to lectio brevior are also what we would expect to see. If we were, however, to analyze VL76 by singular readings then the scribal habits would look much different. According to the singular readings method VL76A adds one word thereby adding more than omitting and contradicting Royse’s theory concerning lectio brevior (see appendix). Here we see again that, when possible, determining scribal habits based on a known Vorlage is to be preferred and gives the clearest picture of scribally created readings. Tab. 2: Variants by Test Passage Words

Add

Om

Sub

Nons

Orth

ErrCorr GraphErr Total

Scribe Rom A



















Scribe  Cor B Eph





































Titus



















Total

,

















 Conclusions concerning the whole of Codex Sangermanensis Some final thoughts concerning the scribes of Codex Sangermanensis are in order. The scribes of 0319 have an introductory understanding of Greek. They can prob-

 3 See Royse (2008) 79.

232  Alan Taylor Farnes ably sound out words. 0319A and 0319B can even perhaps translate basic vocabulary and sentence structure. How much Greek knowledge does one need to understand that τοῖς ἐν Ῥώμῃ can be translated as a relative clause and that qui therefore is a good translation of the phrase (Rom 1.15)? Similarly, if one knows Latin well, is it difficult to identify that in caritate Dei aligns with ἀγαπητοῖς θεοῦ (Rom 1.7)? The scribes of 0319 can pronounce Greek and have a basic understanding of vocabulary and syntax but they lack a deep vocabulary in Greek. 0319B somehow knows enough Greek to know that εκδαπανησησομαι (2 Cor 12.15) is an incorrect reading and properly corrects it. As for 0319’s nonsense readings, to a novice at Greek τιαρεδωκεν (Rom 1.28, Scribe A) and τυγητοι (Titus 3.3, Scribe B) may seem like just as good Greek words as παρεδωκεν and στυγητοι. Additionally, ὡραῖοι is a rare word being only used twice in the New Testament (Matt 23.27; Rom 10.15; other reflexes of the same word can be found in Acts 3.2 and Acts 3.10) so it is understandable, when we remember that copying was actually dictating one’s own aloud reading since scribes would read aloud as they copied and that this scribe reads Latin better than Greek, that an error due to faulty hearing due to a different pronunciation could result in ωπαιοι (Rom 10.15) because the scribe pronounced ὡραῖοι mistaking the rho for a p sound.74 We can be confident that the scribes of 0319 are not fluent in Greek but most of what they lack is simply a deficiency of vocabulary. This codex gives great insight into the scribal habits of scribes who are familiar with the language which they are copying and the habits of scribes who are unfamiliar with the language they are copying. The scribes of 0319, who were unfamiliar with the language being copied, copied the text surprisingly well. They did not add or omit a single word. Such exactness is remarkable. They did, however, create numerous egregious nonsensical readings that would be an obvious error to anyone who knew Greek. So we see that scribes that are unfamiliar with a language will copy the text very well when the text is clear and easy to understand. When judgments must be made, however, such as a difficult correction or an obscured text, a scribe who does not know the language will make egregious errors. On the other hand, scribes who know the language make a different set of errors. These errors will be less egregious and less obvious. Scribes who know the language are more prone to omit rather than to add and to take liberties with the text by, for example, aligning the text more closely with a version with which they are familiar, as did VL76B. Again, even though the scribes of VL76 copied in Latin, their scribal habits are exactly what we would expect from scribes who  4 See Metzger–Ehrman (Oxford 20054) 254–7 for this cause of error in transmission of the text.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  233

know the language whereas the scribes of 0319 provide fresh new insight into the scribal habits of scribes who do not know the language. Due to the status which Codex Sangermanensis enjoys as the earliest duplicate manuscript and one of the relatively few bilingual manuscripts, I hope that it can receive a complete treatment in the near future that performs similar experiments to this study but for the whole of the manuscript instead of by test passages. In the end, our analysis was greatly enhanced by access to the Vorlage Codex Claromontanus. Had we not had access to the Vorlage but rather analyzed this codex based on singular readings alone, then our picture of these codices would look slightly different. While the singular readings method approaches a correct picture of these scribes it is not the most accurate method.

234  Alan Taylor Farnes

 Appendix – Tables

Words in Omissions

Average Words

























A (Singulars)75



























B (Variants)



























B (Singulars)



























Transpositions

Omissions

Net Words Lost Per Significant Variant

Average Words



Net Words Lost

Words in Additions

Net Words Gained / Lost in Substitutions

Additions

A (Variants)

Substitutions

Significant Variants

Words in Omissions / Words in Additions

Tab. 3: 0319 and 0320’s Variant Readings

 75 This row is intended to contrast the actual scribal habits against what the scribal habits would have been had we not had access to the Vorlage. Put another way, this row shows the results as following Royse’s method. This enables us to analyze the difference between analyzing these manuscripts as copies and analyzing these manuscripts on their own as Royse does. Of course, to be quite fair, Royse did not have access to the Vorlagen of the manuscripts which he studied. Likewise, this row shows what the scribal habits would have looked like if we did not have access to the Vorlage.

The Scribal Habits of Codex Sangermanensis in Greek and Latin  235

Other errors due to Graphical Confusion

Orthographic Variants

Percentage

Nonsense Variants

Percentage

Nons. Variants/N– A Page

Nons. Variants/, words









%

. .



.%

.

.

A (Singulars)



X76

X

 .%

. .



.% .

.

B (Variants)









.%

.

.



.%

.

.

B (Singulars)



X

X

 .%

.

.

 .%

.

.

Orth. Variants/, Words

Errors due to Corrections in the Vorlage

A (Variants)

Orth. Variants/N–A Page

Total Variants

Tab. 4: Total Variants

 76 Data is not given in this field because this field analyzes the manuscript as if there were no known Vorlage and therefore we would not have access to the corrections in the Vorlage.

236  Alan Taylor Farnes

Variants

Significant Variants/N–A Page

Total Variants/N–A Page







.

.

.

.

A (Singulars)







.

.

.

.

B (Variants)



,



.

.

.

.

B (Singulars)



,





.



.

Total Variants/, Words

Words

A (Variants)

Significant Variants/, Words

N–A Lines

Tab. 5: Error Rates

Additions Words in Additions Average Words Omissions Words in Omissions Average Words

VLA (Variants)           – – 

VLA (Singulars)        X     

VLB (Variants)        X   – –. 

VLB (Singulars)        X     

Transpositions

Net Words Gained / Lost Per Significant Variant

Net Words Gained / Lost

Net Words Gained / Lost in Substitutions

Substitutions

Words in Omissions / Words in Additions

Significant Variants

The Scribal Habits of Codex Sangermanensis in Greek and Latin  237

Tab. 6: Variant Readings of VL76

238  Alan Taylor Farnes

Total Variants

Errors due to Corrections in the Vorlage

Other errors due to Graphical Confusion

Orthographic Variants

Percentage

Orth. Variants / , Words

Nonsense Variants

Percentage

Nons. Variants / , Words

Tab. 7: Total Variants

VLA (Variants)









%

.







VLA (Singulars)



















VLB (Variants)







 .%

.







VLB (Singulars)



















Words

Adjustment Factor

Variants

Variants / , words

Tab. 8: Error Rates

VLA (Variants)



.



.

VLA (Singulars)



.





VLB (Variants)

,

.



.

VLB (Singulars)

,

.



.

John Bradley

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  Synopsis For many years, the hypogeum of the Aurelii has been considered as a tomb belonging to a group of early Christians or Gnostics. This paper sets out to demonstrate that this conclusion is erroneous relying, as it has, on making comparisons with much later, Christian, iconography rather than examining the evidence in the light of funerary practice contemporary or antecedent to the construction of the tomb. The result is that, far from the tomb being evidence of aberrant Christianity, the commissioners of the tomb were members of a professional collegium with mutual ties arising from that membership rather than any religious affiliation. The outcome throws considerable light on the composition of such organisations.

 Introduction The hypogeum of the Aurelii was discovered on the corner of Viale Manzoni and Via Luzzati on the Esquiline hill, Rome in 1919 during the construction of an underground garage and was excavated by Goffredo Bendinelli between 1920 and 1921.1 Bendinelli discovered a number of brick stamps in the vicinity dated to the early 3th century.2 This chronology accorded with the stylistic evidence of the frescoes which suggested a date around AD 240.3 The tomb’s position, within the Aurelianic walls, meant that it had a comparatively short active life and consequently, when discovered, most of the decoration was intact. The tomb consisted  1 For a summary of the discovery and subsequent acquisition by the state and transfer to the Pontificia Commissione di Archeologia Sacra (PCAS) see Giuliani (2011) 31–45. 2 Bendinelli (1922) 510. 3 Wirth (1934) 86 suggested a date of AD 240 while Himmelmann (1975) 9 thought a slightly earlier date of AD 220 as did Jastrębowska (1981) 37–38. The latter considered the Villa Piccola beneath S. Sebastiano to be c. AD 240 and given the similarities between that and the hypogeum of the Aurelii the author believes that the hypogeum of the Aurelii is likely to date to approximately the same time. https://doi.org/10.1515/9783110632590-015

240  John Bradley of three cubicula (fig. 1), of the two subterranean ones cubiculum B, together with its small vestibule, was substantially complete; whereas cubiculum A was damaged during both discovery and when an extension portal was constructed in the rear wall sometime c. AD 240–270. The upper chamber only survived to a height of about 1.2 metres, but it too had been richly decorated when first built.

Fig. 1: Cross–section through hypogeum of the Aurelii. Bendinelli (1922) tav. 1.

In the upper chamber, Bendinelli found the image of two nude figures alongside a tree with a snake.4 This prompted him to suggest that this was a representation of Adam & Eve and the tomb was therefore Christian in origin, despite its early date, though he acknowledged that questions remained over the iconography. 5 Josef Wilpert, a catholic priest and archaeologist, who in 1903 had produced a magnum opus recording many of Rome’s catacomb paintings, explained the unusual iconography on the grounds that the frescoes were the result of Gnostic beliefs.6 In addition to the ‘Adam & Eve’ there were other images that were supposed to prove a Christian inspiration. The figure of a bearded male sitting on a hillside, dressed in tunic and pallium, seemingly reading to a group of sheep and goats

 4 Wilpert (1924) tav. I. Bisconti (2011) tav. 11. 5 Bendinelli (1922) 436. 6 Wilpert (1924) 8–17.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  241

was thought to represent Christ.7 In particular, a green Latin cross on the right hand wall of the vestibule8 and a fine set of eleven full–length, approximately two–thirds size, portraits on three sides of cubiculum A interpreted as a procession of the twelve apostles.9 Bendinelli initially thought that this so–called ‘Large Procession’ had comprised twelve figures with one having been destroyed to make a portal for the tomb’s extension.10 By the time excavation was complete, however, Bendinelli realised that there had only ever been eleven figures; the portal had not destroyed a supposed twelfth figure, but had destroyed an original arcosolium.11 Consequently, it was most unlikely that they represented the apostles of Christ. Most commentators, however, followed Wilpert’s opinion that the decoration was inspired by some form of nascent Christianity and, although there were differences of opinion over what kind of non–orthodox Christianity was practised, there was little question that Christianity was the basis for the frescoes. The Christian origin was challenged by the Nikolaus Himmelmann in 1975. He argued that the so–called ‘Adam & Eve’ was unlike contemporary representations such as that found at Dura Europas or the later images found in large numbers in catacomb wall paintings and on sarcophagi.12 The male figure did not look especially ashamed and the snake and tree were to one side, rather than between, the two figures. The ‘Shepherd’ figure does look like Christ in Mediaeval representations, but in early Christianity, he was invariably depicted as youthful and beardless.13 Furthermore, the Latin cross upon which Wilpert had placed much emphasis, transpired to be no more than part of a garland and was an accident of preservation rather than a ‘crypto–Christian’ emblem of a persecuted sect.14 In short, a Christian interpretation was based upon parallels that were either uncanonical, anachronistic or mistaken. Paul Corby Finney, in 1980, also called into doubt the prevailing theories by challenging the Gnostic interpretation.15 Yet, despite such scepticism, the idea of the Aurelii being part of the early Christian world remained dominant and is repeated to this day.16

 7 Wilpert (1924) tav. XII. Bisconti (2011) tav. 19. 8 Wilpert (1924) 2, tav. Xa. Bisconti (2011) tav. 51. 9 A summary of most interpretations may be found in Pergola (2011). 10 Bendinelli (1920) 131. 11 Bendinelli (1922) 317. 12 Himmelmann (1975) 10–13. 13 E.g. the statue of Christ the Good Shepherd now in the Vatican, Vat. Mus. Pio Cristiano, 28590. 14 Himmelmann (1975) 24. 15 Finney (1980) 442–7. 16 E.g. Coarelli (2007) 201.

242  John Bradley One of the few contemporary commentators to have rejected a religious foundation is Gian Luca Grassigli of the University of Perugia. In a series of publications concerning the rise of the concept of otium he has placed the Aurelii in the realm of the aristocratic retreat into the world of philosophical study.17 This interpretation too, however, is not without difficulty as, although practically all information on the Aurelii is lacking, we do have one concrete piece of information, a mosaic inscription in the floor of cubiculum A, which unequivocally states that they were freedmen and women and not members of the élite. Furthermore, the majority of the imagery is urban in context and gives little or no indication of some self–imposed exile to the country in order to study philosophy.

 The Inscription The hypogeum is so–called after the floor mosaic (fig. 2) which reads thus18: AVRELIO ONESIMO AVRELIO PAPIRIO AVRELIAE PRIME VIRG AVRELIVS FELICISSIMVS FRATRIS ET COLIBERT B M F

Since Bendinelli’s original excavation report the tomb’s inscription has been expanded to read: Aurelio Onesimo Aurelio Papirio Aureliae Prime virg(ini) Aurelius Felicissimus fratri(bu)s et co(n)libert(is) b(ene) m(erentibus) f(ecit)19

Which may be translated as: ‘Aurelius Felicissimus made (this tomb) for (his) well deserving brothers and fellow–freedmen: Aurelius Onesimus, Aurelius Papirius and Aurelia Prima, virgin’.

 17 Grassigli (2000), (2002) and (2011). 18 ICUR VI.15931 = EDCS 16201053 = EDB 9222. 19 Bendinelli (1922) 320 and 425–6. For the most recent apparatus see Giovagnoli (2011) 229.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  243

Fig. 2: Mosaic inscription, cubiculum A. Bisconti (2011) tav. 41.

244  John Bradley We must not take, however, this information at face value. As the word fratris cannot be realistically read as a genitive singular it has been rendered as fratribus i.e. the dative plural to accord with the dative case of the dedicatees. This interpretation, as a dative plural, is derived from the known use of mensis for mensibus of which there are many examples.20 However, the author has found no unambiguous instance of fratribus being abbreviated to fratris in the same manner as mensibus. Modern inscription databases allow exhaustive searches of known inscriptions in minutes that would previously have taken days. There are many abbreviated forms of fratribus,21 but in more than 80,000 inscriptions of the Epigraphica Database Roma, this is the only instance of fratribus being abbreviated to fratris. Studies in Latin morphology on the other hand demonstrate changes in Latin vowel sounds reflected in the spelling.22 From at least the 3th century AD grammarians such as Terentius Maurus records the long ‘e’ seen in fratres becoming confused with the short ‘I’.23 Therefore, fratris may plausibly be read as a nominative plural, something that would not clash with the other abbreviations in the last line and therefore the inscription ought to be revised to read thus: Aurelio Onesimo Aurelio Papirio Aureliae Prime virg(ini) Aurelius Felicissimus Fratrs et co(n)libert(i) b(ene) m(erentibus) f(ecerunt)

Which may be translated as: ‘Aurelius Felicissimus, brothers and fellow–freedmen made (this tomb) for the well deserving Aurelius Onesimus, Aurelius Papirius and Aurelia Prima, virgin’.

The implication of this change, slight though it is, is that the deceased are commemorated not by one individual, but by many. A group that defined itself as brothers, including a female. That is not the only information that may be teased  20 E.g. EDR 030285 = EDCS 52602891 on an altar found near Rome dated to the 3rd century. The use of mensis being in apposition to annis and diebus for the age of the deceased makes the abbreviation unambiguous. 21 E.g. the long inscription CIL XI.1147 = EDR 130843 now in MAN Parma (no inv. nr.) uses fratribus in full together with the abbreviations fratrib(us), frat(ribus) and fr(atribus), but not fratri(bu)s. 22 Adams (2013) 38–43. Clackson–Horrocks (2007) 273. 23 The date for Terentianus Maurus is uncertain, but is in the range of AD 150–350, Beck (1993) 10.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  245

from this seemingly mundane inscription. The female is defined as virgo and consequently has been thought to be a young girl. Indeed one of the three graves within cubiculum A, that on the left hand side, is that of a child no more than nine years old (fig. 3).24

Fig. 3: Basic layout of pictorial schemes in cubiculum A using current nomenclature.

The decoration above it, however, has a distinctly masculine bias i.e. (1) a bearded ‘Shepherd’ referred to above; (2) a young male on horseback attended by a group of men (referred to as the adventus scene)25 and (3) another group of men seen coming out of a city to meet the rider (the cityscape)26. There are no feminine references on the left wall whereas that opposite, containing an adult grave, has decoration that features a woman in three different places i.e. (1) attending, but not partaking of, a banquet;27 (2) in attendance on two seemingly deceased persons28 and (3) standing beside a loom (these last two in the upper and middle registers of the so–called ‘Homeric’ scene). While virgo is usually associated with female children that is not exclusively the case. There are a number of examples where women in their late teens may be referred to as virgines.29 If the decoration

 24 Bendinelli (1922) 316. Poe (2007) 26, fn 84; 43, fn 128. Poe highlights this point, but does not take the implications further. Borg (2013) 112 and 252 also associates this child–sized arcosolium with Aurelia Prima. 25 Wilpert (1924) tav. XX. Bisconti (2011) tav. 23. 26 Wilpert (1924) tav. XX. Bisconti (2011) tav. 24a–b. 27 Wilpert (1924) tav. XVII. Bisconti (2011) tav. 35. 28 Bisconti (2010) fig. 31. 29 CIL VI.9342; CIL VI.24032 = EDR 102064; CIL VI.33473 = CLE 1882 = Dessau, ILS 7771 = EDR 152970 = EDCS 23900587. The phrase, adulta virgo, is used by Tac. Hist. 3.33 and Gell. NA 7.8.3

246  John Bradley is bespoke for the deceased, a point that will be discussed shortly, then it seems highly unlikely that a young girl would be commemorated by male oriented decoration. Nor would an adult male be commemorated by images of women in prominent positions above their grave. Consequently, a consideration of the sizes of the arcosolia in relation to the adjacent wall decoration strongly suggests that the woman, Aurelia Prima, was adult, but not married and the child was male. Thus, a reappraisal of the inscription gives us the first clue that the Aurelii were all members of some form of group that considered themselves fratres and those fratres included women and children.

 Association between Decoration and Deceased. I posited above that the decoration in cubiculum A, above the arcosolia, was directly related to the deceased within those arcosolia. Such a connection is not always obvious. In many cases decoration lacks inscription or is of a generic character which may make such a link circumstantial. However, there are instances where decorative schema similar to that of the hypogeum of the Aurelii shows a direct connection with the commemorated deceased. The most visible is the 1st century BC Tomb of Eurysaces in Rome where the frieze around the monument directly relates to the deceased’s role as a baker.30 A more modest memorial of about the same time commemorated a Greek physician in Rome called Patron who is depicted with his family in an upper register of decoration in a manner similar to that in cubiculum A.31 The deceased Patron (with servant) travels in one direction while the rest of the family move in the opposite direction. In both instances surviving inscriptions allow us to clearly associate decoration with deceased; a habit that continues into the 4th century with the tomb of Trebius Iustus in Rome32 and the less well–known tomb of Flavius Maximianus in Sicily where a soldier is painted in the spandrel of the arcosolium while the inscription advises that the deceased was a soldier.33 Instances of direct association in fresco decoration and those commemorated is less common than in funerary stele due, in large part, to the less durable nature  for a mature, but unmarried, woman, a phrase that continued in use until at least the 4th century (Amm. Marc. 14.6.11 and 16.7.10). 30 Petersen (2006) 118–9. 31 Martinez (1998). 32 Rea (2004). 33 Agnello (1969).

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  247

of the medium. In the case of the hypogeum of the Aurelii, however, error has been compounded by viewing the frescoes through the eyes of the 4th century and later i.e. the images contained in the hypogeum of the Aurelii have been interpreted with reference to Christian iconography and not to iconography that was contemporary or antecedent to it. Furthermore, commentary has not, in the author’s opinion, taken sufficient cognisance of the one indisputable fact concerning the Aurelii i.e. they were liberti. Freedmen celebrated two key features of their lives by their attainment of citizenship: the chance to form a legally recognised family unit and the work that gave them a position in society. Portrait busts of freedmen and women stare out at us across the centuries advertising their hard–won status: the men in togas, the women wearing a wedding ring, the children, even when deceased, are recognised where possible as freeborn.34 If a family was unattainable then their work was lauded, often in conjunction with their patron. Much work was manual labour that Cicero (Off. 1.150–1) would have looked down upon, but those of more ‘acceptable’ occupations were just as keen to commemorate their profession as their less educated colliberti. Engineers, doctors, midwives, grammatici, scribae, apparitores etc. were few in absolute numbers, but nonetheless they constitute a substantial proportion of the surviving evidence. In her 1992 study of occupations recorded in funerary inscriptions Sandra Rae Joshel identified 306 inscriptions connected with administration i.e. 21% of the total, divided into three groups: financial agents, administrators and secretaries/clerks/copyists.35 The hypogeum of the Aurelii’s cubiculum B in particular is a celebration of the attainment of liberty showing as it does, in the central roof medallion, a ceremony of manumissio vindicta (fig. 4).36 Those who seek answers in the practices of Gnostics and others have described this scene as an ‘initiation’ without saying what the ‘initiation’ is supposed to be. The jurists, Ulp. 40.8, Gai. Inst. 4.16 and Hermogenianus Digestorum libri 40.23 describe how manumission was effected.

 34 For examples of these see especially Kleiner (1977). 35 Joshel (1992) 69, table 3.1. Note this figure excludes members of the imperial bureaucracy which Joshel omitted, Joshel (1992) 16. 36 Jastrzębowska (2013).

248  John Bradley

Fig. 4: Detail of central medallion in cubiculum B. Bisconti (2011) tav. 56.

At this time, the first half of the 3rd century AD, the to–be–freed slave was to be accompanied by his or her owner and a lictor or magistrate who wielded a narrow rod known as festuca or vindicta. The festuca was placed upon the head of the slave and with the regulatory wording a mercantile commodity, i.e. the slave, was made a human being. In addition to the scene in the vault medallion the symbolism of this crucial point in the slave’s life is repeated by multiple images of solitary figures holding a festuca, an image that is very rare outside the hypogeum of

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  249

the Aurelii: I have found only two non–Christian images like this outside the hypogeum whereas there is a total of eighteen within the hypogeum.37 In the upper chamber another important element, the patronus, is commemorated. The image (fig. 5) has been variously described as God or Prometheus creating Mankind. The latter is closely paralleled by a fresco, dated to the late second/early 3rd century, now in the Museum of the Via Ostiensis and I am not the first to draw attention to the similarity.38 However, other than Helga Kaiser–Minn, a Promethean solution has been rejected on the grounds that the hypogeum of the Aurelii figure is unlike the usual bearded, shaggy haired Prometheus.39 Rather, it looks very much like a contemporary male portrait with short–cropped hair and stubble beard. Indeed it bears all the likenesses of a real person in the guise of the mythical Prometheus, in which case it is an example of a genr.e in second and 3rd century tomb decoration sometimes referred to as ‘private deification’ i.e. the deceased are portrayed as mythological characters: a wife as Venus, a young boy as Hercules, a young man as Mars.40 What more appropriate figure could there be than Prometheus, the creator of mankind out of clay, for someone who transformed chattels into men and women through the act of manumission?

 37 The first non–Christian example is in the so–called cubiculum of Miltiades in Area II of the catacomb of S. Callisto. A monochrome drawing of a Lazarus figure beneath the figure with a wand led De Rossi to believe in a Christian inspiration for the decoration. The Lazarus, however, has now completely faded while the remaining, coloured decoration is still clearly visible (autopsy by the author and Dr Eileen Rubery of Cambridge University, 13/9/13). The differential degradation of the image argues strongly for the Lazarus figure to be a later addition when this part of the catacomb was Christianised. There is no other evidence of Christian iconography in this part of the catacomb. De Rossi 1864 II. tav. XXIV, 1; Wilpert (1903) 289, tav. 128, 1. Fiocchi Nicolai et al. (20093) fig. 105. The second example is found in the catacomb of S. Gennaro, Naples. Achelis (1936) 81. While there are Christian images nearby these are probably later additions to what is one of the oldest parts of the catacomb. 38 On the fresco see Squarciapino (1953–1955) 114; on its similarity to the image in the hypogeum of the Aurelii see Kaiser–Minn (1981) 85–91. 39 Himmelmann (1975) 14. 40 The principal work on this phenomenon is Wrede (1981).

250  John Bradley

Fig. 5: Upper chamber, rear wall, right side. Wilpert (1924) tav. I (detail).

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  251

 A Collegium? The use of the word fratres suggests some form of close association to which both deceased and dedicators belonged and this idea is reinforced by images in the two subterranean cubicula. In cubiculum B arcosolia are found set within the rear and the two side walls. In each of these arcosolia there are group images of twelve figures devoid of any apparent form of hierarchy. The rear arcosolium identifies twelve men, the two side arcosolia have groups composed of both men and women. That on the left hand side shows six women and six men alternating. There are no grounds for believing these figures to be marital couples: married couples may sometimes be repeated on sarcophagi to illustrate different aspects or phases of their lives, but they are never shown in the company of other married couples.41 On the right hand side (fig. 6) again there are twelve men and women, but in this instance the symmetry has failed. One figure appears to have been erased to maintain the requisite number of twelve figures (nr. 8a). A male figure (nr. 9) extends his arm to fill the gap thus created,42 it is worth noting that the arm is not stretched out to his female ‘partner’ i.e. the idea that these are marital couples is undermined rather than enhanced by the gesture and the maintenance of groups of twelve took precedence over the strange effect removing one of the figures had on the image. In cubiculum A there is more evidence of the importance of groups of twelve. A banquet scene shows twelve men at a stibadium with three servants and a female figure between figures three and four, but not participating in the convivium itself. Little attention has previously been given to the number of diners. The number is unusually large with very few comparanda (outside the obvious Christian examples). What is also unusual is that there is an even number of diners. In Elżbieta Jastrzębowska’s survey of dining representations there is a marked bias, in scenes with five or more diners, towards there being an odd number (cf. Appendix).43

 41 For example on the ‘biographical’ sarcophagus MNR. Terme 125605. Boymel Kempen (1981) fig. 24. 42 Poe, in her 2007 PhD dissertation on the hypogeum of the Aurelii, considered these figures to be an example of dextrarum iunctio, Poe (2007) 69. 43 Jastrębowska (1979).

252  John Bradley

Fig. 6: Cubiculum B. Right hand arcosolium lunette. Bisconti (2011) tav. 78 modified by author.

This is in large measure due to the habit of giving the central place to the guest of honour as described by Philostratus the Younger (Imagines 396K).44 With an even number, however, there is no central figure, no guest of honour and consequently no hierarchy. In the hypogeum of the Aurelii scene the only evidence of any preference is that given by one of the servants to figure number one sitting in cornu dextro, a position taken by the host.45 A connection appears to exist between an even number of diners and dining related to collegia as may be seen in three examples from the Greek east. The first, a stele from Panormos near Cyzicus, dating to the 1st century BC and now in the British Museum, shows a meeting of an association with six diners entertained at a banquet by musicians.46 A second stele, dating even earlier, to 119 or 104 BC from Triglia, Bithynia is dedicated to a priestess of Apollo and Cybele and shows

 44 Engemann (1982) 241–9. 45 In due course the cornu dextra assumed the place of honour, but that was a 4th century development and was established by the time of Sidonius Apollinarius in the mid–5th century (Sid. Apoll. Epist. 1.11.10–16). 46 British Museum, BM 1890, 0730.1. Harland (2003) 56, fig. 7. Nielsen (2014) 274, pl. 61.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  253

a similar scene of dining and entertainment – this time with ten diners.47 Contemporary with this latter stele comes an inscription from Delos with a consular date of 113 BC.48 This inscription has twelve names, six freeborn, six liberti, who are described as magistri of a group dedicated to Mercury, Apollo, Neptune and Hercules.49 Although such groups are less evident in the west they are not unknown. In Italy, in the town of Minturnae, Latium, a temple was constructed in the reigns of either Augustus or Tiberius.50 Its construction entailed the re–use of a total of twenty–nine slabs that each recorded groups of eleven or twelve persons, inscribed in four groups of three described as magistri, one of which has a consular date of 65 BC. Where there were fewer than twelve names a space was left which would have enabled another name to be added. Groups appear to be composed of socially disparate individuals, freeborn, freed and slave. Professions are not stated, with one exception,51 but there are a number of individuals who are denoted as being socii of the salinatores (salt merchants, 4 No.) or picarii (pitch merchants 5 No.). In addition to the social variation, each group contains either all women or all men except in one instance where two female slaves are grouped with ten freedmen and slaves.52 Other inscriptional evidence shows that groups of twelve were associated with collegia. For example, an inscription found near the Porta Praenestina in Rome shows the purchase of part of a tomb by a freed couple for themselves from a group of twelve socii. M(arcus) Antonius M(arci) l(ibertus)| Philomusus | Pompeia Cn(aei) l(iberta) Zosima | sibi et suis ollaru(m) | decem sepulc{h}rum | partem tertiam dec| mam emit ab soci{e}is | XII.

CIL VI.6150 = Dessau, ILS 7896 = EDCS 19000166. Expansion EDCS. From Ostia we have a memorial to an individual whose posts within his collegium are recorded culminating in his position as iudex inter electos duodecim. What his duties were we cannot tell, but as it is the last in an illustrious career, it was clearly important.

 47 Athens EM 919. Pfuhl and Möbius (1977–9) vol. 2, pl. 332. Harland (2003) 58, fig. 8. Nielsen (2014) 274, pl. 62. 48 CIL I.2 2504 = AE 1910.0010 = EDCS 24700009. 49 The freeborn all have roman trinomina whereas the liberti have only their patron’s praenomen and nomen together with libertination marker, but no cognomen. 50 Johnson (1933) 1–4. 51 A mensor identified in CIL I.2 2702 = EDR 073259. 52 CIL I.2 2683 = EDR 073261.

254  John Bradley T(ito) Flavio T(iti) lib(erto) Hilarioni | Decur(io) coll(egi) fabr(um tignariorum) ex lustro XV (AD 64) | nungento ad subfrag(ia) lustro XVI (AD 69–73) | mag(ister) quinq(uennalis) coll(egi) fabr(um) tignarior(um) |5 lustro XVII AD 74–78) | honorat(us) ex lustro XIIX (AD 79) | censor bis ad mag(istros) creando(s) lust(ris) XIX et XX (AD 84–93) | iudex{s} inter elect(os) XII ab ordine lust(ro) XXII (AD 99–103) | Claudia Ti(berii) f(ilia) Prisca viro optimo |10 et Flavia T(iti) f(ilia) Priscilla patri optimo.

CIL XIV.2630 = EDCS 05800609. Expansion Royden/author.53 Such groups of twelve continue into the 3rd century, though sometimes with what appears to be less parity, as seen in an inscription from the catacomb Ciriaca in Rome: [T(itus) Fl(avius) Her]aclida senior, qui fuit | [inter empt]ores XII loco VII is qui fuerunt | [soci eiu]s monumenti s(upra) s(cripti) in titulo maio[re] | ita u]t s(upra) s(cripti) s(unt) eis donaverunt singula loca qui |5 fuerunt infra scripti | [Fl(avius) Hera]clida iunior emptoris f(ilius) | [eius, qui e]xuperaver(at) de titulo maiore, monmen[tum] | [dila]psum ex vetustate refecit cum maceris | [et rifugi]um fecit et si{i} qua alia intra eum mon[mentum fecit] |10 [pa]riter Fl(avia) Flaviane Heraclides emptoris.

CIL VI.18079 = EDR 032581.54 Expansion EDR. Some inscriptions link men and women within the same group of twelve such as an Augustan inscription dated AD 9, again from Rome, which shows two sets of twelve individuals with women (highlighted in bold) and men freely interspersed thus: Q(uinto) Supicio Camerino C(aio) Poppaeo Sabino co(n)s(ulibus) | curatoribus hisque | L(ucio) Munio Gnomone, Q(uinto) Volumnio Amphione | qui recte soci(i) monumenti sunt |

〈:columna I〉 C(aius) Curtius Theotímus | L(ucius) Marcius Philomusus | M(arcus) Remmius Ìucundus | Sestia Rufa | C(aius) Nautius Amphio | M(arcus) Peccellius Hilarus | Aristia Pithusa | Licinia Epigone | C(aius) Flavius Sosia | C(aius) Petronius Varia | M(arcus) Allius Sabinus | Tiberius |

〈:columna II〉

 53 Royden (1988) nr. 183. 54 Cf. CIL VI.5682 = EDCS 19100085.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  255

Clodia Automate | A(ulus) Valgius Eros | Pollia M(arci) f(ilia) | M(arcus) Livius Anteros | Cordia Nardis | Lutatius Hilarus | Rabiria Eucumene | L(ucius) Marcius Primus | Helvia Helpis | Fannius Pelops | Aelia Thalea | L(ucius) Sestius Phileros |

CIL VI.34004 = EDR 118433. Cap. Mus., NCE 714. Expansion EDR. In the second column, the female and male socii alternate in the same manner as the figures in the side arcosolia frescoes in cubiculum B. A further similarity between these and the hypogeum of the Aurelii is that none displays any overt religious affiliation. The emphasis is on their occupation and/or their membership of a collegium. One final inscription which includes both men and women in a group of twelve includes one woman as a curator: Nominaque adscripserunt decur(ionum) | ea quae i(nfra) s(cripta) s(unt)

〈:columna I〉 T(iti) Satelli Eutychi | T(iti) Satelli Lascivi | Asiatici Caesaris | Besiae Iucundae p(ubli) l(ibertae) | Suavettiae Lacesis

〈:columna II〉 Fr(atrum)

〈:columna III〉 Hermae Fontei | Appulei Ampliati | Cesti Calpeni | Fontei Euaristi | Iuni Maximi Suavettia Amaryll(is), Suavettius Alexan[der] | cu(atores) loc(um) coll(egio) et nym(phis) d(ono) d(ederunt) | [Im]p(eratore) Caesare Ves[pa]siano Ạụ[g(usto) –––] | [T(ito) C]aesare Aug(usti) f(ilio) ̣Vespa[siano ––– co(n)s(ulibus)]

CIL VI.10350 = EDR 121978. Louvre, Ma 1652. Expansion EDR/author. This example, not only has women within the group it also has the letters FR in larger lettering between the two principal columns. The FR has been correctly interpreted as fratrum, but incorrectly interpreted as referring solely to the two Titi Satelli in the first column.55 As there is no example of frater being used as a differential in name lists in the same manner as pater and filius are used for homonyms56 the FR must refer to the group as a whole.

 55 Stenhouse (2002) 114, nr. 57. 56 E.g. CIL XIV.250–2 = EDCS 05700249–51 from Ostia, CIL VI.1060 = EDCS 17600398 and CIL VI.10300 = EDCS 17700507 from Rome, CIL XI.1355 = EDCS 5320028 = EDR 129455 and 129457

256  John Bradley The examples cited above indicate that not only were groups of twelve closely associated over a long period of time (from the late 2nd C BC to at least the 3rd century AD) with the formal structure of some collegia but also that men and women could refer to each other as fratres.57 Furthermore, women could form a constituent part of the collegia. Emily Hemelrijk has recently said that women were almost invisible in the public activities of collegia.58 The evidence, however, points towards, at least some, collegia having parts played by women although they may not always have taken part in banquets alongside the men. Women are evident in the hypogeum of the Aurelii in addition to those we have already looked at. In cubiculum B there are the two women either side of the right hand arcosolium and there is a turbaned figure in the soffit of the rear lower arcosolium.59 This emphasis on groups of twelve informs another mystery – why are there eleven figures in the Large Procession I mentioned earlier? There are in fact two processions: the Large Procession that forms the single largest motif spread over three walls and a Small Procession, also of eleven figures, adjacent to the Banqueting Scene, figure 3 shows the relationship between them. A notable difference between the two processions is that whereas the Large Procession figures are dressed in white the Small Procession figures are dressed in the multi–coloured clothes of the Banqueting Scene and other scenes in the upper register i.e. the hortus conclusus and adventus scenes. This difference has not attracted much attention. Only Poe has commented on the colour schemes used by these groups, arguing that: ‘Garments in a range of hues characterize the blessed [...]’ (i.e. deceased).60 The banquet and the Small Procession, according to Poe, therefore represent activities in the afterlife. This conclusion, however, is not convincing. Evidence from Pompeii shows everyday scenes depicted with individuals wearing

 from Luna and AE 1977, 0265 a and b provide lists of several hundred names yet none mention frater although there are instances of possible siblings being present. 57 On fictive siblings among collegia in the Greek speaking part of the empire see Harland (2005) 491–513. For the west see Bömer (1981) 177. See also CIL VI.8531 and 34026 = EDR 119868 = EDCS 24100370 for western examples. 58 Hemelrijk (2015) 205. In this she concurs with Veyne (1997) 189–91 that women could not be part of a collegium due to the quasi socio–political offices that reflected wider political roles closed to women. 59 Olson (2008a) 53 citing Ulp. 34.2.23.2, states that wearing turbans (mitrae) was restricted to females though they appear to have been worn by men in conjunction with Bacchic worship (Prop. 3.17.30; 4.2.31). 60 Poe (2007) 156–8. Poe’s comparanda include the tomb of Octavia Paulina and the columbarium on the Via Portuense, Rome (now in the MNR. Terme).

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  257

differently coloured clothing.61 Indeed, if anything, the use of coloured clothing indicates a mundane context. Clement of Alexandria’s late 2nd century AD polemic against the wearing of coloured clothing suggests that coloured clothing was the norm (Paed. II, 11). Literary references on the colour of clothing mention that white is used for a variety of special occasions the corollary being that coloured costumes are to be expected in the everyday.62 I propose that the three scenes ought to be read as a sequence: the Banqueting Scene shows an image of life as it was; the Small Procession shows life as it is now, the group bereft of one of its number; and the Large Procession commemorates the deceased in his arcosolium joining a new band of fratres in the hereafter making up the requisite twelve. And what of the child? You will recall that one of the graves was that of a male child no more than nine years old. If this were a collegium which involved women, could it also consider a child among its fratres? Opinion is divided whether or not childhood was considered a distinct phase of life in Roman society.63 Dress such as togas for pre–pubescent girls and bullae for young boys would suggest that they were considered separate, but that does not apply to the realm of work. Our literary evidence is biased towards upper class children who are less likely to start work young. Inscriptional evidence from Italy,64 and papyri from Egypt,65 however, give multiple examples of youngsters being apprenticed early on in life, and the jurist Ulpian 7.7.6.1 considers a slave to have economic value once it has reached the age of five. The type of work would clearly be constrained by their physical and mental development, but there appears to be no reason why a child was not considered to be part of the adult world when it came to work just as it is today in parts of the developing world or Western Europe a hundred and fifty years ago.

 61 E.g. a market scene from the Praedia of Iulia Felix. MAN Naples 9062. Roberts (2013) 50, fig. 35. 62 In addition to the well–known donning of the toga candida during election campaigns Gell. 3.4.1 records that Scipio Africanus wore white while under accusation; Tac. Hist. 2.89 records the wearing of white by primi pili and Hdn. 8.7.2 records that delegates from Italian cities to Pupienus Maximus at Aquileia wore white after the fall of Maximinus Thrax in AD 238. Bennett (2006) 99. 63 Monda (2008) 125–36 argues that the idea of a distinct personality for the child was first expressed by St Augustine in his Confessions. For a general discussion on the definition of childhood in ancient Roman society see Mander (2013) 48–50. For a contrary position see Olson (2008b) 139 who argues that the existence of differing dress between adults and children argues for a distinction between the two stages of life. 64 On children working in Italy and Rome see Petermandl (1997) and Ricci–Salvadori (2008). 65 On the subject of children working in Roman Egypt see Vuolanto (2015).

258  John Bradley

 What Kind of Collegium? To answer this question we need to look at other frescoes within the tomb and search for clues. The rear wall of cubiculum A shows a cityscape with, on the right hand side a scene called the hortus conclusus (fig. 3). Within a walled garden are four figures: two males standing together on the left, a woman standing a little way from them and, on the right, another male figure.66 Given the number and gender of figures in the centre of the hortus, there seems to be little doubt that these represent the deceased and the chief dedicator of the inscription below. On the left side is a large porticus (the so–called ‘forum’ scene) depicted with a group of people ranged around.67 Given the freedman predilection for advertising their occupations it is not inconsistent to consider these as places of work. The work therefore is taking place within the porticus, the crowd in the ‘forum’ scene has a number of individuals highlighted, including one woman,68 which may refer to the deceased, complementing the reference in the hortus conclusus. There were many porticus in Rome that housed bureaucratic offices and some collegia were based in them. For example, the Porticus Octaviae, contained a library upon its inception (Plut. Marc. 30, Suet. Gram. et rhet. 21), and the adjacent Porticus Philippi enclosed the Temple of Hercules Musarum which served as a meeting place for a collegium poetarum.69 While others, e.g. the Porticus Minucia Frumentaria, were associated with administrative functions of the city such as the annona. It is worth noting that there are also porticus shown in the upper chamber with figures prominently displayed. These upper chamber frescoes allow us to narrow the field and come to a conclusion on the work being celebrated by the Aurelii. In particular, I wish to focus on one particular figure with a very particular gesture, a kind of ‘V for Victory’ vertical gesture (fig. 7).

 66 Bendinelli (1922) tav. 11. 67 Wilpert (1924) tav. XXII. Bisconti (2011) tav. 30b. 68 Bendinelli (1922) 357. 69 Schmidt in BNP: III.535, s.v. collegium poetarum. Viscogliosi in LTUR: III.17–19, s.v. Hercules Musarum, Aedes.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  259

Fig. 7: Upper chamber. Right hand wall. Wilpert (1924) tav. IIa (detail).

260  John Bradley When, if at all, this image is discussed it is described as that of a teacher70 or philosopher.71 This didactic interpretation is not likely. Whether teaching in a purely pedagogic sense or as the representation of a ‘philosopher’ there are almost always others in attendance as pupils or admirers, which is not the case here; the only other figure on the same side of the wall is of equal status to the figure making the gesture. The gesture is clearly different in meaning to an oratorical or didactic pose conforming to the description in the Metamorphoses of Apuleius (Apul. Met. 2.21)72 i.e. the first two fingers pointing out, the third and fourth curved into the palm and the thumb extended or where the thumb is pressed onto the two outstretched fingers as described by Fulgentius (Fulg. Virg. cont. M143).73 Confirmation that the vertical ‘V’ pose found in the hypogeum of the Aurelii is different to the oratorical or didactic is evident on sarcophagi decorated with both.74 The didactic horizontal pose described by Fulgentius can be seen in an image from the Louvre (fig. 8) where a child is shown in adult form teaching his playmates. An almost identical scene comes from a sarcophagus in the Vatican (fig. 9) where the figure displays the vertical pose.

 70 Bendinelli (1922) 307, ‘nel gesto del docente’. Wilpert (1924) 12, ‘[...] insegnare di Vangelo’. 71 Bisconti (2004) 21. 72 Ac sic […] porrigit dexteram et ad instar oratorum conformat articulum duobusque infimis conclusis ceteros eminens porrigens et infesto pollice clementer subrigens infit Thelyphron. (‘And so, […] he extended his right arm, shaping his fingers to resemble an orator’s: having bent his two lowest fingers in, stretching the others out and pointing the thumb outwards, gently rising, Thelyphron began’. Trans. Loeb, modified by author). 73 The gesture is assumed by Virgil at the start of his explanation of the Aeneid: itaque compositus in dicendi modem erectis in iotam duobus digitis tertiam pollicem comprimens ita verbis exorsus est (‘and so, posed in a speaking manner with two fingers sticking out in the shape of an iota and pressing together a third, the thumb, he began with [these] words’. Trans. author). Despite the phrase ‘erectis in iotam’ which suggests a vertical gesture the gesture is usually depicted horizontally on sarcophagi and in the Terence manuscripts to signify the character speaking (http://digi.vatlib.it/view/MSS_Vat.lat.3868). 74 E.g. Vat. Mus. Chiaramonti 1256 and 1258. Ewald (1999) 144–5, taf. 13, kat. A23 and A24, respectively; Cagliari cathedral crypt (no inventory number), Ewald (1999) 213–4, taf. 98, kat. I5. Teatini (2011) 201–9, figs 188–92, nr. 42.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  261

Fig. 8: Child sarcophagus. Louvre Ma 1520. Ewald 1999 D7 = H7.

262  John Bradley

Fig. 9: Child sarcophagus. Vat. Mus. Galleria dei Candelabri, 2422. Photograph – author.

The reaction of the child companion is different however, in this instance he is taking down the words of the precocious child. In the same way as Quintillian (Quint. Inst. 11.2.92–104) describes various subtle hand gestures to emphasise points in oratory the gesture of this little boy and the man in the hypogeum of the Aurelii upper chamber denotes dictation. Corroboration of this comes from a statuette found recently in the Templum Pacis of Rome (fig. 10).75 Here the emperor Septimius Severus is shown with the same vertical ‘V’ gesture seen in the hypogeum of the Aurelii upper chamber. Too small to be a cult image it was found close to an area of the Templum Pacis that contained a library associated with the great marble plan of Rome in what was some form of administrative centre, possibly that of the Urban Prefect or a cadastral office.76 In a place that would have had an important role in the imperial administration it is perfectly logical to have the emperor shown dictating laws and rescripts to be copied out and disseminated throughout city and empire.

 75 Spinola (2014) fig. 7a. 76 Meneghini (2010) 17.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  263

Fig. 10: Statuette of Septimius Severus, Templum Pacis. Spinola (2014) fig. 7a.

264  John Bradley In a city the size of Rome, there would have been hundreds of scribae and librarii engaged in the many public and private libraries, archives and administrative organisations. As noted earlier, Joshel’s study of non–imperial occupations in Rome found almost as many deceased engaged in administrative functions as in banausic professions despite the undoubted preponderance of manual workers in a pre–industrial society.77 Furthermore, many women could be and were employed in secretarial posts using a variety of terms i.e. a manu,78 amanuensis,79 libraria80 or notaria grece (sic).81 There were also women who bore the title lectrix82 or anagnostria83 which is probably associated with scribal activity.84 In the 1st century AD Vespasian is said, on the death of his wife, to have resumed his relationship with Caenis, described as liberta et a manu of Antonia (Suet. Vesp. 3), while Juvenal refers to a libraria feeling the wrath of her mistress (Juv. 6.476).85 Closer to the time of the Aurelii Eusebius’ Historia Ecclesiae refers to Origen (c. AD 175– 254) employing: κόραις ἐπὶ τὸ καλλιγραφεῖν̣̔̓ ἠσκημέναις (‘girls trained in beautiful writing’ Euseb. Hist. eccl. 6.23.2) alongside male scribes and copyists. It is not inconceivable that there were also children apprenticed to such work, preparing writing material for their elders. Cicero (Cic. Sull. 42–44) says that during his suppression of the Catiline conspiracy he gave instructions that information concerning the conspiracy was to be copied by all the clerks and distributed.86 A short while later he talks of scribae mei implying that there were public clerks in addition to his own personal staff,

 77 Joshel (1992) 69, table 3.1. 78 CIL VI.7373 = EDR 141842 = EDCS 18600087; 9540 = Dessau, ILS 7397 = EDCS 19200253; 9541 = EDR 160219. 79 CIL VI.9542. 80 CIL VI.3979 = EDR 119454; 8882; 9301 = EDR 137602; 9525 = EDR 078376; 37802. 81 CIL VI.33892 = EDR 000884 = EDCS 24100405. 82 CIL VI.8786 = EDCS 18700127; 33473 = EDR 152970 = EDCS 23900587. 83 CIL VI.33830; 34270. 84 On the subject of educated women at work both as scribes and pedagogues see Treggiari (1976). 85 The term used by Juvenal, libraria, is usually translated as ‘wool–maid’ due to a scholiast’s gloss of lanipendia. For an analysis of this and restitution of the reading libraria see Haines– Eitzen (1998) 638. 86 sed statim describi ab omnibus librariis […] atque edi populo Romano imperavi. […] prius etiam edituri indicium fuerint scribae mei […] (‘but straight away I ordered (the evidence) to be copied by all the copyists and published for the Roman people […] even before the evidence would have been published by my secretaries’. Trans. author). Although magistrates generally used their household staff to assist in their official duties this distinction of Cicero’s suggests that there were, in addition, state employed librarii.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  265

elsewhere Cicero (Cic. Cat. 4.15 and Mur. 42) refers to scribae as an ordo.87 Cassius Dio (Cass. Dio 75.4.5) describing events a few years before the construction of the hypogeum mentions a collegium of grammatici playing an important part in the apotheosis of Pertinax in AD 193. Rather than the decoration of the hypogeum being inspired by early Christianity, I submit it is fully consistent with what we know of freedman art and the presence of administrators either public or private in the city of Rome. Finally, I wish to call upon one further piece of evidence that has recently been discovered known as the ‘Altar of the Scribes’ (fig. 11). In this altar, dated to the second quarter of the 1st century AD, the dedicatees are described as scribae to the curule aediles. They are freeborn and shown working in conjunction with a number of assistants with wax tablets and rotuli. Below them is a crowd of clients who, literally, look up to their superiors. Among this crowd are individuals who also clutch writing paraphernalia and so are also involved in the work of the scribae though their position is clearly a subservient one. Within this group there is at least one female and two children. I believe the evidence provided by the hypogeum of the Aurelii decoration, rather than being associated with nascent Christianity, was created by and for a group of such as those found on the ‘Altar of the Scribes’, administrative assistants dictating and copying their superiors’ instructions to enable the smooth running of an empire.

 87 On ordines below the rank of Equestrian see Cohen (1984) and Tran (2006) esp. 341–2.

266  John Bradley

Fig. 11: ‘Altar of the Scribes’. MNR. Terme, 475113. Photograph – author.

The Hypogeum of the Aurelii: a Collegiate Tomb of Professional Scribes  267

 Appendix Tab. 1: Numbers of Diners in Funerary Images.

No. of diners. No. of examples.

% of total.





%





%





%





%





%





%





%





%





%





%





%



Even No.

Odd No.



%

Odd %

          

%



Even %

– 

Total







%

%

≤ diners







%

%

> diners







%

%

Veronica De Duonni

The Library and the scriptorium of the Abbey of Montevergine in the 12th and 13th Century: Presences and Absences Talking about Montevergine’s book heritage is a difficult and challenging task. Two fundamental elements must be taken into account: first of all its dispersal and the scarcity of material caused by the historical events that struck the congregation, and secondly the absence of systematic studies on the subject that can provide a complete picture. Furthermore, the fact that the lack of many elements, preventing us from lighting up an active scriptorium between the 12th and 13th centuries, still raises doubts and questions about the actual origin of the manuscripts. Indeed, for the testimonies still in situ and for those stored in other libraries, there is little reliable information about their original belonging to the Verginian scriptorium, and for many of them a late confluence in the library is suspected.1 In the library of Montevergine, it is still possible to see the manuscript of the De vita et obitu sancti Guilielmi Confessoris et heremite. The content of this manuscript, since the 16th century, has been the object of study by the verginians themselves, and then by scholars who over time have investigated on the foundation of the Congregation. It is in this manuscript that the main source to recreate the first period of the abbey can be traceable. Moreover, worthy of a mention are the parchment ms. 2 which contains the Admonitio ad filium spiritualem of Basilius, the Liber adhortationum sanctorum patrum egiptiorum ad profectum monachorum of Evágrio Monaco and the Sermones ad monachos of Caesarius of Arles, the ms. 3, Rituale et Missarum Romanum, and ms. 4, a Sacramentarium.2 To this we can add that, outside the Library of Montevergine, in 1904, two manuscripts produced in the scriptorium of Montevergine were found. They are the Evangelarium Vat. Lat. 5100 and the Adhortationes SS. Patrum and some Collationes of Cassiano, ms. Vat. Lat. 7606. The study of the inventory manuscripts preserved in the Vatican Library, a long project of reorganisation started by Rain-

 1 Cf. One must be aware of the many fires, especially the one that happened between 1405 and 1406, that destroyed not only the Church but also the administrative registers, the codes and the parchments. See Mongelli (1956) I.4. 2 Montevergine, Biblioteca statale annessa al monumento nazionale di Montevergine, mss. 1, 2, 3, 4. Cf. Mongelli (1959) and Panarelli (2004). https://doi.org/10.1515/9783110632590-016

270  Veronica De Duonni aldi and carried on throughout 200 years, has pointed out how these two manuscripts have become part of the Vatican Library in two different moments. The codex Vat. Lat. 5100 was already present in the inventory done in 1636, and is therefore included among the modern acquisitions, while the Vat. Lat. 7606 is just part of the tenth volume of the Inventarium, that was written between 1876 and 1878 by de Rossi.3 The few survived records of manuscripts, compared to a probable wider book production are not only due to incidental causes, such as the numerous fires that damaged the abbey, but also to the sad events that the congregation had to endure: the most critical periods in the history of Montevergine, i.e. the time of the Commandry (1430–1588); the deletions made during Napoleonic period and after the union of the nation that moved the assets of the Library to the Grande Archivio di Stato in Naples to have them returned to Montevergine only in 1926.4 It seems surprising that Pennacchini in 1937 counted 200 volumes in Montevergine,5 data that he got from Zigarelli.6 Even if we included the 16th–17th centuries manuscripts, we would not reach that number. Unfortunately, in Montevergine no catalogue has survived to witness the presence of a librarian’s concern, as it has been documented, during the previous centuries, for the monasteries of Bobbio or Nonantola or Fonte Avellana, the chronologically closest one to Montevergine.7  3 Città del Vaticano, Biblioteca Apostolica Vaticana, mss. Vat. Lat. 5100 and 7606. Cf. Vian– D’Aiuto (2011). 4 The first Commandry started from 1430 to 1515 and was directed by cardinals, the second one, lasted until 1588, and happened when the abbey was united to the lodging of SS. Annunziata in Naples. See Mongelli (1962); Decree of February the 13th, 1807. This measure, while suppressing the religious orders of the rules of St. Bernard and St. Benedict and their affiliations during the kindgdom, had an exception in the Article 5 for the abbeys of Montecassino, Cava de’ Tirreni and Montevergine, whose libraries and the heritage not only had to be preserved but also increased in the custody of some religious people. See Cuomo (1973). Law of February the 17th, 1861 followed by the one of June the 16th 1862. The final measure about the suppression of the religious orders was the law n. 3036 of July the 7th, 1866. Actually, during the time of the suppression measures, the organic law of the Archives n. 1379 of November the 12 th 1818 was still active. It devoted all its five titles to the ‘Archives of Cava, Montecassino and Montevergine’, that were considered as sections of the Grande Archivio of Naples; that was the reason why its assets would not have had to be return to Naples, as happened for Cava and Montecassino. The codices of Montevergine are part of the second inventory, named ‘Catalogo dei codici in pergamena che si conservano nel Grande Archivio di Napoli 1875’. Cf. Trinchera (1872) 57; Improta (2013) 261–308. 5 See Pennacchini (1937) 128 and Mongelli (1959). 6 See Zigarelli (1860) 18. 7 About the library collections of Bobbio we know of a catalogue drawn up in the 9th century and published for the first time by Muratori. In the manuscript 2248 of the University Library of

The Library and the scriptorium of the Abbey of Montevergine  271

The presence of a scriptorium and a library must therefore be hypothesised thanks to the aid of the few hints that can be found inside the documentary proves. However, before considering the documents, a glance at the content of ms. 1 can be helpful. The first news about the presence of a scriptorium, or rather about the absence of it, is drawn from the De vita et obitu sancti Guilielmi Confessoris et heremite. We know Guglielmo da Vercelli (1085ca–1142) had initially been ignorant about the sacrae litterae.8 He only learned the Psalm 109 from an unknown Ruggiero in Melfi, and since then he became an expert in sacred writings.9 After this, Guglielmo lived as a voluntary penitent,10 a practice that strongly characterized his spiritual experiences between the 11th and the 12th century, as an itinerant preacher,11 from 1111 to around 1113; and as a hermit, when he climbed the top of the mount Partenio. Guglielmo suggested to monks and priests who had gathered around him to live by a norm anachoretica and he said: [...] meum est consilium, fratres, ut propriis manibus laborantes, victum et vestitum nobis et quod pauperibus erogemus acquiramus et, in statutis horis convenientes divina celebramus offi-

 Bologna we can find the oldest inventory of the Nonantola Abbey books dating back to the 11th century. It is, specifically, the donation made by Rudolph I, abbot between 1002 and 1035. It is also known, for the same monastic house, another inventory from 1331, which lists 185 codices. The library collections of Fonte Avellana had a substantial increase during the 12th century, which was witnessed by the two inventories, more or less contemporary, contained in Vat. Lat. 484. About the monastic chronicles, I analyze only what is narrated in the Chronicle monasterii Casinensis. Desiderio’s book commission is in fact documented by the three lists present in the Chronicle, two of the Treasury, respectively referring to the years 1058–1071 and 1071–1087, which list only the books with precious bindings, and a general catalogue, referred to the manuscripts produced from 1058 to 1071. See. Muratori (1740) III, coll. 817–24, Becker (1885) 64–73, Gullotta (1955), Leclercq (1957) 168–72, Hoffmann (1980) III.18, 384; III.74, 456–7; III.63, 444–6, Cavallo (1987), Zironi (2004). 8 Panarelli (2004) 8. 9 Panarelli (2004) 8. 10 Meersseman (1968); Andenna (1983–1984) 2. From the biography we learn that Guglielmo da Vercelli, during the pilgrimage in Galicia, wearing una contentus clamide, nudis etiam pedibus (f. 3v), ate only bread and water (f. 4r), commissioned a blacksmith two iron circles ut eorum unus ventrem pectus alter circumdet (f. 6v). After a while, he wore a breastplate and a helmet, spent long nights standing on just one foot and slept on the ground. The previous historiography asserted on the contrary the Benedictine origin of the community and of Guglielmo da Vercelli, who wore the monastic dress since 14 years. Cf. De Palma (1932); Acocella (1942); Mongelli (1960) and Mongelli (1971). 11 Cf. Meersseman (1965) and Houben (1995).

272  Veronica De Duonni cia (my advice is to work with our hands and to dress simply and to give our belongings to the poor).12 The instructions of Guglielmo were not different from those given by the Institutio eremitarum, provided by Pier Damiani to his confreres of Fonte Avellana.13 For example, the norm anachoretica included the rule to welcome fratres coming from the voluntary penitential experience, and the refusal of the title of abbot for the regent of the community, as it was set out in chapter XXVIII of Opusculum XV.14 However, the orientation suggested by the saint from Vercelli, a perfect life system based, to use an expression from Andenna, on the concept of penitentia–humilitas, was put under question by the priests who had joined him and who did not accept the absence of a differentiation ‘between the officium and the status, between priests and laics’ (1983–1984, 106). In fact, they complained about the precepts dictated by Guglielmo about the work done with bare hands and poverty principles, they asked the saint to erect a church and buy books and vestments. Then, to satisfy their demands, escorted by one companion, he went to Bari where he found what was necessary to officiate.15 This situation is not surprising, as the absence of a scriptorium in a new– born and – most importantly – hermit community seems to be a common condition. The first texts, then, even if not explicitly mentioned in the story, were not produced in the verginian scriptorium but come from Puglia. Tropeano, a monk and verginian scholar, states that ‘it is not an easy task to demonstrate that the current liturgical codices kept in the Library of Montevergine and dating back to the 12th century are the ones purchased by St. Guglielmo’ (1970, 4). It is the subsequent adherence to the Benedictine Rule by the community which, according to Tropeano, gave the input for the creation of a scriptorium and a library on the Partenio mountain. Certainly books, tablets and stili were instruments for the normal furnishings of a monastic cell, according to the rule of St. Benedict. Furthermore, we should see in the many documents containing applications for a

 12 Panarelli (2004) 14. 13 Andenna (1983–1984) 103. St. Romuald also wandered from hermitage to hermitage. He, after having founded his own hermitage, abandoned it, in order to pursue other foundations The foundations of the saint, like those created by Guglielmo, continued their life and expansion without being under the control of a main one. Only in the following period, there was a house– mother (Camaldoli and Montevergine) and a monastic congregation. See Penco (1983) 199–202 and 231–9. 14 Petri Damiani, coll. 335–64. 15 Panarelli (2004) 14.

The Library and the scriptorium of the Abbey of Montevergine  273

price exemption for the passage and the transportation of animals some evidence that the monks used to stock up on parchment.16 Unfortunately, as said, the existence of a scriptorium finds no direct confirmation in this material. In the documents there is never a specific mention of the production of manuscripts and there is no evidence for armarii, librarii, copyists or even a pictor or miniaturist, that could have been the protagonists of any organized scriptorium.17 However, in some documents it appears the signing of frates as scriptores. In 1194 there is news of a friar Matteo (he didn’t sign the document, he only wrote the text pro futuri temporis memoria);18 moreover, in the document of Abbot Donato of 1216 among the fifty subscriptions we find certain a frater Landulfus humilis et monachus scriptor19 (humble friar Landulfus amanuensis monk) and in a text contanined a parchment dating back to 1403 a frater Barthomoleus Saxo de Mercuriano scriptor monasterii (friar Bartholomeus Saxo from Mercogliano, amanuensis of the monastery) is mentioned, and he also put his signature among the subscriptions.20 It is then clear that the presence of a scriptorium cannot be excluded and it is probably to be linked to the administrative and economic needs of the monastery under the increasing rise of the congregation, and of its social, political and economical role, which was already fully affirmed in the early 13th century21. A scriptorium where false documents were also produced with not only the aim of raising the prestige of the monastery’s origins, but also of strengthening and justifying rights, possessions and privileges backdating them to the Norman times, when they were strongly challenged by Frederick II, who recognized as legitimate only those already existing at the time of William I.22  16 Tropeano (1970) 4–5. 17 Cf. Alexander (1992); Cavallo (2007); Flores D’Arcais–Crivello (2010); Pani (2015), Somma (2015) and Zanichelli (2015). 18 Facsimile in the Archivio di Stato of Naples sez. microfilm work number 4 (14.4.59). Now the support is obsolete and the parchment is visible through a paper reproduction used in the Scuola di Archivistica, Paleografia e Diplomatica dell’Archivio di Stato of Naples, that I have obtained thanks to archivist Salemme. 19 Archivio di Montevergine, parchment. 1297. Cf. Mongelli (1956–1962) II.59 and Colamarco (2008). 20 Archivio di Montevergine, parchment 3970. Cf. Mongelli (1956–1962) V.25. 21 Colamarco (2008) 142–3. 22 Many documents of the first years of the abbey are deemed false. De Meo has already affirmed: ‘Le carte di Montevergine mi son tutte sospette’ (I have some doubts about the documents of Montevergine), see Di Meo (1805) X.45; Carlone (1979); Carlone (1984). About the suppositions of Carlone, we have some questions made by Pratesi who does not agree with his

274  Veronica De Duonni Although we have not found any reference to the manuscript production, ‘through the yellowed parchments’, as Mongelli (1962), 9 says, ‘we can see the whole life of a distant world’. It is through them that we will try to bring out the book heritage of the Montevergine library, during the first centuries. In the parchments we indeed find mentions of some manuscripts that are currently lost but that were frequently used during legal disputes. A first loss, caused by the devotion to the founder, can be deduced from what we can read at the bottom of a document dated back to 1137: ‘what is lacking in these papers I took for myself, me friar Fabrizio Marena, as a sign of devotion because these have been in the hands of Saint William’.23 Donations in particular have enriched the library collection of the abbey. It is true that soils or tenements especially were granted to Montevergine, but in some cases, even donations by churches are documented. For example, the Lapio Lord Ruggiero, son of Aldoino, qui fuit ortus ex genere Lortmannorum (descendant of the lineage of the Normans), in 1130 gave the monastery of Montevergine quod costructum est in loco ubi monte Virigine dicitur et aqua Columbi nominatur (that was built on the mountain called ‘vergine’ and in a place called ‘acqua Columbi’) the church built on the river Calore dedicated to Santa Maria. In the document it is written that it is donated with all its assets, including vestments and books, which unfortunately are not specified.24 We can find more data about the nature of the texts that circulated in that period in a document written in Benevento in 1216.25 We know that that year, Judge Pietro Malanima of Benevento, having oppressed the monastery of Saints Philip and James26 several times, requested the payment of some debts. The issue was first delegated by the Pope to the bishop of Ariano; then it was entrusted to Gregorio, the rector of Benevento, who, however, before pronouncing the verdict, had gone to Rome for the Fourth Lateran Council. Malanima took advantage of this delay and demanded pledges as payment from the church of St. James and then refued to return them peacefully. The pledges were: twelve books and some  reasoning because the researcher has the tendency to generalize some acronyms (‘f.’ and ‘c.’) set to the back of the documents. Cf. also Pratesi (1989) 54 and 26. 23 Archivio di Montevergine, parchment 237. 24 Archivio di Montevergine, parchment 189: Offeruimus […] una hecclesia vocavulo Sancte Marie […] cum omnibus rebus et casis et stabilibus et mobilibus qui pertinentes de predicta hecclesia sunt et cum omnibus libri et ornatum de predicta hecclesia, cf. Tropeano (1978) II.378–81, n. 190, october 1130; Mongelli (1956–1962) I.69, nr. 189–90. The church of S. Maria di Lapio is present among the possessions listed in the seal of Celestine III of November 4, 1197. 25 Archivio di Montevergine, parchment 1405. See Mongelli (1956–1962) II.85. 26 Mongelli (1972); Rotili (1989) and Bolognese (2013) 235–6.

The Library and the scriptorium of the Abbey of Montevergine  275

vestments, specifically two homilies, two passionaries, one missal, two manualia, two rationalia, one psalter, one octaticum and two breviaries. 27 However, there is a document that concerns directly the verginian abbey in 1241.28 It is the narration of the seizure of the monastic treasure, guarded in the castle of Mercogliano, by the notary Martino of Airola, prosecutor of the imperial curia, who stretched out a detailed list of what was found. In this long list, an interesting testimony that often gives us the description of the objects, among the sacred vessels we find also a testa evangelia cum opere de Limogis (evangeliary probably with a precious binding). The document does not reveal any vexatious intention towards the monastery and it seems to be more of a precautionary measure awaiting further directives.29 There is only one manuscript in the treasure, but this should not lead us to think about the existence of a single text present at that time, since in the list of Montecassino treasure as well are mentioned only the texts that have a valuable binding, as in this case. Finally, Gubitosa, as a result of the urgent needs of the abbey, gave 125 ounces of gold to the abbot of the monastery.30 During the Ecumenical Council in 1274 in Lyon, the abbot urgently needed a substantial sum of gold for the necessary costs of the journey and the stay and a tax on all the vassals to try to regain the necessary31 was even imposed. Unfortunately, in May 1278 it was required that the monastery would repay the borrowed sum, but having no way to satisfy immediately the request, it was forced to give as pledge its most precious owned objects: three gilded silver crosses, two silver censers – one of them made of gold –, six gilded silver chalices, two silver candlesticks, two silver basins, a silk miter with pearls and precious stones, a silkem and golden cope, a silver arm, a silver pyramid with gilded silver

 27 Archivio di Montevergine, parchment 1405: libros duodecim et quatuor pannos de altaris/ cohopertam et toaliam unam. Libri autem sunt isti duo omelie, duo passonaria, messale unum, duo manualia, duo racionalia, salterium unum et octaticum unum, breviaria duo. 28 Archivio di Montevergine, parchment 1907. Cf. Mongelli (1956–1962) II.212. 29 Cf. Panarelli (2008). He believes that the seizure must be understood as a demonstration at a tormented period that had opposed the emperor and Gregory IX and resulted in the naval battle near the island of Giglio. Another hypothesis is formulated by Mongelli, according to whom the seizure of Montevergine treasure was related to repairs of the imperial castle of Avellino, and occurred between 1239–1240. In this period, Mercogliano and the Casale of Montevergine were subjected to in the expenses. See Mongelli (1963) 6–7. 30 Archivio di Montevergine, parchment 2326. See Mongelli (1956–1962) III.108–9. See Mongelli (1965) I.270. 31 See Scandone (1955) 169.

276  Veronica De Duonni base and silver decorations and a testa evangelia unum de argento decoratum et epistola Pauli de argento decoratas (evangelary with a silver binding and the Pauline epistles with silver decorations). Fortunately, the document tells that, after a short time, Gubitosa insisted to get the money back and she returned the pledges back. This unstable economic situation at the end of the 13th century became more and more serious. Perhaps one may think this situation seems to justify what is narrated by De Luciis.32 He reports the news of a donation from 1329 by Ottone da Extendardo, who gave fifty gold ounces to the monks in order for them to purchase books and vestments. This information, even if it is not confirmed by the documents, indicates that at that time, there was no active scriptorium because it is explicitly said that the texts are to be bought. Having clarified the situation of the texts that were only reported in some documents (and about some of them we have found sometimes their purpose), conclusively it is worth having a look at other documents, which instead give us information about the loss mainly caused by the numerous fires. An example is narrated in an act from 1408. Fra Giovanni da Naples, prior of the church of Santa Maria in Penta, gives the notary Andrea Toccarello a land in the farmhouse of Cervito; a concession that the monastery is forced to make, because all the houses owned by the Ospedale of Montevergine have been burned during the past two years, cum certis Libris ecclesie Montis Virginis (with some books of the church of Montevergine).33

 32 De Luciis (1619) f. 85r. 33 Archivio di Montevergine, parchment 4027. Cf. Mongelli (1956–1962) V.41–42.

Olivia Montepaone

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius The Dutch humanist Hadrianus Junius (1511–1575)1 is generally considered the first scholar to have identified the original Greek title of Seneca’s satire. In the Animadversa, published in 1556 (1.17),2 Junius quotes the passage of Cassius Dio which reports the existence of a σύγγραμμα called Ἀποκολοκύντωσις (Cass. Dio 60.35),3 and he recognizes that work as the text known until then only as Ludus de morte Claudii, the title reported by the manuscripts. Aside from this small paragraph in the Animadversa Junius also wrote a series of Annotationes on the Apocolocyntosis, which were published in 1557 (a year after the Animadversa) in the edition of Seneca’s Opera Omnia curated by the Italian humanist and heterodox thinker Celio Secondo Curione (1503–1569).4 In these notes Junius discusses corrupt passages of the satire, suggesting emendations and producing conjectures, and bases his observations on the collatio of the previous printed editions with a manuscript he had in hand: the Sancti Amandi codex.5 The traditional identification of this codex with the Valentianensis 411 (V) is the subject matter of the present paper. The testimonia of Seneca’s satire are 48 in total: 3 main manuscripts, dated between 8th and 12th century, and 45 recentiores, all descending from two of the older codices, i. e. the Sangallensis 569 and the Londiniensis Add. 11983.6 Codex V seems to have produced no copies; it is a manuscript of the 9th century, belonging

 1 This paper evolves from research conducted for a previous article published in Quaderni di Storia, cf. Montepaone (2016). On Junius cf. Heesakkers (1997) and van Miert (2011). 2 Hadriani Iunii Hornani medici Animadversorum libri sex, omnigenae lectionis thesaurus in quibus infiniti pene auctorum loci corriguntur et declarantur, nunc primum et nati et in lucem aediti. Eiusdem De coma commentarium, Basel 1556. Reedited in Gruter, J. (1604) Lampas sive fax artium liberalium, hoc est thesaurus criticus, Frankfurt, IV, 318–482, and then by van Arckel, C. (1707) Animadversa eiusdem De coma commentarium, ab autore innumeris locis emendata et insignibus supplementis locupletata. Accedit Appendix Hadriani Iunii ad Animadversa sua, nunc primum ex clariss. viri autographo in lucem edita, Rotterdam, which also includes previously unpublished chapters of the Animadversa. 3 From chapter 29 of Book 60 the text is, as well known, actually an epitome by Xifilinus. On the matter of the title and Cass. Dio’ passage cf. Roncali (1990). 4 On Curione cf. Biondi (1985) and Kutter (1955). 5 Curione (1557) 731. 6 Cf. Roncali (1990) vi–viii. https://doi.org/10.1515/9783110632590-017

278  Olivia Montepaone to the monastery of St. Amande near Valenciennes and today it is kept at the municipal library of Valenciennes. The library acquired all books and manuscripts of the monastery because of the expropriations of monastic possessions in 1789: the manuscripts from St. Amande represent the oldest portion of a great number of codices acquired this way.7 St. Amande was a Benedictine monastery founded in 639, which gradually built up a vast and varied collection of manuscripts; codex 411 was created during the office of the abbot Hucbaldus and bears the following notation of the 15th century: Liber monasterii Sancti Amandi in Pabula ordinis divi Benedicti.8 The Apocolocyntosis is written at ff. 90r–105r. At a closer look however, Junius’ annotations of 1557 show several important discrepancies in the reported readings, since in many cases Junius relays in the Annotationes a variant, that does not correspond to the reading of the Valenciennes manuscript. Even though there are also many ‘correct’ readings in Junius’ notes, that coincide with those of V, the inconsistencies on the other hand are so significant to bring into question the identification of codex V with Junius’ manuscript. The following is a complete list of all the different readings. Tab. 1: Different readings of V and Junius’ notes. V

Iunii Annotationes ()

iuratores (f. r; § .)

iurato res

gaudium publicum (f. r; § .)

publicum gaudium

impresserant (f. r; § .)

Impresserunt

solatum illo (f. r; § .)

sola tum illo

exprime (f. v; § .)

Exprome

sed qua (f. v; § .)

sede qua

notorem (f. v; § .)

noto rem

ΜΟΡΟΥΙΑΑΤΟΥΧΗΙΝ (f. v; § .)

Μωρὸς κληρουχεῖν ~ ἀποχὴν

tantum Iovi (f. v; § .)

in tantum Iovi

vos mera pamalia (f. v; § .)

Vos me rapamalia

de tot actibus iuris (f. v; § .)

de tot acribus viris

 7 A rich bibliography on the history of the library of Valenciennes as well as that of the monastery is offered on the website of the library, https://bibliotheque.ville–valenciennes.fr/iguana/ www.main.cls?surl=histoire; concerning cod. V cf. also Giancotti (1963) 57–67. 8 The manuscript is digitized and freely accessible on the Gallica website: http://gallica.bnf.fr/ ark:/12148/btv1b84525817.r=ludus%20de%20morte%20claudii?rk=107296;4.

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  279

V

Iunii Annotationes ()

nam etiamsi forme agrece nescit ego scio ENTYCONTONIKNNAIHC (f. v; § , –)

Nam etiamsi nescit φόρμιγγος ego scio ἐντύνων τὸ καλλινίκε ἡρακλῆς

quøi (f. r; § .)

Quoi

senatorum (f. v; § .)

Sonatorum

ΜΕΤΑΛΩΧΟΡΙΚΩ (f. v; § .)

μεγαλογορικω

cordatus iano (f. r; § .)

cordatus homo

EIRHKAMEN ΣΥΝΧΑΙΤΡΩΜΕΝ (f. r; § .)

ἡρήκαμεν συγχαίρωμεν

percrebuit (f. r; § .)

percrebruit

imperatus (f. r; § .)

imparatus

ea condonat (f. r; § .)

Aeaco donat

Generally there are quite a few letter–changes, such as exprome, which Junius reports instead of exprime (§ 7.1), or sonatorum instead of a very clear senatorum in V9 (§ 12.1). An interesting case is represented by quoi: Junius states that he can clearly read quoi in his manuscript, which he regards as an older form of qui (741; § 11.1); however, V unmistakably reads quo with a dot below o and an i added above (f. 100r). The most significant cases are perhaps the Greek passages: first of all, even though the Greek of the Valentianensis 411 is corrupt, it remains clearly readable, and this conflicts with what Junius says more than once in the Annotationes;10 secondly there are very evident differences between the readings of V and those reported by Junius. The Greek proverb pronounced by Augustus in his discourse against Claudius’ deification is a particularly significant example: the correct form is nam etiam si sura mea Graece nescit ego scio: ἔγγιον γόνυ κνήμης (§ 10.4); V reads the corrupt nam etiamsi forme agrece nescit ego scio ENTYCONTONIKNNAIHC (f. 99v), and Junius states his manuscripts bears ‘something similar to’ Nam etiamsi nescit φόρμιγγoς ego scio ἐντύνων τὸ καλλινίκε ἡρακλῆς (740).

 9 Curione (1557) 739 and 741. 10 Cf. Curione (1557) 731. The passages are discussed below.

280  Olivia Montepaone Very interesting is the reading of what is supposed to be μεγάλῳ χορικῷ, introducing the anapestic nenia (§ 12.3): while cod. V bears ΜΕΤΑΛΩΧΟΡΙΧΩ11 (f. 101v) Junius reports μεγαλογορικω and therefore proposes to write μεγαληγορία (741). Junius’ conjecture clearly derives from the reading of γ instead of χ, and this cannot have come from the confusion of the two letters by Junius himself. In addition to this, Beatus Rhenanus had already suggested ἐπιτάσει χορικῶς based on what he could read in his own manuscript in 152912 and, even though Junius does not care to mention Rhenanus’ hypothesis in his note to the passage, it seems even more unlikely that he would have failed to recognize and mention the word χορικός in his codex, when the adverb χορικῶς had already been suggested by another scholar. Finally there is an exceptional discrepancy, which I did not report in the chart because of the uncertainty of the manuscript reading: Junius states at 741 that his codex reads Narcissus duci iusserat (§ 13.4) but V has a corrupt text with a marginal note which is not easy to read and gives an even more fallacious variant. At f. 103r V reads narciosum duci cusserat with symbols of correction above osum and above duci; the marginal notes seem to spell narcius ducius but the ink is faded in the last two letters of the second word and the r in the first word is blotted. In this case, the text of the printed editions already presented a correct version, Narcissus duci iusserat, of which there was no need to doubt, and it is likely that Junius did not wish to give account of the nonsensical text of his codex. It is, however, noteworthy that Junius makes no mention of the marginalia and simply reports the text as said, not concerning himself with this particular sentence but mainly with the preceding list of names (Claudius’ and Narcissus’ victims).  11 I report the reading of V as given by modern editors (cf. Roncali [1990] 19), but I am actually more inclined to read METAAΩΧORIΧΩ as α and λ are extremely similar in the writing of cod. V, and often confused by the scribe in the Greek passages: the loop in α is not circular but very thin and angular, drawn rather as two short diagonal lines connected at the end, so that the letter is almost identical to λ, especially if the lower diagonal line is very slight. In this passage what is supposed to be a λ seems to have such narrow slight loop, therefore leaning towards the letter α rather than λ. 12 Beatus Rhenanus is the first scholar to produce an annotated edition of Seneca’s satire since the editio princeps of 1513 had no commentary: in 1515 Rhenanus’ first edition of the Ludus with commentary surrounding the text was published in Basel but at this stage the German humanist appears to have had nothing else aside from the text of the princeps to work on. In 1529 a new manuscript appeared, which Rhenanus calls codex Wissemburgensis (unfortunately the manuscript is lost, cf. Spaltenstein–Petitmengin [1979]) and a new edition of the satire, included in the Erasmian Opera Omnia is published that same year, with a new version of Rhenanus’ notes, based on the findings of this codex. This 1529 edition served as base for Curione’s own work on Seneca: Rhenanus’ new annotations on the satire are fully included in the 1557 edition.

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  281

In addition to these discrepancies there are also several peculiar lessons in cod. Valentianensis, unique if compared to what the printed editions read, which Junius fails to report. Following is a list of the lessons of V alongside the reading of the latest printed edition of the satire, i. e. that of Beatus Rhenanus’ edition of 1529;13 the interpolated passages are marked with an asterisk.14 Tab. 2: Lessons of V and Beatus Rhenanus’ 1537 edition. V

Rhenanus 

om. (f. r; § .)

Asinio Marcello Acilio Aviola coss.*

iussoque (f. r; § .)

Visoque

quid huic et respondit invides? (f. v; § .)

quid huic invides? et respondit

om. (f. r ; § .)

Tum ille*

dicunt (f. v; § .)

Ducunt

quod autem (f. r; § .)

quid autem

 13 I was able to see the 1537 edition, which is a reprint of the 1529 edition: (1537) L. Annaei Senecae opera, et ad dicendi facultatem et ad bene vivendum utilissima, per Des. Erasmum Roterod. ex fide veterum codicum, tum ex probatis auctoribus, postremo sagaci nonnunquam divinatione sic emendata, ut ad genuinam lectionem minimum desiderare possis. Adiecta sunt scholia D. Erasmi Roterodami et Beati Rhenani, illius in bonam partem operis, huius in Ludum de morte Claudii Caesaris, Basel. 14 This list, contrary the previous one, is a selection of the most relevant cases, as the number of omissions is much larger than the discrepancies. In order to give a complete view, all of the other lapses are listed here, including some obvious errors of V, that may not have been mentioned by Junius because manifestly incorrect. Junius fails to report the variant at § 1.1 quaesiverit of cod. V (f. 90r) whereas Rhenanus’ edition had quaesierit; § 1.2, quid viderit present in cod. V (f. 90v) vs. Rhenanus’ quod viderit; § 2.2 intellegi in V (f. 91r) vs. Rhenanus’ intelligi; § 2.4 curru in V (f. 91v) vs. Rhenanus’ cursu; § 2.4 noctis in V (f. 91v) vs. Rhenanus’ nocti; § 4.1 Filaetus in V (f. 93r) vs. Rhenanus’ et laetus; intentus in V (f. 93r) vs. Rhenanus’ intentas; § 5.1 sint acta in cod. V (f. 94r) vs. Rhenanus’ sunt acta; § 5.3 pererraverat in cod. V (f. 94v) vs. Rhenanus’ penetraverat; § 6.1 quae cum illo in cod. V (f. 95r) vs. Rhenanus’ quae cum ipso; § 6.1 vides in cod. V (f. 95r) vs. Rhenanus’s audis; § 6, 1 multa alia in V (f. 95v: it is actually a correction of milia in alia with two dots below m and i and an a written above) vs. Rhenanus’ multa milia; § 6, 2 quod decollare of cod. V (f. 95v) vs. Rhenanus’ quo decollare; § 8.3 stulte stude in V (f. 97v) vs. Rhenanus’ stulte studere; § 8.3 qui Romae inquis in V (f. 97v) vs. Rhenanus’ quia Romane inquit; § 9.1 volo ut servetis in V (f. 97v) vs. Rhenanus’ volo servetis; Ianus pateris in V (f. 98r) vs. Rhenanus’ Ianus pater; § 9.5 deus sit in V (f. 99r) vs. Rhenanus’ deus fiat; § 11.1 nesti in V (f. 100r) vs. Rhenanus’ nescis; § 11.5 socerum in V (f. 100v) vs. Rhenanus’ consocerum; §11.5 vocationem in V (f. 101r) vs. Rhenanus’ vacationem.

282  Olivia Montepaone

V

Rhenanus 

om. (f. r; § .)

Nec post boletum opipare medicamentis plus cibi sumpsit*

quid dicebat (f. v; § .)

quod diceret

illi (f. v; § .)

Ei

ut quid (f. v; § .)

et quid

precidet imperum (f. v; § .)

praecidit ius imperii

duas sicilias (f. v; § .)

duas amitas suas

magnum pompeium et l sillanum (f. r; § .)

Pompeium Magnum Antoniae ex Petina et L. Syllanum Octaviae ex Messalina*

om. (f. v; § .)

dominus domini*

om. (f. v; § .)

ecce extemplo*

om. (f. r; § .)

Exterritus oculos…defendert*

om. (f. v; § .)

Non umquam Sisiphum onere relevari*

Some lapses are quite significant, as for example those of the several interpolations: no manuscript of the Apocolocyntosis reports any of these interpolated passages, which were introduced with the editio princeps of 1513 (probably by the first editor himself, Caius Sylvanus Germanicus).15 In the 1529 edition Beatus Rhenanus, who was in possession of a manuscript of the satire, the codex Wissemburgensis,16 reports and comments each and every passage that is absent in his manuscript. It is important to point out that Junius admits to have used Rhenanus’ 1529 edition to compare the text of the satire with his own manuscript, and also that he often criticizes the German humanist in the praefatio as well as in the notes for his hastiness in correcting the text.17 While in the above–discussed case of the Greek text at § 12.2, ἐπιτάσει χορικῶς was a conjecture, which Junius may not have mentioned because he disagreed with Rhenanus, the absence of the interpolations in Rhenanus’ codex is ‘manuscript evidence’, thoroughly examined by the German scholar, and it seems therefore very curious that Junius should choose not to discuss them at all.18

 15 Cf. Sabbadini (1919). 16 Cf. n. 14 and 15. 17 Curione (1557) 737. 18 Especially because there are three other interpolations, which are in fact discussed by Junius. About Sauromatas et si qui ultra glacialem Boream incolunt Barbari introduced at § 3.3 after Britannos Junius notes that neither Rhenanus’ nor his codex bear this sentence, which is clearly to be regarded as interpolated and therefore removed from the text (738). In the case of the very

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  283

Aside from the interpolations there are a few other noteworthy omissions on Junius’ part. duas sicilias, found in V (f. 99v) where the printed editions read duas amitas suas, is clearly corrupt (the correct form is duas Iulias, § 10.4), and it may not be strange that Junius failed to report it. But other lessons, such as quid dicebat (f. 95v) instead of quod diceret or ut quid (f. 99v) instead of et quid would have been interesting variants to note and comment. Some of these unreported readings actually appear in the 1557 edition as printed marginalia (usually preceded by ‘al.’), introduced by Curione without any further comment. The traditional identification of codex Iunii with codex Valentianensis is mainly based (apart from the aforementioned existence of variants common to both manuscripts) on the fact that Junius refers to his manuscript as ‘Sancti Amandi codex’, and V bears the colophon of the Saint–Amande monastery. It is however worth noticing that in the 1557 edition in neither of the two introductions to the Annotationes Junius refers to his manuscript as ‘Sancti Amandi codex’. The two introductions are a letter and a praefatio (735–737): Junius lived in Haarlem while he was working on the satire and writes a letter – dated July 1557 – ‘Ioanni Chauco consuli Ultraiectino’. The addressed person is Johan van Cuyck (1500?– 1566) burgomaster of Utrecht and humanist19 with whom Junius often exchanged codices; several letters testify to this activity.20 In this letter of 1557 Junius thanks his friend for having sent the manuscript of the satire, which he is now directing back to him, and refers to the codex exclusively as ‘libellus’ or ‘exemplar manuscriptum’. Chaucus is depicted by Junius not only as a great collector of manuscripts and emendator of classical texts, but he is also described through the traditional image of the ‘savior’ who brought out of the libraries codices which were damaged by woodworms and neglect (736). The only instance where Junius refers to his manuscript as ‘codex Sancti Amandi’ is after the two prefaces, preceding the textual notes: ‘duae illae litterae S. A. Sancti Amandi codicem vetustum innuunt, quo usi in castigando sumus’ (737).21 The notation ‘S. A.’ does not in fact accompany all lessons cited from the  long interpolation at the end of Augustus’ speech Cogitate patres conscripti..principium dedit (§ 11.3) Junius does not seem to consider the passage spurious but he states that it does not appear in his manuscript, just as Rhenanus had noted in his codex (741). As for Ille autem patrono plura blandiri volebat quem Mercurius iterum festinare iussit, et virga morantem impulit introduced in § 13.3 after venire nos nuncia, Junius states that his codex omits the phrase and suspects it to be spurious (741). 19 Cf. Regt (1930) 353. 20 For Junius’ correspondence cf. Heesakkers–van Miert (2010); and particularly the letter in Scheltema (1839) 22–23. 21 ‘Those two letters S. A. stand for Saint Amande’s manuscript, which we used in correcting’.

284  Olivia Montepaone manuscript within the Annotationes: Junius also uses generic expressions such as ‘manuscriptum exemplar’ or ‘codex noster’.22 The identification of Junius’ codex with the Valentianensis 411 appears for the first time in K. R. Fickert’s edition of Seneca’s Opera omnia of 1845. At that time the knowledge about the Apocolocyntosis manuscripts was much more restricted than nowadays, and up until Fickert in fact only one of the three main testimonia, namely the Sangallensis 569, was known to scholars. Fickert reports in his edition the new discovery of the Valentianensis and says that the manuscript has been examined by Theodor Oehler.23 Oehler (1810–1843)24 collated many Valenciennes manuscripts, amongst which was also cod. 411, and sent his observations to Fickert. This is Fickert’s statement in vol. II of his edition, published a year before the volume containing the Apocolocyntosis: Cum enim de ludo edendo desperassem fere propter codicum vetustorum et integrorum inopiam, iucundissime mihi accidit, ut Theodorus Oehler Moeno–Francofurtanus, qui in Museo Rhenano de codicibus manuscriptis nuper non pauca disseruit ipsius diligentiam et eruditionem probantia, Parisiis d. XXVIII Martii mensis ad me daret litteras, quibus mihi nuntiavit Valentiani esse codicem illius libelli sec. IX scriptum, operam suam in eo excutiendo simul pollicitus.25

In the preface to vol. III, which contains the satire, Fickert discusses at length the textual status of the Apocolocyntosis, and specifies concerning codex V: De ludo Senecae emendatius edendo spes, quam ex Theod. Oehleri litteris conceperam (vol. II praef. p. XVIII) decepta fere est, cum cod. Valentianensis, quem paucis ante mortem suam mensibus cum editione Tauchnit. accuratissime contulit Th. Oehler, alius non sit quam cod.  22 It is important to note that the variants introduced by generic expressions are not the only ones that report different readings from V: in both cases, where Junius uses the notation S. A. and also when he uses codex noster or similar expressions, there are equally discrepancies as well as correct readings. 23 Fickert (1845) III.viii. 24 Nothing is known about Oehler’s life aside from the dates of birth and death. 25 ‘When I had almost no hope of producing a better edition of the Ludus because of the lack of old and complete manuscripts, it happily occurred that Theodor Oehler from Frankfurt, who lately discussed quite a lot about old manuscripts in Rheinisches Museum proving his own diligence and erudition, sent me a letter from Paris on March 28th, announcing that there was a manuscript of the satire in Valenciennes, written in the IXth century; at the same time he promised me that he would examine it’. The reference to Oehler’s contribution in Rheinisches Museum probably alludes to what was published in 1842 (Rh. Mus. 1, 130–43) in the section Mittheilungen aus und über Handschriften: the articles in this section are all written by Friedrich Ritschl, but in the introduction Ritschl firstly thanks Theodor Oehler for communicating the contents of some Latin manuscripts.

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  285

St. Amandi (cod. Iunii), olim ab Hadr. Iunio usurpatus, nec suppleat lacunas aut Graeca corrupta melius concipiat.26

Fickert then reports Oehler’s concise description of the contents of codex V. I have been able to see Oehler’s handwritten notes on the Valenciennes manuscripts, which are kept as ‘Auszüge aus Handschriften der Bibliothek Valenciennes’ at the Universitätsbibliothek in Frankfurt amongst the ‘Nachlässe’ from different scholars.27 Following is a transcription of what Oehler wrote about the Apocolocyntosis: „Senecę LUDUS De MORTe CLAUDii „Quid actum sit in caelo ante diem III. iđ. octobŕ.† anno nono initio scłi felicissimi volo memoriae tradere. x 15 1/3 f. – illum eaco donat is menandro liberto suo tradidit ut a cognitionibus esset. (Hucusque = Ed. Tauchnitz. 12° V 1832 p. 214–226) Damnabis numquam longum post tempus amicum Mutavit mores sed pignora prima memento. Reliqua et sequens pag. vacant. †

Nomina consulum desunt in Cd.

Particularly interesting is the annotation in brackets, which seems to state that the text of cod. V is identical to the editio Tauchnitziana (with the exception of the nomina consulum which were absent in the manuscript, as Oehler notes with an asterisk at the end of the page): as Fickert specified, Oehler used this edition of 1832 for his collation.28 The text of this edition is in actual fact very distant from that of the Valentianensis, since it is clearly derived from the textus receptus of the printed editions without any use of specific manuscripts. In an article of 1888 in the journal Breslauer philologische Abhandlungen Otto Rossbach declares that Oehler’s collation either was not very precise or had been wrongly used by Fickert.29 Rossbach therefore carries out a new collation and notices the discrepancies between the lessons reported by Junius and those of codex

 26 ‘The hope of a better edition of Seneca’s Ludus, which I had built up from Theodor Oehler’s letter (vol. II praef. p. XVIII), was deceived since the cod. Valentianensis, accurately collated by Oehler a few months before his death with the Tauchnitz edition, is none other than cod. Sancti Amandi (cod. Iunii), once exploited by Hadrianus Junius, and it does not fill up holes nor reports a better version of the corrupt Greek sections’. 27 Ms. Ff. Th. Oehler III, 3, Universitätsbibliothek Johann Christian Senckenberg, Frankfurt am Main. 28 Tauchnitz (1832). 29 Rossbach (1888) 25–31.

286  Olivia Montepaone V. Rossbach states that the codex used by Junius was in fact the Valentianensis, but that the Dutch scholar did not report the lessons in the way they appear in the Valentianensis; clearly Rossbach did not regard the discrepancies as significant enough to doubt the identification of codex Iunii with V, which he never puts in question. Rossbach suggests that Junius could have seen yet another manuscript in addition to V: the one referred to with the initials S. A. would be the Valentianensis, while the one mentioned in the preface would be the manuscript sent to Junius by Chaucus, which would have created the confusion. Finally Rossbach states that, what should really be reconsidered is Junius’ credibility in reporting variants: ‘non exigui momenti est quaerere quae H. Iunii fides sit in afferendis librorum vetustorum lectionibus’.30 The identification between Junius’ manuscript and V also leads scholars to recognize Junius as the author of two annotations written in the margins of the Valentianensis (f. 95r), which report the correct version of two Greek passages of the satire (§ 5.4, Od. 9.39 and Od. 9.40). Scholars unanimously ascribe the handwriting to the 16th century; Junius does in fact correct the two passages in the Annotationes of 1557, and these are recognized today in modern apparatuses as Junius’ emendations. If we wish to see the two manuscripts as one and the same, we must picture that, apart from Junius making numerous errors in reporting variants, the codex was brought out of the monastery, where it was safely guarded, by Chaucus or someone at Chaucus’ order; it must then have been sent to Junius, who read it and wrote the two marginal notes (and only these two notes, of all the passages he emendated), and then sent it back to Chaucus, who was able to have it put back in the monastery. At this point, there is reason to suggest that Junius was in fact reading a copy of the Valentianensis. As already mentioned, the variants found in the Annotationes confirm that Junius’ manuscript is related to V31. The elements that could

 30 ‘It would not be without importance to investigate if H. Junius is reliable in reporting readings from old manuscripts’, Rossbach (1888) 29. 31 Readings peculiar to V and reported by Junius are for example: § 3.2 mensis in V (f. 91v) while all other manuscripts have mensibus; § 4.1 subtemina in V (f. 92v) vs. subtemine/subtegmitra/subtegmina in other codices; § 8.2 oro per quod in V (f. 97r) fully spelled out vs. the abbreviated version in the other codices; § 9.3 fama nimium of V (f. 98r) and L vs. famam mimum; § 9.4 adhuc found only in V (f. 98r) vs. ad hunc/ad hoc; § 11.1 regnatum in cod. V (f. 100r) vs. regnat; § 11.3 ad summa in cod. V (f. 100v) vs. ad summam/at summam; § 14, 1 cedit subscriptionem in V (f. 104r) vs. edit subscriptionem; § 14, 3 si uni dii laturam fecissent of cod. V (f. 104v) vs. siumdiu laturam fecissent/si unius dii laturam fecissent; § 14, 4 illi ex veteris in V (f. 104v) vs. ulli ex veteribus/ulli ex veternis; § 15.2 adiudicatur c. Caesar in cod. V (f. 105r) vs. adiudicatur c. caesari.

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  287

suggest the presence of a copy are the aforementioned case of quoi, which clearly seems a correction mistaken by the scribe for an addition: the small dot below the o may not have been seen or may have been misunderstood and the i written above the o was regarded as an added letter instead of a substitution. The lesson iurato res instead of iuratores is also significant: precisely such small detail – the separation of two syllables – is an easy mistake to make while rapidly copying a text, especially since in V the letters are quite large and well spaced. Similar to this is publicum gaudium instead of gaudium publicum, a simple inversion of the two words, which can easily have occurred while copying the text. Another element which points towards a copy is the case of Narcissus duci iusserat: the fact that Junius does not mention the marginal notations here (nor anywhere else, even though V has several marginal corrections), may suggest that in his manuscript there were none, because the scribe would not have copied the corrections as marginalia of course, but he would have made the indicated alterations to the text (just as he did for quoi). Most significant is perhaps the Greek. Junius states that the Greek portions of text in his codex are almost unreadable: particularly the statement ‘dum mihi cum scriptoris Graece haud dubie imperiti manu (quae in imitandis Graecis characteribus mira litterarum portenta invexerat) certamen acre suscipitur’32 (Praef. 735) seems to suggest a complete confusion of the characters, as well as ‘usqueadeo confusa est ab imperito scriba Graecarum literarum textura, ut […]’.33 (Ann. 740). This conflicts with the actual state of the Greek passages in the Valentianensis, where (as visible in the digitized manuscript online) the Greek letters are quite clear. It appears that whoever wrote the manuscript used by Junius was not able to imitate the Greek letters and produced a very poor copy. A letter dated 1557, kept at the Universitätsbibliothek in Basel, documents the mailing of Junius’ Annotationes to the editor Herwagen and explains how Junius intended his work to be published.34 After two pages of apologizing for having sent his work late, Junius describes with great precision how he wishes his notes to appear. First of all he talks about graphical aspects, such as what he wants printed in italics – and he reports one of the contradictory lessons as an example, percrebruit instead of percrebuit in cod. V (§ 13.5; f. 103r). Afterwards Junius states

 32 ‘While I was battling against the hand of the scribe undoubtedly ignorant of the Greek (who introduced extraordinary monstruosities in trying to copy the Greek characters)’. 33 ‘The structure of the Greek letters has been so confused by the ignorant scribe that […]’. 34 G2 I 15:1:59–60, Universitätsbibliothek Basel.

288  Olivia Montepaone that ‘in Castigationibus S. A. significare codicem vetustum e Bibliotheca Sancti Amandi nobis communicatum quo usi fuimus’.35 This letter allows us therefore to confirm without any doubt that Junius himself wanted to identify his manuscript as codex Sancti Amandi and also that the codex ‘had been communicated’ to Junius through Chaucus, i.e. that Junius did not make himself the journey to the monastery. The use of the verb communicare with reference to the manuscript is quite interesting: it seems to suggest, rather than an actual exchange of codices, a transmission or communication of what was contained inside the manuscript.36 Junius used the same verb in the preface to the Annotationes (737): ‘Porro ubi iam codices impressos cum manuscripto exemplari (quod Ioannes Chaucus noster, vir a Gratiis et Musis factus, et adiuvandis melioribus literis natus, nobis communicavit) contulissem […]’.37 In the letter to Herwagen Junius also says: ‘epistolam vero ad Io. Chaucum meam praefigi velim’. The Dutch humanist specifies that he wishes his letter to

 35 ‘In the Castigationes the two letters S. A. stand for the old manuscript we used which was transferred to us from the library of Saint Amand’. (italics mine). 36 Professor G. Benedetto pointed out another interesting occurrence of the verb communicare in Bonaventura Vulcanius’ preface to his Nili Archiepiscopi Thessalonicensis De primatu Papae Romani libri duo, Ludguni Batavorum 1595, *4: ‘Cum itaque tandem Nili codicem e Bibliotheca Vaticana fideliter transscriptum nactus essem, eumque amicis aliquot meis gravibus doctis et piis viris communicassem, hortati sunt [...]’ (‘As I came across Nilus’ codex from the Vatican Library, accurately transcribed, and I communicated it to some serious, learned and pious friends, they urged […]’). The context is similar to that of Junius’ letter, and what seems to be implied here is precisely the exchange of a transcript, not of the original manuscript, as the expression ‘fideliter transscriptum’ clearly states. Another very interesting use of communico appears in Janus Gruterus’s praefatio to his edition of Seneca’s Opera Omnia in 1594: Gruterus gives a precise account of the manuscripts he was able to see, and after mentioning five Palatine manuscripts he states that a sixth codex ‘mihi comunicarunt Fratres minores qui Coloniae’ (Gruterus [1594], praefatio IIII). This last codex, deriving from a monastic library, is isolated from the others, which were personally consulted by Gruterus at the Palatine library, and is also different from the subsequent manuscript that Gruterus discusses, which was sent to him by a personal friend: in the latter case Gruterus uses the verb ‘transmisit’, which appears to imply that the actual codex, and not a copy of it, had travelled to him. Gruterus is very precise throughout the entire edition in reporting variants from other manuscripts or humanist conjectures, never failing to account for different readings of a passage, and therefore his use of distinct verbs concerning the codices (communico vs. transmitto) can be regarded as conscious and as a significant clue to how he came in possession of them. 37 ‘Moreover when I collated the printed editions with the manuscript (which our Ioannes Chaucus, a man made by the Graces and the Muses and born to give aid to the letters, reported to us) […]’. (italics mine)

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  289

Chaucus to be used as preface (as it actually appears in the edition), and insists that Chaucus is given a prominent role within the edition. The following sentences are unfortunately not easy to read because they have been corrected and overwritten: In liminari vero facie operum Senecae post Coelii Curionis ut fit mentionem seorsim fieri mentionem velim diligentiae utriusque nostrum ac primum Io. Chauci hoc modo vel simili: Controversiarum libri ex editione Io. Chauci Ant. F. Ita enim illi viri erudito visum est. Deinde mei item ἀποκολοκύντωσις Senecae seu Ludus de morte Claudii ex restitutione Hadriani Junii cum eiusdem annotationibus. Item sententiae ad veteris codicis lectionem repositae ab eodem.38

Junius wishes that Chaucus’ name appear before his own and alongside Curione’s, a fact which again indicates that Chaucus had an important role in getting hold of the manuscript and in interpreting it; the preface of 1557 puts the same emphasis on Chaucus. It is therefore possible to suggest that Chaucus did not simply hand a manuscript to Junius, but that he had a copy made for his friend and only such copy of V left the monastery. Junius’ request to give priority to Chaucus was not respected by the editor Herwagen. The edition is presented as Seneca’s Opera Omnia, edited by Celio Secondo Curione, and the edition of the Apocolocyntosis appears as Curione’s work only, to which Renanus’ and Junius’ notes are a mere addition: the only mention of Chaucus is the one Junius makes in his prefatory letter. Curione disputes Junius’ discovery of the Greek title, claiming that he had made the same finding and that he had told Herwagen about it long before Junius (732): in the effort to emphasize his own claim, Curione completely overshadows the Dutch humanist and leaves no space to Chaucus. The letter Junius sent to Herwagen is undoubtedly ambiguous – especially with regard to the use of the verb communicare – and the way the Basel edition was put together during the summer of 1557 seems to add further confusion, introducing Curione, who appears before Junius within the edition. Ironically, Curione and his contributions have been almost forgotten and now very rarely appear in critical apparatuses of the Apocolocyntosis, while Junius is regarded not just as the first to have suggested the Greek title found in Cass. Dio, but also as  38 ‘I wish that at the end of Seneca’s works after mentioning Celio Curione, both of our works should be mentioned and especially first Ioannes Chaucus’ in this or a similar manner: the Controversiarum libri ex editione Io. Chauci Ant. F. This is how the erudite man [scil. Chaucus] wishes it to be. Afterwards my own Senecae Apocolocyntosis seu Ludus de morte Claudii ex restitutione Hadriani Junii cum eiusdem annotationibus. Item sententiae ad veteris codicis lectionem repositae ab eodem’.

290  Olivia Montepaone the only author of the several emendations to the text of the satire that were in actual fact proposed by both him and the Italian humanist in this 1557 edition.39 In conclusion, it is certain that, as Rossbach already noticed, there are several textual discrepancies between Junius’ readings and those of cod. V, which I believe to be significant enough to doubt the identification of codex Iunii with codex Valentianensis. To this one must add the practical difficulties in bringing the manuscript out and then putting it back into the Saint–Amande monastery. The hypothesis of a copy made on Chaucus’ orders and then sent to Junius can solve such problems. The fact that Junius refers to his manuscript in several occasions as codex vetustus cannot be considered proof that he had the original manuscript, or an old copy of it: as is well known, humanists tend to be extremely imprecise with chronological notations in reference to manuscripts and vetus or even vetustissimus may very well apply to manuscripts dating back less than a century, or even simply be used for a handwritten text in contrast to the printed version;40 in addition to this and most importantly, humanists were generally keen to emphasize the possession of old testimonia, which would increase the worth of their own annotations, giving greater substance to them. It is therefore not at all strange that Junius avoids saying that his manuscript was in fact a copy and not the original version. What remains to be explained are the two marginal notes correcting the Greek verses in the exchange between Claudius and Hercules at § 5.4, in f. 95r of the codex. In this case there is only room for speculation. The corrections may be the result of a later revision of the codex, made after the 1557 edition was published, so that the reviser may have thought to introduce as marginalia the two passages which had already been corrected at his time: codex V shows two other revising hands (although dated not long after its first composition),41 and a subsequent revision would not have been unusual.42 It is of course also possible that another humanist, reading the Apocolocyntosis in codex V, independently came to the same solution, as often happens: Curione in the 1557 edition claims to have reached the same conclusion as Junius concerning these verses because of the  39 Alongside the claim to the discovery of the title there are several other corrections to the text that Curione declares to have made before his colleague, particularly regarding the Greek passages: Curione states that he was using a similar manuscript to Junius’, which still had traces of the Greek and therefore allowed him to come to the same conclusions as Junius did in several cases (732). 40 Cf. Rizzo (1973). 41 Cf. Roncali (1990) viii. 42 The fifteenth–century ex libris may as well be regarded as sign of a later ‘revision’ of the monastic possessions.

Apocolocyntosis, codex V and the manuscript of Hadrianus Junius  291

traces of Greek he saw in his own codex.43 This possibility raises the question of whether or not the marginal annotations in V predate the copy made for Junius: if they were already present in the manuscript before Junius’ copy was made, they would have been included in the copy and this would also mean that Junius took credit for emendations he did not make himself. Junius gives us no clue as to what was written in his codex in these passages because in the Annotationes he merely states ‘extat hic versus Odysseae nono᾿ and then ‘hic versus praecedentem ordine sequitur’ (followed by Latin translations and explanations of the passages)44; from what we have seen in the case of the other Greek readings reported in the Annotationes, even if the scribe making Junius’ copy found the emendations in the margins of the manuscript, he would not have been able to reproduce a good version of them. That another humanist before Junius could have so ingeniously corrected the Greek, without publishing or giving some kind of account of his conjecture anywhere, nor seeking any kind of recognition, seems a less likely scenario than that of a later hand, acting after such corrections were already made public.

 43 Curione (1557) 733: ‘hunc et sequentem Graecum versum ex manuscripto codice reposuimus’. 44 Curione (1557) 738.

Cristiana Roffi

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age: the Example of Germany The aim of my paper is to investigate the rescript of the episode of Narcissus and Echo written by Ovid (Ov. Met. 3.339–510) in the work published in 1545 by Jörg Wickram; this text is a remake of the corrupted work made by Albrecht von Halberstadt, dating back to a period between the late twelfth and early 13th century. The selection of this topic was particularly innovative because, although the reception of Ovid’s Metamorphoses in different areas of Europe (for example, in France and Italy) was the main subject of numerous studies, in Germany this author has been little examined. The texts written by Albrecht and Wickram themselves were investigated with particular accuracy, in order to discredit the idea of a period of German artistic decay. The first contributions to the interpretation of von Halberstadt’s poetics are to be found in Haupt, who in 1843 published the essay Die vorrede Albrecht von Halberstadt with no commentary; in 1851 J. Grimm edited three works about the German author. It was not possible to re–edit the text of von Halberstadt until 1855; during that year, in the ducal archive of Oldenburg two fragments, which dated back to the late 13th century, were found. Then, in 1966 something unexpected happened: during a review of the documents from the catalogue of Oldenburg, another fragment was surprisingly discovered by Last (C). The opinion already supported by Leverkus, who confirmed the hypothesis of the Metamorphoses’ transmission through a medieval manuscript dismembered in Oldenburg at the end of ‘600, was therefore confirmed by the discovery of fragments C, D and E. The influence of Ovid on German literature, which in the 9th century achieved a vital impetus thanks to Karl Bestrebungen, a Kaiser from Saxony interested in classics, did not leave special marks; in the many works realized by copyists we do not find any allusion to this author, while we are aware that the Aeneid and the Ecloge written by Virgil were appreciated and read with extraordinary attention. In Germany, the manuscripts related to Ovid’s tradition began to be mentioned and listed in the catalogues of libraries from the 9th century (for example, copies of the Ars and Metamorphosis were found in the monastery of Reichenau, the Heroides were discovered in Murbach, and an extensive index of works was found in the library of Blaubeuren). Only during the 16th century special attention

https://doi.org/10.1515/9783110632590-018

294  Cristiana Roffi was paid to Ovid’s Metamorphoses: Johannes Boemus composed a text entitled Metamorphosin Ovidii Teutonicam Facio, but it was never printed. However, thanks to the Ottonian dynasty, something changed: in the 10th century many classical texts began to be studied with more dedication than in other centuries, even though the lack of catalogues (not properly transcribed) does not help to shed light on the contentious status quaestionis. Between the late tenth and early 11th century, the Metamorphosis achieved a widespread transmission and, until the end of Middle Ages, Ovid was a distinguished author. In southern Germany and France, for example, the commentary written by the Bavarian Benedikt Beuren, dating back to the 11th century, is a clear sign of the importance of Ovid: Manegold Lautenbach (Manegaldus), an Alsatian scholar, who in 1086 taught in Bavaria, lectured his students about the Latin writer. In addition, we can observe that in Germany the reception area of Ovid’s works has been influenced by the near Provence: Heinrich von Veldeke’s Eneit, for example, was inspired by the Roman d’Eneas. It is highly probable that Albrecht von Halberstadt used Chrétien de Troyes’s works as an indirect source for the episode of Philomela, whom both poets wrongly called Philomena, highlighting the use of the French model. At the end of the 11th century, then, Ovid was officially included in scholastic anthologies. In the 12th century, the author is also mentioned in the works of Wernher von Elmendorf, a composer of didactic poems whose purpose was to teach the nobility the proper manners to be used in society, and Wolfram von Eschenbach, author of some lyrical and epic poems as the well–known romance Parzival. During the 13th century Ovid also began to be translated into vernacular language. During the Middle Ages, as claimed by Lehmann, the author was ‘der führende Verherrlicher des Weltgenusses, der Dichter irdischen fleischlichen der Liebe, der Erotiker’.1 Soon, the poet became a milestone for the courtly poetry, promoting the best conditions for the birth of a new vision of love. It is known that Ovid influenced the medieval theory about love, also in Germany. The theme of minne, which we find in the work of von Halberstadt, has many elements to be dated back to Ovid, including the symbology, the effects and the manifestations of love, perceived as a state that cannot be overlooked. For example, Albrecht von Halberstadt used both the terms minne and Liebe to indicate a passion with tragic effects, not considering the term minne as a service to the Lady. Even the poet Konrad von Würtzburg, whose production enhances the ideals of chivalry, judged Ovid as the proto–theorist of courtly love. In his Trojaner Krieg, he made  1 Lehmann (1927) 8.

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age  295

a list of five effective ways to seduce a woman: visus (‘Anblick’), colloquium (‘Gespräch’), tactus (‘Berührung’), oscula (‘Küssen’), coitus (‘sexuelle Vereinigung’). A few words on von Halberstadt and his family: he belonged to the dynasty of Thuringia, which was founded by Ludwig the Bearded (who died in 1055); in 1129 his son Ludwig II took the power, then married Jutta, the sister of Barbarossa, in 1150. This marriage gave him five children: Jutta, Ludwig, Friedrich, Heinrich Raspe and Hermann. Ludwig II was an austere father, but a strong admirer of arts and also a skilled strategist: during his first years of ruling, he built an enduring relationship with Ludwig VII, King of France and adversary of Barbarossa. The first phase of the literary production at the Thuringian court began in 1190, when Heinrich von Veldeke, Herbort von Fritzlar and Albrecht von Halberstad translated ancient texts into vernacular, thus propagated them. During the second period of the poetic production, the poets Walther von der Vogelweide and Wolfram von Eschenbach were popular and appreciated. The Metamorphoses written by Albrecht von Halberstadt are addressed to Albrecht’s promoter and protector, Hermann von Thuringia, who ruled from 1190 to 1217; his Pfalzgrafschaft took place in December 1190, when his brother died and left the power to him. For a more exhaustive dating, we cannot forget the various references contained in the Eneit by Heinrich von Veldeke and the Troja Roman by Herbort von Fritzlar, both written and commissioned under the protection of the Landgrave. As De Boor2 claimed, the ‘Thuringian court became the main centre for the development of literary interests: es ist weit einleuchtender, dass Landgraf Hermann zu dem Auftrag solcher Vorgeschichte im Zusammenhang bald mit und nach der Vollendung der Eneit gegeben hat. Um 1210 gehört Thüringen der Hof zu den anerkannten Zentren höfischen Lebens und Interessen Literarischer’. Of the work of von Halberstadt remained only five fragments, found in the so–called manuscript of Oldenburg, dismembered at the end of ‘600 and presumably dating back to a period between the late 12th or early 13th century. Here they are: – Fragment A, illustrating the description of the episode of Tereus and Philomela (Ov. Met. 6.440–79/ALBR. 5.1–144); – Fragment B, depicting the episodes of Peleus and Thetis (Ov. Met. 9.221– 65/ALBR. 5.126–227), Midas and Laodomedontis (Ov. Met. 11.194–220/ALBR. 5.72–125);

 2 De Boor (1953) 47–52.

296  Cristiana Roffi –

– –

Fragment C, which contains the narration of the events of Iphis and Anaxarete and the consequent protagonist’s transformation into a stone (Ov. Met. 14.698–758/ALBR. 5.25–144); Fragment D, which omits lines 708/27 and 772/95 of the eleventh book of the Metamorphoses; Fragment E, which includes vv. 245/82 and 350/69, corresponding to the thirteenth book of the Metamorphoses.

The language used by Albrecht von Halberstadt is the dialect of Thuringia, the German region where Albrecht lived after his birth in the city of Halberstadt, in Saxony. The Landgrave Hermann of Thuringia urged him to accomplish the work; his text, which belongs to the first period of German literature, differs from other classical rewritings, because it is faithful to the original content of the Latin text. Albrecht von Halberstadt’s work, despite its innovation, did not receive the appreciation of his contemporaries; the failure might have been caused by the many mythological and pagan elements included in the text, and by the difficulty to translate such a noble and dignified Latin into an average register that could be understood by both the public and the royal entourage. Because of this, the author was not mentioned in the chronicles written by his contemporaries; on the other hand, both Gottfried von Straßburg and Heinrich von Veldeke were quoted in the official registers belonging to the entourage of Hermann of Thuringia, whose aim was to transmit the knowledge of Roman history, since the Roman empire was perceived as the natural predecessor of the German one. In addition, the insertion of several Ovid’s episodes in Heinrich von Veldeke’s Eneit, including Dido’s monologue during her sleepless night, which has many correspondences with the one of Byblis, and the dialogue between the queen of Carthage and her sister, which refers to the conversation between the nurse and Myrrha, confirms the interest of the Landgrave in the Latin author and his texts, contextualized in the Medieval era. For example, the description of Aeneas and Lavinia’s marriage is connected to the theme of the minne and to the Herrschaft of Aeneas and, symbolically, to that of Hermann of Thuringia. A particularly interesting episode for medieval exegetes was the one of Narcissus. In the 14th century, Robert Holkot wrote a commentary entitled Moralia super Ovidii Metamorphoses, but the best–known work of this period was undoubtedly Thomas Walensis’ Moralitates, contained in two manuscripts by the 14th century found in Troyes.3 The first allegorical interpretations of the famous story dates back to the 12th century, thanks to the Policraticus, written by John of  3 The rewritings of Piramo and Tisbe’s episode achieved a widespread diffusion.

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age  297

Salisbury, in which Narcissus is perceived as a symbol of vanitas. In France, Arnolphe d’Orleans interpreted this figure as an example of the serious consequences of not loving in a proper manner. The ‘Vulgate Commentary’ was composed in 1250 in France, probably at Orléans. And it was in France that the story of Narcissus achieved great popularity thanks to the versions of Bertran de Paris, Bernart Ventador and Guillaume de Lorris. Whereas in Germany, the well–known version of Heinrich von Morungen (13th century) combined the Minnesang’s theme with Narcissus’ character:4 the young man looks at his reflection, source of joy and life, and falls desperately in love with it5 (Lied 32, Strophe 3). For what concerns Wickram’s work, Heinzmann6 asserts that Wickram added many supplements to Albrecht’s work, based on a good knowledge of the literature of that period. Actually, his rewriting is full of quotations from other works, all revisions of the classics: the translation of Heinrich Stainhöwel De claris mulieribus of Boccaccio; the version written by Hieronymus Ziegler, inspired by De casibus virorum illustrium of Boccaccio; the Odyssey of Simon Schaidenreissers; the Aeneade of Thomas Murners and the Narrenschiff of Sebastian Brants.7 As for Wickram’s work, five editions were printed: – the first one, dating back to the 1545, was printed by Ivo Schöffer, who was active from 1531 to 1555 and followed scrupulously the orders of Charles V. He also printed the Metamorphoses of Frontinus (1532–1537), works of Caesar (1532), Tacitus (1533), and Celsus (1532, 1537); – the second one, dating back to 1551, was printed again by Ivo Schöffer in Mainz;

 4 See Bartsch (1861) lvii. 5 Jâ enwil ich niemer des eralten, / swenne ich si sihe, mir sî von herzen wol. / Sît si herzeliebe heizent minne, / so enweiz ich, wie diu liebe heizen sol (Lied 11, Strophe 2/3). The most detailed definition of the terms is that by Walther von der Vogelweide (Lied 27): Minne ist minne, tuot sie wol; tuot sie wê, sô heizet sie niht rehte minne [...] Minne ist zweier herzen wuenne, / teilent sie gelîche, sost diu minne dâ. 6 Heinzmann (1969) 1ff. 7 See Rückert (1997) 21. Heinzmann (1969) 17 claimed that Wickram succeeded in breathing new life into the 16th century: ‘Wickrams Erweiterungen geben über das rein Stoffliche hinaus einen Einblick in das Denken und Reflektieren des 16. Jahrhunderts. Es hat sich gezeigt, daß Wickram an vielen Stellen den Geist des 16. Jahrhunderts klar zum Ausdruck bringt, sowohl in der Modernisierung sozialer Beziehungen zu Beginn der Zeit des Überseehandels als auch in der rationalistisch–didaktischen Deuteungsweise seiner Zeit’. Regarding Boccaccio’s texts, cf. Heinzmann (1969) 38–47. The prologue itself, addressed to ‘an den freundlichen Leser’ has a connection with the De claris mulieribus.

298  Cristiana Roffi –

– – –

the third one, dating back to 1581; in this version the dedication of the author in which Wickram elaborates on his poetic intentions is omitted. However, there is also a long dedication (of nine pages length) written by the publisher; Sigmund Feyerabend to Nicolaus Reussner Lauwingen, rector of the University, followed by a of Ovid’s biography and a brief preface; the fourth one, dating back to 1609, printed in Frankfurt by Johann Saurn; the last one, dating back to 1631 and printed in Frankfurt by Gottfried Tampach.

In addition to these five copies, in literature there is evidence of other three editions, whose authenticity is questioned by scholars; they were printed in Frankfurt respectively in the years 1551, 1625, and 1641. The last two are missing in any catalogue and in the official register of editions made in Frankfurt; the edition of 1551 is probably the Mainz version dating back to the same year. The title stresses that the work has been copied by that of Albrecht von Halberstadt and through the work of Wickram gebessert und mit der Figuren Fabeln gezirt (3.9). The intent of the work is to modernize the text of Albrecht von Halberstadt, in order for it to be easily understood by both the young and the old (10.10–14). After the publication of Meyster Albrecht’s prologue, Wickram apologizes to the reader if the verses seem difficult to understand (10.4–7), and because of this, he says he will try to make them more comprehensible to the public (10.10–11). As the author himself declares, his intention is to mock the pagan ceremonies and the imaginary world created by Ovid: ire heydnische Cerimonien verlachett verspottet (11.13). In addition, thanks to both a philological and a cultural–historical research, the study described above highlighted that these German translations tried to adapt the original texts to the public to whom they were presented and read. Referring to von Halberstadt, the aim of his work was to acclaim the majesty and superiority of the king and his royal family. For what concerns Wickram, the main purpose of the author was to modernize the text of Albrecht von Halberstadt, in order to simplify the understanding of the original text. In conclusion, following the categorizations of the poetic text identified by Holmes (The versification: the forms of translation and the translation of the forms, in Contemporary Translation Theories, edited by S. Nergaard, Milan 1995), both Albrecht von Halberstad and Wickram could be classified as ‘analogic translators’, because they adapted the original text to their own poetic tradition. Furthermore, the accurate investigation of the German texts compared to the Latin one stressed the following differences: – type of metra;

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age  299

– –

exegetical inserts and omissions; tendency to moralization.

As regards the metra, Rücker stressed that the transposition from Latin hexameters to octosyllabic verses implies that Wickram’s text surpassed the original work, because the vulgarizer deletes some episodes or expands on others that he finds interesting. In Narcissus and Echo’s episode, for example, 406 octosyllabic verses correspond to 171 hexameters. Comparing the fragments of von Halberstadt and Wickram’s text and looking at the respective number of verses, Bolte8 already highlighted that the 144 verses of Fragment A amount to 144 in Wickram’s text, while the 279 verses of the Fragment B amount to 265 in Wickram’s version. According to Runge,9 Wickram’s aim is to lengthen the verses (‘die verse zu strecken’); short terms are replaced with longer phrases, articles are changed into possessive pronouns. The complete list of adaptations made to the original text was conducted by Heinzmann,10 who paid special attention to genealogical contents and to misinterpretations originated from the complex language used by Albrecht. Concerning the exegetical insertions and omissions, complex surnames and epithets derived from geographical or divine names are often simplified by the German author, who probably did not understand the Latin model; on the other hand, we ought not to forget that Wickram worked for an audience lacking erudite background. At the beginning of the Metamorphosis, the Latin author describes the Milky Way and Jupiter’s palace, which Ovid compares to that of Augustus (Ov. Met. 1.168–76). The description is omitted in Wickram’s, as well as the depiction of Augustus’ palace; the German writer mentions only the existence of a himelische (n) Stross and the gods heading to this meeting (W. 1.321–6). However, the rerum gestarum accomplished by the emperor (Ov. Met. 1.199– 205/W. 1.385–94) and the history of Ariadne and Theseus (only mentioned in Ovid; Ov. Met. 8.169–173) are emphasized thanks to the inclusion of the Minotaur’s episode. Here is a more detailed list of the verses in which the adjectives are omitted: – adeunt pariter Cephisidas undas (1.369) is translated with the sentence Zu eynem wasser fluß sie giengen (1.677);

 8 Bolte (1906) xxv. 9 Runge (1908). Bartsch (1861) cxxxvi analyzes the use of the expletive elements, in order to mark the rythm of the verses: ‘oft sind Stellen von zwölf oder mehr Zeilen hinter einander bei Wickram lesbar, an denen wenig zu ändern ist als die eingeschobenen Flickwörter, mit denen Wickram die fehlenden Senkungen zu ergänzen suchte’. 10 See Heinzmann (1969) and Rücker (1997) 93–275.

300  Cristiana Roffi – – – – – – – –

Pylios […] in agros (2.684) with the equivalent uff grüner weyd (2.1451); Dictaeque rura (3.3) turns into uber Meer (3.1); Castalio […] antro (3.14) becomes Und den berg abhin kommen was (3.53); Martius anguis (3.32) transposes into Eyn grosser Serpent (3.82); Illirycos […] fines (4.568) interchanges in ferre landt (4.1071); Gnosiacae […] carinae (8.144) in seinem schiff (8.266); Scythica […] sagitta (10.588) translated with pfeil (10.1086); Aegeo […] in aequore (11.663) with uff dem meer (11.1143).

On the other hand, omissions affect the following events: – the transformation of Meleager’s sisters (8.526–46). The metamorphosis into a bird of Meleager’s sister (8.537–41) is connected in Wickram’s version with the story of Philemon and Baucis (VIII.611–724); – Dionysus’ cycle and Pentheus’ death (511–733). Led by Agave, Pentheus’ mother, the Bacchae forced trapped Pentheus down from the treetop, ripped off his limbs, his head, and tore his body into pieces. In addition, Ovid denotes the association between the image and its source, points to the central component of Narcissus’ myth, the relationship and separation of the body and the reflected image, and thus establishes the irony of the situation. The way in which visual and auditory reflections both lead to the illusion of non–existent presences and sentences is a singular feature which is totally omitted in Wickram’s work. The expunctions involve the stories of Theseus and Acheoloo (8.547–73) and the transformation of the Naiades in Echinades (8.573–610); the omissions regarding the thirteenth and fourteenth book of Ovid’s work were probably missing in the original text of Albrecht von Halberstadt. The episodes describe the Trojan cycle and, in particular: – Aeneas’ departure from Circe (14.1–74); – the Cercopi (14.91–100); – Aeneas’ meeting with the Sibyl (14.101–51); – Caieta’s funeral (14.441–4); – the description of Aeneas’ ships (14.527–65); – Ardea (15.566–80). In the macro–structure of the work, the lack of part of the poem, the one often called ‘Little Aeneid’ (its reasoning is that this avian metamorphosis is structurally parallel to another avian metamorphosis, that of the Memnonides from the pyre of Memnon, the event which closes the ‘Little Iliad’ – Ov. Met. 13.600–20) is

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age  301

certainly explained due to the determination of the Landgrave to consolidate the development of an unpopular historical subject. In addition, the author, in order to make his work more comprehensible, tried to connect the Latin myths to German traditional legends, in which fantastic creatures are often involved. By doing so, the introduction of female characters, including Liriope and Eco, had the aim to replace the classical figure of the nymph with a waldtjunckfraw, a so–called young girl who lives in the forest and whose respective male figures can be found in those of satyrs and sileni, elderly rustic (daimones) in the train of the god. The terms nympha/nymphae are translated by Wickram with the German word wasser frawen11 (Ov. Met. 2.238/W. 2.497), Jungfrawen (Ov. Met. 2.445/W. 2.936). On the other hand, the relationship with the woods is stressed by the use of waltfrawen (Ov. Met. 3.314/ W. 3.766), Waldt junckfraw (Ov. Met. 3.357/W. 3.871), walt Gottin (Ov. Met. 3.456/ W. 3.1113), wasser Göttin (Ov. Met. 4.302/ W. 4.567). The Nereides, belonging to the aquatic nymphs, are described as Morwunder (Ov. Met. 1.302/W. 1.569) or Meer Frawen (Ov. Met. 13.899/ W. 13.1223). What is more, both Ovid and Wickram applied a very different range of adjectives to emphasize the ephebic beauty of Narcissus, including the mixture of red and white, generally used for descriptions of women. The characterization of Narcissus begins with his portrayal as the archetypal elegiac figure, proud and unattainable. The seduction of the young is deliberately ambiguous: Narcissus is an adolescent whose body has no definite shape that could fully determine the genre. The candor used by Ovid to describe the beauty associated with virginity re–appears in the episode of Philomela written by Wickram: haub weiß als der schne (W. II.909).12 As evident in the examples given, Wickram’s application of the adjectives employed by Ovid in descriptions regarding the beauty of various characters is remarkable.13 Narcissus’ attractiveness is accentuated by Wickram in vv. 1013–5 of the first book: zweimolen schoener wardt / Als ire liechten wengling zart / Mit rosenfarb vermischet worden. The white arm (weissen arm, v. 1061) is also used to describe the paleness of the young boy: his face is depicted as ‘the colour of milk’ (angsichts milchfar, v. 1029); the beauty of the mouth is emphasized by the German adjective rothen. On the contrary, the adjective red is recurrent to describe Narcissus’ mouth, comparing it to a ruby (W. 1.1498–1491).  11 Regarding the word Wasser Göttin and its connection with nympha see Ernout–Meillet (20014) 374, 453. 12 What is more, there are a lot of comparisons between stars and the moon: An schön den andren vor thet gohn / Gleich wie der Mon dem morgenstern / Und wie die Sonn dem Mon so fern / vorgehnt / so weit ir schöne gar / Fürtraff der andren Jungfrawen schar (W. 2.1536–1540). 13 Rücker (1997) 240: ‘mit seiner Schilderung der schönen Dame folgt er also nicht einer Mode seiner Zeit, sondern es handelt sich um eine spielerische Einlage’.

302  Cristiana Roffi Is it possible to analyze the use of both colours in two other episodes of the Metamorphoses: the bath of Diana (W. 3.430–3) and the representation of Circe (W. 14.421–3): Tab. 1: Use of red and white colours in Ov. Met. 3.430 –3 and 14.421 –3W. .–

Ov. Met. -–

W. .–

Ov. Met. .ff.

Eyn wunderlich farb so schon / Gleich die morgen röt auff thut ghon / Weiß und roth durcheynander zogen / Verfloesset wie der Regenbogen.

Qui color infectis adversi solis ab ictu / nubibus esse solet aut purpureae Aurorae / is fuit in vultu visae sine veste Dianae.

Bei deinen augen man ich dich / Welche leuchten eym Rubin glich / Darzu bei deiner jugent schon.

Per, o tua lumina, dixit / quae mea ceperunt, perque hanc, pulcherrime, formam.

The combination of white and red, mixed as they were part of a rainbow (verfloesswet wie der Regenbogen, W. 3.430–3) is missing in Ovid’s text, as well as the comparison with the sparkle of a ruby (W. 14.422).

According to Ehrismann,14 these adjectives, especially used to indicate the female beauty, demonstrate that the use of a repertoire of formulaic descriptions derived from medieval German stories.15 For instance, we find the same expressions in the Nibelungenlied, the Tristan of Gottfried von Straßburg and Wolfram von Eschenbach’s Parzifal, as well in the Eneit of Veldeke; contemporary authors of Albrecht von Halberstadt, whose poetry has certainly influenced the descriptions of Wickram’s Metamorphoses.16 By doing so, it is undeniable Wickram’s familiarity with such beauty attributes. Considering Narcissus’ description, in two other novels, Goldfaden and Der irr Reitende Pilger, we found the same golden

 14 Ehrismann (1927) 124. For example, see Nibelungenlied, v. 282.2: ir rocapsen varwe. 15 Köhn (1930) 20–23: ‘In der höfischen Dichtung, sowohl in der Lirik als in der Epik, wie auch in den Heldenepen sind zwei Bilder besonders häufig, der Vergleich mit der Sonne und der Rose. (...) Dieselben Bilder wiederholen sich bei fast allen höfischen Dichtern in allen Formen vom ausgeführten Vergleich mit Sonne, Mond und Sternen, besonders dem Morgenstern’. 16 In Tristan the main character is Isolde. Her mother is described as a daz vroliche morgenrot (10886), Isolde is ir sunnen (10887). Her hair is blonde: so gelich und alse einbaere / was ir han dem golde (1098), the eyes shine as a hawk (1099) and are smooth like a mirror (spiegelglas, 1100). What is more, in Parzifl (3.30.2–25), a different range of adjectives is applied, and it includes the mixture of red and white: munt durchliuhtec rôt (5), von snêwîzem beine (11) and blanke hant (25). See Köhn (1930) 2–105; Ennen (1985) e Rinn (1996).

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age  303

hair, red mouth, white teeth and the association with Apelles17 included in Wickram’s Metamorphoses. In this way, Narcissus’s story validates the employment of descriptions already extant in Albrecht von Halberstadt’s work.18 Another aspect one should take into account is the period in which both Wickram and Albrecht set their work. The tale exposed by Albrecht ends with Augustus: an das zil / das zu Augustus kamm (v. 68). In addition, the prologue followed a Christian perspective, establishing a ‘Heilsgeschichte’: Do wolt geboren werden / erscheinen und der erden / Christus unser heilant / Von Gott dem vatter gesant / Von eyner junkfrawen geboren / Darzu sunderlich ausserkoren (vv. 75 –80). The advent of Christianity, evaluated as a period of peace, is also present in the last book of Wickram: Als nun die Zeit was so fridsam / Unser Troster uff Erden kam / Von eyner Junkfraw reyn und zart / Das Wort ewig geboren Wardt / So erlöst hat uns von dem Todt (15.521–5). The mention of the same themes included in the prologue clarify that Wickram’s aim was to create a ‘Ringkomposition’:19 the author prefigured the salvation of individuals made possible by the birth of Christ.

 17 Goldfaden 5.80.5–7: Ihr gantzer leib hett von Apelle / nit zierlicher gemalt werden / moegen […]; Der irr reitende Pilger: 908–909: Ja wann Apelles diss alssammen / Gmalt het e 4094–4096: Diss alles stuend gemalet so schon / Als wans Apelles selb hett gton / Der aller maler maister was. In the novel Knabenspiel, the main character Concordia is described with a red mouth (lustige(n) rubin farb, 3.63.13) and pink cheeks (ire wanglein schon mit kleynen grublein bekleydet / lieblich rosiniert; 3.63.11). In addition, the girl is a schönen zuchtigen junckfrawen (3.65.14); the use of the term zuht has a symbolic value in the courtly vocabulary, see Ehrismann (1927) 248–53. As regards the connection between beauty and kindness, both belonging to the main characters of epic clashes as Tristan and Parzifal, see Ehrismann (1927) 189, under the term schöne unde guot. In the novel Goldfaden we found similar deeds: Lewfrid, the main character, falls in love with the king’s daughter Angliana, described as below (5.79.31–33/5.79.33–80): Die wenglin mit rosenfarb geziert / das mündlin einem Rubin gleich an der farb. 18 Rücker (1997) 245–6: ‘in den Roman Von guten und bösen Nachbarn finden sie jedoch auffallend viele Stellen, an denen in sehr ausfrührlicher Weise Elemente aus der antiken Mythologie eingearbeitet sind. […] Die Beispiele zeigen, daß Wickram an einigen Stellen in seine Romane aus Ovid geschöpft hat. Daher kann man folgern, daß er auch die höfischen Schönheitsschilderungen in Albrechts Ovid Verdeutschung schon so vorgefunden und Ähnliches an entsprechender Stelle in seine Romane eingebaut hat, ebenso wie er zum Beispiel höfische Elemente aus übersetzten Romanen wie dem Pontus übernommen hat’. The theory that Wickram used Petrach’s texts for his descriptions is not plausible, considering his elementary knowledge of Latin. 19 Bräuer (1990) 161: ‘mit diesem Ausdruck der Friedenssehnsucht ist ein Gedankenbogen um das Werk gespannt, der inmitten der waffenklirrenden Literatur des Zeitraums eine ebensolche Ausnahmestellung einnimmt wie die unmittelbare Übertragung aus dem Leteinischen’.

304  Cristiana Roffi Analyzing the rhetorical figures in the Latin text, including the polyptoton with variation of diathesis and anaphora, we could notice that in the German language the Latin lexica are expressed through the repetition of certain terms and have taken advantage of the necessity of the German language to express the subject. The same requirement is used to increase the frequency of the personal pronoun of the first and second singular person. In addition, the author emphasizes the peculiarities of the language thanks to a phonic perseverance that emphasizes the confusion; the requirement of an expressed subject, essential for the German language, increases the frequency of first and second person singular pronouns, repeated 42 times, and of the possessive one, which occurs 8 times. In addition, throughout the entire episode Wickram does not succeed in translating the correspondence between the terms corpus/vox. In the book 3 of the Metamorphoses, we found the expressions related to vox six times: 3.339; 3.356; 3.362; 3.381; 3.385; 3.494. Wickram barely translated these lexes: the adjective vocalis of verse 356 is omitted, as well as the translation of the Latin terms imago vocis (v. 385), a clause that in Ovid connects two episodes since it recalled the iunctura imago formae (v. 416), through which the Latin author introduces the reflection of Narcissus. Echo prefigures the intertextual design of Narcissus’ episode; her ardent verbal exchange with her beloved covers the entire gamut of the intertextual phenomenon, ranging from an exact reproduction of the original meaning, a clear parallelism in signification, to its radical inversion into the total opposite. Keeping in mind the nuances of sonus and verba, we can see that there is a subtle difference in the depiction of the original utterance and Echo’s response to it. She waits for ‘sounds’ to which she sends back ‘words’. The soni are the raw materials that Echo shakes into sua verba. For example, v. 501 offers an onomatopoeic depiction of an echo, with the correction of the vowel in hiatus giving a fading effect to the second vale. Regarding the tendency to moralization, the author often omits the use of expressions that are clearly bound to the erotic sphere and does not give space to the physical dimension of love. In addition, he does not remark the ephebic nature of Narcissus’ sexuality, which attracts both males and females. What is more, in Ovid’s version, the author introduces a pederotic vision of love, through the mention of the multi illum iuvenes [...] cupiere null illum iuvenes [...] tetigere (vv. 353–5), and the hymn of praise to the gods of vengeance made by another rejected young man: Wickram, who lived in a culture in which homosexual relationships were not allowed by the ethics of the time, could not accept these references and, therefore, preferred ignoring all the allusions to a relationship between two young people of the same sex. Even the gesture, which plays a fundamental part in the episode of Narcissus and Echo, is ignored by the German text: the hands

The Textual Transmission of Ovid’s Metamorphoses during the Medieval Age  305

and the pale chest are completely omitted. The explicit references to the neck and arms of the young man (vv. 389–90) of the Latin edition, are absent not only in Wickram’s text, but also in similar descriptions that can be found in the version of Albrecht von Halberstadt; in this way, the translators deliberately avoid mentioning body parts connected with sexuality. On the contrary, Wickram emphasizes the impossibility of the love between Narcissus and his reflection, declaring that no one should love the way he did. The romantic love described in medieval epic texts yields to the virtual absence of eroticism; morality and reason are the two most important factors in assuring bourgeois security within society; these attitudes translate into the typical virtues of his characters: ‘Wohlverhalten, Gesittetheit und Unauffälligkeit’. The remark made by the German author advises his audience against a passion which does not bring fruits; what is more, Narcissus refuses to start any social relationship. In conclusion, Wickram’s Metamorphoses represented a rescript of Narcissus and Echo’s episode, rather than a property translation; Wickram paid little attention to the style of the original text, focusing mainly on the transmission of the structure of the episode, rather than trying to reproduce the Latin wordplays. The comparison between the work of Wickram and Albrecht von Halberstadt showed that Wickram was not skilled in the Latin language; he admitted his lack of classical background in Ov. Met. 5.21–24.20 The conjecture that, at least scholarly, he could read Latin is only presumed from the prologue of the Sieben Hauptlastern, in which Wickram pretends to have studied it:21 So hab ich auch wenig Latein / Gstudiert drum ich inn meinem Leben. To elucidate the question, Heinzmann22 suggested a derivation from German sources, rather than from Latin texts; Wickram used German sources in conjunction with Albrecht von Halberstadt’s original text, instead of directly consulting ancient sources. However, it is still plausible that Wickram had an elementary knowledge of Latin, learned during his youth.23 In addition, Wickram’s ignorance of Latin and his provincial milieu

 20 As highlighted by Bartsch (1851) cxxxi–iii: ‘es war des Lateins nicht kundig. Die Versicherung aus Wickrams eigenem Munde, dass er kein Latein verstehe ist für uns von Wichtigkeit, weil wir dadurch sein Verhältniss zu Albrecht und auch das Verhältniss Albrechts zu Ovid genauer bestimmen können’. 21 See also Kleinschmidt (1982) 259. 22 Heinzmann (1969) 20. 23 See Cannon–Geary (1980) 145 and Wagähll (1992) 90–99: ‘neben den Alten Latein – und Pfarrschulen hatten sich im Mittelalter in den Städten Schreibschulen entwickelt, die Lesen, Schreiben und Rechnen in der Volkssprache lehrten und sich durch höhere Praxisorienthiertheit gegenüber den Lateinschulen auszeichneten. Wir wissen nicht, welche Schulen Wickram besuchte, aber es ist anzunehmen, daß er eine Schule besuchte, die den oben beschreiben glich,

306  Cristiana Roffi turned out ironically to have favoured his development of independent themes and his excessive self–confidence in the handling of those. Nowhere was the gap between vernacular practice and Neo–Latin theory more significant than in the realm of prose fiction. To conclude, the differences introduced by the German author could have been produced by a lack of direct examination of ancient sources but, as significant studies suggested (above all, see Rücker [1997]), the comprehension of the Latin text would have bene possible for Wickram thanks to the presence of a glossed manuscript and to the help of a school teacher who probably simplified the translation of Ovid’s Metamorphoses.24 The existence of this co–worker is not entirely improbable: Bolte25 already asserted the providential help of ‘eines befreundeten Schulmeisters’.

 denn wenig überzeugend scheint die Annahme, er sei Autodidakt gewesen. Die Stellung des Vaters hatte warscheinlich die Erziehung des Sohnes ermöglicht. Wohlabende Bürgerfamilien schickten ihre Kinder zum Privatunterricht, aber auch hier wurde eine praktisch ausgerichtete Ausbildung der von den Pfarr oder Lateinschulen angebotenen theoretischen vorgezogen’. 24 Runge (1908) 13: ‘es sollte danach – auch in Anbetracht der vielen groben Fehler, die der vermeintliche Ratgeber stehen lässt doch erwogen werden, ob nicht Wickram selbst den lateinischen Text einsah, ohne dass eine Mittelsperson ihm half. Sicher entscheiden lässt sich die Sache nicht, zumal da auch die Möglichkeit besteht, dass die von Wickram benutzte Handschrift des Albrechtschen Textes von den erhalten Bruchstücken abwich oder lateinische Glossen enthielt. Diese Berater hätte dann mindestens im 6. 8. 11. 13. Buche, also innerhalb einer grössen Partie, seine Hand bei Einzelheiten und Kleinigkeiten im Spiele gehabt’. 25 Bolte (1906) xxvii.

Bibliography Abbreviations AE BNP CCCA 3 CCGS CCSL CEG CIL CLE Cod. Iust. CSEL CTh Davies APF Dessau ILS DNP EDB EDCS EDR EM Athens HGP ICUR ICURns

IG ILLRP Louvre, Ma LTUR MAN Naples MAN Parma MGH MNR Terme

(1888–) L’Année épigraphique, Paris. Cancik., H., Schneider, H., Landfester, M. et al. (eds.) (2002–2010) Brill’s New Pauly. Encyclopaedia of the Ancient World, Leiden–Boston. Vermaseren, M.J. (1977) Corpus Cultus Cybelae Attidisque (CCCA), III, Italia– Latium, Leiden. (1977) Corpus Christianorum Series Graeca, Turnhout. (1953–) Corpus Christianorum Series Latina, Turnhout. Hansen, P.A. (1983) Carmina Epigraphica Graeca saeculorum VIII–V a.Chr.n., Berlin–New York. Mommsen, T. (1863–) Corpus Inscriptionum Latinarum, Berlin. Buecheler, F. (1895–1897) Carmina Latina Epigraphica, 2 vols., Leipzig. Krueger, P. (1877) Corpus Iuris Civilis, Codex Iustinianus, II, editio stereotypa, Berlin. (1866–) Corpus Scriptorum Ecclesiasticorum Latinorum, Vienna. Krueger, P., Mommsen, T. (1905) Theodosiani Libri XVI cum Constitutionibus Sirmondianis, I, pars posterior, textus cum apparatu, Berlin. Davies, J.K. (1971) Athenian Propertied Families, 600–300 B.C., Oxford. Dessau, H. (1892–1916) Inscriptiones latinae selectae, Berlin. Cancik, H., Schneider, H. (eds.) Der Neue Pauly. Enzyklopädie der Antike, Stuttgart–Weimar, 21 vols., 1996–2003 Epigraphic Database Bari. Epigraphischen Datenbank Clauss–Slaby. Epigraphic Database Roma. Epigraphical Museum, Athens. Historical Greek Pronuntiation in Caragounis, C.C. (2004). De Rossi, G.B. (1857–) Inscriptiones Christianae urbis Romae septimo saeculo antiquiories, Rome. (1922–1992) Inscriptiones christianae urbis Romae septimo saeculo antiquiores, nova series, I–X, Pontificio Istituto di Archeologia Cristiana, Vatican City. (1973–) Inscriptiones Graecae, consilio et auctoritate Academiae Litterarum Regiae Borussicae editae, Berlin. Degrassi, A. (1957–63) Inscriptiones Latinae liberae rei publicae, Florence. Musée du Louvre, Antiquités grecques et romaines. Steinby, E.M. (1993–2000) Lexicon Topographicum Urbis Romae, 6 vols., Rome. Museo Archeologico Nazionale di Napoli. Museo Archeologico Nazionale di Parma. Pertz, G.H. et al. (1826–) Monumenta Germaniae Historica inde ab a. c. 500 usque ad a. 1500, Hanover–Berlin etc. Museo Nazionale Romano, Terme di Diocleziano.

https://doi.org/10.1515/9783110632590-019

308  Bibliography Mus. Cap. NCE Nuovo catalogo epigrafico, schedario ms. presso la sezione epigrafica dei Musei Capitolini. NA28 Nestle, E. and E., Aland, B. and K., Karavidopoulos, J., Martini, C.M., Metzger B.M. (201228) Novum Testamentum Graece, 28th Revised Edition Edited by the Institute for New Testament Textual Research Münster/Westphalia under the direction of Strutwolf, H., Stuttgart. OCT Oxford Classical texts OLD Glare, P.G.W. (ed.) (20122) Oxford Latin Dictionary, Oxford. PCBE IV Pietri, L., Heiljmans, M. (2013) Prosopographie Chrétienne du Bas–Empire 4: La Gaule Chrétienne (314–614), I–II, Paris. PL Migne, J.P. (1844–1855) Patrologiae Cursus Completus, Series Latina, Paris. PLRE II Jones, A.H.M., Martindale, J.R., Morris, J. (1980) The Prosopography of the Later Roman Empire, II, AD 395–527, Cambridge. RE Wissowa, G.G., Kroll, W., Witte, K., Ziegler, K. (eds.) (1839–1980) Paulys Realencyclopädie der Classischen Altertumswissenschaft, neue Bearbeitung unter Mitwirkung zahlreicher Fachgenossen, I–XXIV, I A–X A, Suppl. I–XV, Register, Stuttgart–München. RICIS 2 Bricault, L. (2005) Recueil des Inscriptions Concernents les Cultes Isiaques (RICIS), II, Paris. Tod 2 Tod, M.N. (1948) A Selection of Greek Historical Inscriptions, II, Oxford. TLL (1894–) Thesaurus Linguae Latinae, Munich. TR Textus Receptus Vat. Mus. Musei Vaticani, Vatican city. VL Sabatier, P. (1949–) Vetus Latina: die Reste der altlateinischen Bibel, neu gesammelt und hrsg. von der Beuron, E. (Freiburg).

Online Databases BerlPap DIGIVATLIB CPP EDB EDCS EDH EDR MQDQ LDAB LGGA

MP3 TLL

Berliner Papyrusdatenbank < https://berlpap.smb.museum/ > Digital Vatican Library < https://digi.vatlib.it/ > Catalogue of Paraliterary Papyri < http://cpp.arts.kuleuven.ac.be/searchform. html > Epigraphic Database Bari < http://www.edb.uniba.it/ > Epigraphische Datenbank Clauss–Slaby < http://www.manfredclauss.de/ > Epigraphic Database Heidelberg < http://edh–www.adw.uni–heidelberg.de/ > Epigraphic Database Roma < http://www.edr–edr.it/ > Musisque Deoque< http://www.mqdq.it/public/ > Leuven Database of Ancient Books < https://www.trismegistos.org/ldab/ > Lexicon of Greek Grammarians of Antiquity Mertens–Pack3, Catalogue des papyrus littéraires grecs et latins < http:// cipl93.philo.ulg.ac.be/Cedopal/MP3/dbsearch_en.aspx > Thesaurus Linguae Latinae online < http://www–1degruyter–1com–1thesauruslinguaelatinae.emedien3.sub.uni–hamburg.de/view/db/tll>

Bibliography  309

References Abert, H. (1924) ‘Lampros’, RE 12.1, 587. Achelis, H. (1936) Die Katakomben von Neapel, Leipzig. Acocella, C. (1942) ‘La Congregazione Verginiana (1119–1879)’, Il Sacro Speco di San Benedetto di Subiaco 48, 29–33. Acosta Hughes, B. (2010) Arion’s Lyre. Archaic Lyric into Hellenistic Poetry, Princeton. Adam, J.A. (19632) The Republic of Plato, II, Cambridge. Adams, J.N. (2013) Social Variation and the Latin Language, Cambridge. Adkin, N. (1992) ‘Hieronymus Ciceronianus: the “Catilinarians” in Jerome’, Latomus 51.2, 408– 20. Adkin, N. (1999) ‘Jerome’s Vow “Never to Reread the Classics”: Some Observations’, REA 101, 161–7. Agnello, S.L. (1969) ‘Pitture cimiterali inedite di Siracusa’, Akten des VII Internationalen Kongresses für christliche Archäeologie, Vatican City, 327–31. Ahrens, H.L. (1855) Bucolicorum Graecorum Teocriti, Bionis, Moschi reliquiae. Accedentibus incertorum idylliis, I, Leipzig. Ahrens, H.L. (18612) Bucolicorum Graecorum Theocriti, Bionis, Moschi reliquiae. Accedentibus incertorum idylliis, Leipzig. Aland, K. (1994) Kurzgefasste Liste der Griechiscen Handschriften des Neuen Testaments, Berlin. Aland, K., Aland, B. (19872) The Text of the New Testament, transl. Rhodes, E.F., Grand Rapids. Alessandri, S. (1982) ‛Alcune osservazioni sui segretari ateniesi del IV sec. a.C.’, ASNP 12, 7– 70. Alexander, J.J. (1992) Illuminators and Their Methods of Work, London. Allen, D. (2000) ‛Changing the Authoritative Voice: Lycurgus “Against Leocrates”’, ClAnt 19, 5– 33. Allen, T.W. (1931) Homeri Ilias, Oxford. Andenna, G. (1983) ‘Guglielmo da Vercelli e Montevergine: Note per l’interpretazione di una esperienza religiosa del XII secolo nell’Italia meridionale’, in Fonseca, C.D. (ed.) L’esperienza monastica benedettina in Puglia, I, Congedo, 87–118. Anderson, C.A. (1995) Athena’s Epithets. Their Structural Significance in the Plays of Aristophanes, Stuttgart–Leipzig. Anderson, G. (1993) The Second Sophistic: A Cultural Phenomenon in the Roman Empire, London–New York. Anderson, W.S. (1995) ‘Liber, Child and Freedman’s Free Son’, Arethusa 28.2, 151–64. Andorlini, I. (1992) ‘I papiri e la tradizione medievale nella ricettazione dei testi medici tardoantichi’, in Garzya, A. (ed.) Tradizione e ecdotica dei testi medici tardoantichi e bizantini. Atti del Convegno Internazionale (Anacapri 1990), Naples, 13–27. Andorlini, I. (1993) ‘L’apporto dei papiri alla conoscenza della scienza medica antica’, in Haase, W. (ed.) Aufstieg und Niedergang der römischen Welt (ANRW), 2.37.1, Berlin–New York, 458–562. Andorlini, I. (2003) ‘L’esegesi del libro tecnico: papiri di medicina con scolî e commenti’, in Papiri filosofici. Miscellanea di studi, IV, Florence, 9–29 [reprinted in Andorlini (2017)]. Andorlini, I. (ed.) (2004a) Testi Medici su papiro, Atti del Seminario di studio (Firenze 2002), Florence.

310  Bibliography Andorlini, I. (2004b) ‘Un ricettario da Tebtynis: parti inedite di PSI 1180’, in Andorlini, I. (ed.) (2004a) Testi Medici su papiro, Atti del Seminario di studio, Firenze, 3–4 giugno 2002, Florence, 81–118. Andorlini, I. (2006) ‘Il “gergo” grafico ed espressivo della ricettazione medica antica’, in Marcone, A. (ed.) (2006) 142–67 [reprinted in Andorlini, I. (2017)]. Andorlini, I. (2007) ‘Prescription and Practice in the Greek Medical Papyri from Egypt’, in Froschauer, H., Römer, C. (eds.) Zwischen Magie und Wissenschaft. Ärzte und Heilkunst in den Papyri aus Ägypten, Vienna, 23–33 [reprinted in Andorlini, I. (2017)]. Andorlini, I. (2017) πολλὰ ἰατρῶν ἐϲτι ϲυγγράμματα. Scritti sui papiri e la medicina antica, ed. Reggiani, N., Florence. Andorlini, I. (2018) ‘Tipologia testuale e linguaggio tecnico nelle ricette su papiri: tre casi di ricette conservati dai PSI’, in Reggiani, N. (ed.) (2018a), 113–27. Argentieri, L. (1998) ‘Epigramma e libro. Morfologia delle raccolte epigrammatiche premeleagree’, ZPE 121, 1–20. Argentieri, L. (2007) ‘Meleager and Philip as Epigram collectors’, in Bing, P., Bruss, S. (eds.) Brill's Companion to Hellenistic Epigram Down to Philip, Leiden, 147–64. Arrighetti, G. (1968) ‘Il POx XIII 1611. Alcuni problemi di erudizione antica’, SCO 17, 76–98. Arrighetti, G. (1987) Poeti, eruditi e biografi, Pisa. Astarita, L.M. (1993) La Cultura nelle «Noctes Atticae», Catania. Baesecke, G. (1908) ʻHerbort von Fritzlar, Albrecht von Halberstadt und Heinrich von Veldekeʼ, Zeitschrift für deutsches Altertum und Literatur 50, 366–82. Baesecke, G (1909) ʻDie Datierung Albrecht von Halberstadtʼ, Zeitschrift für deutsches Altertum und Literatur 51, 163–74. Bagordo, A. (1998) Die antiken Traktate über das Drama, Stuttgard–Leipzig. Bailey, S. (1965) Cicero’s Letters to Atticus, II, Cambridge. Bakola, E. (2010) Cratinus and the Art of Comedy, Oxford. Balashov, Y. (1994) ‘Should Plato’s Line Be Divided in the Mean and Extreme Ratio?’, AncPhil 14.2, 283–95. Baldwin, B. (1975) Studies in Aulus Gellius, Lawrence. Barbis Lupi, R. (1994) ‘La paragraphos: analisi di un segno di lettura’, in Bülow–Jacobsen, A. (ed.) Proceedings of the 20th International Congress of Papyrologists (Copenhagen 1992), Copenhagen, 414–7. Barbour, R. (1981) Greek Literary Hands: A. D. 400–1600, Oxford. Barchiesi, A., Rosati G. (2007) Ovidio, Metamorfosi, (libri III–IV), II, Milan. Barigazzi, A. (1952) ‘Sopra alcuni epigrammi ellenistici’, Hermes 80, 494–6. Barker, A. (1989) Greek Musical Writings, II, Cambridge. Barker, A. (1994) ‘Ptolemy's Pythagoreans, Archytas, and Plato's Conception of Mathematics’, Phronesis 39.2, 113–35. Barker, A. (2003) ‘Early Timaeus Commentaries and Hellenistic Musicology’, in Sharples, R.W., Sheppard, A. (eds.) Ancient Approaches to Plato’s Timaeus, London, 73–87. Barker, A. (2007) The Science of Harmonics in Classical Greece, Cambridge. Barker, A. (2014) ‘Pythagorean Harmonics’, in Huffman, C.A. (ed.) A History of Pythagoreanism, Cambridge, 185–203. Barrow, R.H. (1973) Quintus Aurelius Symmachus, Prefect and Emperor. The Relationes of Symmachus A.D. 384, Oxford. Bartsch, K. (1861) Albrecht von Halberstadt und Ovid im Mittelalter, Leipzig.

Bibliography  311

Barwick, K.B. (1958) ‘Zyklen bei Martial und in den kleinen Gedichten des Catull’, Philologus 102, 284–318. Bastianini, G., Gallazzi, C., Austin, C. (2001) Posidippo di Pella. Epigrammi (P.Mil.Vogl. VIII 309), Milan. Bastianini, G., Haslam, M., Maehler, H., Montanari, F., Römer, C., Stroppa, M. (eds.) (20122) Commentaria et Lexica Graeca in Papyris reperta, I, 1.4, Aristophanes, Berlin–Boston. Battegay, R. (1970) Narzißmus und Objektbeziehungen. Über das Selbs zum Objekt, Bern. Battezzato, L. (2003) Tradizione testuale e recezione letteraria antica della tragedia greca, Atti del convegno Scuola Normale Superiore, Pisa, 14–15 giugno 2002, Amsterdam. Beck, J.W. (1993) Terentianus Maurus De Syllabis, Göttingen. Becker, G. (1885) Catalogi bibliothecarum antiqui, Bonn. Behaghel, O. (1909) ʻZum deutsche Ovid von 1210ʼ, Anzeiger für deutsches Altertum und deutsche Literatur 33, 313–4. Bellocchi, M. (2009) ‘Gli oracoli in esametri di Aristofane come testimonianza di poesia oracolare nell'Atene del tardo V secolo a.C.’, RFIC 137, 23–40. Bendinelli, G. (1920) ‘Roma. Ipogeo con pitture, scoperto presso il viale Manzoni’, NSA 26, 123–41. Bendinelli, G. (1922) ‘Il monumento sepolcrale degli Aureli al Viale Manzoni in Roma’, Monumenti antichi pubblicati per cura della Reale Accademia Nazionale dei Lincei 28, 289–520. Bennett, J. (2006) ‘New Evidence from Ankara for the collegia veteranorum and the albata decursio: In Memoriam J. C. Mann’, AS 56, 95–101. Bernhardy, G. (1822) Eratosthenica, Berlin. Bernsdorff, H. (2006) ‘The Idea of Bucolic in the Imitators of Theocritus. 3rd–1st cent. B.C.’, in Fantuzzi, M., Papanghelis, T.D. (eds.) Brill’s Companion to Greek and Latin Pastoral, Leiden–Boston, 167–208. Beta, S. (2016) Il labirinto della parola: enigmi, oracoli e sogni nella cultura antica, Torino. Beudel, P. (1911) Qua ratione Graeci liberos docuerint, papyris, ostracis, tabulis in Aegyptus inventis illustratur, Diss. Münster. Biles, Z. (1999) ‘Eratosthenes on Plato Comicus’, ZPE 127, 182–8. Biondi, A. (1985) ‘Celio Secondo Curione’, in Dizionario Biografico degli Italiani 31, 443–9. Bird, G.D. (2010) Multitextuality in the Homeric Iliad. The Witness of the Ptolemaic Papyri, Washington DC. Bisconti, F. (2004) ‘Ipogeo degli Aureli. Alcune riflessioni e qualche piccola scoperta’, RAC 80, 13–38. Bisconti, F. (2010) ‘Lavori nelle catacombe: il lutto, Circe e S. Paolo’, RAC 86, 25–52. Bisconti, F. (ed.) (2011) L’ipogeo degli Aureli in Viale Manzoni: restauri, tutela, valorizzazione e aggiornamenti interpretativi, Vatican City. Blanck, H. (1992) Das Buch in der Antike, Munich. Blanck, H. (2008) Il libro nel mondo antico, pref. di Canfora, L., ed. rivista e aggiornata a cura di Otranto, R., Bari. Blass, F. (18982) Die attische Beredsamkeit, I–III.2, Leipzig. Blok, J.H., Lambert, S.D. (2009) ‛The appointment of Priests in Attic gene’, ZPE 169, 95–121. Blum, R. (1977) Kallimachos und die Literaturverzeichnung bei den Griechen, Frankfurt. Bo, D. (1965–1966) Lexicon Horatianum, I–II, Hildesheim. Boegehold, A.L. (1995) The Law–courts at Athens: Sites, Buildings, Equipment, Procedure and Testimonia, The Athenian Agora 28, Princeton.

312  Bibliography Bolognese, I.L. (2013) The Monastery of Montevergine. Its Foundation and Early Development (1118–1210), Diss. University of Leeds. Bolte, J. (1905) Georg Wickrams Werke. Siebenter band (Ovids Metamorphosen, Buch 1–8), Tübingen. Bömer, F. (1969) P. Ovidio Naso. Metamorphosen. Kommentar, Buch I–III, I, Heidelberg. Bömer, F. (1981) Untersuchungen über die Religion der Sklaven in Greichenland und Rom. Erster Teil: die Wichtigsten Kulte und Religionen in Rom und im Lateinischen Westen, Wiesbaden. Bonati, I. (2016) ‘L’etichettatura del farmaco: radici antiche di una tradizione millenaria’, in Reggiani, N. (2016a) 43–78. Borg, B.E. (2013) Crisis and Ambition. Tombs and Burial Customs in Third–Century CE Rome, Oxford. Borgo, A. (2005) Il ciclo di Postumo nel libro secondo di Marziale, Naples. Bouchard, E. (2016) Du Lycée au Musée. Théorie poétique et critique littéraire à l’époque hellénistique, Paris. Boudon–Millot, V. (2008) ‘The Library of a Greek Scholar in the Roman Empire: New Testimony from Galen’s Recently Discovered Peri Alupias’, in Cilliers, L. (ed.) Asklepios. Studies on Ancient Medicine, Bloemfontein, 7–18. Boudon–Millot, V., Jouanna, J. (2010) Galien: Ne pas chagriner, Paris. Bovie, S.P. (1959) The Satires and Epistles of Horace, Chicago. Bowersock, G.W. (1969) Greek Sophists in the Roman Empire, Oxford. Boymel Kampen, N. (1981) ‘Biographical Narration and Roman Funerary Art’, AJA 85, 47–58. Bredehorn, K. (1999) ‘Aus Fuldischen Handschriften. Codex Waldeccensis: Fragmente einer griechisch–lateinischen Bibelhandschrift (olim Arolsen–Mengeringhausen)’, Archiv für mittelrheinische Kirchengeschichte 51, 455–514. Brisson, L. (1992) Platon, Timée, Critias, Paris. Brisson, L. (1998) Le Même et l’Autre dans la Structure ontologique du Timée de Platon: un Commentaire systematique du Timée de Platon, Sankt Augustin. Brugnoli, G. (1965) ‘Donato e Girolamo’, VetChr 2, 139–49. Bühler, J. (1925) Die Hohenstaufen. Nach zeitgenössischen Quellen, Leipzig. Bultrighini, U. (2014) ‛Eschine e la phéme in giudizio’, RCCM 56, 317–30. Buongiovanni, C. (2009) ‘Marziale, libro X. Gli epigrammi 1 e 2 tra poesia e politica’, Athenaeum 97.2, 507–26. Buongiovanni, C. (2011) ‘Parlare a nuora perché suocera intenda: Marziale e l’ambiguo destinatario dell’epigramma 10.103’, in Perelli, R., Mastandrea, P. (eds.) Latinum est et legitur. Metodi e temi nello studio dei testi latini, Amsterdam. Buongiovanni, C. (2012) Gli epigrammata longa del decimo libro di Marziale. Introduzione, testo e commento, Pisa. Buonopane, A. (2009) Manuale di Epigrafia latina, Rome. Burkert, W. (1972) Lore and Science in Ancient Pythagoreanism, transl. Minar, E.L., Cambridge [or. ed. (1962) Weisheit und Wissenshaft: Studien zu Phytagoras, Philolaos, und Platon, Nürnberg]. Burton, J.B. (1995) Theocritus’s Urban Mimes: Mobility, Gender and Patronage, Berkeley. Butler, H.E. (1958) The Institutio Oratoria of Quintilian, Cambridge–Mass. Butler, S. (2002) The Hand of Cicero, London. Butler, S. (2011) The Matter of the Page: Essays in Search of Ancient and Medieval Authors, Madison.

Bibliography  313

Cain, A. (2009) The Letters of Jerome. Asceticism, Biblical Exegesis, and the Construction of Christian Authority in Late Antiquity, New York. Cairns, F. (1984) ‘Theocritus’ First Idyll: The Literary Programme’, WS n.f. 18, 89–113. Cajetani, C. (1853) Petri Damiani Opusculum XV, in PL 145, 335–64. Calame, C. (2011) ‘Funerary Gold Lamellae and Orphic Papyrus Commentaries. Same Use, Different Purpose’, in Edmonds, R.G. (ed.) ‘Orphic’ Gold Tablets and Greek Religion: Further Along the Path, Cambridge, 203–18. Callu, J.–P. (1982) Symmaque, Lettres, tome II (Livres III–V), Paris. Cambiano, G. (1992) ‘La nascita dei trattati e dei manuali’, in Cambiano, G., Canfora, L., Lanza, D. (eds.) Lo spazio letterario della Grecia antica, I.1, Rome, 525–53. Camerarius, J. (1530) Theocriti syracusani Idyllia triginta sex, latino carmine reddita, Helio Eobano Hesso interprete, Hagenau. Cameron, A. (1993) The Greek Anthology from Meleager to Planudes, Oxford. Cameron, A. (2011) The Last Pagans of Rome, Oxford. Campbell, D.A. (1991) Greek Lyric, III, Cambridge–London. Cancik, H. (1967) ʻSpiegel der Erkenntnisʼ, AU 10, 42–53. Canevaro, M. (2013) The Documents in the Attic Orators: Laws and Decrees in the Public Speeches of the Demosthenic Corpus, Oxford. Canfora, L. (1993) ‘La biblioteca e il museo’, in Cambiano, G., Canfora, L., Lanza, D. (eds.) Lo spazio letterario della Grecia antica, I.2, Rome, 11–29. Canfora, C. (1995) ‛Le collezioni superstiti’, in Cambiano, G., Canfora, L., Lanza, D. (eds.), La ricezione e l’attualizzazione del testo, II, Rome, 90–250. Canfora, L. (2002) Il copista come autore, Palermo. Canfora, L. (2018) Per una storia delle biblioteche, Bologna. Cannon–Geary, I. (1980) The Bourgeoisie Looks at Itself: The Sixteenth Century in Germany Literary Histories of the Nineteenth Century, Göppingen. Canobbio, A. (2008) ‘Epigrammata longa e breves libelli. Dinamiche formali dell’epigramma marzialiano’, in Morelli, A. (2008), 169–93. Caragounis, C.C. (2004) The Development of Greek and the New Testament: Morphology, Syntax, Phonology, and Textual Transmission, Tübingen. Carey, C. (2015) ‛Solon in the Orators’, Trends in Classics 7, 110–128. Carletti, C. (1986) Iscrizioni cristiane a Roma. Testimonianze di vita cristiana (secoli III–VII), Florence. Carletti, C. (2008) Epigrafia dei cristiani in Occidente dal III al VII secolo. Ideologia e prassi, Bari. Carlone, C. (1979) ‘Il problema dei falsi ed alcune presunte dipendenze verginiane’, Samnium LII, 78–102. Carlone, C. (1984) Falsificazioni e falsari cavensi e verginiani nel sec. XIII, Altavilla Silentina. Caroli, M. (2007) Il titolo iniziale nel rotolo librario greco–egizio, Bari. Casson, L. (2003) Biblioteche del mondo antico, Milan. Cavallera, F. (1922) Saint Jérôme. Sa vie et son oeuvre, Première Partie, Louvain–Paris. Cavallo, G. (1986) ‘Conservazione e perdita dei testi greci: fattori materiali, sociali, culturali’, in Giardina, A. (ed.) Tradizione dei classici, trasformazioni della cultura, Rome–Bari, 83–172. Cavallo, G. (1987) ‘Dallo scriptorium senza biblioteca alla biblioteca senza scriptorium’, in Alessio, G.C., Pugliese Carattelli, A. et al. (eds.) Dall’eremo al cenobio: la civiltà monastica in Italia dalle origini all’età di Dante, Milan, 329–422.

314  Bibliography Cavallo, G. (1989) ‘Testo, libro, lettura’, in Cavallo, G., Fedeli, P., Giardina, A. (eds.) Lo spazio letterario di Roma antica, II, La circolazione del testo, Rome, 307–41. Cavallo, G. (1994) ‘Discorsi sul libro’, in Cambiano, G., Canfora, L., Lanza, D. (eds.) Lo spazio letterario della Grecia antica, I, 3. I Greci e Roma, Rome, 613–47. Cavallo, G. (ed.) (1975) Libri, editori e pubblico nel mondo antico. Guida storica e critica, Rome– Bari. Cavallo, G. (ed.) (2002) Le biblioteche nel mondo antico e medievale, Rome. Cavallo, G. (2007) ‘Le scriptorium médiéval’, in Jacod, C. (ed.) Lieux de Savoir. Espace et communautés, Paris, 537–55. Centrone, B. (2014) ‘The Pseudo–Pythagorean Writings’, in Huffman, C.A. (ed.) A History of Pythagoreanism, Cambridge, 315–40. Ceresa–Gastaldo, A. (1988) Gerolamo, Gli uomini illustri, Florence. Cholmeley, R.J. (19192) The Idylls of Theocritus, London. Ciccarelli, I. (1997) ‘Ovidio Tristia 4, 10 e i topoi della sphragís’, Aufidus 32, 61–92. Cingano, E. (1990) ‘L’opera di Ibico e di Stesicoro nella classificazione degli antichi e dei moderni’, AION (filol) 12, 189–224. Citroni, M. (1970) ‘Un proemio di Marziale: I, 3’, in Studia Florentina Alexandro Ronconi sexagenario oblata, Rome, 81–91. Citroni, M. (1975) M. Valerii Martialis Epigrammaton Liber Primus, Introduzione, testo critico e commento, Florence 1975. Citroni, M. (1986) ‘Le raccomandazioni del poeta: apostrofe al libro e contatto col destinatario’, Maia 38, 111–46. Citroni, M. (1988) ‘Pubblicazione e dediche dei libri in Marziale’, Maia 40, 3–39. Citroni, M. (1990) ‘I destinatari contemporanei’, in Cavallo, G., Fedeli, P., Giardina, A. (eds.) Lo spazio letterario di Roma antica, III, La ricezione del testo, Rome, 53–116. Citroni, M. (1995) Poesia e lettori in Roma antica: Forme della comunicazione letteraria, Rome– Bari. Citroni, M. (2002) ‘L’immagine della Spagna e l’autorappresentazione del poeta negli epigrammi di Marziale, in Urso, G. (ed.) Hispania terris omnibus felicior. Premesse ed esiti di un processo di integrazione, Atti del convegno internazionale, Cividale del Friuli, 27–29 settembre 2001, Pisa, 281–301. Clackson, J., Horrocks, G. (2007) The Blackwell History of the Latin Language, Oxford. Classen, C.J. (1985) ‘Martial’, Gymnasium 92, 329–49. Clausi, B. (1995) ‘Storia sacra e strategia retorica. Osservazioni sull’uso dell’“exemplum” biblico nell’Adversus Iovinianum di Gerolamo’, CrSt 16, 457–84. Coarelli, F. (2007) Rome and Environs: An Archaeological Guide, Berkeley. Cody, J.V. (1976) Horace and Callimachean Aesthetics, Brussels. Cohen, B. (1984) ‘Some Neglected Ordines: The Apparitorial Status–Groups’ in Nicolet, C. (ed.) Des ordres à Rome, Paris, 23–60. Colamarco, T. (2008) ‘Il cosiddetto Statuto dell’Abate Donato’, in Borraccini, R.M., Borri, G. (eds.) Virtute et Labore: Studi offerti a Giuseppe Avarucci per i suoi settant’anni, Spoleto, 132–50. Colwell, E.C. (1969a) ‘Hort Redivivus: A Plea and a Program’, in Colwell, E.C. (ed.) Studies in Methodology in Textual Criticism of the New Testament, Leiden, 148–71. Colwell, E.C. (1969b) ‘Method in Evaluating Scribal Habits: A Study of P45, P66, P75’, in Colwell, E.C. (ed.) Studies in Methodology in Textual Criticism of the New Testament, Leiden, 106– 124.

Bibliography  315

Condello, F. (2007) ‘Riordinare una biblioteca orale: Omero ciclico, Omero girovago e il problema delle “doppie attribuzioni”’, in Andrisano, A.M. (ed.) Biblioteche del mondo antico, Rome, 13–35. Conte, G.B. (1974) Memoria dei poeti e sistema letterario, Turin. Conte, G.B. (2017) Stealing the Club from Hercules. On Imitation in Latin Poetry, Berlin–Boston. Coopeland, R. (1995) Rhetoric, Hermeneutics, and Translation in the Middle Ages Academic Traditions and Vernacular Texts, Cambridge. Corazza, F. (2016) ‘New Recipes by Heras in P. Berol. Möller 13’, ZPE 198, 39–48. Cornford, F.M. (1937) Platos Cosmology: The Timaeus of Plato, London. Cornford, F.M. (1987) ‘Mathematics and Dialectic in the Republic VI–VII’, in Bowen, A.C. (ed.) Selected Papers of F.M. Cornford, New York–London–Garland, 37–52. Corssen, P. (1887) Epistularum Paulinarum codices Graece et Latine scriptos: augiensem boernerianum claromontanum examinavit inter se comparavit ad communem originem revocavit, Kiel. Coulson, F.T. (1991) The ‘Vulgate’ Commentary on Ovid’s Metamorphoses. The Creation Myth and the Story of Orpheus, Toronto. Courcelle, P. (1969) Late Latin Writers and Their Greek Sources, transl. Wedeck, H.E., Cambridge–Mass. Cribiore, R. (1996) Writers, Teachers and Students in Greco–Roman Egypt, Atlanta. Cucchiarelli, A. (2015) Orazio, l’esperienza delle cose (Epistole, Libro I), Venice. Cugusi, P. (1985) Aspetti letterari dei carmina latina epigraphica, Bologna. Cugusi, P. (2007) Per un nuovo corpus dei Carmina Epigraphica. Materiali e discussioni, Rome. Cuomo, G. (1973) Le leggi eversive del secolo XIX e le vicende degli Ordini religiosi della provincia di Principato Citeriore, Mercato San Severino. Curione C.S. (1557) L. Annaei Senecae philosophi opera quae extant omnia C. S. Curionis cura castigata, Basel. Curran, J. (2012) ‘Virgilizing Christianity in Late Antique Rome’, in Grig, L., Kelly, G. (eds.), Two Romes, Rome and Constantinople in Late Antiquity, Oxford–New York, 325–44. Curtius, E.R. (1953) European literature and the Latin Middle Ages, transl. Trask, W.R., Princeton (repr. 2013). Dale, A.M. (1968) The Lyric Metres of Greek Drama, Cambridge. Dau, A. (1887) De M. Valerii Martialis libellorum ratione temporibusque, Diss. Rostock. Davies, M., Finglass, P.J. (2014) Stesichorus. The Poems, Cambridge. Davis, G. (1991) Polyhymnia: The Rhetoric of Horatian Lyric Discourse, Berkeley–London. De Luciis, O. (1619) Supplemento alla Historia di Montevergine, ms. De Palma, E. (1932) ‘Intorno alla leggenda de vita et obitu S. Guilielmi confessoris et heremite’, Irpinia IV, 59–75. De Rossi, G.B. (1864) La Roma sotterranea cristiana descritta ed illustrate, Rome. Debut, J. (1986) ‘Les documents scolaires’, ZPE 63, 251–78. Degani, E. (1993) ‘L’epigramma’, in Cambiano, G., Canfora, L., Lanza, D. (eds.) Lo spazio letterario della Grecia antica, I, La produzione e la circolazione del testo, II, L’ellenismo, Rome, 197–233. Degni, P. (1998) Usi delle tavolette lignee e cerate nel mondo greco e romano, Messina. De Jong, I.J.F., Sullivan J.P. (1994) Modern Critical Theory and Classical Literature, Leiden–New York–Cologne. Del Grande, C. (1946) Ditirambografi. Testimonianze e frammenti, Naples. Della Bona, M.E. (2013) ‘Gare simposiali di enigmi e indovinelli’, QUCC 2, 169–182.

316  Bibliography Depew, M., Obbink, D. (2000) Matrices of Genre: Authors, Canons and Society, Cambridge– Mass. Detienne, M. (1988) Les savoirs de l'écriture en Grèce ancienne, Lille. Dettori, E. (2000) Filita di Cos. Testimonianze e frammenti grammaticali, Rome. Di Benedetto, V. (1971) Euripide. Teatro e società, Turin. Di Giovine, C. (2003) ‘Marziale e I componimenti diffamatorii. Tipologie e forme dell’autodifesa’, BStudLat 33, 84–99. Di Luzio, A. (1969) ‛Papiri omerici d’epoca tolemaica e la costituzione del testo dell’epoca arcaica’, RCCM 11, 3–152. Di Marco, M. (2009) ‘Un epigramma–indovinello su tavoletta cerata (P. Lit. Lond. 63 = T. British Museum inv. 29527 = Pack2 1765 = LDAB 5055)’, ZPE 169, 85–87. Di Meo, A. (1805) Annali critico diplomatici del Regno di Napoli nella mezzana età, Naples. Di Tullio, D. (1915) Gli studii sulla comedia nell’età Alessandrina e l’opera di Eratostene «Περὶ τῆς ἀρχαίας κωμῳδίας», Rome. Dickey, E. (2006) Ancient Greek Scholarship: a Guide to Finding, Reading, and Understanding Scholia, Commentaries, Lexica, and Grammatical Treatises, from their Beginnings to the Byzantine Period, Oxford–New York. Diehl, E. (1925) Anthologia Lyrica, V, Leipzig. Diels, H. (1898) ‘Die Elegie des Poseidippos aus Theben’, SPW, 847–58. Dilts, M.R. (1997) Aeschines. Orationes, Leipzig. Dorandi, T. (2007) Nell’officina dei classici. Come lavoravano gli autori antichi, Rome. Dorjahn, A.P. (1927) ‛Poetry in Athenian Courts’, CP 32, 85–93. Dorjahn, A.P., Fairchild, W.D. (1972) ‛Extemporaneous Elements in Some Minor Attic Orators’, CB 48, 62–63. Dörrie, H. (1967) ‘Echo und Narcissus (Ovid, Met. 3, 341–510). Psychologische Fiktion in Spiel und Ernst’, AU 10, 54–75. Dörrie, H. (1981) ‘Formula Analogiae: an Exploration of a Theme in Hellenistic and imperial Platonism’, in Blumenthal, H.J., Markus, R.A. (eds.) Neoplatonism and Early Christian Thought, London, 35–39. Dörrie, H. (1987) Der Platonismus in der Antike, I, Stuttgart–Bad Cannstatt. Doufour, M. (1932) Aristote. Rhétorique, Paris. Dover, K.J. (1968) Aristophanes. Clouds, Oxford. Dover, K.J. (1971) Theocritus’s Select Poems, Glasgow. Dover, K.J. (1978) Greek Homosexuality, Cambridge. Dreher, J.P. (1990) ‘The Driving Ratio in Plato’s Divided Line’, AncPhil 10, 159–72. Drerup., E. (1898) ‛Ueber die bei den attischen Rednern eingelegten Urkunden’, Jahrbücher für Klassische Philologie Supplementband 24, 221–366. Driscoll, J., Pierazzo, E. (2016) Digital Scholarly Editing: Theories and Practices, Cambridge. Dué, C. (2001) ‛Achilles’ Golden Amphora in Aeschines’ Against Timarchus and the Afterlife of Oral Tradition’, CPh 96, 33–47. Durrbach, F. (1932) Lycurgue. Contre Léocrate. Fragments, Paris. Ebers, G. (1873) ‘Papyrus Ebers’, ZÄS 11, 41–45. Edmonds, J.M. (1912) Theocritos, Moschus, Bion, Cambridge–Mass. Edquist, H. (1975) ‘Aspects of Theocritean Otium’, Ramus 4, 101–14. Ehrismann, O. (1927) Ehre und Mut, Âventiure und Minne. Höfische Wortgeschichten aus dem Mittelalter, Munich.

Bibliography  317

Engemann, J. (1982) ‘Der Ehrenplatz beim antiken Sigmamahl’, in Jenseitvorstellungen in Antike und Christentum. Gendenkschrift für Alfred Stuiber. JbAC Ergänzungsband 9, Münster, 239–50. Ensoli, S., La Rocca, E. (eds.) (2000) Aurea Roma. Dalla città pagana alla città cristiana, Rome. Epp, E.J. (1976) ‘The Eclectic Method in New Testament Textual Criticism: Solution or Symptom?’ HTR 69.3/4, 211–57. Erasmus D. (1537) L. Annaei Senecae opera, et ad dicendi facultatem et ad bene vivendum utilissima, per Des. Erasmum Roterod. ex fide veterum codicum, tum ex probatis auctoribus, postremo sagaci nonnunquam divinatione sic emendata, ut ad genuinam lectionem minimum desiderare possis. Adiecta sunt scholia D. Erasmi Roterodami et Beati Rhenani, illius in bonam partem operis, huius in Ludum de morte Claudii Caesaris, Basel. Erbse, H. (1969–1977) Scholia Graeca in Homeri Iliadem, I–V, Berlin. Ercoles, M. (2013) Stesicoro: le testimonianze antiche, Bologna. Ewald, B.C. (1999) Der Philosoph als Leitbild. Ikonographische Untersuchungen an Römischen Sarkophagreliefs, Mainz. Eymann, H.S. (1996) Epistula ad Romanos, VL 21, Freiburg. Fabbrini, D. (2007) Il migliore dei mondi possibili. Gli epigrammi ecfrastici di Marziale per amici e protettori, Florence. Facchini Tosi, C. (1998) Anneo Floro, Storia di Roma, La prima e la seconda età, Bologna. Fairclough, H.R. (1999) Virgil Eclogues Georgics Aeneid I–VI, revised by Goold, G.P., Cambridge. Fantuzzi, M. (1998) ‘Textual Misadventures of Daphnis. The Pseudo–Theocritean Id. 8 and the Origins of the Bucolic Manner’, in Harder, A., Regtuit, R.F., Wakker, G.C. (eds.) Genre in Hellenistic Poetry: Third Hellenistic Workshop on Hellenistic Poetry (Groningen 28–30 August 1996), Groningen, 61–79. Fantuzzi, M. (2006) ‘Theocritus’ Constructive Interpreters and the Creation of a Bucolic Reader’, in Fantuzzi, M., Papanghelis, T.D. (eds.) Brill’s Companion to Greek and Latin Pastoral, Leiden–Boston, 235–62. Fassino, M., Prauscello, L. (2001) ‘Memoria ritmica e memoria poetica: Saffo e Alceo in Teocrito Idilli 28–30 tra ἀρχαιολογία metrica e innovazione alessandrina’, MD 46, 9–37. Fedeli, P. (1989) ‘I sistemi di produzione e diffusione’, in G. Cavallo, P. Fedeli, A. Giardina (eds.) Lo spazio letterario di Roma antica, II, La circolazione del testo, Rome, 343–78. Fedeli, P. (1994) Orazio Flacco, Le epistole, Rome. Fedeli, P. (1996) ‘A un libro che se ne va (Hor. epist. 1,20)’, in Santini C., Zurli, L. (eds.) Ars Narrandi, Scriti di narrativa antica in memoria di Luigi Pepe, 11–25. Fedeli, P. (2012) Orazio: tutte le poesie, Milan. Ferrari, F. (2000) ‘Commentari specialistici alle sezioni matematiche del Timeo’, in Brancacci, A. (ed.) La filosofia in età imperiale. Le scuole e le tradizioni filosofiche, Atti del colloquio (Roma 17–19 giugno 1999), Naples, 169–224. Ferrari, F. (2004) ‘Posidippus, the Milan Papyrus and Some Hellenistic Anthologies’, Classic@ 1, 1–14. Fickert, K.R. (1845) L. Annaei Senecae opera, Leipzig. Finglass P.J. (2015) ‘Reperformances and the Transmission of Texts’, Trends in Classics 7, 259– 76. Finney, P.C. (1980) ‘Did Gnostics Make Pictures?’ in Layton, B. (ed.) The Rediscovery of Gnosticism. Vol. 1: The Valentinians, SHR, Supplements to Numen 41, Leiden, 434–54.

318  Bibliography Fiocchi Nicolai, V., Bisconti, F., Mazzoleni, D. (20093) The Christian Catacombs of Rome: History, Decoration, inscriptions, transl. Stella, C.C., Touchette L.–A., Regensburg. Fisher, N. (2001) Aeschines. Against Timarchus, Translated with Introduction and Commentary, Oxford. Flores D’Arcais, F., Crivello, F. (eds.) (2010) Come nasce un manoscritto miniato. Scriptoria, tecniche, modelli e materiali, Atti del convegno tenutosi a Milano, Università Cattolica del Sacro Cuore–Biblioteca Ambrosiana, 6–7 marzo 2008, Modena. Floridi, L. (2007) Stratone di Sardi. Epigrammi, Alessandria. Foertsch J., Morton A.G. (1997) ‘In Search of Gender Neutrality: Is Singular They a Cognitively Efficient Substitute for Generic He?’, Psychological Science 8.2, 106–11. Ford, A. (1999) ‛Reading Homer from the Rostrum: Poems and Laws in Aeschines’ Against Timarchus’, in Goldhill, S., Osborne, R. (eds.) Performance Culture and Athenian Democracy, Cambridge, 231–56. Fournet, J.–L., Magdelaine, C. (2001) ‘Liste de vins aromatisés à usage médical’, in Andorlini, I. (ed.) Greek Medical Papyri I, Florence, 164–70. Fowler, D. (1995) ‘Martial and the book’, Ramus 24.1, 31–58. Fraenkel, E. (1957) Horace, Oxford. Frampton, S. (2016) ‘What to do with the books in the De finibus’, TAPhA 146.1, 117–47. Frampton, S. (2019) Empire of Letters: Writing in Roman Literature and Thought from Lucretius to Ovid, Oxford. Fränkel, H. (1945) ʻOvid. A poet between two worldsʼ, Berkeley–Los Angeles. Fraser, P.M. (1970) ‘Eratosthenes of Cyrene: Lecture on a Master Mind’, PBA 56, 175–207. Frede, H.J. (1962) Epistula ad Ephesios, VL, XXIV, Freiburg. Frede, H.J. (1964) Altlateinische Paulus–Handschriften, VLSup, IV, Freiburg. Frede, H.J. (1966) Epistulae ad Philippenses et ad Colossenses, VL, XXIV, Freiburg. Frede, H.J. (1973) Ein Neuer Paulustext und Kommentar, VLSup, VII, Freiburg. Frede, H.J. (1975) Epistulae ad Thessalonicenses, Timotheum, Titum, Philemonem, Hebraeos, VL, XXV, Freiburg. Fritzsche, A.Th.H. (ed.) (1868) Theocriti Idyllia, I, Leipzig. Fröhlich, U. (1995) Epistula ad Corinthios I, VL.22, Freiburg. Fürst, A. (2016) Hieronymus. Askese und Wissenschaft in der Spätantike, Freiburg. Fusi, A. (2011) ‘Marziale e il fantasma di Scorpo. Nota a 10.48.23’, in Perelli, R., Mastandrea, P. (eds.) Latinum est et legitur. Metodi e temi nello studio dei testi latini, Amsterdam, 261– 80. Gaiser, K. (1994) La dottrina non scritta di Platone: Studi sulla fondazione sistematica e storica delle scienze nella scuola platonica, Milan [or. ed. (19631, 19682) Platons ungeschriebene Lehre: Studien zur systematischen und geschichtlichen Begrundung der Wissenschaften in der platonischen Schule, Stuttgart–Klett]. Gallavotti, C. (1951) Lingua, tecnica e poesia negli Idilli di Teocrito, Rome. Gallavotti, C. (1966) ‘Le coppe istoriate di Teocrito e di Virgilio’, PP 21, 421–36. Gallavotti, C. (19933) Theocritus. Quique feruntur bucolici graeci, Rome. Gamberale, L. (2013) San Gerolamo intellettuale e filologo, Rome. Gamillscheg, E., Harlfinger, D., Hunger, H. (1981) Repertorium der griechischen Kopisten, 800– 1600, III, Vienna. Garofalo, I., Lami, A. (2012) Galeno: L’anima e il dolore. De indolentia, De propriis placitis, Milan.

Bibliography  319

Gardthausen, V. (1967) Ammianus Marcellinus, Rerum gestarum libri quae supersunt, editio stereotypa editionis primae (MDCCCLXXIV), I–II, Stuttgart. Garulli, V. (2004) ‘Rassegna di studi sul Nuovo Posidippo (1993–2003)’, Lexis 22, 291–340. Gasti, F. (2016) ‘Aspetti della presenza di Cicerone nella tarda antichità latina’, in De Paolis, P. (ed.) Cicerone nella cultura antica Atti del VII Simposio Ciceroniano (Arpino 8 maggio 2015), Cassino, 27–54. Geerdts H.J. (1952–1953) ʻDas Erwachen des bürgerlichen Klassenbewusstseins in den Romanen Jörg Wickramsʼ, Wissenschaftliche Zeitschrift der Friedrich–SchillerUniversität Jena 2, 117–124. Geus, K. (1995) ‘Anmerkungen zur “Musiktheorie” des Eratosthenes’, in Antike Naturwissenschaft und ihre Rezeption, V, Trier, 49–62. Geus, K. (2002) Eratosthenes von Kyrene. Studien zur hellenistischen Kultur–und Wissenschaftsgeschichte, Munich. Giancotti F. (1963) Ricerche sulla tradizione manoscritta delle Sentenze di Publilio Siro, Florence. Giangrande, G. (1978) ‘Preliminary Notes on the Use of Paragraphos in Greek Papyri’, Museum Philologicum Londinense 3, 147–51. Giannarelli, E. (2007) ‘Le forme del comico in Girolamo’, in Mazzucco, C. (ed.) Riso e Comicità nel Cristianesimo antico. Atti del Convegno (Torino, 14–16 febbraio 2005), Alessandria, 291–304. Gignac, F.T. (1976) A Grammar of the Greek Papyri of the Roman and Byzantine Periods, 2 vols., Milan. Giordano Ramponi, A. (1982) Sulpiciae Conquestio (Ep. Bob.37), Bologna. Giordano, D. (19902) Chamaeleontis Heracleotae Fragmenta, Bologna. Giovagnoli, M. (2011) ‘L’Apparto epigrafico dell’ipogeo degli Aureli’, in Bisconti, F. (ed.) (2011), 229–32. Giuliani, R. (2011) ‘L’ipogeo degli Aureli: quasi un secolo di tutela, dalla scoperta ai giorni nostri. Con un’appendice di Cecilia Proverbio’ in Bisconti, F. (ed.) (2011), 31–50. Gleason, M. (2006) ‘Greek Cities under Roman Rule’, in Potter, D.S. (ed.) A Companion to the Roman Empire, Malden, 228–49. Godel, R. (1964) ‘Réminiscences de poètes profanes dans les lettres de St–Jérôme’, MH 21, 65– 70. Goltz, D. (1974) Studien zur altorientalischen und griechischen Heilkunde. Therapie– Arzneibereitung–Rezeptstruktur, Wiesbaden. Goulet, R. (1989) ‘Adraste D’Aphrodise’, in Goulet, R. (ed.) Dictionnaire des Philosophes Antiques, I, Paris, 56–57. Gow, A.S.F. (19522) Theocritus. Edited With a Translation and Commentary, 2 vols., Cambridge. Grant, J.N. (1989) Editing Greek and Latin texts: Papers given at the Twenty–Third Annual Conference on Editorial Problems, University of Toronto, 6–7 November 1987, New York. Grassigli, G.L. (2000) ‘Il regno della villa. Alle origini della rappresentazione della villa tardoantica’, Ostraka 9.1, 199–226. Grassigli, G.L. (2002) ‘L’ipogeo degli Aureli: tra trascendenza e identità pagana’, in Iconografia 2001, studi sull’immagine: Atti del Convegno (Padova, 30 maggio –1 giugno 2001), Rome, 405–18. Grassigli, G.L. (2011) Splendidus in Villam Secessus: Vita quotidiana, cerimonali e autorappresentazione del dominus nell’arte tardoantica, Naples. Gray, C. (2015) Jerome, Vita Malchi, Introduction, Text, Translation, and Commentary, Oxford.

320  Bibliography Graziosi, B. (2008) ‘The Ancient Reception of Homer’, in Hardwick, L., Stray, C. (eds.) (2008) 26–37. Gregory, C.R. (1900) Textkritik des Neuen Testamentes, Leipzig. Grenfell, B.P., Hunt A.S. (1919) The Oxyrhynchus Papyri, XIII, London. Griesbach, J.J. (1793) Symbolae Criticae, Halle. Grilli, A. (2004) ‘Alla scuola di san Gerolamo’, in Urso, G. (ed.) Dall'Adriatico al Danubio. L'Illirico nell'età greca e romana, Pisa, 385–94. Gruter, J. (1594) L. Annaeus Seneca a Mureto correctus et notis illustratus. Accedunt seorsim Animadversiones in quibus praeter omnes passim omnium huius superiorisque aevi doctorum hominum emendationes interpretationesque, quamplurima loca supplentur, confirmantur, corriguntur, illustrantur ope M.SS. quae in Bibliotheca Electoris Palatini: Iani Gruteri opera, Heidelberg. Gruter, J. (ed.) (1604) Lampas sive fax artium liberalium, hoc est thesaurus criticus, Frankfurt. Gryson R. (1999) Altalateinische Handschriften Manuscrits Vieux Latins: Répertoire descriptif, VL.1/2A, 2 vols., Freiburg. Guichard, L.A. (2007) ‘Acertijos de uso escolar en papiro, tablilallas y ostraca’, in Delgado, J.A., Pordomingo, F., Stramaglia, A. (eds.) Literatura y escuela en Grecia Antigua. Actas del Simposio International. Salamanca 2004, Cassino, 224–36. Guidorizzi, G. (1984) ‘Gli epigrammi papiracei di epoca imperiale’, in Atti del XVII Congresso Internazionale di Papirologia. Napoli 19–26 Maggio 1983, II, Naples, 313–17. Gullotta, G. (1955) Gli antichi cataloghi e i codici della abbazia di Nonantola, Vatican City. Gunderson, E.T. (2009) Nox philologiae: Aulus Gellius and the fantasy of the Roman library, Madison. Gurd, S. (2010) ‘Verres and the scene of rewriting’, Phoenix 64.1/2, 80–101. Gutzwiller, K.J. (1996) ‘The Evidence for Theocritean Poetry Books’ in Harder, M.A., Wakker, G.C., Regtuit, R.F. (eds.) Theocritus, Groningen, 119–42. Gutzwiller, K.J. (1998) Poetic Garlands. Hellenistic Epigrams in Context, Berkeley–Los Angeles. Gutzwiller, K.J. (2005) The New Posidippus. A Hellenistic Poetry Book, Oxford, Oxford University Press. Gutzwiller, K.J. (2006) ‘The Bucolic Problem’, CPh 101.4, 380–404. Hadorn, R. (1984) ‘Narziß. Der Mythos als Metapher von Ovid bis heute‘, Würzburg. Hagendahl, H. (1958) Latin Fathers and the Classics. A Study on the Apologists, Jerome and other Christian Writers, Göteborg. Haines–Eitzen, K. (1998) ‘“Girls Trained in Beautiful Writing”: Female Scribes in Roman Antiquity and Early Christianity’, Journal of Early Christian Studies 6.4, 629–46. Haines–Eitzen, K. (2000) Guardians of Letters: Literacy, Power, and the Transmitters of Early Christian Literature, Oxford. Halperin, D. (1983) Before Pastoral. Theocritus and the Ancient Tradition of Bucolic Poetry, Yale. Hansen, M.H. (1975) Einsangelia. The Sovereignty of the People’s Court in Athens in the Fourth Century B.C. and the Impeachment of Generals and Politicians, Odense. Hanson, A.E. (1996) ‘Introduction’, in Youtie (1996) xv–xxv. Hanson, A.E. (1997) ‘Fragmentation and the Greek Medical Writers’, in Most, G.W. (ed.) Collecting Fragments / Fragmente sammeln, Göttingen, 289–314. Hanson, A.E. (1998) ‘Galen: Author and Critic’, in Most, G.W. (ed.) Editing Texts / Texte edieren, Göttingen, 22–52. Hanson, A.E. (2002) ‘Papyrology: A Discipline in Flux’, in Most, G.W. (ed.) Disciplining Classics / Altertumwissenschaft als Beruf, Göttingen, 191–206.

Bibliography  321

Hanson, A.E. (2010) ‘Doctors’ Literacy and Papyri of Medical Content’, in Horstmanshoff, M. (ed.) Hippocrates and Medical Education, Leiden–Boston, 187–204. Hardie, A. (1983) Statius and the Silvae. Poets, patrons and epideixis in the Graeco–Roman world, Leeds. Hardie, P. (2002a) ‘Lucretius and the delusions of Narcissus’, MD 20, 71–89. Hardie, P. (2002b) Ovid’s Poetics of Illusion, Cambridge. Hardie, P., Moore, H. (2010) Classical Literary Careers and their Reception, Cambridge, 197– 208. Hardwick, L., Stray, C. (eds.) (2008) A Companion to Classical Receptions, Oxford. Harland, P.A. (2003) Associations, Synagogues, and Congregations: Claiming a Place in Ancient Mediterranean Society, Minneapolis. Harland, P.A. (2005) ‘Familial Dimensions of Group Identity: “Brothers” (ΑΔΕΛΦΟΙ) in Associations of the Greek East’, JBL 124, 491–513. Harmon, A.M. (1921) Lucian, III, Cambridge–Mass. Harris, E.M. (1985) ‛The Date of the Trial of Timarchus’, Hermes 113, 376–80. Harris, E.M. (2014) The Rule of Law in Action in Democratic Athens, Oxford Harrison, S.J. (1988) ‘Deflating the Odes: Horace, Epistles 1.20’, CQ 38.2, 473–6. Harrison, S.J. (2001) Texts, Ideas and the Classics: Scholarship, Theory and Classical Literature, Oxford. Harrison, S.J. (2007a) Generic Enrichment in Vergil and Horace, Oxford. Harrison, S. J. (2007b) ‘Horatian Self–representations’, in (id.) (ed.) The Cambridge Companion to Horace, Cambridge, 24–35. Haslam. M. (1997) ‛Homeric Papyri and Transmission of the Text’, in Morris, I., Powell, B. (eds.) A new Companion to Homer, Leiden–New York–Köln, 55–100. Haupt, M. (1843) ʻDie Vorrede Albrecht von Halberstadtʼ, Zeitschrift für deutsches Altertum und Literatur 3, 289–92. Heath, T. (19562) The Thirteen Books of Euclid’s Elements, 3 vols., Oxford. Heesakkers C.L. (1997) ‘Hadrianus Junius’, in Chomarat, J., Nativel, C. (eds.) Centuriae Latinae, Geneva, 449–55. Heesakkers C.L., van Miert D. (2010) ‘An Inventory of the Correspondence of Hadrianus Iunius (1511–1575)’, Lias 37, 109–268. Heinzmann, G. (1969) Albrecht von Halberstadt und Jörg Wickram. Studien zu einer Rekonstruktion von Albrechts Metamorphosen, Munich. Helm, K. (1927) ʻHeinrich von Morungen und Albrecht von Halbertstadtʼ, Beiträge zur Geschichte der deutschen Sprache und Literatur 50, 143–5. Hemelrijk, E.A. (2015) Hidden Lives, Public Personae. Women and Civic Life in the Roman West, Oxford. Hexter. R., Selden D.L. (1992) Innovations of antiquity, New York. Hilberg, I. (1912) Sancti Eusebii Hieronymi Epistulae, II, in Corpus Scriptorum Ecclesiasticorum Latinorum, LV, Leipzig. Hiller, E. (1870) ‘Der Πλατωνικός des Eratosthenesʼ, Philologus 30, 60–72. Himmelmann, N. (1975) ‘Das Hypogäum der Aurelier am Viale Manzoni’, in Akademie der Wissenschaften und der Literatur 7, 7–26. Hirschberg, B. (1919) Darstellung der Frau in den Romanen Jörg Wickrams und Untersuchung des kulturgeschichtlichen Wertes der Schilderungen, Diss. Greifswald. Hirzel, R. H. (1963) Der Dialog: ein literarhistorischer Versuch, Hildesheim [or. ed. Leipzig (1895)].

322  Bibliography Hoffmann, H. (1980) Chronica Monasterii Casinensis, in MGH, Scriptores 34, Hanover. Holder, A. Keller, O. (1894) Pomponi Porfyrionis, Scholia antiqua in Q. Horatium Flaccum, Wagner. Holtz, L. (1977) ‘A l'école de Donat, de saint Augustin à Bède’, Latomus 36.2, 522–38. Holwerda, D. (1952) ‘De novo Chamaeleontis studiorum testimonio’, Mnemosyne 5, 228–31. Holwerda, D. (1977) Scholia in Aristophanem, I.31, Scholia vetera in Nubes, Groningen. Houben, H. (1995) ‘La predicazione’, in Musca, G., Sivo, V. (eds.) Strumenti, tempi, luoghi di comunicazione nel Mezzogiorno normanno–svevo, Proceedings of the eleventh Norman– Swabian Days, Bari, 26–29 ottobre 1993, Bari, 253–73. Houghton, H.A.G. (2016) The Latin New Testament: A Guide to its Early History, Texts and Manuscripts, Oxford. Howley, J. (2017) ‘Book–Burning and the Uses of Writing in Ancient Rome: Destructive Practice between Literature and Document’, JRS 107, 213–36. Howley, J. (2018) Aulus Gellius and Roman Reading Culture. Text, Presence, and Imperial Knowledge in the Noctes Atticae, Cambridge. Huffman, C.A. (2005) Archytas of Tarentum, Cambridge. Huffman, C.A. (2014) A History of Pythagoreanism, Cambridge. Hunter, R. (1999) Theocritus: A selection. Idylls 1, 3, 4, 6, 7, 10, 11 and 13, Cambridge. Hutchinson, G.O. (2008) Talking Books: Readings in Hellenistic and Roman Books of Poetry, Oxford. Ihm, M. (1895) Damasi Epigrammata, Leipzig. Immisch, O. (1911) Zu Martial, Hermes 46, 481–517. Improta, A. (2013) ‘Tra sopravvivenza e distruzione: I codici dell’archivio di Stato di Napoli, Bartolomeo Capasso e lo studio della miniatura a Napoli alla fine del XIX secolo’, Archivio storico per le province napoletane 131, 261–308. Irigoin, J. (1975) ‘Les Bucoliques de Theocrite. La Composition du Recueil’, QUCC, 19, 27–44. Irigoin, J. (1996) ‘L’écriture grecque du Codex de Béze’, in Parker, D.C., Amphoux, C.B. (eds.) Codex Bezae: Studies from the Lunel Colloquium, June 1994, Leiden, 3–13. Isnardi Parente, M. (1998) ‘L’Accademia antica e le sue vicende filosofiche’, RSF 53, 215–34. Jacob, A. (2017) ‘Demotic Pharmacology: An Overview of the Medical Manuscripts in the Papyrus Carlsberg Collection’, in Reggiani, N. (ed.) (2018a), 52–79. Jastrzębowska, E. (1979) ‘Les scenes de banquet dans le peintures et sculptures des IIIe et IVe siècles’, RecAug 14, 3–90. Jastrzębowska, E. (1981) Untersuchungen zum christlichen Totenmahl aufgrund der Monumente des 3. Und 4. Jahrhunderts unter der Basilika des Hl. Sebastian in Rom, Frankfurt. Jastrzębowska, E. (2013) ‘Darstellungen der “manumissione vindicta” im Hypogäum der Aurelier in Rom’, Boreas 36, 53–64. Jauss, H.R. (1970) ‘Literary History as a Challenge to Literary Theory’, transl. Bezenger, E., New Literary History 2.1, 7–37. Johannsen, N. (2006) Dichter über ihre Gedichte. Die Prosavorreden in den Epigrammaton libri Martials und in den Silvae des Statius, Göttingen. Johnson, J. (1933) Excavations at Minturnae, II, Inscriptions, part I: Republican Magistri, Rome. Johnson, W. (2010) Readers and Reading Culture in the High Roman Empire: A Study of Elite Communities, Oxford. Jones, C.P. (1986) Culture and society in Lucian, Cambridge–Mass.

Bibliography  323

Joshel, S.R. (1992) Work, Identity, and Legal Status at Rome: A Study of the Occupational Inscriptions, Norman–London. Jouan, F., Van Looy, H. (2000) Euripide, VIII.2, Fragments, Béllerophon–Protésilas, Paris. Jouanna, J. (2012) Greek Medicine from Hippocrates to Galen. Selected Papers, Leiden–Boston. Jung (2006) Marathon und Plataiai: zwei Perserschlachten als “lieux de mémoire” im antiken Griechenland, Göttingen. Junius, H. (1556) Hadriani Iunii Hornani medici Animadversorum libri sex, omnigenae lectionis thesaurus in quibus infiniti pene auctorum loci corriguntur et declarantur, nunc primum et nati et in lucem aediti. Eiusdem De coma commentarium, Basel. Kahlos, M. (1994) ‘Fabia Aconia Paulina and the Death of Praetextatus – Rhetoric and Ideals in Late Antiquity (CIL VI 1779)’, Arctos 28, 13–25. Kahlos, M. (1995) ‘The Restoration Policy of Vettius Agorius Praetextatus’, Arctos 29, 39–47. Kahlos, M. (1997) ‘Vettius Agorius Praetextatus and the Rivalry between the Bishops in Rome in 366–367’, Arctos 31, 41–54. Kahlos, M. (2008) ‘Bene Currunt... The Image of Good Pagans in fourth– and fifth century Christian Literature’, in Alenius, K, Fält, O.K., Mertaniemi, M. (eds.) Imagology and Cross– cultural Encounters in History, in Series Studia Historica Septentrionalia 56, Rovaniemi, 63–69. Kahlos, M. (2009a) ‘Ritus ad solos digitos pertinens (Lact. inst. 5.19.29): A Caricature of Roman civic religion in Lactantius’ Institutiones divinae’, in Brakke, D., Jacobsen, A., Ulrich, J. (eds.) Jews, Christians and Pagans in Antiquity –Critique and Apologetic, Aarhus, 283– 302. Kahlos, M. (2009b) ‘The Rhetoric of Tolerance and Intolerance – From Lactantius to Firmicus Maternus’, in Jacobsen, A., Kahlos, M., Ulrich, J. (eds.) Changes and Continuities in Christian Apologetic, Early Christianity in the Context of Antiquity, Aarhus, 79–95. Kahlos, M. (2010a) ‘Refuting and Reclaiming Monotheism: Monotheism in the Debate between “Pagans” and Christians in 380–430’, in Mitchell, S., Van Nuffelen, P. (eds.) Monotheism between Pagans and Christians in Late Antiquity, Leuven–Walpole–Mass., 167–79. Kahlos, M. (2010b) Vettio Agorio Pretestato. Una vita senatoriale nella transizione, Rende (CS). Kahlos, M. (2011) ‘The Importance of Being a Pagan’, in Brown, P., Lizzi Testa, R. (eds.) Pagans and Christians in the Roman Empire. The Breaking of a Dialogue (IVth–VIth Century A.D), Münster, 187–92. Kaiser–Minn, H. (1981) Die Erschaffung des Menschen auf den spätantiken Monumenten des 3. und 4. Jahrhunderts, Münster. Kartschoke, D. (1982) ʻBald bracht Phebus seinen Wagen… Gattungsgeschichtliche Überlegungen zu Jörg Wickams Nachbarn Romanʼ, Daphnis 11, 717–41. Kaster, R.A. (1988) Guardians of Language: The Grammarian and Society in Late Antiquity, Berkeley. Kelly, J.N.D. (19982) Jerome: His Life, Writings and Controversies, Worcester–London. Kendeffy, G. (2015) Lactantius as Christian Cicero, Cicero as a Shadow–like Instructor, in Altman, W.H.F. (ed.) Brill’s Companion to the Reception of Cicero, Leiden–Boston, 56–92. Kenney, E.J. (1982) The Cambridge History of Classical Literature, II, Latin Literature, Cambridge. Kenyon F.G. (1900) ‘Codex Bezae. Codex Bezae Cantabrigiensis Quattuor Evangelia et Actus Apostolorum complectens Graece et Latine, sumptibus Academiae phototypice repraesentatus (Cantabrigiae, MDCCCXCIX)’, JThS 1.2, 293–9. Kenyon F.G. (19122) Handbook to the Textual Criticism of the New Testament, London.

324  Bibliography Kerdelhue, A. (1991) ʻEconomie, politique et culture dans la Thuringe d’Hermann 1er (1155– 1217)ʼ, in Buschinger, D., Spielwok, W. (eds.) Economie, politique et culture au Moyen Âge: Actes du Colloque Paris, 19 et 20 mai 1990, 103–18, Amiens. Keulen, W. (2008) Gellius the Satirist: Roman Cultural Authority in the Attic Nights, Leiden. Kiessling, T. (1826) Theocritus, Bion et Moschus. Graece et latine, I, London. Kilpatrick, R.S. (1986) The Poetry of Friendship. Horace, Epistles I, Edmonton. Kleiner, D.E.E. (1977) Roman Group Portraiture, New York. Klingner, F. (19593) Q. Horatius Flaccus Opera, Leipzig. Knochenhauer, Th., Menzel, K. (1871) Geschichte Thüringens zur Zeit des ersten Landgrafenhauses (1139–1247), Aalen. Koehler, B. (1767) Notae et emendationes in Theocritum. Accedit specimen emendationum in scriptores arabicos, Lubeck. Köhn, A. (1930) Das weibliche Schönheitideal in der ritterlichen Dichtung, Leipzig. Kokott, H. (1978) Literatur und Herrschaftsbewusstsein. Wertstrukturen der vor–und frühhoefischen Literatur. Vorstudien zur Interpretation mittelhochdeutscher Texte, Frankfurt. Kolb, A., Fugmann, J. (2008) Tod in Rom. Grabinschriften als Spiegel römischen Lebens, Mainz. Körte, A. (1924) ‘Literarische Texte mit Ausschluß der christlichen’, APF 7, 114–60, 225–58. Kranz W. (1961) ‘Sphragis, Ichform und Namensiegel als Eingangs– und Schlußmotiv antiker Dichtung’, RhM 104.1, 3–46 and RhM 104.2, 97–124 Krass, A. (2009) ‘Der zerbrochene Spiegel. Minnesang und Psychoanalyse: Das Narzisslied Heinrichs von Morungen’, in Bronfen, E., Goebel, E. (eds.) Narziss und Eros. Bild oder Text?, Göttingen, 77–100. Krevans, N. (2010) ‘Bookburning and the Poetic Deathbed: The Legacy of Virgil,’ in Hardie, P., Moore, H. (eds.) Classical Literary Careers and their Reception, Cambridge, 197–208. Kutter S. (1955) Celio Secondo Curione. Sein Leben und sein Werk (1503–1569), Basel– Stuttgart. Kwapisz, J. (2013) ‘Were There Hellenistic Riddle Books?’, in Kwapisz, J., Petrain, D., Szymanski, M. (eds.) The Muse at Play, Berlin–Boston, 149–67. La Penna, A. (1997) ‘Un'eco di Meleagro nei Carmina di Orazio?’, Maia 49, 255–6. La Rocca, E., Parisi Presicce, C. (eds.) (2010) Musei Capitolini. Le sculture del Palazzo Nuovo, I, Rome. Labarbe, J. (1949) L’Homère de Platon, Liège. Labourt, J. (1949) Saint Jérôme, Lettres, I, Paris. LaFrance, Y. (1986) Pour interpréter Platon. La Ligne en République VI, 509d–511e. Bilan analytique des études (1804–1984), Montréal–Paris. Lana, I. (1949) La Satira di Sulpicia. Studio critico, testo e traduzione, Turin. Lang, P. (2013) Medicine and Society in Ptolemaic Egypt, Leiden–Boston. Lanni, A.M. (1997) ‛Spectator Sports or Serious Politics? οἱ περιεστηκότες and the Athenian Lawcourt’, JHS 117, 183–9. Lape, S. (2006) ‛The psychology of prostitution in Aeschines’ speech against Timarchus’, in Faraone, C.A., McClure, L.K. (eds.) Prostitutes and courtesans in the ancient world, Madison, 139–60. Lapini, W. (1994) ‘Noterella su Orazio (epist. 1, 20)’, InvLuc 15–16, 147–58. Lasserre, F. (1975) ‘L'èlègie de l'huître (P. Louvre inv. 7733v)’, QUCC, 145–76. Last, M. (1966) ʻNeue Oldenburger Fragmente der Metamorphosen–Übertragung des Albrechts von Halberstadtʼ, Oldenburger Jahrbuch 65, 41–60.

Bibliography  325

Lawall, G. (1967) Theocritus’ Coan Pastorals. A Poetry Book, Washington D.C. Leclercq, J. (1957) ‘Un ancien catalogue des manuscrits de Font Avellane’, RBen 67, 168–72. Lee, G. (1990) The Poems of Catullus, Oxford. Leggeri, A. (2018) ‘Dalla cacografia del medico alla ricetta elettronica’, in Reggiani, N. (ed.) (2018a), 406–10. Legrand, Ph. (19463) Bucoliques Grecs, I, Paris. Lehmann, E.L. (1931) Antike Martialusgaben, Diss. Jena. Leone, D. (1972) ‛Appunti per la storia del testo di Eschine’, AFLM 5/6, 12–43. Leverkus, W. (1859) ʻAlbrecht von Halberstadt: Fragmenteʼ, Zeitschrift für deutsches Altertum und Literatur 11, 358–74. Lewis, C., Short, C. (1879) A Latin Dictionary, Oxford. Lindsay, M.W. (1903) The ancient editions of Martial, Oxford. Lizzi Testa, R. (2009) ‘Legislazione imperiale e reazione pagana. I limiti del conflitto’, CrSt 31, 385–409. Lloyd–Jones, H. (2003) ‘All by Posisippus?’, in Accorinti, D., Chuvin, P. (eds.) Des gèants à Dionysos: Mélanges de mythologie et de poésie grecques offerts à Francis Vian, Alessandria, 277–80. Lloyd–Jones, H., Parsons, P. (1983) Supplementum Hellenisticum, Berlin–New York. Lougovaya, J. (2012) ‘New Epigrams’, in Ast, R., Cuvigny, H., Hickey, T.M., Lougovaya, J. (eds.), Papyrological Texts in Honor of Roger S. Bagnall, Durham 187–90. Lowe, A.E. (1925) ‘Some Facts about Our Oldest Latin Manuscripts’, CQ 19.3/4, 197–208. Lowe, A.E. (1928) ‘More Facts about Our Oldest Latin Manuscripts’, CQ 22.1, 43–62. Lowe, A.E. (1950) Codices Latini Antiquiores: A Palaeographical Guide to Latin Manuscripts prior to the Ninth Century, Oxford. Lowrie, M. (1997) Horace’s Narrative Odes, Oxford. Lübben, A. (1865) ʻNeues Bruchstück von Albrecht von Halberstadtʼ, Die Deutsche Vierteljahrsschrift für Literaturwissenschaft und Geistesgeschichte 10, 237–45. Luebeck, A. (1872) Hieronymus quos noverit scriptores et ex quibus hauserit, Leipzig. Lundon, J. (2004) ‘P. Oxy VIII 1088: problemi e proposte’, in Andorlini (ed.) (2004) 119–30. Luppe, W. (2002) ‘EinLeipzigerEpigramm–Papyrus (P. Lips. inv. 1445 verso). Mit Tafeln XXII– XXIV’, APF 48, 197–206. MacCormack, S. (2013) ‘Cicero in Late Antiquity’, in Steel, C. (ed.) The Cambridge Companion to Cicero, Cambridge, 251–305. MacDonald, C. (1995) Cicero Pro Murena, Edited with introduction, notes and vocabulary, London. Machado, C. (2011) ‘Roman Aristocrats and the Christianization of Rome’, in Brown, P., Lizzi Testa, R. (eds.) Pagans and Christians in the Roman Empire. The Breaking of a Dialogue (IVth–VIth Century A.D), Münster, 493–516. Macleod, M.D. (1974) Luciani opera, Oxford. Maehler, H. (2000) ‘L’évolution matérielle de l’hypomnèma jusqu’à la basse époque: le cas du POxy. 856 (Aristophane) et PWürzburg 1 (Euripide)’, in Goulet–Cazé, M.O. (ed.) Le commentaire entre tradition et innovation, Paris, 29–36. Malcovati, E. (1977) ‘Licurgo. Orazione contro Leocrate e frammenti’, in Marzi, M., Leone, P., Malcovati, E. (eds.) Oratori attici minori, Iperide. Eschine. Licurgo, I, Turin. Malik, P. (2017) P.Beatty III (P47): The Codex, Its Scribe, and Its Text, Leiden. Mander, J. (2013) Portraits of Children on Roman Funerary Monuments, Cambridge. Manetti, D., Roselli, A. (1982) Ippocrate, Epidemie, Libro sesto, Florence.

326  Bibliography Maravela, A. (2010) ‘Vina Fictitia from Latin into Greek: The Evidence of the Papyri’, in Evans, T.V., Obbink, D.D. (eds.) The Language of the Papyri, Oxford, 253–66. Marcone, A. (ed.) (2006) Medicina e società nel mondo antico, Atti del convegno di Udine (2005), Florence. Marconi, G. (ed.) (2008) Il fanciullo antico. Soggetto tra formazione e religio, Atti della Giornata di studio, Isernia, 14 Novembre (2007), Alessandria. Marganne, M.–H. (1980) ‘Une étape dans la transmission d’une préscription médicale: P. Berlin Möller 13’, in Pintaudi, R. (ed.) Miscellanea papyrologica, Florence, 179–83. Marganne, M.–H. (1981) Inventaire analytique des papyrus grecs de médecine, Geneva. Marganne, M.–H. (1984) ‘La ‘collection médicale’ d’Antinoopolis’, ZPE 56, 117–21. Marganne, M.–H. (2004) Le livre medical dans le monde gréco–romain, Liège. Marganne, M.–H. (2007) ‘La terminologie de la librairie dans la Collection Hippocratique’, in Boudon–Millot, V., Guardasole, A., Magdelaine, C. (eds.) La science médicale antique. Nouveaux regards. Études reunites en l’honneur de Jacques Jouanna, Paris, 315–35. Marincola, J. (2006) ‛Herodotus and the poetry of the past’, in Dewald, C., Marincola, J. (eds.) The Cambridge Companion to Herodotus, Cambridge, 13–28. Marolla, G. (2017) ‘Presenza di classici in Girolamo, epist. 123 a Geruchia’, VetChr 54, 127–41. Marshall, P.K. (1968) Noctes Atticae, Oxford. Martano, A. (2012) ‘Chamaeleon of Heracles Pontica: The Sources, Text and Translation’, in (id.), Matelli, E., Mirhady, D. (eds.) Praxiphanes of Mytilene and Chamaeleon of Heraclea. Text, translation, and discussion, New Brunswick, 157–338. Martin, V., De Budé, G. (19522) Eschine. Discours, I, Contre Timarque. Sur l’ambasserie infidéle, Paris. Martindale, Ch., Thomas, R. (2006) Classics and the Uses of Reception, Oxford. Martinez, J.–L. (1998) ‘Reliefs et inscriptions trouvés avec la tombe de Patron’, in Blanc, N. (ed.) Au Royaume des Ombres. La peinture funéraire antique, IVe siècle avant J.–C., IVe siècle après J.–C., Paris, 85–95. Martínez–Maza, C. (2003) ‘Devotas mistéricas en la Roma bajoimperial’, Aevum 78, 53–68. Marzullo, B. (1953) ‛Aeschines, in Tim. 128’, Maia 6, 68–75. Massa Positano, L. (1960) Jo. Tzetzae commentarii in Aristophanem, Groningen–Amsterdam. Massaro, M. (2009) ‘Una terza via: epigrafia e letteratura in parallelo (l'Alcesti di Euripide e i CLE)’, in Gómez Font, X., Fernández Martínez, C., Gómez Pallarès, J. (eds.) Literatura epigráfica. Estudios dedicados a Gabriel Sanders, Zaragoza, 225–51. Massaro, M. (2013a) ‘L’impaginazione delle iscrizioni latine metriche e affettive’, in RPPA 85, 365–413. Massaro, M. (2013b) ‘Radici orali di convergenze tra epigrafia e letteratura nel linguaggio funerario (poetico o affettivo)’, in Fernández Martínez, C., Limón Belén, M., Gómez Pallarès, J., del Hoyo Calleja, J. (eds.) Ex officina. Literatura epigráfica en verso, Sevilla, 253–74. Massaro, M. (2014) ‘Te, lapis, obtestor...: Le vicende di un distico sepolcrale’, in Pistellato, A. (ed.) Memoria poetica e poesia della memoria. La versificazione epigrafica dall’antichità all’umanesimo, Venice, 65–102. Mathiesen, T.J. (1999) Apollo’s lyre: Greek Music and Music Theory in Antiquity and the Middle Ages, Lincoln–London. Matthaios, S. (2011) ‘Eratosthenes of Cyrene: Readings of his ‘Grammar’ Definition’, in (id.), Montanari, F., Rengakos, A. (eds.) Ancient Scholarship and Grammar, Berlin–New York, 55–85.

Bibliography  327

Mayer, R.G. (1994) Horace: Epistles 1, Cambridge. Mayer, R.G. (2012) Horace: Odes 1, Cambridge. McCutcheon, R.W. (2015) ‘Silent Reading in Antiquity and the Future History of the book’, Book history 18, 1–32. McDonnell, M. (1996) ‘Writing, Copying, and Autograph Manuscripts in Ancient Rome’, CQ 146, 469–91. McGann, M.J. (1963) ‘Vinnius Valens, Son of Vinnius Asina?’, CQ 13, 258–9. McGill, S. (2012) Plagiarism in Latin Literature, Cambridge. Meersseman, G.G. (1965) ‘Eremitismo e predicazione itinerante dei secoii XI e XII’, in Violante, C. (ed.) L’eremitismo in Occidente nei secoli XI e XII, Proceedings of the Second International Study Week, Mendola, 30 agosto–6 settembre 1962, Milan, 164–81. Meersseman, G.G. (1968) ‘I penitenti nei secoli XI e XII’, in I laici nella ‘societas Christiana’ dei secoli XI e XII, Proceedings of the Second International Study Week, Mendola, 21–27 agosto 1965, Milan, 306–39. Melmoth, W. (1957) Pliny. Letters, Cambridge–Mass. Mendels, J., Spuler, L. (1959) ʻLandgraf Hermann von Thüringen und seine Dichterschuleʼ, Die Deutsche Vierteljahrsschrift für Literaturwissenschaft und Geistesgeschichte 33, 361–88. Meneghini, R. (2010) ‘Le biblioteche pubbliche di Roma nell’alto impero', in Perrin, Y. (ed.) Neronia VIII: Bibliothèques, livres et culture écrite dans l’empire romain de César à Hadrien, Actes du VIIIe Colloque international de la SIEN (Paris, 2–4 Octobre (2008), Bruxelles, 32–40. Merli, E. (2006a) ‘Identity and Irony. Martial’s Tenth Book, Horace, and the Tradition of Roman Satire’, in Nauta (2006), 257–70. Merli, E. (2006b) ‘Martial between Rome and Bilbilis’, in Rosen, R.M., Sluiter, I. (eds.) City, Countryside, and the Spatial Organization of Value in Classical Antiquity, Leiden–Boston, 327–47. Metzger, B.M. (19942) A Textual Commentary on the Greek New Testament, Stuttgart. Metzger, B.M., Ehrman, B.D. (20052) The Text of the New Testament: Its Transmission, Corruption, and Restoration, Oxford. Micaelli, C. (1979) L’influsso di Tertulliano su Girolamo. Le opere sul matrimonio e le seconde nozze, Augustinianum 19, Rome, 415–29. Michelakis, P. (2002) Achilles in Greek Tragedy, Cambridge. Michie, J. (1965) The Odes of Horace, the Centennial Hymn, New York. Milne, H.J.M. (1927) Cataloghi of the Literary Papyri in the British Museum, London. Molinos Tejada, M.T. (1990) Los dorismos del Corpus bucolicum, Amsterdam. Monaco, M. (2000) ‛La tradizione manoscritta di Eschine tra i papiri e i codici medievali’, Aegyptus 1.2, 3–98. Monda, S. (2008) ‘Personalità del bambino e rappresentazione dall'infanzia in Roma antica’, in Marconi (2008), 125–36. Monda, S. (2012) Ainigma e Γρῖφος. Gli antichi e l’oscurità della parola, Pisa. Mongelli, G. (ed.) (1956–1962) Abbazia di Montevergine. Il regesto delle pergamene, Rome. Mongelli, G. (1959) I codici dell’Abbazia di Montevergine, Montevergine. Mongelli, G. (1960) ‘La spiritualità di San Guglielmo da Vercelli di fronte a quella cluniacense’, in Spiritualità cluniacense, Atti del II Convegno storico internazionale (Todi, 12–15 ottobre 1958), 286–307. Mongelli, G. (1962) L’archivio dell’abbazia di Montevergine, Rome.

328  Bibliography Mongelli, G. (1963) Montevergine come potenza economica nei primi secoli della sua storia, Avellino. Mongelli, G. (1965) Storia di Montevergine e della Congregazione verginiana, Avellino. Mongelli, G. (1971) ‘L’origine benedettina della congregazione verginiana’, RBen 81, 259–81. Mongelli, G. (1972) ‘Il monastero verginiano dei SS. Filippo e Giacomo in Benevento’, Samnium 45, 9–49. Montana, F. (2006) ‘L’anello mancante: l’esegesi ad Aristofane tra l’antichità e Bisanzio’, in Avezzù, G., Scattolin, P. (eds.) I classici greci e i loro commentatori, Rovereto, 17–34. Montana, F. (2011) ‘The Making of Greek Scholiastic Corpora’, in Montanari, F., Pagani, L. (eds.) From Scholars to Scholia. Chapters in the History of Ancient Greek Scholarship, Berlin– New York, 105–89. Montana, F. (2012) ‘P. Oxy. XXXV 2737’, in Bastianini, G. et al. (20122), 157–82. Montana, F. (2015) ‘Hellenistic Scholarship’, in Montanari, F., Matthaios, S., Rengakos, A. (2014) 60–183. Montanari, F. (1989) ‘Chamaeleon’, in Corpus dei papiri filosofici greci e latini, I.1, Florence, 403–18. Montanari, F. (1997) ‘Dionysios Musikos’, in DNP, III, 638–9. Montanari, F. (1998) ‘Zenodotus, Aristarchus and the Ekdosis of Homer’, in Most, G.W. (ed.) Editing texts = Texte edieren, Göttingen, 1–21. Montanari F., Matthaios, S., Rengakos A. (eds.) (2014) Brill’s Companion to Ancient Greek Scholarship, 2 vols., Leiden. Montanari, F., Pagani, L. (2011) From Scholars to Scholia: Chapters in the History of Ancient Greek Scholarship, Berlin. Montepaone, O. (2016) ‘Apocolocyntosis Harlemensis: un codice Senecano tra XV e XVII secolo’, QS 84, 207–45. Montes Cala, J.G. (2013) ‘Poetología y metapoesía en Teócrito: sobre el motivo de los cantores itinerantes en el Idilio VII’, Euphrosyne 41, 63–83. Montfaucon de, B. (1708) Palaeographia Graeca, Paris. Moog–Grünewald, M. (1979) Metamorphosen der “Metamorphosen”: Rezeptionsarten der ovidischen Verwandlungsgeschichten in Italien und Frankreich im 16. und 17. Jahrhundert, Heidelberg. Mordine, M.J. (2010) ‘Sine me, liber, ibis: the poet, the book and the reader in Tristia 1.1’, CQ 60. 2, 524–44. Morelli, A.M. (1994) ‘Sul papiro di Ossirinco LIV 3723. Considerazioni sui caratteri dell'elegia erotica ellenistica alla luce dei nuovi ritrovamenti papiracei’, RFIC 122, 385–421. Morelli, A.M. (ed.) (2008) Epigramma longum. Da Marziale alla tarda antichità. From Martial to Late Antiquity, Atti del convegno internazionale, Cassino 29–31 maggio 2006, Cassino. Moulton, J.H. (1906) A Grammar of New Testament Greek, 2 vols., Edinburgh. Munari, F. (1960) Ovid im Mittelalter, Stuttgart–Zurich. Muratori, L.A. (1740) Antiquitates italicae medii aevi, Milan. Nauta, R.R. (2002) Poetry for Patrons: Literary Communication in the Age of Domitian, Leiden. Nazzaro, A.V. (1979) ‘Figure di donne cristiane: la vedova’, in Uglione, R. (ed.) Atti del II Convegno nazionale di studi su: la donna nel mondo antico, Torino 18–19–20 aprile 1988, Turin, 197–219. Nestle, E. (1901) Textual Criticism of the Greek New Testament, transl. Edie, W., New York. Newlands, C. (2009) ‘Statius’ Prose Prefaces’, MD 61, 229–56. Nicholls, M. (2010) ‘Parchment Codices in a New Text of Galen’, G&R 57, 378–86.

Bibliography  329

Nielsen, I. (2014) Housing the Chosen: The Architectural Context of Mystery Groups and Religious Association in the Ancient World, Turnhout. Nisbet, G. (2013) Greek Epigram in the Roman Empire: Martial’s Forgotten Rivals, Oxford. Nisbet, R.G.M. (1959) ‘Notes on Horace, Epistles 1’, CQ 9, 73–76. Nisbet, R.G.M. (1995) Collected Papers on Latin Literature, Oxford. Nisbet, R.G.M., Hubbard, M.E. (1970) A Commentary on Horace’s Odes: Book I, Oxford. Nobili, M. (2008) ‘Rus, seu potius domus. Note critiche agli epigrammi di Marziale a Giulio Marziale (4, 74; 7, 17)’, in Morelli (2008), 327–71. North, H. (1952) ‛The Use of Poetry in the Training of the Ancient Orators’, Traditio 8, 1–33. Notargiacomo, S. (2009) Medietà e proporzione: due concetti matematici e il loro uso da parte di Aristotele, Milan. Nünlist, R. (1998) Poetologische Bildersprache in der frühgriechischen Dichtung, Stuttgart– Leipzig. Nünlist, R. (2009) The Ancient Critic at Work, Cambridge. Nutton, V. (1995) ‘The Medical Meeting Place’, in van der Eijk, P.J., Horstmanshoff, H.F.J., Schrijvers, P.H. (eds.) Ancient Medicine in Its Socio–Cultural Context, Amsterdam–Atlanta, I, 3–26. Ober, J., Strauss, J.B. (1990) ‛Drama, Political Rhetoric and the Discourse of Athenian Democracy’, in Winkler, J.J., Zeitlin, I. (eds.) Nothing to Do with Dionysos? Athenian Drama in its Social Context, Princeton, 237–70. Oikonomides, A.N. (1974) Abbreviations in Greek: Inscriptions, Papyri, Manuscripts, and Early Printed Books, Chicago. Olding, G. (2007) ‛Myth and writing in Aeschines’ Against Timarchus’, in Cooper, C. (ed.) Politics of Orality, Orality and Literacy in Ancient Greece, VI, Leiden–Boston, 15–170. Oliensis, E. (1995) ‘Life after Publication: Horace, Epistles 1.20’, Arethusa 28.2, 209–24. Oliensis, E. (1998) Horace and the Rhetoric of Authority, Cambridge. Olson, K. (2008a) Dress and the Roman Woman, London. Olson, K. (2008b) ‘The Appearance of the Young Roman Girl’, in Edmondson, J., Keith, A. (eds.) Roman Dress and the Fabrics of Roman Culture, Toronto, 139–57. Orlandi, S. (2011) ‘Gli ultimi sacerdoti pagani di Roma: analisi della documentazione epigrafica’, in Brown, P., Lizzi Testa, R. (eds.) Pagans and Christians in the Roman Empire. The Breaking of a Dialogue (IVth–VIth Century A.D), Münster, 425–66. Orlowsky R., Orlowsky U. (1992) Narziß und Narzißmus im Spiegel von Literatur, Kunst und Psychoanalyse: vom Mythos zur leeren Selbstinszenierung, Munich. Otranto, R. (2000) Antiche liste di libri su papiro, Rome. Page, D. (1962) Poetae Melici Graeci, Oxford. Panarelli, F. (2004) Scrittura agiografica nel mezzogiorno normanno. La vita di san Guglielmo da Vercelli, Galatina. Panarelli, F. (2008) ‘Il mondo monastico e Federico II: il caso di Montevergine’, in Houben, H., Vogeler, G. (eds.) Federico II nel Regno di Sicilia Realtà locali e aspirazioni universali, Atti del Convegno internazionale di studi, Barletta, 19–20 ottobre 2007, Bari, 189–220. Pani, L.E. (ed.) (2015) Teoria e pratica del lavoro nel monachesimo altomedievale, Atti del Convegno internazionale di studio (Roma–Subiaco, 7–9 giugno 2013), Spoleto. Parker, D.C. (1992) Codex Bezae: An Early Christian Manuscript and Its text, Cambridge. Parker, D.C (2008) An Introduction to the New Testament Manuscripts and Their Texts, Cambridge.

330  Bibliography Parker, D.C. (2009) review of Scribal Habits in Early Greek New Testament Papyri by Royse, J.R., BASP 46, 255–8. Parker, D.C. (2012) Textual Scholarship and the Making of the New Testament, Oxford. Parker, L.P.E. (1997) The Songs of Aristophanes, Oxford. Parpulov, G.R. (2012) ‘The Bibles of the Christian East’, in Marsden, R., Matter, E.A. (eds.) The New Cambridge History of the Bible, II, From 600–1450, Cambridge. Parroni, P. (1984) ‘Nostalgia di Roma nell’ultimo Marziale’, Vichiana 13, 126–34. Pasquali G. (1949), ‘Biblioteche’, in Enciclopedia italiana di scienze, lettere ed arti, VI, Rome 942–7. Pasquali, G. (19522) Storia della tradizione e critica del testo, Florence. Paul, H. (1843) ʻDie Vorrede Albrechts von Halberstadtʼ, Zeitschrift für deutsches Altertum und Literatur 3, 289–91. Payne, M. (2001) ‘Ecphrasis and Song in Theochritus’ Idyll I’, GRBS 42, 263–87. Pearcy, L.T. (1994) ‘The Personification of the Text and Augustan Poetics in Epistles 1.20’, CW 87.5, 457–64. Pease, A.S. (1919) ‘The Attitude of Jerome towards Pagan Literature’, TAPhA 50, 150–67. Pecere, O. (2010) Roma antica e il testo. Scritture d’autore e composizione letteraria, Rome– Bari. Peirano, I. (2014) ‘Sealing the Book, the Sphragis as Paratext’, in Jansen, L. (ed.) The Roman Paratext, Frame, Texts, Readers, Cambridge, 224–42. Pellizer, E. (1984) ʻL’eco, lo specchio e la reciprocità amorosa. Una lettura del mito di Narcisoʼ, QUCC 46, 21–35. Penco, G. (1983) Storia del monachesimo in Italia: Dalle origini alla fine del medioevo, Milan. Pennacchini, L.E. (1937) Elementi di Archivistica: nozioni raccolte ad uso delle scuole di paleografia, diplomatica ed archivistica dei RR. Archivi di stato e di concorsi nelle varie amministrazioni pubbliche, Rome. Pergola, A. (2011) ‘Il quadrante delle interpretazioni’, in Bisconti, F. (2011), 81–125. Perilli, L. (2005) ‘«Il dio ha evidentemente studiato medicina». Libri di medicina nelle biblioteche antiche: il caso dei santuari di Asclepio’, in Naso, A. (ed.), Stranieri e non cittadini nei santuari greci, Atti del Convegno internazionale, Florence, 427–510. Perilli, L. (2006) ‘Asclepio e Ippocrate, una fruttuosa collaborazione’, in Marcone, A. (ed.) (2006) 26–54. Perilli, L. (2007) ‘Conservazione dei testi e circolazione della conoscenza in Grecia’, in Andrisano, A.M. (ed.) Biblioteche del mondo antico. Dalla tradizione orale alla cultura dell’Impero, Rome, 36–71. Perilli, L. (2017) Galeni vocum Hippocratis glossarium / Galeno. Interpretazione delle parole difficili di Ippocrate, Berlin. Perlmann, S. (1964) ‛Quotation from Poetry in the Attic Orators of the Fourth Century’, AJPh 85, 155–72. Perrone, S. (ed.) (2009) Commentaria et Lexica Graeca in Papyris reperta, II.4, Comoedia et mimus, Berlin. Perrone, S. (2010) ‘Paralleli comici nell’esegesi a commedia su papiro’, in Montana, F. (ed.) Aner Polytropos. Ricerche di filologia greca antica dedicate dagli allievi a Franco Montanari, Rome, 85–103. Petermandl, W. (1997) ‘Kinderarbeit im Italien der Prinzipatszeit. Ein Beitrag zur Sozialgeschichte des Kindes’, Laverna 8, 113–36.

Bibliography  331

Peters, U. (1981) Fürstenhof und höfische Dichtung. Der Hof Hermanns von Thüringen als literarisches Zentrum, Konstanz. Petersen, L.H. (2006) The Freedman in Roman Art and History, Cambridge. Petrie, A. (1922) Lycurgus: The Speech Against Leocrates, Cambridge. Petrucci, F.M. (2012) Teone di Smirne. Expositio rerum mathematicarum ad legendum Platonem utilium, Sankt Augustin. Pfeiffer, R. (1968) History of Classical Scholarship from the Beginnings to the End of the Hellenistic Age, Oxford. Pfuhl, E., Möbius, H. (1977–1979) Die Ostgriechischen Grabreliefs, Mainz. Pharr, C., Sherrer Davidson, T., Brown Pharr, M. (1952) The Theodosian Code and Novels and the Sirmondian Constitutions, a Translation with Commentary, Glossary, and Bibliography by Pharr, C. in collaboration with Sherrer Davidson, T. and Brown Pharr, M., Princeton. Pighi, G.B. (1944) ‘Commentariolus electorum ad Aurelium Iosephum Amatucci’, Aevum 18, 14– 51. Pinto, P.M. (2003) Per la storia del testo di Isocrate. La testimonianza d’autore, Bari. Poe, A. C. (2007) The Third–Century Mausoleum (“Hypogeum”) of the Aurelii in Rome: Pagan or Mixed–Religion Collegium Tomb, Diss. Rutgers University. Pommerening, T. (2018) ‘„Wer weiß was?” Heilkundliches Wissen und Wissenstransfer zur Zeit der Pharaonen’, in Reggiani, N. (ed.) (2018a). Pordomingo, F. (1994) ‘Sur les premières anthologies d'épigrammes sur papyrus’, in Bülow– Jacobsen, A. (ed.) Proceedings of the 20th International Congress of Papyrologists. Copenhagen 22–29 Augusto 1992, Copenhagen, 326–31. Pordomingo, F. (2001) ‘El epigramma griego en la scuela’, in González Castro, J.F., Vidal, J.L. (eds.) Acta del X Congreso Espanol de Estudios Clásicos, III, Historia antigua, Humanismo, tradiciòn clásica, Didáctica e Instrumenta Studiorum, Madrid, 707–20. Pordomingo, F. (2007) ‘Vers une caractérisation des anthologies sur papyrus’, in Palme, B. (ed.) Akten des 23. Internationalen Papyrologen Kongresses, Wien, 22–28 Juli 2001, Vienna, 549–57 Pordomingo, F. (2013) Antologiás de época helenística en papiro, Florence. Porena, P. (2005) ‘Problemi di cronologia costantiniana. L’imperatore, Vettius Rufinus e il senato’, AnTard 13, 205–46. Pòrtulas, J. (2012) ‘Cult Poetry in Archaic Greece’, in (id.), Riu, X. (eds.) Approaches to Archaic Greek Poetry, Messina, 221–48. Pratesi, A. (1989) ‘Divagazioni di un diplomatista sul Codice Diplomatico Verginiano’, in Tropeano, M.P. (ed.) La società meridionale nelle pergamene di Montevergine. I Normanni chiamo gli Svevi, Montevergine, 11–42. Price, L. (2012) How to Do Things with Books in Victorian Britain, Princeton. Puelma, M. (1997) ‘Epigramma: osservazioni sulla storia di un termine greco–latino’, Maia 49, 189–213. Quaglia, R. (2007) ‘Presenze di Omero nei frammenti della commedia antica’, Maia 59, 239– 62. Raffa, M. (2013) ‘The Debate on logos and diāstema in Porphyry’s Commentary on Ptolemy’s Harmonics’, Greek and Roman Musical Studies 1, 243–52. Rea, R. (ed.) (2004) L’ipogeo di Trebio Giusto sulla Via Latina. Scavi e Restauri, Vatican City Readeon, B.P. (1991) The form of Greek romance, Princeton. Reardon, B.P. (1971) Courants littéraires grecs des IIe et IIIe siècles après J.C., Paris.

332  Bibliography Rebenich, S. (1992) Hieronymus und sein Kreis. Prosopographische und sozialgeschichtliche Untersuchungen, Stuttgart. Rebenich, S. (2002) Jerome. The Early Church Fathers, London–New York. Rebenich, S. (2009) ‘Christian Asceticism and Barbarian Incursion: The Making of a Christian Catastrophe’, Journal of Late Antiquity 2.1, 49–59. Reed, J.D. (2010) ‘Idyll 6 and the Development of Bucolic after Theocritus’, in Clauss, J.J., Cuypers, M. (eds.) A Companion to Hellenistic Literature, Malden–Oxford, 238–50. Reggiani N. (2010) ‘Dalla magia alla filologia: documenti su libri e biblioteche nell’Antichità’, Papyrotheke 1, 97–135. Reggiani, N. (2015) ‘Le unità di misura dei liquidi nei papiri: questioni di varietà, astrazione, uso’, MBAH 33, 131–56. Reggiani, N. (ed.) (2016a) Medica–Papyrologica. Specimina di ricerca, Parma. Reggiani, N. (2016b) ‘Tra ‘sapere’ e ‘saper fare’: il problema della standardizzazione delle unità di misura dei liquidi nella testimonianza dei papiri greci d’Egitto’, in Reggiani (ed.) (2016a) 107–46. Reggiani, N. (2017) Digital Papyrology I: Methods, Tools and Trends, Berlin–Boston. Reggiani, N., (ed.) (2018a) Parlare la medicina: fra lingue e culture, nello spazio e nel tempo, Atti del Convegno Internazionale (Parma, 5–7 Settembre 2016), con la collaborazione di Bertonazzi, F., Florence. Reggiani, N. (2018b) ‘Prescrizioni mediche e supporti materiali nell’Antichità’, in Reggiani, N. (ed.) (2018a), 128–44. Reggiani, N. (2018c) ‘The Digital Corpus of the Greek Medical Papyri, or New Ways of Conceiving Digital Editions of Papyri’, in (id.) (ed.) (2018) Digital Papyrology II: Case Studies on the Digital Edition of Ancient Greek Papyri, Berlin–Boston, 3–61. Reggiani, N. (2019a) ‘Ancient Doctors’ Literacies and the Digital Edition of Papyri of Medical Content’, Classics@ 15, forthcoming. Reggiani, N. (2019b) ‘The Corpus of Greek Medical Papyri Online and the Digital Edition of Ancient Documents’, in Nodar, A., Torallas, S. (eds.) Proceedings of the 28th International Congress of Papyrology (Barcelona 2016), Barcelona, forthcoming. Regt, W.M.C. (1930) ‘Johan van Cuyck’, in Nieuw Nederlandsch Biografisch Woordenboek 8, 353. Reynolds, L.D., Wilson, N.G. (19873) Copisti e Filologi. La tradizione dei classici dall’antichità ai tempi moderni, Padua. Ricci, C., Salvadori, M. (2008) ‘Eucharis, virgo docta et erudita. Puellae di spettacolo a Roma, tra educazione e formazione’, in Marconi (2008), 145–66. Ritschl, F. (1843) ‘Mittheilungen aus und über Handschriften’, RhM 1, 130–43. Rizzo, S. (1973) Il lessico filologico degli umanisti, Rome. Robbins, E. (1999a) ‘Lamprokles’, in DNP, VI, 1088. Robbins, E. (1999b) ‘Kydias’, in DNP, VI, 957. Robert, L. (1938) Études d’Epigraphie et Philologie, Paris. Roberts, C.H., Skeat, T.C. (1987) The Birth of the Codex, London–Oxford. Roberts, P. (2013) Life and Death in Pompeii and Herculaneum, London. Rocconi, E. (2010) ‘Dionysius [12] Musicus’, in Lexicon of Greek Grammarians of Antiquity (LGGA). Rolfe, J.C. (1927) Gellius: Attic Nights, Cambridge–Mass. Rolfe, J.C. (1940) Ammiani Marcellini rerum gestarum libri quae supersunt, transl. Rolfe, J.C., II, Aberdeen.

Bibliography  333

Rolfe, J. C. (1970) Suetonius: in Two Volumes with an English Translation, Cambridge. Roman, L. (2006) ‘A history of lost tablets’, ClAnt 25.2, 351–88. Roncali, R. (1969) ‛Lista dei manoscritti di Eschine, Licurgo, Lisia’, AFLB 14, 381–403. Roncali, R. (1990) L. Annaei Senecae Apocolocyntosis, Leipzig. Rosati, G. (1983) Narciso e Pigmalione: illusione e spettacolo nelle Metamorfosi di Ovidio, Florence. Ross, W.D. (1961) Aristotle, De Anima, Oxford. Rossbach, O. (1888) ‘De Senecae philosophi librorum recensione et emendatione’, Breslauer philologische Abhandlungen 2.3, 1–96. Rossi, L. (2001) The Epigrams Ascribed to Theocritus. A Method Approach, Leuven. Rossi, L.E. (2000) ‘La letteratura alessandrina e il rinovamento dei generi letterari della tradizione’, in Pretagostini, R. (ed.) La letteratura ellenistica. Problemi e prospettive di Ricerca, Atti del Colloquio Internazionale, Università di Roma «Tor Vergata», Roma 29–30 aprile 1997, Rome, 149–61. Rotili, M. (1989) ‘Insediamenti verginiani nel Sannio’, in Tropeano, M.P. (ed.) La società meridionale nelle pergamene di Montevergine. I Normanni chiamano gli Svevi, Montevergine, 221–34. Rowe, G.O. (1972) ‛A Problem of Quotations in Demosthenes’ Embassy Speech’, TAPhA 103, 441–49. Royden, H.L. (1988) The Magistrates of the Roman Professional Collegia in Italy from the First to the Third Century A.D., Pisa. Royse, J.R. (1995) ‘Scribal Tendencies in the Transmission of the Text of the New Testament’, in Ehrman, B.D., Holmes, M.W. (eds.) The Text of the New Testament in Contemporary Research: Essays on the Status Quaestionis, Grand Rapids, 239–52 = Royse, J.R. (20132) ‘Scribal Tendencies in the Transmission of the Text of the New Testament’, in Ehrman, B.D., Holmes, M.W. (eds.) The Text of the New Testament in Contemporary Research: Essays on the Status Quaestionis, Leiden, 461–78. Royse, J.R. (2007) Scribal Habits in Early Greek New Testament Papyri, Atlanta. Royse, J.R. (2008) ‘The Early Text of Paul (and Hebrews)’ in Hill, C.E., Kruger, M.J. (eds.) The Early Text of the New Testament, Oxford, 177–203. Rücker, B. (1997) Die Bearbeitung von Ovids Metamorphosen durch Albrecht von Halberstadt und Jörg Wickram und ihre Kommentierung durch Gerhard Lorichius, Göppingen. Runge O. (1908) Die Metamorphosen–Verdeutschung Albrechts von Halberstadt, Berlin. Sabbadini R. (1919) ‘Il testo interpolato del Ludus di Seneca’, RIFC 47, 338–45. Saenger, P. (1982) ‘Silent Reading: Its Impact on Late Medieval Script and Society’, Viator 13, 367–414. Saibene, M.G. (2011) ʻHermann di Turingia, politico e mecenate, alla luce della poesia corteseʼ, Filologia Germanica 3, 225–56. Salgado, O. (2002) ‘France and the Transmission of Latin Manuscripts’, in Sandy, G. (ed.) The Classical Heritage of France, Leiden, 25–46. Salomone Gaggero, E. (1981) Floro, Epitome di Storia Romana, Milan. Sanders, G. (1981) ‘Le dossier quantitatif de l’épigraphie latine versifiée’, AC 50, 707–20. Sanz Morales, M. (1991) ‛Variantes textuales en una cita homérica de Licurgo: Ilíada XV 494– 499 en Contra Leócrates 103’, Emerita 59, 63–80. Sanz Morales, M. (2001) ‛El Homero des Esquines’, AC 70, 49–67. Sartori, F. (1985) ‘La ricchezza di Stella e Violentilla, Index 13, 201–21 = in (id.), Dall’Italía all’Italia, II, 137–61.

334  Bibliography Sbardella, L. (2000) Filita. Testimonianze e frammenti poetici, Rome. Scandone, F. (1955), ‘I comuni del Principato Ultra (in provincia di Avellino) all’inizio della dominazione angioina’, Samnium 28, 162–71. Schaefer, A. (18852) Demosthenes und seine Zeit, I–II, Leipzig. Scheltema P. (1839) Hadriani Iunii epistolae selectae nunc primum editae, Amsterdam–Leipzig. Scherf, J. (2008) ‘Epigramma longum and the arrangement of Martial’s book’ in Morelli, A.M. (2008), 195–213. Schmid, W. (1984) ‘Spätantike Textdepravationen in den Epigrammen Martials’, in (id.), Ausgewählte philologische Schriften, Berlin–New York, 400–44. Schmitz, T. (2007) Modern Literary Theory and Ancient Texts, Malden–Oxford–Carlton. Scholl, R. (2002) Der Papyrus Ebers. Die größte Buchrolle zur Heilkunde Altägyptens, Leipzig. Schorn, S. (2012) ‘Chamaeleon: Biography and Literature Peri tou deina’, in Martano, A., Matelli, E., Mirhady, D. (eds.) Praxiphanes of Mytilene and Chamaeleon of Heraclea. Text, Translation, and Discussion, New Brunswick, 411–44. Scodel, R. (2007) ‛Lycurgus and the State Text of Tragedy’, in Cooper, C. (ed.) Politics of Orality. Orality and Literacy in Ancient Greece, Leiden–Boston, 129–54. Scorza, G. (1932) ‘Il peripatetico Cameleonte’, Rivista indo–greco–italica 18, 1–48. Scrivener, F.H.A. (1883) A Plain Introduction to the Criticism of the New Testament, London. Seek, O. (1883) Q. Aurelii Symmachi quae supersunt, in MGH, VI.1, Berlin. Seo, J.M. (2009) ‘Plagiarism and poetic identity in Martial’, AJPh 130.4, 567–93. Sgobbi, A. (2006) ‘Stesich. fr. 274 Davies: un inno di Stesicoro ad Atena?’, Acme 59.3, 285–99. Shackleton Bailey, D.R. (1993) Martial. Epigrams, Cambridge–London. Shipley, F.W. (1904) Certain Sources of Corruption in Latin Manuscripts, New York. Simonetti, M. (1957) Tirannio Rufino, Apologia, Alba. Sissa, G. (1999) ‛Sexual Bodybuilding: Aeschines against Timarchus’, in Porter, J. (ed.) Constructions of the Classical Body, Ann Arbor, 147–68 Smith, N.D. (1996) ‘Plato’s Divided Line’, AncPhil 16, 25–46. Solmsen, F. (1942) ‘Eratosthenes as Platonist and Poet’, TAPhA 63, 192–213. Somma, M.C. (2015) ‘…Cum ingenti toedio et labore (Chronicon Casauriense, RIS II, 2, coll. 797–798) scriptoria, biblioteche ed archivi nei monasteri benedettini altomedievali’, in Pani, L.E. (ed.) (2015), 89–118. Sonnino, M. (2010) Euripidis Erechthei quae exstant, Florence. Sonny, A. (1896) Ad Dionem Chrysostomum analecta, Kiev. Souter A. (1905) ‘The Original Home of Codex Claromontanus (DPAUL)’, JThS 6, 240–2. Spaltenstein F., Petitmengin P. (1979) ‘Beatus Rhenanus éditeur de l’Apocoloquintose et le Codex Wissenburgensis’, RHT 9, 315–27. Sparagna, S. (2014) ‘Il XII libro di Marziale e la metapoetica dei luoghi’, in La biblioteca di ClassicoContemporaneo 1, 4–15. Spinola, G. (2014) ‘I ritratti dei poeti, filosofi, letterati e uomini illustri nelle biblioteche romane’, in Meneghini R., Rea R. (eds.) (2014) La Biblioteca Infinita: i luoghi del sapere nel mondo antico, Rome, 155–75. Squarciapino, M.F. (1953–1955) ‘Pitture di una tomba sulla via Ostiense’, BCAR 75, 109–15. Stackmann, K. (1967) ʻDie Auslegung des Gerhard Lorichius zur „Metamorphosen”– Nachdichtung Jörg Wickrams. Beschreibung eines deutschen Ovid–Kommentars aus der Reformationszeitʼ, Zeitschrift für deutsche Philologie 86, 120–60. Stama, F. (2014) Frinico. Introduzione, traduzione e commento, Heidelberg. Starr, R.J. (1987) ‘The Circulation of Literary Texts in the Roman world’, CQ 37, 213–23.

Bibliography  335

Steffen, V. (1964) Chamaeleontis fragmenta, Warsaw. Stenhouse, W. (2002) The Paper Museum of Cassiano dal Pozzo, Series A, Antiquities and Architecture, VII, Ancient Inscriptions, London. Stephanus, H. (1579) Theocriti aliorumque poetarum Idyllia. Eiusdem epigrammata. Simmiae Rhodii Ouum, Alae, Securis, Fistula. Dosiadis Ara. Omnia cum interpretation latina, Paris. Stephens, S. (2006) Ptolemaic Pastoral, in Fantuzzi, M., Papanghelis, T.D. (eds.) Brill’s Companion to Greek and Latin Pastoral, Leiden–Boston, 91–118. Strecker, K. (1884) De Lycophrone Euphronio Eratosthene comicorum interpretibus, Greifswald. Strout, D., French, R. (1941) ‘Phrynichos (8)’, in RE 20.1, 920–25. Strutwolf, H. (2012) ‘Scribal Practices and the Transmission of Biblical Texts: New Insights from the Coherence–Based Genealogical Method’, in Kloppenborg, J.S., Newman, J.H. (eds.), Editing the Bible. Assessing the Task Past and Present, Atlanta, 139–60. Swain, S. (2004) ‘Bilingualism and Biculturalism in Antonine Rome’, in Holford–Strevens, L., Vardi, A. (eds.) The Worlds of Aulus Gellius, Oxford, 3–41. Swanson, R. (1998) New Testament Greek Manuscripts: Variant Readings Arranged in Horizontal Lines against Codex Vaticanus: Acts, Sheffield. Swanson, R. (2001) New Testament Greek Manuscripts: Variant Readings Arranged in Horizontal Lines against Codex Vaticanus: Romans, Pasadena. Swanson, R. (2005) New Testament Greek Manuscripts: Variant Readings Arranged in Horizontal Lines against Codex Vaticanus: 2 Corinthians, Pasadena. Syme, R. (1958) Tacitus, 2 vols., Oxford. Syndikus, H.–P. (20013) Die Lyrik des Horaz: Eine Interpretation der Oden, Darmstadt. Szelest, H. (1980) ‘Ut faciam breviora mones epigrammata, Corde… Eine Martial–Studie’, Philologus 124, 99–108. Taddei, A. (2012) Licurgo. Contro Leocrate, Milan. Tarrant, R.J. (1997) Aspects of Virgil’s Reception in Antiquity, in Martindale, C. (ed.) The Cambridge Companion to Virgil, Cambridge, 56–72. Tauchnitz, K. (1832) L. Annaei Senecae Philosophi Opera omnia ad optimorum librorum fidem accurate edita, V, Leipzig. Teatini, A. (2011) Repertorio dei sarcofagi decorati della Sardegna Romana, Rome. Thesleff, H. (1961) An Introduction to the Pythagorean Writing of the Hellenistic Period, Åbo. Thierry, J.J. (1963) The Date of the Dream of Jerome, VChr 17, 28–40. Thomas, R. (1992) Literacy and Orality in Ancient Greece, Cambridge. Timpanaro, S. (1986) Per la storia della filologia virgiliana antica, Rome. Tischendorf, C. (1832) Codex Claromontanus siue Epistulae Pauli Omnes, Leipzig. Torre, C. (2000) Il matrimonio del sapiens: ricerche sul de matrimonio di Seneca, Genoa. Tosi, R. (1994) ‘La lessicografia e la paremiografia in età alessandrina e il loro sviluppo successivo’, in Montanari, F. (ed.) La philologie Grecque à l’époque Hellénistique et Romaine, Vandœuvres–Geneva, 143–97. Tosi, R. (1998a) ‘Gli Ateniesi μολγοί: da Aristofane ad Eratostene’, SemRom 1.1, 123–36. Tosi, R. (1998b) ‘Appunti sulla filologia di Eratostene di Cirene’, Eikasmós 9, 327–46. Totelin, L.M.V. (2009) Hippocratic Recipes. Oral and Written Transmission of Pharmacological Knowledge in Fifth–and Fourth–Century Greece, Leiden–Boston. Tran, N. (2006) Les Membres des Associations Romains. Le rang social des collegiati en Italie et en Gaules, sous le Haut–Empire, Rome. Treggiari, S. (1976) ‘Jobs for Women’, AJAH 1, 76–104.

336  Bibliography Treu, K. (1996) Die Griechischen Handschriften des Neuen Testaments in der UdSSR: Eine systematische Auswertung der Texthandschriften in Leningrad, Moskau, Kiev, Odessa, Tbilisi und Erevan, Berlin. Trinacty, C. (2012) ‘The Fox and the Bee: Horace’s First book of the Epistles’, Arethusa 45, 57– 77. Trinchera, F. (1872) Degli archivi napoletani, Naples. Trojahn, S. (2002) Die auf Papyri erhaltenen Kommentare zur Alten Komödie. Ein Beitrag zur Geschichte der antiken Philologie, Munich–Leipzig. Tropeano, M.P. (1970) La biblioteca di Montevergine nella cultura del Mezzogiorno, Naples. Tropeano, M.P. (ed.) (1978) Codice DiplomaticoVerginiano, Montevergine. Trout, D.E. (2001) ‘The Verse Epitaph(s) of Petronius Probus: Competitive Commemoration in Late–Fourth–Century Rome’, NECJ 28.3, 157–76. Trout, D.E. (2013) ‘‘Being Female’: Verse Commemoration at the Coemeterium S. Agnetis (Via Nomentana)’, in Harrison, C., Sandwell, I., Humfress, C. (eds.) Being Christian in Late Antiquity: A Festschrift for Gillian Clark, Oxford, 215–34. Trout, D.E. (2014) ‘Vergil and Ovid at the tomb of Agnes: Constantina, Epigraphy, and the Genesis of Christian Poetry’, in Bodel, J., Dimitrova, N. (eds.) Ancient Documents and their Contexts. First North American Congress of Latin and Greek Epigraphy (2011), Leiden–Boston, 263–82. Trout, D.E. (2015a) Damasus of Rome. The Epigraphic Poetry, Oxford. Trout, D.E. (2015b) ‘Napkin art: Carmina contra paganos and the difference satire made in fourth century Rome’, in Salzman, M.R., Sághy, M., Lizzi Testa, R. (eds.) Pagans and Christians in the Late Antique Rome. Conflict, Competition, and Coexistence in the Fourth Century, New York, 213–31. Turner, E.G. (1977) The Typology of the Early Codex, Philadelphia. Turner, E.G. (19872) Greek Manuscripts of the Ancient World, edited by Parsons, P.J., London. Tzagalis, C.C. (2007) ‛CEG 594 and Euripides’ Erechteus’, ZPE 162, 9–13. Uguaglia, M. (2011) ‘Alcune osservazioni sull’uso di λόγος in Aristotele’, in Reppellin, F.F., Micheli, G. (eds.) La scienza antica e la sua tradizione, 4. Seminario di studi (Gargnano, 13–15 ottobre 2008), Milan, 81–118. Usher, S. (1999) Greek Oratory. Tradition and Originality, Oxford. Valckenaer, L.C. (1781) Theocriti, Bionis et Moschi carmina bucolica. Graece et Latine, Leiden. van Arckel, C. (1707) Animadversa eiusdem De coma commentarium, ab autore innumeris locis emendata et insignibus supplementis locupletata. Accedit Appendix Hadriani Iunii ad Animadversa sua, nunc primum ex clariss. Viri autographo in lucem edita, Rotterdam. Van der Valk, M. (1964) Researches on the Text and Scholia of the Iliad, I–II, Leiden. van Leeuwen, J. (1898) Aristophanis Nubes cum prolegomenis et commentariis, Leiden. van Miert D. (ed.) (2011) The Kaleidoscopic Scholarship of Hadrianus Junius (1511–1575). Northern Humanism at the Dawn of the Dutch Golden Age, Leiden. Van Sickle, J. (1975) ‘Epic and Bucolic (Theocritus, Id. VII, Virgil Ec. I)’, QUCC 19, 45–72. Van Sickle, J. (1976) ‘Theocritus and the Development of the Conception of the Bucolic Genre’, Ramus 5.1, 18–44. Van Sickle, J. (1980) ‘The Book–roll and Some Conventions of the Poetic Book’, Arethusa 13.1, 5–42. Vardi, A. (2001) ‘Gellius against the Professors’, ZPE 137, 41–54. Veltri, G. (2006) Libraries, Translations, and ‘Canonic’ Texts. The Septuagint, Aquila and Ben Sira in the Jewish and Christian Traditions, Leiden–Boston.

Bibliography  337

Verrengia, A. (1999) Dione di Prusa. In Atene, sull’esilio (or. XIII), Naples. Veyne, P. (1997) The Roman Empire, transl. Goldhammer, A., Cambridge–Mass. Vian, P., D’Aiuto, F. (2011) Guida ai fondi manoscritti, numismatici, a stampa della Biblioteca Vaticana, I, Dipartimento Manoscritti, Vatican City. Vinge, L. (1967) The Narcissus Theme in Western European Literature up to the Early 19. Century, Gleerup. Vix, J.L. (2016) ‘Ælius Aristide et la comédie’, in Pernot, L., Abbamonte, G., Lamagna, M. (eds.) Ælius Aristide écrivain, Turnhout, 375–92. Von Staden, H. (1997) ‘Inefficacy, Error and Failure: Galen on dokima pharmaka aprakta’, in Debru, A. (ed.) Galen on Pharmacology. Proceedings of the 5th International Galen Colloquium (Lille 1995), Leiden–New York–Cologne, 59–83. Von Staden, H. (1999) ‘Rupture and Continuity: Hellenistic Reflections on the History of Medicine’, in van der Eijk, P.J. (ed.) Ancient Histories of Medicine. Essays in Medical Doxography and Historiography in Classical Antiquity, Leiden–Boston, 143–88. Vottero, D. (1989) Lucio Anneo Seneca, Questioni Naturali, Turin. Vox, O. (1991) ‘Poetesse di Teocrito’, in De Martino, F. (ed.), Rose di Pieria, IX, Bari, 198–220. Vulcanius, B. (ed.) (1595) Nili Archiepiscopi Thessalonicensis De primatu Papae Romani libri duo, Leiden. Vuolanto, V. (2015) ‘Children and Work. Family Strategies and Socialisation in the Roman and Late Antique Egypt’, Acta Instituti Romani Finlandiae 42, 97–112. Wallace, R.W. (2003) ‘An Early Fifth–Century Athenian Revolution in Aulos Music’, HSPh 101, 73–92. Warenda, A. (1993) ‘They’, Writing Across the Curriculum 4, 99–107. Warrior, V.M. (2006) Roman religion, New York. Watson, J.S. (1986) Cicero on Oratory and Orators, Carbondale. Wehrli, F. (19692) Die Schule des Aristoteles, IX, Phainias von Eresos, Chamaileon, Praxiphanes, Basel–Stuttgart. Wendel, C. (1914) Scholia in Theocritum vetera, Leipzig. Wenzel, H. (1974) ʻFrauendienst und Gottesdienst. Studien zur Minne–Ideologieʼ, Philologische Studien und Quellen 74, 98–117. West, D. (1995) Horace Odes 1: Carpe Diem, Oxford. West, M.L. (1982) Greek Metre, Oxford. West, M.L. (2007) Indo–european Poetry and Myth, Oxford. West, S. (1967) The Ptolemaic Papyri of Homer, («Papyrologica Coloniensia» 3), Cologne– Opladen. Westcott, B.F., Hort F.J.A (1882) The New Testament in the Original Greek: Introduction and Appendix, New York. Wettstein, J.J. (1751) Novum Testamentum Graecum, 2 vols., Amsterdam. Wheeler, A.L. (1924) Ovid, Tristia–Ex Ponto, with an English translation, London. White, J.W. (1912) The Verse of Greek Comedy, London. Whitmarsh, T. (2001) Greek literature and the Roman Empire: The Politics of imitation, Oxford. Wilamowitz–Moellendorf, U. (1900) ‘Die Textgeschichte der griechischen Lyriker’, Abhandlungen der königlichen Gesellschaft der Wissenschaften zu Göttingen. Philologisch–historische Klasse 4.3, 84–5. Wilamowitz–Moellendorf, U. (1906) Die Textgeschichte der griechischen Bukoliker, Berlin. Williams, G. (1992) ‘Representations of the Book–Roll in Latin Poetry: Ovid, Tr. 1, 1, 3–14 and Related Texts’, Mnemosyne 45.2, 178–89.

338  Bibliography Williams, M.H. (2006) The Monk and the Book. Jerome and the Making of Christian Scholarship, Chicago–London. Wilpert, J. (1903) Roma Sotterranea. Le pitture delle catacombe romane, Rome. Wilpert, J. (1924) ‘Le pitture dell’ipogeo di Aurelio Felicissimo presso il Viale Manzoni in Roma’, in Atti della Pontificia Accademia Romana di Archeologia, Memorie 1.2, 5–43. Wilson, N.G. (1967) ‘A chapter in the History of scholia’, CQ 17, 244–56. Winkler, J.J. (1985) Auctor and actor: A Narratological reading of Apuleis’s Golden Ass, Berkeley. Winsbury, R. (2009) The Roman Book. Books, Publishing and Performance in Classical Rome. Classical Literature and Society, London. Winterton, R. (1667) Poetae Minores Graeci, Hesiodus, Theocritus, Moschus, Bion Smyrn., Simmias Rhod., Musaeus, Theognis, Phocylides... Et aliorum incertorum autorum: quibus subjungitur, eorum potissimum quae ad philosophiam moralem pertinent, Cambridge. Wirth, F. (1934) Römische Wandmalerei vom Untergang Pompejis bis ans Ende des 3. Jahrhunderts, Darmstadt. Wolfer, E.P. (1954) Eratosthenes von Kyrene als Mathematiker und Philosoph, Zürich– Groningen. Woolf, G. (2003) ‘The city of letters’, in Edwards, C., Woolf, G. (eds.) Rome the Cosmopolis, Cambridge, 203–21. Woolf, G. (2013) ‘Introduction: Approaching the Ancient Library’, in König, J., Oikonomopoulou, K., Woolf, G. (eds.) Ancient Libraries, Cambridge, 1–20. Wouters, A. (1969) ‘Le PSI II 144: texte biographique ou grammatical?’, CE 44.88, 325–33. Wrede, H. (1981) Consecratio in formam deorum. Vergöttlichte Privatpersonen in der römischen Kaiserzeit, Mainz. Youtie, L.C. (1996) The Michigan Medical Codex, ed. Hanson, A.E., Atlanta [= (ead.) (1986– 1987) ‘The Michigan Medical Codex’, ZPE 65, 123–49; 66, 149–56; 67, 83–95; 69, 163–9; 70, 73–103]. Zadorojnyi, A.V. (2013) ‘Libraries and paideia in the Second Sophistic: Plutarch and Galen’, in König, J., Oikonomopoulou, K., Woolf, G. (eds.) Ancient Libraries, Cambridge, 377–400. Zalateo, G. (1961) ‘Papiri scolastici’, Aegyptus 41, 160–234. Zanichelli, G. (2015) ‘La strutturazione del lavoro all’interno dello scriptorium’, in Pani, L.E. (ed.) (2015), 119–39. Zanker, P. (1966) ʻ“Iste ego sum”. Der naive und der Bewußte Narzißʼ, BJ 166, 152–70. Zetzel, J.E.G. (1973) ‘Emendavi ad Tironem: Some notes on classical scholarship in the second century A.D.’, HSPh 77, 225–43. Ziegler (ed.) (1844) Theocriti Carmina, Tübingen. Zigarelli, G. (1860) Viaggio storico artistico al real santuario di Montevergine con una breve descrizione dei paesi che si scovrono da quelle alture, e degli uomini che vi si distinsero in ogni ramo, Naples. Ziolkowski, T. (2004) Ovid and the Moderns, New York. Zironi, A. (2004) Il monastero longobardo di Bobbio. Crocevia di uomini, manoscritti e culture, Spoleto. Zuntz, G. (1953) The Text of the Epistles: A Disquisition upon the Corpus Paulinum, London. Zuntz, G. (19752) Die Aristophanes–Scholien der Papyri, Berlin.

Index Index of Ancient, Medieval and Early Modern Names Achilles1 50, 78, 80--82, 117 Adrastus of Aphrodisia 146 (n. 8), 147 Aelius Aristides. 131 (n. 25), 134, 137 Aelius Donatus 4, 94 Aeschines 3, 75, 78---85 Alcaeus 121 Ambrosius 101 (n. 48) Anthologia Graeca 2, 15, 16, 107, 115--6, 119, Antiphanes 107 Antonia Fabia Paulina 189---98 Antonius 28 Apuleius, Lucius 260 Aristophanes 74, 125---41 Aristotle 74, 158--9 Aristoxenus 148---51 Arnobius 95 Arnolphe d'Orleans 297 Arruntius Stella 62--63 Artemidorus of Tarsus 115---6 Asclepiades 122 Athena (Pallas) 14, 21, 99 (n. 41), 126--40 Athenaeus 108, 112 (n. 4), 121--2, 128 (n. 15) Atticus, Titus Pomponius 28 Augustus 2, 18--19, 21, 35, 43---44, 170, 253, 279, 283 (n. 18), 293, 399, 303 Bartholomeus Saxo 273 Beatus Rhenanus 280---2 Benedict, St. 270--2, 278 Bertran de Paris 297 Bion 112--3, 116 Caesar, Gaius Iulius 18---19, 101, 255, 297 Caesarius of Arles 270 Caesellius 54--55 Calcidius 146

 1 Names and sources are indicized following the usage of contributors. https://doi.org/10.1515/9783110632590-020

Callimachus 2, 15--17, 22, 128 (n. 17), 136, 138 Camerarius 119 Cassius 30--31, 48 Cassius Dio, 48 (n. 8), 265, 277 Cato (Elder) 36, 98--99 Catullus 29, 32, 34, 48, 69 Celio Secondo Curione 11, 277, 289 Chamaeleon 126--131, 134--41 Charchinus (Charchinos), 74 Charles V 297 Chaucus, Ioannes 283, 286--90 Choerilus 16 Cicero, Marcus Tullius 4, 20, 27--29, 34--25, 48, 54--55, 69, 91--103, 247, 264--5 Claudianius 175, 176--8 Claudius 11, 277--80, 285, 289---90 Claudius, Quintus 54--55, 63 Clearchus 146 Cleobulus 108 Cleophon (Kleophon) 74 Columella 70 Crantor 146, 164 Cyprian 92 de Lorris, Guillaume 297 Demosthenes 3, 75--80, 85 Diocletian 190 Diodorus Siculus 86 Diogenes Laertius 125, 129, 173 Diomedes 99 (n. 41) Dionysius of Halicarnassus 132, 173 (n. 28) Domitian 3, 60--72 Domitius Apollinaris 62 Donato, abbot 273 Dosiadas 112

344  Index Ennius 36 Erasmus 280---1 Eratosthenes 7--8, 125--141, 142--65 Euclides 149, 152, 163 Eumolpus, King 73 Euphorion 14 Euripides 3, 74--75, 87--89, 129 (n. 17), 130, 132 (n. 29), 137 (n. 44) Eurysaces 246 Eusebius of Caesarea 91, 264 Feyerabend, Sigmund 298 Flavius Maximianus 246 Florus 102 Frontinus 20--21, 62--63, 297 Frontinus, Sextius Iulius 63 Galen 52, 173 (n. 30), 174--85 Gellius 2---3, 47--58 Geruchia 96 Giovanni da Napoli, fra 276 Gottfried von Straßburg 296 Gruterus, Janus 288 Guglielmo da Vercelli, St. 271 Harpocration 74, 85, 125 Heras of Cappadocia 181--2 Hesychius 111, 119---20 Hieronymus Ziegler 297 Hippocrates 168 (n. 5, 8), 171, 173, 174 (n. 34), 177--8, 186 (n. 77) Holkot, Robert 298 Homer 3, 23, 75--89, 117, 129 (n. 17), 132 (n. 29), 173 (n. 29), 178, 188 (n. 83) Horace 2, 13---24, 25--46, 179 Janus 21--22 Jerome, St. 4, 91--103 Johann Saurn 298 Junius, Hadrianus 11, 277--91 Juvenal 264 Lactantius 95 Lamprocles 7, 121---41 Landulfus 273 Latron 69 Lauwingen, Reussner 298 Livy 92, 96 Lucan 59, 94

Lucian 2--3, 47--58 Ludwig II 295 Ludwig the Bearded 295 Ludwig VII 295 Lycophron 135--6 Lycurgus 3, 73--89 Macrobius 9, 198--90 Maecenas 35, 42--43 Marius 20--21, 25, 101 Martial, Marcus Valerius I, 3, 50--72, 179 (n. 51) Mela, Titus Pomponius 70 Meleager 2, 16--17, 24 Mesomedes 108 Minucius Felix 95 Moschus 112--3, 116 Munatius Gallus 67 Murners, Thomas 297 Nerva 60, 63 Ottone da Extendardo 276 Ovid 9, 12, 32---33, 34 (n. 9), 38, 43, 48, 101 (n. 48), 198, 293---306 Patroclus 78---82 Paul, St. 92--95, 103, 204--8, 218--19, 276 Paula 92 Paulinus of Nola 95 (n. 22) Petronius Probus 196 Philip of Opus 173 Philitas 108, 120--2 Philostratus the Younger 252 Photius 85 (n. 67) Phyrnicus 7, 127---41 Pier Damiani 272 Pindar 14, 123, 129 (n. 17), 138 (n. 45) Plato 7--8, 48, 79, 93, 115, 129, 144---65, 173, 174 (n. 34) Plautus, Titus Maccius 9, 21, 35--36, 94, 101, 196 Pliny (Elder) 19, 171 Pliny (Younger) 48, 62--63, 65 Plutarch 77 (n. 23), 85 (n. 67), 98, 99 (n. 33), 102, 107, 129--9, 146, 164, 258 Pollux 87, 132 Porphyry 146, 148--9, 160, 163

Index  345

Posidonius 146 Praetextatus, Vettius Agorius 9, 189--99 Proclus 146, 164 Protagoras 93 Quintilian, Marcus Fabius 21, 39 (n. 32), 64, 65 (n. 16) 69, 173 (n. 28), 262 Rudolph I 272 Rufinus of Antiochia 105 (n. 4) Rufinus of Aquileia 93--94, 98, 103 Rufus (rhetorician) 132 Sallust, Gaius 21, 94 Sappho 107, 121 Schöffer, Ivo 297 Sebastian Brants 297 Seneca (Elder) 69, 98, 101 Seneca (Younger) 9, 11, 69, 98 (n. 33), 278--91 Septimius Severus 262---3 Sidonius Apollinaris 64 (n. 16), 252 (n. 45) Stainhöwel, Heinrich 297 Stesichorus 131--4, 138---41 Strabo 69, 170, 172 Stratocles (Stratokles) 87--88 Straton of Sardis 105 Suetonius, Gaius 43---44, 49 (n. 8), 58, 98 (n. 35), 258, 264 Tacitus, Publius Cornelius 48 (n. 8), 100 (n. 43), 246 (n. 29), 257 (n. 62), 297 Tampach, Gottfried 298 Terentius Maurus 244 Terentius, Publius 196

Tertullian 4, 96, 103 Theocritus 6, 109 (n. 22), 111--23 Theon of Smyrna 7, 144, 166 Theophrastus 73, 85 Thersites 50 Tiberius 253 Timarchus 77--79, 84 Trajan 60, 63--65 Trebius Iustus 246 Tzetzes, Ioannes 127 (n. 11), 130 (n. 19), 137--8 Ulysses 99 (n. 41) Valerius Maximus 102 Varro, Marcus Terentius 49, 51--52, 171 Varus, Publius Quintilius 29 Ventador, Bernart 297 Vertumnus 21-22 Vespasian 264 Vinnius Valens 18--21, 35 Virgil 4, 20--21, 36, 45, 48, 94 (n. 16), 99-101, 116, 198 (n. 29), 260 (n. 73), 293 von Halberstadt, Albrecht 12, 293--306 von Morungen, Heinrich 297 von Thuringia, Hermann 296 von Veldeke, Heinrich 296 Vulcanius, Bonaventura 288 Walensis, Thomas 296 William I 273 Xenocrates 146 (n. 8), 159 (n. 48), 164 (n. 56)

346  Index

Index of Ancient Sources Adrastus .– .



Aelius Aristides Schol. Ael. Ar. Or. ..–

Ammianus Marcellinus ..  (n. ) ..  (n. )

, 

Aeschines Or.  . . . . . . . . .  .   

–, , ,   (n. ),   (n. ) ,  ,      , ,   (n. ) , ,   

Anthologia Palatina . . . . . 14.45 14.60

,   (n. )    (n. ) 107 (n. 13) 107 (n. 13)

Apollonius of Rhodes Argon .

 (n. )

Apuleius Met. .. .

 (n. ) 

Architas VS  B

 (n. )

Aristophanes Ach. –

 (n. )

Schol. Aeschin. . Vita    Aeschylus Ag.  Pers.  Ambrosius Hexam. ..

 (n. )

 (n. )  (n. )  (n. )

Av. 

 (n. )

 (n. )

Schol. Ar. Av. 

 (n. )

 n. 

Eq. 

 (n. )

Lys. 

 (n. )

 (n. )

Index  347

Nub.  –    Schol. Ar. Nub. 

 (n. )  , , , , , , ,  

, , , , , 

Schol. Anon. Rec. Ar. Nub. a  (n. )

Schol. Tzetz. Ar. Nub.   (n. ),  Vesp. –

 (n. )

Ausonius Epist. .

 (n. )

Bible Ps. .



New Testament Acts . .

 

Cor .–  .

, ,  

 Cor .– 

, 

     , , –



De An. b –

 (n. )

Eph .–  .–  .–  . .–  .– 

Eth. Nic. ..

 (n. )

Matt .

Rh. ..b



Rom .– 

Aristoteles Ath. Pol. . – 



Aristoxenus Herm. .–. .–  .– 

 (n. )  (n. )  (n. )

Athenaeus . . . . .c

 (n.)  (n. )    (n. )

. . . . 1.19 1.20 1.20b 1.20c 1.21 1.23 1.24 1.25

, , , ,  , ,  ,  ,  ,  214, 216 219–20 216, 218 216 216, 218 228 218, 220, 228 217– 8

348  Index . 1.28 1.29 1.30 1.32 10.15 Titus .– – . .a .b .– . . . . . . a b  VL . Caesar BCiv. .. Callimachus Aetia  Pf. Ep.  . Fr. . Hymn. . Cassius Dio . ..

– 218– 9, 232 212 214, 228 229 232

     , ,   –,  –   

 (n. )





, , ,  

 (n. )

 (n. )

 

Catullus   

, ,  (n. )  (n. )  (n. )

Cicero Att. .. .. .. ..

  (n. )   (n. )

Brut. .

 (n. )

Cat. .. .

 (n. ) 

Cato 



De or. .



Font. 



Inu. ..



Mur.  

 

Off. .– 



Phil. .



Pis. 

 (n. )

Planc. 

 (n. )

Index  349

Sull. –



Tusc. .



Verr. II.. II..

  (n. )

Clement of Alexandria Paed. II.  Cornutus ND .

 (n. )

Demosthenes Or. .  .  . . . . . . . . .

 (n. )   (n. ) , , ,  n. ,   (n. )   (n. ) , ,   (n. )   (n. )  (n. )  (n. )

 S

 (n. )

Etymologicum Magnum Fr.   (n. ) Euclides El. I def.  V def. 

 (n. ) 

Euripides IA 

 (n. )

Rh.  

 (n. )  (n. )

Vit. Eur. .– 

 (n. )

Eusebius Hist. Eccl. ..



Eusthatius Fr. 

 (n. )

Eutocius Sph. Cycl. .– .

 (n. )

Dio Chrysosthomus .  (n. )

Festus p.  L.



Diogenes Laertius .  (n. ) .–   (n. )

Florus .



Diodorus ..–  Eratosthenes Fr.  S





Ep. It.   .  Frontinus Str. .

  (n. )



350  Index Fulgentius Myth.  Virg. Cont. M Gaius Inst. . Galen Antid. . Comp. Med. Gen. . ..–  K



Harpocration κ  ε δ  μ 

  (n. )  (n. )  (n. )



Hermogenes Id. 



Herodianus ..

 (n. )

Hesiod Th. 

 (n. )

Hippocrates Epid. ..



Homer Il. . . .–  . .–  . . .  .–  .–  .

 (n. )  (n. )   (n. )     (n. ),  ,  , , , ,   (n. ), .





 

Comp. Med. Sec. loc. .. , ,  Diff. resp. ..–  K



Gloss.  p.  K

 (n. )

Indol.  - 

  

In Hp. Epid. .. – 

 (n. )

Gellius NA . . .. . .. . . . . .

 (n. )  (n. )  (n. )   (n. )  ,   (n. )  (n. )  (n. )

Hymn. Aphr. – 

 (n. )

Od. . . . . . . .

 (n. )  (n. )    (n. )  (n. )  (n. )

Index  351

Horace Ars P. –    –  Carm. . . . . . . . . . . . . . . . Epist. . . . . . . .

   ,  (n. )

      , , , , ,  , , ,       , ,,,  

,  (n. )  (n. ) , , , , ,  (n. )   (n. )  , –, ,  – , – , –, 

Epod. .



Sat. .. ..–  .. –  ..

   (n. ) 

Iamblichus De communi mathematica scientia .– .  (n. ) Ibicus Fr. 

 (n. )

Ilias Latina .–

 (n. )

Isocrates Or. .–  .

 (n. )  (n. )

Jerome Adu. Rufin. . . .

 (n. )  

Ep.  .  . . .  . . . . . . .  . . . 123.13 123.14 123.15 123.16

  ,  ,  (n. )  (n. )    (n. )  (n. )   (n. )  (n. )  (n. )  (n. ) , , ,  , ,  –   103 98, 99 99 100–1

352  Index In Ezech.  Praef.

 (n. )

In Hab. .



in Is. ..

 (n. )

Vir. Ill. 



Junius Ann.  

 

Juvenal .



Lucian Ind.    

,   , ,  

Lycurgus Leoc.             .   –   

 (n. )  (n. )  (n. ),      (n. )  (n. )  (n. )   (n. )   (n. ) ,     

             Martial Ep. . . . . . . . . . . . . . . . . . . . . . . . . . . .  . .  . .

            (n. ) 

 (n. ),  (n. ) Preface  (n. )  (n. )  (n. )    (n. )  (n. )    (n. )  (n. )   (n. )  (n. )  (n. )  (n. )   (n. ),  n.   (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )

Index  353

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

 (n. )  (n. )  (n. )  (n. )  (n. ) ,  ,  , ,   ,  ,  , ,  ,  , ,   –    (n. )  (n. ),   (n. )  (n. ),   ,   – , ,  –    (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )  (n. )

Xen. 

 (n. )

Orosius Hist. .

 (n. )

Ovid Met. .–  .–  . . .–  . .–  . . . . . .–  . . . . . .–  . . . . .– .–  .–  . .–  .–  .–  .–  .–  .–  . .–  . .–  . .–  .–  .–  .–  .–  .–  .– 

            ,                                 

354  Index .–  .– 

 

Tr. ..

, 

Rem. 

 (n. )

Philitas of Cos Fr. 

, 

Photius Bibl. 

 (n. )

Phrynicus Fr.  K.– A.

 (n. )

Pindar Ol. .

.– e   .b

 (n. )  (n. ),   (n. )  (n. )

Symp. c

 (n. )

Tht. e

 (n. )

Tim. b b d a– b b– d

 (n. )  (n. )  (n. ) , – , 

Plautus Amph. 

 (n. )

 (n. )

Cist. 

 (n. )

Schol. Pind. Ol. .

 (n. )

Trin. – 

Nem. .

 (n. )

Plato Ion. d

 (n. )

Leg. a

 (n. ),  (n. )

Phlb. b

 (n. )

Prm. d

 (n. )

Resp. .d .d– d

 (n. ) , 



Pliny the Elder HN . ..– 

 

Plutarch Alex. Fort. .. ff.

 (n. )

De an. Proc. b– f

 (n. )

De mus. d– e

 (n. )

Mar. 

 (n. )

Index  355

Marc. 



Mor. b

 (n. )

X or. b b

,  (n. )  (n. )

Pollux .– 



Porphyry In Pt. Harm. .–  .– .

 (n. )  (n. )

Proclus In Tim. ..– .

 (n. )

Theol. Plat. ..– 

 (n. )

Propertius .. ..

 (n. )  (n. )

Quintilian Inst. .. .. ..–

 (n. )  (n. ) 

Rufinus Apol. Adu. Hier. . . Sallust Iug. .

  (n. )



Seneca the Elder Contr. .. Suas. . . Seneca the Younger QNat. a



 (n. )  (n. )

 (n. )

Sidonius Apollinaris Carm. .–   (n. ) Epist. ..– 

 (n. )

Sophocles Aj. 

 (n. )

Stesichorus Chil. I.. – 



Stobaeus ..



Strabo ...–  .

  (n. )

Suda ε  θ .

 (n. )  (n. )

Suetonius Gram. et rhet. 



Vit. Hor. . .

 

356  Index Vesp.  .

  (n. )

Tirtaeus Fr. W



Tertullian Castit. .

 (n. )

TrGF –  Radt  Kannicht

 – 

Monog. .

 (n. )

Ulpian ... ... .

  (n. ) 

Valerius Maximus .



Vegetius .



Theophrastus Char. .

 (n. )

Virgil Aen. . .

 (n. ) 

Tacitus Hist. . . .. Theocritus Id.  .   .    Theon of Smyrna .– . .– . .–  .– . .– . .– . .– . .– . .– 

 (n. )  (n. )  (n. )

– ,  , ,     (n. )  –,  – –

 (n. )  (n. )  –  (n. ) – , , -    (n. )

G. .–  .–  Xenophontes Oec. . Symp. .

,  

 (n. )

 (n. )

Index  357

Index of Material Sources PAPYRI P. Ant. s.n. = Theocritus’ Pa 112–114 P. Ant. 3.186 185 P. Berl. Möller 13 180, 181 (n. 62) P. Bodmer II + CBL BP XIX + Inv. Nr. 4274/4298 = NT P66 213 P. Ebers 168, 169 and n. 10, 170, 176, 181 (n.62) P. Fuad. inv. 222 84 (n. 65) P. Heid. Lit. 2 + P. Hib. 1.22 + P. Grenf. 2.4 = Iliad Þ12 82 (n. 56) P. Lips. 1445 109 P. Lit. Lond. 63 5–6, 105–10 P. Lit. Lond. 132 182 (n. 63) P. Louvre inv. 7733 108 (n. 17) P. Mert. 1.12 177 P. Mich. 17.758 184 P. Mil. Vogl. 8.309 115 and n. 23 P. Oxy. 1.15 109 P. Oxy. 4.671 109 P. Oxy. 8.1088 175, 182 (n. 64) P. Oxy. 13.1611 7, 125 (n. 1), 131 (n. 23), 132 and n. 31, 134 (n. 33), 135, 136 and n. 39, 140, 141 and n. 54 P. Oxy. 15.1800 85 (n. 67) P. Oxy. 17.2064 + 50.3548 = Theocritus’

O. Claud. 2.220 175 (n. 39)

Po 114 P. Oxy. 35.2737 125 (n. 1) P. Oxy. 54.3726 109 P. Oxy. 74.4968 = NT P127 203 (n. 8) P. Ross. Georg. 1.14 109 P. Ryl. 1.29 179 PSI 1.17 109 PSI 2.144 125 (n. 1) PSI 6.718 179, 180, 183 (n. 66) PSI 10.1180 169, 181 (n. 62), 182 (n. 64) P. Sorb. 2328 87 P. Stras. inv. 1016 109 P. Turner 39 125 (n. 1) Perg. Louvre 6678 et Rainer 114 and n. 18

CIL: CIL I.2 2504 = AE 1910.0010 = EDCS 24700009 253 (n. 48) CIL I.2 2683 = EDR 073261 253 (n. 52) CIL I.2 2702 = EDR 073259 253 (n. 51) CIL VI.1060 = EDCS 17600398 255 (n. 56) CIL VI.1777 = Dessau, ILS 1258 189 (nn. 5 and 8–9), 190 (nn. 10–11) CIL VI.1778 189 (n. 5), 190 (n. 11) CIL VI.1779 = Dessau, ILS 1259 (Roma) 9, 189 (n.2 and 5), 190 (n. 11 and 15), 191 CIL VI.3979 = EDR 119454 264 (n. 80) CIL VI.5682 = EDCS 19100085 254 (n. 54) CIL VI.6150 = Dessau, ILS 7896 = EDCS 19000166 253

OSTRAKA O. Bodl. 2.2174 109

INSCRIPTIONS Carletti (1986): Nr. 23 197 (n. 27) Nr. 24 197 (n. 27) Nr. 25 197 (n. 27) Nr. 26 197 (n. 27) Nr. 27 197 (n. 27) Nr. 29 197 (n. 27) Nr. 31 197 (n. 27) Nr. 33 198 (n. 28) Nr. 49 197 (n. 27) Nr. 52 197 (n. 27) Nr. 91 197 (n. 27) Nr. 92 197 (n. 27) Nr. 103 197 (n. 27) Nr. 115 197 (n. 27) Nr. 124 197 (n. 27) Carletti (2008): Nr. 66 198 and n. 28 Nr. 69 198 (n. 28) CEG: CEG 594 88

358  Index CIL VI.7373 = EDR 141842 = EDCS 18600087 264 (n. 78) CIL VI.8531 and 34026 = EDR 119868 = EDCS 24100370 256 (n. 57) CIL VI.8786 = EDCS 18700127 264 (n. 82) CIL VI.8882 264 (n. 80) CIL VI.9301 = EDR 137602 264 (n. 80) CIL VI.9342 246 (n. 29) CIL VI.9525 = EDR 078376 264 (n. 80) CIL VI.9540 = Dessau, ILS 7397 = EDCS 19200253 264 (n. 78) CIL VI.9541 = EDR 160219 264 (n. 78) CIL VI.9542 264 (n. 79) CIL VI.10300 = EDCS 17700507 from Rome 255 (n. 56) CIL VI.10350 = EDR 121978. Louvre, Ma 1652 255 CIL VI.18079 = EDR 032581 254 CIL VI.24032 = EDR 102064 246 (n. 29) CIL VI.33473 = CLE 1882 = Dessau, ILS 7771 = EDR 152970 = EDCS 23900587 246 (n. 29), 264 (n. 82) CIL VI.33830 264 (n. 83) CIL VI.33892 = EDR 000884 = EDCS 24100405 264 (n. 81) CIL VI.34004 = EDR 118433. Cap. Mus., NCE 714 255 CIL VI.34270 264 (n. 83) CIL VI.37802 264 (n. 80) CIL VI.41341a 196 (n. 21) CIL IX.329 = Dessau, ILS 5557a = AE 1999, 511 (Canusium) 190 (n. 12) CIL XI.1147 = EDR 130843 244 (n. 21) CIL XI.1355 = EDCS 5320028 = EDR 129455 and 129457 from Luna and AE 1977, 0265 a and b 255 (n. 56) CIL XIV.250–2 = EDCS 05700249–51 from Ostia 255 (n. 56) CIL XIV.2630 = EDCS 05800609 254 EDR: EDR 030285 = EDCS 52602891 244 (n. 20) ICUR: ICUR II.1 = Ihm (1895), nr. 84 198 (n. 29) ICUR VI.15931 = EDCS 16201053 = EDB 9222 242 (n. 18)

ICURns: ICURns III.8470 198 (n. 28) ICURns V.13196 198 and n. 28 IG: IG I.2 91.11 172 (n. 24) IG II.2 47 172 IG II.2 457 87 (n. 93) IG II.2 1533 172 ILLRP: ILLRP 819 = CIL I.2 3146 196 (n. 21) MANUSCRIPTS: Basel, Universitätsbibliothek Basel G2 I 15:1:59–60 287 (n. 34) Bologna, Bologna University Library Ms. 2248 270 (n. 7) Cambridge, Cambridge University Library Codex Bezae = University Library, Nn. 2. 41 = Dea or 05 203 (n. 8) Cambridge, Trinity College Codex Augiensis = Trinity College, B.17.1 = 010 206, 210 Dresden, Saxon State Library Codex Boernerianus = Saxon State Library, A 145b = 012 206, 210, 219–20 Florence, Biblioteca Medicea Laurenziana Laur. XXXII 52 = Theocritus’ Cod. G 113, 116 (n. 28) Laur. Conv. Soppr. 66 = Tzetzes’ Cs 137 Frankfurt am Main, Universitätsbibliothek Johann Christian Senckenberg Ms. Ff. Th. Oehler III, 3 285 (and n. 27) London, British Library Lond. Add. 11983 277 Ms. Burney 95 = Crippsianus (A) 86 (n. 81) Marburg, Hessisches Staatsarchiv

Index  359

Codex Waldeccensis = Hessisches Staatsarchiv Best. 147 Hr 2,2 = NT 0320 = VL83 205–6, 234 Milan, Biblioteca Ambrosiana Ambr. B 75 sup. (Gr. 104) = Theocritus’ C 113 Ambr. C 222 inf. (Gr. 886) = Theocritus’ K = Tzetzes’ Amb 116 (n. 28), 137 Ambr. G 32 sup. (Gr. 390) = Theocritus’ A 116 (n. 28) Ambr. L 39 sup. = Aristophanes’ M 126 Modena, Biblioteca Universitaria Estense α U 5.10 = Aristophanes’ E 126-8 (and n. 9), 140 (n. 53) Montevergine, Archivio – Biblioteca Statale di Montevergine Ms. 1 11, 269 and n. 2, 271 Ms. 2 11, 269 and n. 2 Ms. 3 11, 269 and n. 2 Ms. 4 11, 269 and n. 2 Parchment 189 274 (n. 24) Parchment 237 274 and n. 23 Parchment. 1297 273 and n. 19 Parchment 1405 274 and n. 25, 275 and n. 27 Parchment 1907 275 and n. 28 Parchment 2326 275 and n. 30 Parchment 3970 273 and n. 20 Parchment 4027 276 and n. 33 Mount Athos, Great Lavra Monastery Λ.173. = NT 1646 223 Napoli, Biblioteca Nazionale Gr. II F 25 = Aristophanes’ Np 126-8, 140 (n. 53) Oldenburg, Staatsarchiv Manuscript of Oldenburg = Best. 297 E 22 + ohne Sign. 12, 293, 295 Oxford, Bodleian Library Ms. Bodl. Auct. T.II.08 86 (n. 81)

Paris, Bibliothèque Nationale de France Codex Claromontanus = Par. Gr. 107 AB = 06 = VL75 10, 204–38 Par. Gr. 2726 = Theocritus’ D 113 Par. Gr. 3002 80 (n. 43) Ravenna, Biblioteca classense Gr. 429 = Aristophanes’ R 126-8 (and n. 13) Saint Petersburg, National Library of Russia Codex Sangermanensis = National Library of Russia, Gr. 20 = Dabs1 = NT 0319 = VL76 10, 201–38 St. Gallen, Stiftsbibliothek Sang. 569 277, 284 Vatican City, Vatican Library Urb. Gr. 141 = Tzetzes’ U 137 Vat. Gr. 38 = Theocritus’ T 116 (n. 28) Vat. Gr. 42 = Theocritus’ E 116 (n. 28) Vat. Gr. 913 = Theocritus’ H 113 Vat. Lat. 484 270 (n. 7) Vat. Lat. 5100 11, 269, 270 and n. 3 Vat. Lat. 7606 11, 269, 270 Valenciennes, Bibliothèque municipal Valentinianensis 411 11, 277–91 Venice, Biblioteca Nazionale Marciana Marc. Gr. 307 = Theon’s A 143 (n. 4), 144, 148 (n. 19) Marc. Gr. 474 = Aristophanes’ V 126-8 (and n. 13)