Dynamic Epigraphy: New Approaches to Inscriptions 9781789257892, 9781789257908, 1789257891

This volume, with origins in a panel at the 2018 Celtic Conference in Classics, presents creative new approaches to epig

259 85 4MB

English Pages 232 [233]

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents
List of figures
List of abbreviations
Editor’s acknowledgements
Chapter 1 Introduction: thoughts on the nature of inscriptions
Chapter 2 Towards a theoretical model of the epigraphic landscape
Chapter 3 Materializing epigraphy: archaeological and sociolinguistic approaches to Roman inscribed spindle whorls
Chapter 4 Written to be (un)read, written to be seen: beyond Latin codes in Latin epigraphy
Chapter 5 Epigraphic strategies of communication: the visual accusative of Roman Republican dedications of spoils
Chapter 6 Inscribing the artistic space: blurred boundaries on Romano-British tombstones
Chapter 7 When poetry comes to its senses:* inscribed Roman verse and the human sensorium
Chapter 8 Lassi viatores: poetic consumption between Martial’s Epigrams and the Carmina Latina Epigraphica
Chapter 9 Epigraphy and critical fabulation: imagining narratives of Greco-Roman sexual slavery
Recommend Papers

Dynamic Epigraphy: New Approaches to Inscriptions
 9781789257892, 9781789257908, 1789257891

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Dynamic Epigraphy

Dynamic Epigraphy New Approaches to Inscriptions

Edited by

Eleri H. Cousins

Oxford & Philadelphia

Published in the United Kingdom in 2022 by OXBOW BOOKS The Old Music Hall, 106–108 Cowley Road, Oxford, OX4 1JE and in the United States by OXBOW BOOKS 1950 Lawrence Road, Havertown, PA 19083 © Oxbow Books and the individual contributors 2022 Paperback Edition: ISBN 978-1-78925-789-2 Digital Edition: ISBN 978-1-78925-790-8 A CIP record for this book is available from the British Library Library of Congress Control Number: 2021951952 All rights reserved, with the exception of Chapter 7. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical including photocopying, recording or by any information storage and retrieval system, without permission from the publisher in writing. Chapter 7 is an Open Access article, distributed under the terms of the Creative Commons 3.0 Unported Licence http://creativecommons.org/licenses/by/3.0/. A copy may be downloaded from http://books.casematepublishing.com/Dynamic_Epigraphy.pdf

Printed in the United Kingdom by Short Run Press Typeset in India by Lapiz Digital Services, Chennai. For a complete list of Oxbow titles, please contact: UNITED STATES OF AMERICA UNITED KINGDOM Oxbow Books Oxbow Books Telephone (01865) 241249 Telephone (610) 853-9131, Fax (610) 853-9146 Email: [email protected] Email: [email protected] www.oxbowbooks.com www.casemateacademic.com/oxbow Oxbow Books is part of the Casemate Group

Front cover: Latin inscription at the base of the obelisk of Theodosius, situated at the hippodrome in Istanbul, Turkey, Wikipedia. Back cover: Epitaph of Atellia Cleunica. Museo Arqueológico Municipal de Cartagena (Spain). Photo © M. Cristina de la Escosura.

Contents List of figures�����������������������������������������������������������������������������������������������������������������������������vi List of abbreviations�������������������������������������������������������������������������������������������������������������� viii Editor’s acknowledgements��������������������������������������������������������������������������������������������������������x 1. Introduction: thoughts on the nature of inscriptions���������������������������������������������� 1 Eleri H. Cousins 2. Towards a theoretical model of the epigraphic landscape������������������������������������ 17 Kelsey Jackson Williams 3. Materializing epigraphy: archaeological and sociolinguistic approaches to Roman inscribed spindle whorls�������������������������������������������������������������������������������� 39 Alex Mullen 4. Written to be (un)read, written to be seen: beyond Latin codes in Latin epigraphy������������������������������������������������������������������������������������������������������������������������ 65 M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano 5. Epigraphic strategies of communication: the visual accusative of Roman Republican dedications of spoils�������������������������������������������������������������������������������� 95 Fabio Luci 6. Inscribing the artistic space: blurred boundaries on Romano-British tombstones������������������������������������������������������������������������������������������������������������������� 121 Hanneke Salisbury 7. When poetry comes to its senses: inscribed Roman verse and the human sensorium��������������������������������������������������������������������������������������������������������������������� 143 Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz 8. Lassi viatores: poetic consumption between Martial’s Epigrams and the Carmina Latina Epigraphica������������������������������������������������������������������������������������������ 177 Alessandra Tafaro 9. Epigraphy and critical fabulation: imagining narratives of Greco-Roman sexual slavery��������������������������������������������������������������������������������������������������������������� 201 Deborah Kamen and Sarah Levin-Richardson

List of figures Fig. 1.1: PH monogram, St Andrews (view towards St Salvator’s Chapel)  Fig. 1.2: PH monogram, St Andrews (view towards North Street) Fig. 2.1: The funeral monument of James Lumsden of Airdrie, c. 1598, Crail Kirkyard  Fig. 2.2: Detail of heraldic panel from Lumsden monument  Fig. 2.3: Detail of monogram on east side of Lumsden monument Fig. 2.4: Crail Kirkyard with early modern funeral monuments numbered  Fig. 2.5: The Lumsdaine of Innergellie monument in Kilrenny Fig. 3.1: Spinning with a distaff, drop spindle and whorl Fig. 3.2: Inscribed spindle whorl from Autun (number 6) Fig. 3.3: Find-spots of imperial-period inscribed spindle whorls  Fig. 3.4: Image of funerary stele depicting a woman holding a goblet, spindle, and distaff from cemetery Pont-l’Évêque, Autun  Fig. 3.5: Replicas of spindle whorls numbers 6, 8, 13, made for LatinNow by Potted History Fig. 4.1: Epitaph of Atellia Cleunica  Fig. 4.2: Epitaph of Cn. Atellius Toloco  Fig. 4.3: Stela of Pompeia Nereis  Fig. 4.4: Tabula mensa podiale of Alexander  Fig. 4.5: Stela of Boderus Fig. 4.6: Stela of M. Caecilius Flavinus and M. Caecilius Flavus Fig. 4.7: Stela of Annia Buturra Fig. 4.8: Stela of Λ(ουκίῳ) Αἰλίῳ Μελιτίνῳ Fig. 4.9: Sample of four different funerary inscriptions with analogous textual structures  Fig. 4.10: Distribution map of the find spots of the four inscriptions in Fig. 4.9 Fig. 4.11: Monument of Piedra Escrita  Fig. 5.1: Dedication by L. Quinctius Flamininus in Praeneste Fig. 5.2: Dedication by M. Fulvius Nobilior in Rome Fig. 5.3: Votive dedication by M. Claudius Marcellus in Rome Fig. 6.1: Tombstone of Marcus Aurelius Nepos, Chester, early third century AD Fig. 6.2: Tombstone of Flavia Augustina and Family, York, mid-third century AD

1 2 19 23 24 28 32 44 46 47 54 56 74 75 76 77 79 80 81 81 85 86 89 101 106 111 125 129

List of figures Fig. 6.3: Tombstone of Vellibia Ertola, Corbridge, late third or early fourth century AD Fig. 6.4: Tombstone of Sextus Simil[…], Chester, second or third century AD Fig. 6.5: Tombstone of L. Valerius Aurelius, Gloucester, second century AD Fig. 6.6: Tombstone of Philus, Cirencester, first century AD Fig. 6.7: Tombstone of Mamma, Carvoran, third century AD Fig. 7.1: CIL IV 1595 

vii

133 135 136 138 139 152

List of abbreviations AE L’Année Épigraphique Corpus Inscriptionum Latinarum CIL CILA Corpus de Inscripciones Latinas de Andalucia CLE  Bücheler, Franz, and Ernst Lommatzsch, eds. 1895–1926. Carmina Latina Epigraphica Corpus Signorum Imperii Romani CSIR EAGLE Europeana Network of Ancient Greek and Latin Epigraphy (https:// www.eagle-network.eu) Epigraphic Database Bari (https://www.edb.uniba.it) EDB Epigraphic Database Rome (http://www.edr-edr.it/default/index.php) EDR Fouilles de Delphes FD Jacoby, Felix, ed. 1923–. Die Fragmente der griechischen Historiker FGrH Epigraphic Database Heidelberg (https://edh-www.adw.uni-heidelberg. HD  de/home) Hispania Epigraphica HEp Hispania Epigraphica Online Database (http://eda-bea.es) HEpOl Inscriptiones Christianae Vrbis Romae ICVR IG Inscriptiones Graecae Decourt, Jean-Claude, ed. 2004. Inscriptions grecques de la France IGF Moretti, L. 1968–1990. Inscriptiones Graecae Urbis Romae IGUR Instrumenta Inscripta Latina IIL ILAlg Inscriptions latines d’Algérie ILCV Diehl, Ernst, ed. 1925–1967. Inscriptiones Latinae Christianae Veteres Vives, José. 1971–1972. Inscripciones latinas de la España romana ILER ILGIL Pfahl, Stefan F. 2012. Instrumenta Latina et Graeca inscripta des Limesgebietes von 200 v. Chr. bis 600 n. Chr Benzina ben Abdallah, Zeïneb. 1986. Catalogue des Inscriptions Latines ILPBardo  Paiennes du musée du Bardo Degrassi, Attilio, ed. 1957–1963. Inscriptiones Latinae Liberae Rei Publicae ILLRP Dessau, Hermann, ed. 1892–1916. Inscriptiones Latinae Selectae ILS Wuilleumier, Pierre, ed. 1963. Inscriptions latines des Trois Gaules ILTG Inscriptions latines de la Tunisie ILTun Inscr. It. Inscriptiones Italiae IRCP d’Encarnação, José. 1984. Inscrições romanas do conventus Pacensis IRMN Castillo, Carmen, Joaquin Gómez-Pantoja, and Maria Dolores Mauleón, 1981. Inscripciones romanas del Museo de Navarra

List of abbreviations

ix

IRT  Reynolds, Joyce and J.B. Ward-Perkins, eds. 2009. Inscriptions of Roman Tripolitania, enhanced electronic reissue by Gabriel Bodard and Charlotte Roueché (http://inslib.kcl.ac.uk/irt2009/) MRR Broughton, T.R.S. 1951–1952. The Magistrates of the Roman Republic PacHum Searchable Greek Inscriptions at the Packard Humanities Institute (https://epigraphy.packhum.org) RIB Roman Inscriptions of Britain Riese, Alexander. 1894. Anthologia latina: sive poesis latinae supplementum Riese, AL RIG  Recueil des inscriptions gauloises RRC Crawford, Michael H. 1974. Roman Republican Coinage SEG Supplementum Epigraphicum Graecum Suppl. It. Supplementa Italica Kovács, Péter and Ádám Szabó, eds. 2009. Tituli Aquincenses TitAq

Editor’s acknowledgements This volume has its origins in the 11th Celtic Conference in Classics panel “(Un)Set in Stone: Fresh Approaches to Epigraphic Material” at St Andrews in 2018. I want to thank all the panel participants, including those who were unable to contribute to this volume, for the lively, stimulating conversations that took place there, which planted the roots for many of the points I discuss in the introduction. My gratitude goes as well to the School of Classics at the University of St Andrews for their financial support of the panel’s invited speakers, and to Alice König and Myles Lavan for invaluable editorial advice in the early stages of the volume’s preparation. Contributors were writing and rewriting their chapters at the height of the COVID-19 pandemic, and I am tremendously grateful to them for their perseverance under a range of very difficult circumstances – and for their patience with my own pandemic-related snafus. Finally, it is likely this volume would not have been completed – and certainly not on time! – without the support of George Watson, whose eagle eye is a boon to copy-editing, and whose company keeps me above water.

Chapter 1 Introduction: thoughts on the nature of inscriptions Eleri H. Cousins

In the cobbled pavement in front of the university chapel of St Salvator’s in St Andrews sits a monogrammed PH (Figs 1.1 and 1.2). An observer stationed for a day within sight of it would see passersby engaging with it in a variety of different ways. Some people might walk by, on, or over it without any seeming awareness. Others would seem to avoid it deliberately – leaping over it, veering round it, walking at a distance but with a careful glance. Others still might seem to stop and gaze at it deliberately, in groups or by themselves, pointing or not, accompanied by human guides or book ones. The reason for this range of behaviours lies in the monogram’s role in the mythos of St Andrews student life. It marks the spot where, supposedly, the sixteenth-century Protestant martyr Patrick Hamilton was burnt at the stake. As such, undergraduates are told that the monogram bears a curse: anyone who steps on Hamilton’s initials while still a student will undoubtedly fail their Fig. 1.1: PH monogram, St Andrews (view towards St degree. Hence the way the monogram Salvator’s Chapel). Photo © Myles Lavan.

2

Eleri H. Cousins

Fig. 1.2: PH monogram, St Andrews (view towards North Street). Photo © Myles Lavan.

shapes the flow of human bodies in front of St Salvator’s: the contortions to avoid it by students, the spotlighting of it by tourists, the complete indifference from academic staff. For the latter, the monogram may exist more as concept than as physical text; as a lecturer at the University in the 2010s, I knew from my students that the initials existed somewhere, but must have walked over them countless times before I finally realized where that somewhere was. Hardly surprising: the monogram’s design and organic integration into the cobbles mean that the letters can serve as background visual noise if one is not already alert to their presence and meaning. The papers collected in this volume are deliberately diverse in their approaches to epigraphic material, together seeking to push the boundaries of epigraphy as a discipline and to demonstrate the analytical fruits of interdisciplinary approaches to inscriptions. The nature of the material they explore is equally diverse, though perhaps particularly focused on the Roman world; however, the emphasis throughout is on using their various case studies not as an end in themselves, or to illuminate any one period or place, but as a means of exploring broader methodological and theoretical issues to do with how we use inscriptions as evidence. Together, they offer a vision of a more dynamic epigraphy: both in the sense of inscriptions themselves, and how we study them. I have started this introduction with the St Salvator’s PH in

1.  Introduction: thoughts on the nature of inscriptions

3

part as a nod to the origins of this volume in a panel1 at the 2018 Celtic Conference in Classics, held in St Andrews. But the monogram also serves as an excellent opening gambit for exploring the dynamics of text in the material world, and for the intertwined themes and questions concerning the nature of epigraphy and our study of it that are raised by the chapters collected here. What is involved in an embodied experience of epigraphy? How do epigraphic texts merge with their surroundings and blur the lines between word, image, and landscape? How do the capacities of epigraphic language enable or hinder epigraphic communication? And to what extent are inscriptions perceived rather than read, and what roles do storytelling and intertextuality play in the conveyance of an inscription’s meaning(s)? These are all questions that are offered up by individual chapters in this volume, but especially by the interplay between them. This introduction, therefore, seeks to draw out this interplay, highlighting what I believe makes the chapters in this volume not only vibrant and exciting individually, but also how they come together as more than the sum of their parts to illustrate the ongoing potential of inscriptions and the way in which we think about them.

Epigraphy as an academic genre Historians have long mined inscriptions as historical sources, of course, but until fairly recently studying inscriptions for themselves was almost entirely the preserve of epigraphers. As similar sub-disciplines – for instance, numismatics or pottery studies – can bear witness, a need for specialist expertise can be a double-edged sword. The scholarship that results is usually extraordinarily rigorous within the bounds of the field and completed to a high technical standard. Yet the results, to outsiders, can often feel dry, perhaps even barren, and their capacity to contribute to wider debates can be opaque. Meanwhile, the high specialist technical barrier means that scholars from outside the sub-discipline rarely engage with the material in more than a passing way – and if they do engage in-depth, often leave themselves far too easily open to the charge of dilettantism. What results, then, is two parallel academic conversations, hyper- and hypo-specialized, neither of which do full justice to the material. Most of the chapters in this volume deal with inscriptions on stone, and here the problems have been particularly acute. These are extraordinary objects, with complex social functions and myriad methodological ramifications, but our traditional ways of dealing with them, both in publications and in museums, have rendered their “scope for the imagination” difficult to grasp. The printing requirements of traditional corpora, as Kelsey Jackson Williams (Chapter 2) and others have noted, all but erase the vital material component of inscriptions, reducing them to sanitized texts. Conversations with museum curators, meanwhile, almost uniformly stress the 1 ‘(Un)set in Stone: Fresh Approaches to Epigraphic Material’, 11th Celtic Conference in Classics, St Andrews, 2018.

4

Eleri H. Cousins

deep and understandable difficulties of displaying inscriptions in a way that catches the eye, let alone the sustained attention, of even the most committed museum-goer. As a result, epigraphic collections are often left to moulder in outdoor spaces, or, as in the case of a recent redisplay in a regional museum near my own university, removed from view almost entirely. Making inscriptions “interesting” can only partly be solved even by displays (the Galleria Lapidaria of the Capitoline Museums is a stand-out example) that go above and beyond in interpreting the language, the text, and the meaning of individual inscriptions to visitors, for inscriptions’ interest lies not only in their content (relatively easy to convey), but in the very idea of them (much more difficult).2 The challenges that render inscriptions difficult to appreciate as objects in their own right, alongside the constraints of traditional corpora, have contributed, then, to an academic environment that has tended to veer away from engagement with that idea of epigraphy. Epigraphic publications have traditionally focused more on the what rather than the why of inscriptions – beginning and ending with the reconstruction of the text, the establishment of linguistic and onomastic parallels, the tracing of individuals’ careers and family lineages, etc. Similarly to the identification and categorization of other forms of material culture, this is necessary preliminary work, but all too often undertaken as an end in itself, or as a prelude to using epigraphic texts solely as documentary evidence, divorced from archaeological or material context. All this has been changing in recent years, however. Scholarship on epigraphy, from all times and places but perhaps especially from the ancient Mediterranean and Greco-Roman world, has been undergoing a revolution of sorts, driven above all by an increasingly interdisciplinary interest in inscriptions as objects, as texts, and as objects/texts. This volume is therefore part of a growing wave of academic work that searches for new ways of seeing and analysing epigraphic material that move away from the more traditional forms of epigraphy as an academic genre, while still keeping inscriptions themselves centre-stage as the objects of study. Particularly productive areas of discussion in recent years have included the interplay between image and text on inscriptions,3 the materiality of inscriptions as physical objects,4 and the role of inscriptions in civic landscapes,5 as well as the role of epigram as a genre in ancient literature,6 and many of the chapters here draw or build on aspects of these discrete 2 This is not a trivial problem, nor one whose consequences are confined to the public perception of inscriptions: academic and public hierarchies of value for artefacts are more entwined, and academics’ experiences of museum spaces more formative for scholarship, than many of us would like to admit. No one is putting an inscription at the top of the Louvre’s Escalier Daru – or rather, the inscription on the base of the Winged Victory will never be the reason she is there. 3 Beginning in particular with Newby and Leader-Newby 2007; Squire 2009. 4 E.g., with varying degrees of success, the papers collected in Berti et al. 2017; Petrovic et al. 2018. 5 E.g., Ma 2013; Machado 2019. 6 A vast literature, although for the most part conducted within literary/philological disciplinary bounds, and so usually dealing with physical inscriptions primarily as prelude to examination of their literary counterparts; notable exceptions include the papers collected in Liddel and Low 2013.

1.  Introduction: thoughts on the nature of inscriptions

5

conversations. To date, however, these conversations have largely remained parallel ones, with little cross-fertilization between them. Volumes dedicated to individual threads have given us the chance to deepen each conversation, but at the expense of the dynamism which results from the interactions between threads. This volume, therefore, attempts to break new ground by putting these separate conversations in dialogue with each other, with – as the rest of this introduction will discuss – some constructive and thought-provoking results.

Embodied epigraphy Many of the chapters here are concerned with the embodied experience of epigraphy, and with considering human interaction with inscribed objects from a more or less explicitly phenomenological perspective. In Chapter 7, “When poetry comes to its senses: Inscribed Roman verse and the human sensorium”, Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz use the evocative language of Latin verse inscriptions to take us step by step through a sensory exploration of epigraphy, moving beyond our accustomed focus on sight alone into considerations of sound, touch, smell, taste, and even synaesthesia. The physicality – indeed the literal handling – of inscribed objects is also brought to the fore by Alex Mullen in Chapter 3, “Materializing epigraphy: archaeological and sociolinguistic approaches to Roman inscribed spindle whorls”. Here, the tactile and the physical are especially important, with Mullen exploring how the playful, possibly amatory messages on the whorls intersect with the way users may have rolled the whorls down their thighs to start them spinning, or worn them around their necks when not in use. Meanwhile, in Chapter 2, “Towards a theoretical model of the epigraphic landscape”, Kelsey Jackson Williams draws explicitly on archaeological and anthropological conversations of phenomenology to explore the varying dimensions of human interaction with an inscription in its landscape setting – a theme that is also echoed in the methodologies of M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano in Chapter 4, “Written to be (un)read, written to be seen: Beyond Latin codes in Latin epigraphy”, as well as the perspectives of Fabio Luci in Chapter 5 “Epigraphic strategies of communication: the visual accusative of Roman Republican dedications of spoils”. All three of these chapters speak to the ways in which we “zoom in” and “zoom out” during our engagement with an inscription placed into a wider landscape (physical, for the moment – below I will consider intertextual ones as well). Taken together, then, these discussions highlight a range of points that are crucial to the concept of an “embodied” epigraphy – that is, both the ways in which our experience of inscriptions is an embodied one, and how we can integrate that fundamental fact into our academic discourse. Perhaps the most important thread that emerges is how we conceptualise the sensory experience of inscriptions. We are accustomed to speak of “the viewer” when focusing our attention on a (usually theoretical) human interacting with/perceiving/reading an inscription. Indeed,

6

Eleri H. Cousins

a growing emphasis on the viewer has been at the forefront of some of the most creative work on epigraphy in recent years (riffing on the same turn in art history), and as de la Escosura Balbás et al. point out in Chapter 4, the integration of ideas of the viewer into our discussion of epigraphic monuments marked a key inflection point in thinking about epigraphy as monuments, rather than solely as documentary sources. The concept is powerful. But it is fundamentally focused on sight, and the papers here draw our attention to just how incomplete a framework this is for expressing the totality of human engagement with an inscribed object and its broader context. Indeed, although Jackson Williams, de la Escosura et al., and Luci frame their discussions of epigraphic landscapes in terms of “the viewer”, when their chapters are placed into dialogue with Cenati et al.’s systematic, self-conscious exploration of the role of each human sense in an epigraphic encounter, we can see that they have all in fact transcended the notion of epigraphy mediated through sight alone. Even while their discussions are rooted in what a person sees (and unquestionably sight is the most immediate and obvious sense we use to encounter inscriptions), their phenomenological methodologies, with an emphasis on landscape, perspective, and above all on movement towards, from, and around epigraphic monuments, mean that fundamentally they are not basing their interpretations in the viewpoints of static, disembodied eyes, but rather around human bodies in motion. Perhaps a better term than “viewer” for what is happening here would be “perceiver” – though that still does not do justice to the sorts of active, bodily encounters that are highlighted by Mullen (and which are perhaps easiest for us to grasp when speaking about small, handle-able objects such as spindle whorls as opposed to large-scale stone monuments – but which nonetheless clearly should be placed at the heart of our conceptions for both).

Capacities of epigraphic language The inscriptions dealt with in this volume vary from poetic texts dozens of lines long to laconic messages composed of a single word or a few letters. The authors of each chapter approach these texts and their language in a variety of different ways, but what becomes clear is that the capacity of an epigraphic text to convey meaning is not dependent on its length, and that the qualities of epigraphic language enable mechanisms of communication unique to the medium of inscriptions. This is particularly true of short inscriptions. Often abbreviated or condensed in their language, either shortening words or indeed skipping them altogether, they have not usually been read by scholars for much beyond the simple extraction of information, and certainly not close read. What the chapters here show, however, is that close reading of shorter texts is not merely possibly, but necessary, and that, rather than being a barrier to communication, the characteristics of their language can be seen as “a feature, not a bug”, in that they open up ways of delivering semantically complex messages that are not possible in other forms of writing.

1.  Introduction: thoughts on the nature of inscriptions

7

Let us consider, for example, the phenomenon of epigraphic formulae. Repetitious, conventionalised language is near-ubiquitous in inscriptions – indeed, it is one of the key characteristics of the medium across time and place, from the royal titulatures of Achaemenid monuments to the “Estd.” or “R.I.P.” of the present-day. Familiarity breeds contempt, as the saying goes, and formulaic epigraphic language, through its sheer ubiquity, becomes an afterthought for scholarship. We pay heed instead to the elements that stand out to us as individual; we discuss not the votum solvit libens merito (or more commonly VSLM) on a given Roman votive altar, but the name of the deity, the identity of the dedicator. Unless we choose to reproduce the whole text of the inscription, the dedicatory formula may not even be mentioned, elided over in academic analysis even as its presence lurks behind – is indeed fundamentally implicated in – our very identification of the votive as a votive. And this of course is the key point: formulaic language on inscriptions may render any one particular text seemingly “uninteresting”, but in the aggregate, formulae, far from diluting the effectiveness of inscriptions, are in fact what endow them with social potency. Each time a formula is used, its semantic significance is strengthened, and oscillates between the particular and the collective. So, the repetition of funerary formulae in a cemetery transcends the grief and loss of any individual family, and becomes a networked statement of communal rituals of mourning – and in so doing reinforces the power of any particular headstone to convey in a few words a narrative of particularised loss. The formulae I have used above have been deliberately presented (or reiterated) in abbreviated form, since the formulaic nature of epigraphic language is what primarily enables its abbreviated characteristics. Here too, as several authors explore (most notably de la Escosura Balbás et al. in Chapter 4, Luci in Chapter 5, and Hanneke Salisbury in Chapter 6, “Inscribing the artistic space: Blurred boundaries on RomanoBritish tombstones”), the dynamism of epigraphic communication lies as much – or more – in the form as in the content of the text. As de la Escosura Balbás et al. discuss in Chapter 4, the ability to recognise the meaning encoded in an abbreviation hangs upon shared knowledge of codes and sub-codes on the part of both creator and audience, and often the abbreviation can take on new meanings and connotations of its own, separate from the original, unabbreviated form. (Their modern example of the manipulation and ultimate re-textualization of the emoticon XD in Spanish is particularly thought-provoking in this respect.) When publishing epigraphic texts, we are accustomed to expanding out the abbreviations – to solving the riddle that they present. This is in fact the antithesis of how they are experienced in an epigraphic encounter, where the meaning of a familiar abbreviation is absorbed without recourse to spelling-out. Epigraphic language, in other words, requires a different sort of fluency to spoken language, or literary language: Petronius’ freedman who famously can read his lapidarias litteras needs a very different – but perhaps no less complex – sort of literacy to someone reading Cicero or Virgil.7 7

Petron., Sat. 58.7.

8

Eleri H. Cousins

It might be possible to push this even further, and stress the opportunities for deliberate ambiguity that may result from abbreviated language – or rather, to stress that we should be open to the possibility that an abbreviation can function polyvalently, as it were, rather than needing a single solution. As an illustration, in 1969, Duncan Fishwick, in a classic article, put forward an elegant solution for the systematic expansion of various abbreviations on Romano-British inscriptions into either numen Augusti or Numina Augustorum (distinctions with both theological and chronological implications).8 It surely is the case that the vast majority of the time either one or the other was intended. However, it is worth considering whether we should open ourselves up to the possibility of a multiplicity of meanings, and that abbreviations might stand in for multiple concepts simultaneously, whether deliberately on the part of the carver or commissioner, or as an after effect due to varying reader responses, even if those responses contradict or deviate from the carver’s intention. (For the latter, Death of the Author comes to epigraphy.) Moving beyond the question of possible expansions, and focusing on the dynamics of abbreviation itself, the tendency towards abbreviated language in epigraphy can be taken to two extremes. The first, explored especially by Jackson Williams (Chapter 2), de la Escosura Balbás et al. (Chapter 4), and Salisbury (Chapter 6), is when abbreviated forms cease or almost cease to represent a word at all, and instead become understood as images – transcending the textual entirely. This is a concept that feeds into broader discussions about the interplay between art and text. One particularly interesting dimension of the latter, illuminated by Salisbury’s work on Romano-British tombstones, is how we respond in instances where word and image seemingly deviate from each other in their content and message. For instance, in one tombstone from York (RIB 685; Fig. 6.2), commemorating a family of four, the sculptured relief portraits seem to be mismatched in both age and gender from the individuals listed in the inscription. Our conditioned response here is once again to find a solution: whether that be to establish against the odds a convincing oneto-one correspondence between names and portraits, or to blame the confusion on either artisanal error (this explanation is steeped in preconceptions about provincial culture, as I shall discuss below) or mismatch between a prefabricated relief and a personalised inscription. Salisbury shows how an interpretation can in fact be reached without recourse to a tidy solution, and how the differing capacities of text and image can come together to send complementary messages, not contradictory ones. This is reminiscent of the ways in which, for example, mythological scenes in ancient art exploit the capacities of the visual medium to tell stories both different and differently from those we encounter in literature: for instance, the non-linear visual storytelling that takes place on the metopes of Temple C at Selinunte, where Medusa holds close her offspring, Pegasus, who was born from the blood of her severed neck. 8

Fishwick 1969.

1.  Introduction: thoughts on the nature of inscriptions

9

The second extreme of abbreviation is when the text becomes so abbreviated that it vanishes entirely. Here, too, it is clear that we should lean into the gap rather than pretend it does not exist. Luci, discussing the role of statues as “visual accusatives” in the short dedicatory inscriptions of Roman Republican statue bases, stresses that translations or publications which “restore” the missing text detract rather than add to our understanding, ignoring how an inscription becomes more than its text. This brings us back to the issue of how epigraphic texts are published. We are now well-accustomed, as discussed above, to the notion that traditional corpora erase the material component of inscriptions, elevating the textual above all else. What we perhaps still do not sufficiently recognize, however, is the way that corpora, and the editorial conventions of publishing inscriptions, can fail to do justice to the text as well. Some obviously egregious examples are noted in this volume. On one tombstone discussed by Salisbury, for instance, the DM is deliberately placed to intersect with the name of the deceased, but in the Roman Inscriptions of Britain is published as a separate, initial, line (RIB 3074; Fig. 6.5). Mullen, meanwhile, discusses bilingual inscriptions where the two halves of a single text have been published in two different, linguistically segregated, corpora. But even epigraphic best practices can skew our experience of the capacities of epigraphic language outlined here. The expansion of abbreviations, the provision of missing words, what is or is not included in translations (when provided) – all of these create a very different text to the one confronted on the monument itself. This way in which our experience of inscriptions is almost inevitably mediated through their publication in corpora brings me to the next dimension of our engagement with epigraphy that the papers in this volume continually bring to the fore: the issue of intertextuality.

Intertextuality Intertextual dynamics are most obviously at play, and most explicitly considered, in the chapters that deal with longer texts, in particular those by Cenati et al. (Chapter 7), Jackson Williams (Chapter 2), and Tafaro (Chapter 8). Not coincidentally, the texts that these chapters engage with are also for the most part poetic in nature, with Cenati et al. and Tafaro focusing on Latin verse inscriptions from the Roman period, and Jackson Williams exploring Latin and Scots texts on early modern Scottish funerary monuments. These texts are self-evidently artistically ambitious, rich in allusion and imagery, with obvious affinities with literary poetic practice, and they take part in intertextual landscapes of varying sizes and scopes. Jackson Williams, for example, explores the highly localized intertextuality on the tombstone of James Lumsden in Crail, Fife, where verses in Latin and Scots play off each other on a single monument. Each in isolation emphasizes different aspects of Lumsden’s character, with the Latin verses conjuring a secular image of an educated member of the elite, and the Scots stressing a pious, Calvinist theology; together, they draw a more

10

Eleri H. Cousins

nuanced, multi-faceted picture of Lumsden than either can do alone. Moving beyond the stone itself, the verses are also in intertextual dialogue with similar texts on both contemporaneous and later monuments elsewhere in Fife, knowledge of which would add further dimensions to reader responses. Likewise, Latin verse inscriptions from the Roman Empire engage with an epigraphic landscape of inscribed poetry going back to the Archaic period. But the intertextual dimensions of verse inscriptions are of course not confined to monuments in stone; they cross into the world of literary verse written by hand on ephemeral media such as papyri. The ways in which written epigrams from the Hellenistic period onwards play with the tension between notionally nugatory and ephemeral literary verse and the pretensions to eternity of inscribed stone monuments have been the subject of much attention from scholars of ancient literature. Tafaro in this volume examines this tension as a bi-directional one, exploring not only how Martial riffs on the epigraphic form to construct his witty, tongue-in-cheek authorial persona, but also how analysis of the workings of Martial’s literary games allow us to recognize similar sophisticated dynamics at play in the epigraphy. Tafaro’s chapter also implicitly raises the question of a bidirectional intertextuality between literature and epigraphy not only for us as scholars, but for ancient encounters with inscriptions. Martial’s readers would have recognized the games he was playing from their own encounters with stone epigrams, but likewise at least some of the people commissioning and encountering verse epigrams would have recognized the norms from their knowledge of the literary version. Returning to the sensory, embodied, experience of epigraphy discussed above, we can see a similar dynamic at play in the concept of tasting or eating inscriptions. In their discussion of epigraphy through the lens of the five senses, Cenati et al. are almost apologetic about the absurdity of including taste in their analysis. However, not only do they themselves uncover some highly creative instances of literally eating the written word, Tafaro’s chapter also points out that the metaphor of taste when it comes to text is a common one, with Martial’s epigrams, for example, routinely playing on the double meaning (which works as well in Latin as it does in English) of taste in the sense of eating, and taste in the sense of discernment. The literary examples suggest that perhaps a conception of taste is not so far from epigraphic experience as we might expect. This two-way intertextual street between inscriptions and the written word is also addressed by Jackson Williams, in a way that raises issues for the broader study of all inscriptions, not solely verse ones. A few years after the erection of the Lumsden monument in Crail, its verse inscriptions were reproduced by John Johnston in his book the Heroes ex omni historica Scotici lectissimi, alongside other poems (epigraphic and not) commemorating other Scottish men of note. From this point on, Jackson Williams observes, any educated reader acquainted with Johnston’s book who visited the Crail Kirkyard would have read the inscribed verses in intertextual dialogue with their printed analogues. We can apply this to the issue raised by the volume contributors who are dealing with shorter, non-literary inscriptions of the way in which epigraphic corpora

1.  Introduction: thoughts on the nature of inscriptions

11

and the conventions of supplying missing words and expanding out abbreviations mediate our scholarly experience of epigraphic material. What these papers dealing with longer inscriptions demonstrate is that this dynamic can in fact be framed as an intertextual one. As with the hypothetical reader of Johnston, more often than not our physical encounter with a given inscription – if it comes at all – comes after our initial encounter with a text, an image, or a description on the printed or digital page. The inscribed text on the monument itself becomes in many ways, then, an aide-memoire to the expanded text encountered through the corpus. This has the effect of reducing or erasing entirely the embodied component of an epigraphic encounter – a component which, as I have discussed above, is at the heart of what inscriptions do. However, by being alert to the ways in which our responses are conditioned by the intertextuality between catalogue and object – by self-consciously treating the publication of an inscription not as a straightforward record of the object but as a text unto itself in intertextual dialogue with that object, we can move towards a more mindful negotiation of the relationship between inscribed object, printed or digital epigraphic corpus, and our own scholarship.

Storytelling The various threads I have discussed so far – embodied epigraphy, the nature of epigraphic language, the intertextuality that exists between epigraphic and literary texts – shed light on a much more fundamental question: how exactly do we read inscriptions? A recurrent point in many of the chapters here is the idea that, in keeping with our broader knowledge of reading processes in the ancient world, inscriptions would have been read aloud – and indeed that this is a quality that can be exploited by the genre. Some of the inscriptions explored by Cenati et al., for instance, are clearly playing with this dimension, for example by turning the reader of a funerary inscription into a mouthpiece for the deceased. De la Escosura Balbás et al., meanwhile, point out that out-loud readings would have been a mechanism by which literate – or epigraphically literate – people could convey information to non-literate companions. But what exactly do we imagine is being said in such a reading, especially for short or abbreviated texts? As the sections above have discussed, a significant portion of an inscription’s impact was encoded in non-verbal form – e.g., its shape, material, iconography – or in its textual characteristics – abbreviations and ellipses, the physical placement of texts on a stone, the framing of textual and visual elements. These are factors that would not be conveyed by a straightforward reading of the text alone out loud, and indeed, their impact, for most viewers, was likely to be unvocalized, even subliminal. In other words, while an inscription may theoretically be read (though, already in the case of abbreviated or elliptical texts, more accurately interpreted), an epigraphic monument cannot be, at least not in any literal sense.9 There are shades here of archaeological debates concerning the aptness of “reading” as a metaphor for human engagement with material culture. 9

12

Eleri H. Cousins

This insight raises a further point, which is how normal is it, in fact, to convey an inscription’s meaning, either to oneself or one’s companions, via straightforward reading? Let us use, as a thought experiment, the modern experience of wandering through some space festooned with epigraphy: a historic graveyard, for example, or other sort of commemorative space – Poets’ Corner in Westminster Abbey, perhaps. Our verbal engagement in these instances is mostly summative. We remark, for instance, on the sadness of a tombstone of a child who died young, or point out the memorial of a noted individual. Rather than repeating the information contained in an inscription verbatim, we provide commentary upon it, supplemented by our own emotional palette, knowledge of the inscription’s subject, or ways in which the monument reminds of other things we have known or seen. In other words, we do not read so much as tell stories to ourselves and our companions, stories that emerge from the intersection of the epigraphic object and our relationship to it. This has profound implications for how we understand the social role of inscriptions in past societies, but, as Deborah Kamen and Sarah Levin-Richardson in Chapter 9, “Epigraphy and critical fabulation: Imagining narratives of Greco-Roman sexual slavery”, make clear, it also has potentially revolutionary ramifications for how we conceive of and frame our own engagement with epigraphic material as scholars. In this chapter, Kamen and Levin-Richardson make use of critical fabulation, a methodology (originally developed by Saidiya Hartman for the exploration of the lives of enslaved individuals on the Trans-Atlantic slave route) that uses storytelling to speculate about the human experiences and narratives that lie beyond the strict limits of our documentary evidence. Using manumission inscriptions from Delphi and graffiti from the brothel at Pompeii as their case studies, Kamen and LevinRichardson imagine the events, told from a series of perspectives, that may have led to the creation of each text, and in so doing focus our attention on individuals, most notably enslaved women, whose experience and agency are often erased in our scholarship due to a lack of direct documentary evidence for their voices and actions. This is a methodology that is peculiarly suited to the nature of inscriptions themselves, which, as Kamen and Levin-Richardson discuss (and as I also dwell on elsewhere10), are objects that make visible for us particular moments in time in a longer story that both predates the creation of the inscription and continues on after it. Nevertheless, actively envisioning the undocumented portions of the story is, as Kamen and Levin-Richardson acknowledge, a self-consciously radical approach, blurring the traditional lines between scholarship and fiction in a way that some will find deeply uncomfortable. Yet what we must recognize is that in fact this sort of storytelling, stripped of the genre markers of fiction and endowed with those of academic writing, is in fact at the very heart of what any scholarship on past societies does. However, as Kamen and Levin-Richardson address, our usual speculations and deductions – our storytelling – about ancient material traditionally find their legitimacy by keeping 10

Cousins 2021, 207–208.

1.  Introduction: thoughts on the nature of inscriptions

13

to the bounds of topics deemed acceptable for academic conjecture via unspoken hierarchies of value and importance that prioritize certain types of people and certain sorts of historical questions. So, for instance, we are comfortable speculating about the undocumented lacunae in the careers of ‘important’ (for which read politically or militarily powerful, or simply socially elite) individuals, almost universally men, or we are willing to reconstruct the circumstances that may have led to the ratification of a treaty or the promulgation of an edict. Non-elite or enslaved men or women, however, are not perceived as “changing history”, and therefore we become comfortable with discussing their experiences only when the evidence is secure – when we are “sure”. The upshot of this is that we hold to a higher standard of proof discussion of the very lives that are often, by the nature of the evidence, more uncertain, more in the evidentiary shadows – ultimately resulting in their erasure altogether. The same dynamic is at play in other chapters in this volume, particularly those dealing with questions of gender. The suggestive messages of Mullen’s inscribed spindle-whorls, for example, have universally been read through the lens of a male gaze; however, as she makes clear, the only barrier to an interpretation focused on female agency is our own belief in male involvement as the default, safer interpretation in the face of uncertainty. Likewise, Salisbury stresses that some of the perceived mismatches between iconography and text on RomanoBritish tombstones are resolved if we accept more fluid criteria for the identification of female figures in particular.

Interdisciplinary epigraphy We have seen in the preceding sections how the interplay between chapters in this volume illuminates various aspects of the nature of epigraphy and epigraphic encounters. This is in no small part due to the interdisciplinary nature of the chapters, both individually and collectively. Cenati et al. and Tafaro both take Latin poetry and literary methodologies as their starting point, while Luci and Salisbury turn to art history and visual analysis. Mullen makes a powerful claim for “archaeologizing” epigraphy and employs the concept of translingualism from linguistics to deepen our understanding of the social function of bilingual texts, while Jackson Williams and Kamen and Levin-Richardson draw on methodologies from anthropology and history respectively. De la Escosura Balbás et al. perhaps are the contributors who write most explicitly from the standpoint of epigraphy alone, but do so in order to query the very procedures and assumptions of traditional epigraphic work. Finally, almost all the chapters are phenomenological to a greater or lesser degree – a coincidence, but one which perhaps suggests the general direction of travel in our academic understanding of epigraphic encounters. All of these chapters demonstrate the utility of combining traditional epigraphic analysis with methodologies from other disciplines, but they also show how the benefits of interdisciplinarity come not merely from the provision of analytical tools

14

Eleri H. Cousins

from other disciplines, but more fundamentally through the opening up of other ways of thought. To illustrate this further, I want to delve a little deeper into two intertwined disciplinary threads that run through many of the chapters, and which speak to my own particular research interests: Roman provincial studies and archaeological perspectives. The chapters by de la Escosura Balbás et al., Salisbury, and Mullen are all explicitly focused on material from the western Roman provinces, with Cenati et al. also employing case studies from Roman Britain and other provincial contexts. What feels particularly valuable about the way in which provincial epigraphy is deployed by these authors is how they place provincial evidence unapologetically at the heart of discussions that are fundamentally about much bigger methodological questions, with ramifications for contexts beyond the provincial. This is to a certain degree unusual in classical studies, in particular for provincial objects that do not conform to conventional classical standards of knowledge or aesthetics. Romano-British reliefs like the tombstone of Vellibia Ertola from Corbridge (Fig. 6.3), for instance, are often used to discuss the aesthetics (or aesthetic failures) of provincial art, but very rarely if at all considered as “good to think with” for issues beyond the boundaries of Roman Britain. In her discussion of Ertola’s tombstone, however, Salisbury sidesteps the question of aesthetic success or failure entirely, focusing her discussion solely on its complex dynamics of framing and spatial organization. In so doing, she not only frees the monument from our expectations of what it “should” be evidence for – she also injects fresh perspectives into art historical and epigraphic debates about the function of frames and spatial boundedness more broadly. Similarly, de la Escosura Balbás et al.’s myriad insights into the nature and function of epigraphic language stem from an outlook that refuses to erase the idiosyncrasies of provincial epigraphy and takes seriously the way in which provincial epigraphic language varies from a perceived Latin norm. This brings us back to the interaction between epigraphic text and epigraphic corpora discussed above – or more particularly to the question of translating inscriptions. As with the expansion of abbreviations and missing words, translations of inscriptions tend to smooth out variant language. Deity names are standardized, for example, while “misspellings” or grammatical idiosyncrasies are not reflected in the final translation. As the discussions in this volume of provincial or otherwise non-standard language make clear, however, this sort of interpretative tidying not only obscures a great deal of what makes any given inscription compelling, it also closes off conceptual and methodological doors that would allow us to build a deeper and more thoughtprovoking framework for epigraphy more broadly. To this end, the insistence of Luci, for instance, on literal translations that do not supply missing words, or of Kamen and Levin-Richardson on translations of technically “ungrammatical” language that convey something of the flavour of the original (“fock” for “futues”, for example, in Pompeian graffiti), is especially important. Likewise, Mullen’s translingual approach to the linguistic categorisation of the messages on the Gallic inscribed spindle-whorls, messages that blur the lines

1.  Introduction: thoughts on the nature of inscriptions

15

between Gaulish and Latin, actively celebrates the fluidity and flexibility of language use in provincial contexts. Similarly to Salisbury’s reliefs, where frames are used both to separate and to join, Mullen’s translingual spindle-whorls joyfully confound our imposed expectations about how language functions, broadening our interpretative horizons in the process. Mullen also raises the question of what it means to “archaeologize” epigraphy, and it is here that I would like to conclude. The obvious answer to this question is one of tools and methodology. At a very practical level, Mullen notes the evermore-important role of archaeological techniques and technologies in the study of epigraphy – for example, the increasing use of Reflectance Transformation Imaging (RTI) for the reading of inscriptions, or of Geographic Information Systems (GIS) for the mapping of epigraphic landscapes. At a methodological level, perhaps the most natural assumption is that an archaeological epigraphy would be one that pays particular attention to context, archaeological or otherwise. Context is undoubtedly critical, and it should (but still often does not) go without saying that we need to contextualize inscriptions as archaeological objects. However, as Mullen argues, perhaps the most crucial way in which we can “archaeologize” epigraphy is through outlook, and above all through the embrace of uncertainty that, she and I would both argue, lies at the heart of archaeology as a discipline. Archaeological evidence – inherently fragmentary, difficult to date, constantly changing in response to new discoveries or the revision of old ones – demands a receptiveness to open-endedness, to being continuously rigorous but also continuously flexible in our thinking: to deal with archaeological material is to be comfortable maintaining multiple equally possible, but often mutually exclusive, interpretations of the same evidence simultaneously. A similar need for the embrace of uncertainty underlies many of the issues outlined in the sections above, from our desire to “solve” epigraphic language to the dilemma of how to do justice to the barely documented lives of socially marginalized individuals. Counterintuitively, an uncertain epigraphy, more focused on questions than answers, may lead us to a more vibrant, more textured view of the past. This is just one of the ways in which the interdisciplinary approaches brought together in these chapters might change our outlooks on inscriptions and open up the bounds of epigraphy as a discipline. Similarly, I have undoubtedly outlined in this introduction only a few of the possible connections between the chapters collected here. Intertextuality applies, of course, to our academic discourses as well as to the texts we study. I recognize that it is not uncommon for readers of edited volumes to dip into only one or two papers, whichever ones most align with their broader interests, or are most immediately relevant to their current research questions, and the chapters in this volume will undoubtedly accrue layers of meaning for those readers via their intertextuality with other scholarship in other places. However, I hope that readers will also find, as I have, that the interplay of ideas within this volume also sparks conversations worth having.

16

Bibliography

Eleri H. Cousins

Berti, Irene, Katharina Bolle, Fanny Opdenhoff, and Fabian Stroth, eds. 2017. Writing Matters: Presenting and Perceiving Monumental Inscriptions in Antiquity and the Middle Ages. Matierale Textkulturen 14. Berlin: de Gruyter. Cousins, Eleri H. 2021. “Ritual on the Edge: The Dialectics of Religious Expression on the Frontiers of Roman Britain.” In Dialectics of Religion in the Roman World, edited by Francesca Mazzilli and Dies Van der Linde, 193–213. Potsdamer Altertumswissenschaftliche Beiträge. Stuttgart: Franz Steiner Verlag. Fishwick, Duncan. 1969. “The Imperial Nvmen in Roman Britain.” Journal of Roman Studies 59: 76–91. Liddel, Peter and Polly Low, eds. 2013. Inscriptions and Their Uses in Greek and Latin Literature. Oxford Studies in Ancient Documents. Oxford: Oxford University Press. Ma, John. 2013. Statues and Cities: Honorific Portraits and Civic Identity in the Hellenistic World. Oxford: Oxford University Press. Machado, Carlos. 2019. Urban Space and Aristocratic Power in Late Antique Rome: AD 270–535. Oxford, New York: Oxford University Press. Newby, Zahra and Ruth E. Leader-Newby, eds. 2007. Art and Inscriptions in the Ancient World. Cambridge, New York: Cambridge University Press. Petrovic, Andrej, Ivana Petrovic, and Edmund Thomas, eds. 2018. The Materiality of Text – Placement, Perception, and Presence of Inscribed Texts in Classical Antiquity. Brill Studies in Greek and Roman Epigraphy 11. Leiden: Brill. Squire, Michael. 2009. Image and Text in Graeco-Roman Antiquity. New York: Cambridge University Press.

Chapter 2 Towards a theoretical model of the epigraphic landscape Kelsey Jackson Williams

Since the advent of printing, the publication of epigraphic texts and, accordingly, their study, has been conditioned by the technology of moveable type. Gutenberg’s legacy made it easy to reproduce the text of an inscription, but expensive and timeconsuming to reproduce its more purely visual and spatial aspects.1 Examples of this history of text-centric epigraphic publication can be seen from the earliest printed studies of epigraphy, discussed by William Stenhouse, through the monumental publications of Jan Gruter and his successors in the seventeenth and eighteenth centuries, to the beginnings of the Corpus Inscriptionum Latinarum under the direction of Theodor Mommsen in the nineteenth.2 It is only in comparatively recent history, first with the development of post-letterpress printing technologies and second with the advent of the digital age that the visual and spatial characteristics of an inscription are likely to be recorded with the same precision and fidelity as its textual characteristics. With that change has come an increasing awareness that scholars should consider not only the textual meaning of an inscription, but also its art historical contexts and placement in the built and natural landscapes. Nonetheless, epigraphy remains a highly textual field; while the technologies of reproduction have changed, those of interpretation are only just beginning to catch up.

See generally Febvre and Martin 1958 and Eisenstein 1979. William Stenhouse has discussed the earliest epigraphic collections in detail in his Reading Inscriptions and Writing Ancient History: Historical Scholarship in the Late Renaissance (Stenhouse 2005). The publication of Jan Gruter’s Inscriptiones antiquae totius orbis Romani (Gruter 1603) marked an important watershed moment, as did the later edition of the same work edited by Johann Georg Graevius and his collaborators (Gruter 1707). For more on the still-murky history of epigraphic compendia between Gruter and Mommsen see Jackson Williams 2017, 80. The Corpus inscriptionum Latinarum began publication in 1853, under the supervision of Mommsen, and continues to appear under the aegis of the Berlin-Brandenburgische Akademie der Wissenschaften (see https://cil.bbaw.de/).

1 2

18

Kelsey Jackson Williams

The present chapter has been written with this historical context in mind. Its goal is to lay out a theoretical framework for understanding the totality of text, image, and surroundings in an epigraphic artefact, what might be called “the epigraphic landscape”. To do this, it has drawn on the resources of processual archaeology and landscape phenomenology, as well as more traditional art history and epigraphy, but has applied them – altering and developing their methods in the process – to a specifically epigraphic context. Rather than discussing this framework in the abstract, instead it will be demonstrated through a case study: an example drawn from the neglected corpus of early modern Scottish epigraphy. Using such a monument as a case study demonstrates that the methodology proposed here has applicability well beyond the classical period, however broadly defined, while also highlighting the ongoing reception, continuity, and transformation of classical epigraphic practices in early modern Europe; both classical and post-classical students of epigraphy can benefit from a methodological conversation begun across chronological boundaries. This exploration will begin with the monument itself, the most familiar object of epigraphic enquiry, and gradually work outwards, exploring the ever-widening ripples of meaning that it produces in its human and natural landscapes.

The monument Our case study is a late sixteenth-century funeral monument in the parish kirkyard of the royal burgh of Crail, a once prosperous fishing port at the far eastern tip of the county of Fife, midway down the North Sea coast of Scotland (Fig. 2.1).3 It is a product of Scotland’s post-Reformation Renaissance and an early post-Reformation example of what, in the Scottish context, is typically described as a “mural monument”, a substantial architectural construct, almost always for a funereal purpose and with an epigraphic inscription at its centre, which would be built into the wall of a kirk, kirkyard, or burial aisle.4 The text on our case study makes it clear that it pertains to one James Lumsden of Airdrie and before considering the stone itself it will be useful to recover what we can of the socio-biographical contexts of its subject from other sources. Airdrie is a small estate four miles to the west of the present village of Crail and had been owned by the Lumsden family as early as 1450, when it was erected into a barony along with other lands in Fife and the county of Haddington.5 The first notice of James Lumsden, the subject of the funeral monument, occurs a little over a hundred 3 “Kirk”, “kirkyard”, and analogous terms in Scots are equivalent to the English “church”, “churchyard”, etc., and have been preferred in this chapter. 4 Graham 1960, 212–18, offers a useful taxonomy of early modern funeral monuments, describing mural monuments as “consist[ing] of three main parts – a pedestal, columns and entablature framing a central inscribed panel, and a pediment, the last often topped by a finial. The columns are usually flanked by massive scrolls, and the whole is crowded with Renaissance ornamentation and funerary emblems”. 5 Thomson et al. 1882–1914, vol. 2, no. 402.

2.  Towards a theoretical model of the epigraphic landscape

19

Fig. 2.1: The funeral monument of James Lumsden of Airdrie, c. 1598, Crail Kirkyard. Photo © Kelsey Jackson Williams.

20

Kelsey Jackson Williams

years later when, as a minor and the second son of John Lumsden of Blanerne or of that Ilk, he was retoured heir of tailzie to his cousin (consanguineus) Thomas Lumsden of Airdrie in that barony on 14 January 1566.6 He would have been about 10 or 11 years old at the time. The acquisition of a barony might ordinarily suppose a degree of wealth, but James acquired his estate under less than favourable terms. When he was retoured heir to Airdrie, two widows of former owners were then living: Euphemia Lundin, widow of William Lumsden of Airdrie – who was still alive in 1582 – and Marjory Douglas, widow of Thomas Lumsden of Airdrie, who was living as late as 1591.7 Scottish law reserved part of the estate and its revenues, the “terce”, to these ladies during the terms of their lives.8 As well as this burden, James, upon his inheritance, had also obliged himself to pay substantial sums of money, presumably as dowries, to the three sisters of his cousin and predecessor.9 Despite these financial burdens, James rebuilt the house at Airdrie, overseeing the construction of the present tower house whose armorial panels date its completion to 1588 (RCAHMS 1933, 61–2). Raising funds, however, required making an inroad into the core of the estate. On 25 October 1587 he feued (leased) 16 acres of arable land near Cupar to David Jamesoun, a burgess of that burgh, on 24 June 1591 he sold outright his lands of Gleghorn across the Forth to an Edinburgh merchant, while his lands of Powran were disposed of later that year to a fellow laird.10 By 24 March 1598, a few months before his death, the remainder of the barony of Airdrie had been temporarily granted to Archibald Douglas of Whittinghame, an Edinburgh lawyer who was presumably amongst his chief creditors.11 While his family held onto the remains of Airdrie until its final disposition in 1605, James’s tenure marks the beginning of the estate’s descent into ruin.12 James was a creditor as well as a debtor, but equally unsuccessful on this side of the balance book. At some point before 24 July 1590, he had lent to Jean Lyon, Countess of Angus, the substantial sums of 9,208 pounds Scots and 8,000 merks in two separate bonds. The countess, however, refused to acknowledge the debt and it was left to her kinsman William Douglas, 9th Earl of Angus, to negotiate a bond for repayment with Lumsden. Accordingly, the countess gave James “a grite fair diamant sett in gold” valued at 2,000 crowns, apparently as a partial payment of her debt. Unbeknownst Record Commission 1811–1816, vol. 1, Fife, no. 61. In Scottish law a “retour” was the return drawn up by an assize confirming the ownership of land by the heir of a deceased individual. A “tailzie”, similar to the English entail, was an inalienable settlement of property on a specific, usually agnatic, line of individuals. See Stair 1759, 228–9 (tailzie), 494 (retour). 7 Beveridge 1893, 138, 140. 8 Stair 1759, 287. 9 Beveridge 1893, 144. 10 Thomson et al. 1882–1914, vol. 5, nos. 1676, 1886, 1986. 11 Thomson et al. 1882–1914, vol. 6, no. 692. 12 The final sale in 1605 to William Turnbull of Pittencrieff is recorded in Thomson et al. 1882–1914, vol. 6, no. 1611. 6

2.  Towards a theoretical model of the epigraphic landscape

21

to James, however, the jewel belonged not to the countess, but to the crown, and a few days later on 4 August he and his servant David Ferry were denounced as rebels for not having appeared to answer for their possession of the same. While James eventually delivered the jewel to the king and was exonerated of any wrongdoing, it seems unlikely that he received sufficient compensation from his aristocratic debtor.13 Nor did James’s misfortunes end there. On 27 June 1592 he was one of the chief actors in the Earl of Bothwell’s attempted assault on the king at Falkland Palace.14 His tangled affairs with the Countess of Angus were now used as a weapon against him by the royal party when her new husband, Alexander Lindsay, Lord Spynie, had James put to the horn (declared an outlaw) for debts allegedly owing to her.15 By September he was in royal custody and attempting to parley his way out of a sticky situation by implicating “Lord Hamilton and others of good quality” in the conspiracy, “a matter troubling the King greatly”.16 He was released later in the month, presumably in part due to his willingness to cooperate with the crown.17 While Robin Macpherson assumes he was tortured during his imprisonment, thus in part explaining his eagerness to compromise other supporters of Bothwell, this seems to be supposition only.18 Parallel to this chequered financial and political career, there are indications that Lumsden was a member, at least to some degree, of the hardline Presbyterian faction in the Scottish church, a faction centred on the east of Fife. From the time of the 1584 so-called “Black Acts”, which established the supremacy of the crown and the episcopal order over the Scottish kirk, the theologian Andrew Melville and other like-minded individuals had formed a Presbyterian party in opposition to the royal position.19 Melville himself was rector of the University of St Andrews until his deposition in 1597 and, as Jamie Reid Baxter has established, many of his adherents formed a tightly-knit intellectual and theological circle centred on east Fife.20 Lumsden himself had several links to this circle. His brother Robert had married Isobel Cor, daughter of the leading Presbyterian Clement Cor and herself a friend of the Presbyterian poet Elizabeth Melville, and both James and Robert had interests in Hebridean fishing, interests which would lead the latter, with other hardline Fife lairds, to invest in the ultimately disastrous Lewis Plantation early in the seventeenth century.21 While no direct evidence for Lumsden’s religious leanings is known (beyond what will be presently deduced from his funeral monument), the larger social groups in which he moved suggest a distinctly Calvinist and Presbyterian ethos.22 As such, The relevant documents are quoted at length in Beveridge 1893, 146–8. Salisbury 1883–1976, vol. 13, 465; Bain et al. 1898–1969, vol. 10, 708. 15 Bain et al. 1898–1969, vol. 10, 763. 16 Bain et al. 1898–1969, vol. 10, 771. 17 Bain et al. 1898–1969, vol. 10, 779. 18 Macpherson 1998, 521. 19 For the historical context see Wormald 1991 and Mason and Reid 2014. 20 Reid Baxter 2017. 21 MacCoinnich 2015, 100, 411, and passim; Reid Baxter 2017, 61. 22 Reid Baxter 2017, 64, asserts that James’s “piety is not in question”, based on the same evidence. 13 14

22

Kelsey Jackson Williams

the funeral monument erected over his corpse reflects both the lavish expenditure which characterised one part of his life and, as we shall see, the Presbyterian faith with which it coexisted. This biographical account provides us with a sphere of reference within which we can situate the stone’s subject, though it should be emphasised that while the possibility of recovering such fine-grained biographical detail surely enriches an epigraphic study, it is hardly essential; the same process proposed here could as easily be applied to a carved stone, classical or early modern, whose subject is otherwise entirely absent from the written record. We can now turn to the layers of meaning present in the stone itself: the visual, the symbolic, and the textual. The first is the purely visual appearance of the object, its power as a “material medium” separate from any supra-visual codes.23 The symbolic consists of the meanings that can be extracted from symbols present on the object: in the case of the present example, emblems of mortality, heraldry, monograms, and, indeed, the visual aspects of the texts. Finally, the textual embraces the meanings that a literate person fluent in the appropriate language(s) could extract from an inscription. In this instance, there are multiple textual layers to be explicated dependent upon a reader’s knowledge of Scots and/or Latin. The visual codes move from the general to the minute. Most obviously and essentially, the shape of the Lumsden tomb, its position built into the wall of the kirkyard, the recessed niche below, and the massive pediment above, all characterise it as a funeral monument. They also characterise it as a particularly lavish example of the genre; there were few tombs in east Fife built during the later sixteenth century that could match Lumsden’s in size and ornament. The only immediately comparable example would have been the massive mural monument to Robert Stewart, Prior of St Andrews, dated 1586, in St Leonard’s Chapel, St Andrews, 10 miles to the north.24 Looking more closely, a well-travelled viewer would notice additional visual cues. The symmetrical spires that give the monument its immediately recognisable appearance are characteristic of sixteenth-century Dutch architecture, one of several visual linkages between the burgh of Crail and its trading partners across the North Sea. By contrast, however, the capitals of the lower order of pillars are decisively medieval in character, suggesting a familiarity on the part of the sculptor with the rich medieval architectural heritage of eastern Scotland.25 Taken as a whole, the monument consciously exists between old and new styles, remaining international, novel, and eclectic without rejecting the medieval heritage surrounding it. The symbolic codes of the monument situate it within the Calvinist and noble cultures we have already seen. Prominent to a viewer are the facing heads projecting from the See Tilley 2008, 19–20, for “the material medium of the rock”. RCAHMS 1933, 247. Stewart’s tomb, like Lumsden’s, echoes a medieval canopy tomb with its lower recess partially protected by pillars, but has a smaller, square pediment flanked with scrolls rather than the massive upper range of the Crail monument. 25 RCAHMS 1933, 60, suggests that the capitals have been taken from “early 13th-century models”. 23 24

2.  Towards a theoretical model of the epigraphic landscape

23

interior planes of the frieze. On the left is a spade-bearded firgure apparently meant to represent Lumsden himself, on the right a death’s head. This is a more sophisticated variation on the crossed bones and hourglass so common in Scottish funeral carving: a memento mori, which in this instance has taken the form of a truncated echo of the transi tombs of the later middle ages.26 It represents the ideal “good death” (mors beata) prized by medieval and early modern Christians and is, in turn, a reminder to the viewer to prepare for their own inevitable mortality.27 This reminder of the transience of the flesh and the vanity of the world exists in tension with a symbol representing just such vanity: the heraldic achievement that occupies the central panel of the pediment (see Fig. 2.2). Heraldry is the most Fig. 2.2: Detail of heraldic panel from Lumsden ubiquitous form of symbolic code monument. Photo © Kelsey Jackson Williams. across carved stones in Scotland – indeed, across carved stones in early modern Europe – and performs a variety of functions for the observer aware of its meaning.28 The nature of a heraldic achievement as a visual identifier used by multiple generations of an agnatic kinship group links it both to the individual and to a dynastic history; in other words, the panel in Crail both signifies the position of James Lumsden as the subject of the monument and links him to a deeper history of Lumsdens as members of a gentry kinship group. Parallel with the heraldic achievement are the intricate circular monograms to either side of it and the additional monogram on the eastern side of the monument, facing outwards to the rest of the kirkyard. The monogram on the side is palindromic, reading I L OA L I, i.e., Iames Lumsden of Airdrie, and is echoed in more compressed form by the monograms to the left and right of the heraldric panel each of which reads J L O A with “JAMES LUMSDEN” cut in small capital letters above the left monogram and “DE ARDRIE” above the right. A similar monogram is present on Airdrie House 26 For a concise introduction to emblems of mortality on Scottish tombs see Willsher 2005, 38–41. For the transi tomb tradition see Cohen 1973. 27 See Lahtinen and Korpiola 2018. 28 Thiry 2014.

24

Kelsey Jackson Williams

and it would appear to have served as the personal emblem of James himself, an individual symbolic marker to pair with the dynastic marker of heraldry.29 At this stage we have already reconstructed a rich nexus of meaning present in the monument before even beginning to discuss the textual codes present, that is to say the inscriptions to which a more traditional approach to epigraphy might limit itself. There are Fig. 2.3: Detail of monogram on east side of Lumsden five distinct texts – not counting the three monograms – present on the monument: monument. Photo © Kelsey Jackson Williams. three placed in discrete panels along the length of the frieze and two occupying the large panels below the frieze. All are now significantly eroded and, in parts, entirely illegible but a careful inspection of the monument itself, combined with the readings given by Beveridge in 1893, allows for a partial transcription:30 Frieze, central

Prima decvs thalamos et opes mihi contvlit æt[as] proxima et innumeris avcta pericla malis vivere cvm desii vixi qvod defvit ævi mortalis nobis vita beata dedit l.1 Beveridge reads aer[a] for aet[as]; l. 2 Beveridge reads immeritis for innumeris

Frieze, right

Iames L[v]msden of Ardrie his a[na]gram [l]ord ie[s]vs made man frie died xxiii agvst 1[59]8

Frieze, left

Hic dormit vir pivs et nobi lis iacobvs lvmsden de ard rie qvi obiit 23 Avgvsti an no domini 1598 ætatis suæ 43

Central panel, right

To the savil depart[ed s]ore d[istrenȝeit] A similarly complex monogram founded upon the geometrical proportions of a perfect circle can be seen in the carved details of the laird’s loft at Pitsligo, Aberdeenshire, which dates to the early 1630s (cf. Chernoff 2012, 115 and figure 5.9). 30 In my transcription practice I have followed a slightly simplified version of the Leiden System as summarised in Cooley 2012, 352–5. 29

2.  Towards a theoretical model of the epigraphic landscape my sillie savil that hes sa lang indvre[d] the wretchit woes of wardlie miserie now in thy wayis of welth fvl weil asvrd and [dea]f to wardlike [w]ardlie vanitie [g]o nestle [f]irst be[low] and then go frie frome dvngeone dark and [fearfull] sl[av]er[ie] stand not in dovt bot [b]oldlie go and sie that sight that fvllie al thy hairt c[a]n [gre] [of si]nfvl slovghe qvhan in al thy [ma]gest[ie] [ovt of] the grave til thov fr[om]e dvst be f[rie] [- - -] grap[p]le t[il - - -] [- - -] can his [- - -] [- - - - - -] [- - - - - -]

25

gre, to be in agreement with

1. 1 Beveridge reads d[istravght] for d[istrenȝeit], but the former is not attested in Middle Scots. Distrenȝe in its sense of “to subject to constraint or distress” is the more linguistically plausible reading.

Central panel, left

[ - - -] in his [- - -]st hes grantit the that grace [- - -]e qvha sinneris til imbrace t[h]at cursit [- - -]ad agane [- - -] o pla [- - -] [- - - - - - &c.]

The bilingual nature of the inscriptions immediately presents multiple levels of meaning: one which could be construed by a literate individual fluent only in Scots, another open to a member of the educated Scottish elite with access to Latin, and a third corner case in which we might imagine a foreign visitor, literate in Latin but not in Scots. The Scots poems on the central panels are both the most accessible and most physically present inscriptions, occupying the space broadly at eye-height for an individual standing in front of the monument. While the left-hand poem is too worn to allow for more than a conjecture at its contents, the right-hand poem presents us with more than enough to allow for analysis. It begins with a title, “To the savil departed sore distrenȝeit,” and is in the form of a 14-line sonnet in iambic pentameter following an extremely tight rhyme scheme of ABAB BBBB BB[--] [--]. Its sense is a devout Calvinist turning away from the “wretchit woes” of the world and towards a union with the divine. The divine “magestie” appears at the turn of the poem, between octave and sestet, and it seems plausible that the description of earthly imprisonment in the octave would have been balanced with a description of heavenly freedom in the obliterated final lines. This poem, together with its now illegible companion, which seems to have dwelt on the subjects of grace and salvation, are paired linguistically with the Scots inscription on the right-hand side of the freeze. Lumsden’s name and his date of death are plainly given along with the almost playful anagram of his name: “lord

26

Kelsey Jackson Williams

iesvs made man frie”.31 Collectively, they paint a picture of conventional Calvinist devotion, one that would hold Lumsden up as a pious model for the Scots-literate readers of his parish. The Latin texts offer a distinctly different inflection to this presentation of Lumsden. The Latin memorial inscription on the left side of the frieze roughly echoes the Scots on the right in its statements of Lumsden’s name and the date of his death (adding his age, which is not present in the Scots), but its language – “Here rests a pious and noble man” – emphasises Lumsden’s secular rank in a way that is not present in the Scots. A similar focus on Lumsden’s earthly life can be seen in the four lines of verse that occupy the centre of the frieze. The first two lines contrast the glory of Lumsden’s youthful marriage and wealth with the “numberless evil dangers” of his later life, while the second ends on a more conventional truism that “when I ceased to live, I lived” for “what is missing from the mortal life, a blessed life gives to us”. Again, the fortunes of Lumsden’s secular career are given more prominence in the Latin than in the Scots, offering a more nuanced interpretation for a bilingual than for a monolingual reader. A sufficiently well-read, Latinate contemporary of Lumsden would have recognised even more textual traces and echoes encoded into the face of the monument. The prima and proxima aetas of the first two lines of Latin verse allude back to Aristotle’s three ages of man, particularly in their emphasis on the reversal from the noble idealism of youth to the pessimism and uncertainty of mature age.32 Likewise, the idea of living in Christ only when one had ceased to live in the world was a Christian truism in early modernity for Catholics and Protestants alike. In the years following Lumsden’s death, our hypothetical learned reader might, however, have become aware of a more proximate analogue for the Latin verses. In 1603 the Heroes ex omni historia Scotica lectissimi of John Johnston was published in Leiden for sale in Edinburgh and included a printed version of the same verses that appear on Lumsden’s tomb.33 At the time of its publication Johnston was master of St Mary’s College, St Andrews, and one of the most outspoken hardline Presbyterians in the Scottish church, joining Andrew Melville in challenging the royal power and James VI and I’s episcopal hierarchy.34 The Heroes, while notionally an impartial poetic paeon of praise to famous Scots, had a strong Presbyterian and east coast bias; Lumsden’s inclusion would undoubtedly have strengthened any viewer of the monument’s suppositions as to his religious leanings as well as placing him within

31 Anagrams on names in this way were a common form of literary play in early modern Europe, see Camden 1605, 150–7, and the numerous examples given therein. 32 Aristotle, Rhetoric 2.12–14, 1388b–1390b. 33 Johnston 1603. 34 M’Crie 1824, vol. 2, 284.

2.  Towards a theoretical model of the epigraphic landscape

27

the circle of an influential and controversial national figure, adding an ex-post-facto intertextuality to the monument’s inscriptions.35 The textual codes on this monument can be seen to operate on at least three layers: Scots, Latin, and intertextual. When this is added to the already densely layered series of visual and symbolic codes, we can begin to see the full range of possible interpretation that a viewer could derive purely from the monument itself, while also recognising that only a privileged few would possess the keys necessary to explicate every code present. However, I shall proceed to argue that these many layers of meaning still only represent a proportion of the totality of meaning that we can derive from this, or any other, carved stone.

The stone in its space We should now consider the stone in its space. “Its space” here means both the physical spaces immediately surrounding the stone, but also the spaces it creates by virtue of its placement within those physical spaces. This can best be explained with reference to the theories of Chris Tilley. Tilley’s works on landscape phenomenology, The Materiality of Stone and, more recently, Body and Image offer a theoretical toolbox for opening up new possibilities in epigraphy and it is useful to tease out how some of his ideas can apply in this field.36 Speaking of prehistoric rock art, Tilley posed himself a set of intriguing questions: I wanted to experiment, [he wrote], with a phenomenologically informed kinaesthetic approach to … rock art. In other words, I was interested in what effects the carvings themselves had on my body as someone looking at them: What did I have to do to see the carvings? How did I have to move?37

His conclusion was that “it was not possible to see the carvings in any way that I might wish, or decide”. Fair enough, one might say. If a stone has carving on its front, you cannot view that from the back. But Tilley drew more far-reaching conclusions. “The carvings,” he wrote, “were exerting their own power and influence in relation to what I saw and from where I saw it, and how I saw it. I was no longer a free agent … There was a dialectic at work between the rock itself, and its landscape location, and the positioning of the images carved on it.”38 In many ways, this is an application of David Turnbull’s earlier work on the Maltese megaliths. Turnbull’s point was – once again – that:

35 Reid Baxter 2017, 63, has suggested that Lumsden’s joint donation with Clement Cor of two yards in St Andrews to St Leonard’s College may also have played a role in his celebration by the St Andrean Johnston. 36 Tilley 2004; 2008. 37 Tilley 2008, 16. 38 Tilley 2008, 16.

28

Kelsey Jackson Williams People perform objects of all kinds, but especially buildings, by moving through and around them but buildings also perform people by constraining their movements and by making likely certain kinds of encounters between them and others.39

In other words, the physical monument forces us to move in certain ways if we want to interact with it and that accordingly deforms and reshapes the human spaces around it. This can be usefully applied in the present case study with reference to the map in Figure 2.4 of Crail Kirkyard as it would have appeared around 1725 (the Lumsden monument – no. 11 – is circled). By 1725 – the end of the first wave of monumental building in Crail – the kirkyard was full of elaborate mural monuments, 16 in all, which are here numbered on the west and south walls. In 1598 when the Lumsden monument was constructed, however, this topography would have looked very different. This was only a generation after the Scottish Reformation in 1560, which had dramatically disrupted traditional burial practices. The custom of “kirk burial” was vehemently opposed by the religious establishment, but new habits of exterior burial epigraphy were still very much in

Fig. 2.4: Crail Kirkyard with early modern funeral monuments numbered. © Kelsey Jackson Williams; based on the map by Beveridge and with now-demolished sections of the kirk highlighted in grey (Beveridge 1893, plan facing 67). 39

Turnbull 2002, 135.

2.  Towards a theoretical model of the epigraphic landscape

29

flux depending on the individual locale.40 Not long after the erection of our case study, a mausoleum in nearby Collessie reminded passers-by of this new epigraphic and burial environment with the determined lines: Defyle not Christ’s kirk with your carrion A solemn sait for God’s service prepar’d For praier; preaching and communion Your burial should be in the kirk yard41

What, then, can the Lumsden monument’s place in the landscape tell us? The arrow on the map follows the path that leads from the kirkyard gates to the main door of the kirk itself. This is the path that the inhabitants of the burgh would have trod every Sunday. Now consider the location of the Lumsden monument. When it was constructed, its builders would – we may presume – have had the entire western wall of the kirkyard accessible to them, the other monuments not yet having been erected. Why then did they choose the furthest possible location still in the sight line of the main path? There are two competing motives at work here. The attempted ecclesiastical regulation of burial practices has already been discussed above. This had been an ongoing issue, but one on which the General Assembly – the supreme authority in the post-Reformation Scottish church – had been taking an increasingly strict line. In 1588 – only 10 years before the date of the Lumsden monument – the Assembly had made the latest in a series of instructions, complaining that “albeit inhibitioun hes bein diverse tymes made for avoyding” the abuse of kirk burial, “yet the acts and constitutiouns of the Kirk are daylie brockin”. The new act recommended to the civil power the outlawing of “burial within kirks, and sicklyke erecting of tombis, and laying of troghes in kirkyeards”.42 The problem, as the General Assembly well knew, was that prominent burial monuments were an integral part of Scottish elite culture and one which the elites were loath to part with. The Lumsden monument, as the first major post-Reformation burial monument within the parish of Crail, acted as something of a litmus test for how the local community would respond to the national decree. The way in which James Lumsden’s heirs handled this question was masterful. The monument is placed almost as far away from the kirk door as possible, creating a spatial statement of the family’s pious submission to the decrees of the kirk. At the same time, however, it is visible – albeit at a distance – from the moment a viewer enters the kirkyard gate until they walk through the kirk door itself. The monument has been perfectly poised in the three-dimensional space of the kirkyard so that it simultaneously performs a particular post-Reformation, Calvinist form of religious obedience while still firmly reminding its viewer of the Lumsden family’s power and wealth. Cf. Spicer 2000. Spicer 2000, 149. 42 Kirk of Scotland 1839, vol. 2, 733. 40 41

30

Kelsey Jackson Williams

It seems likely that this careful positioning of the monument played a role in dictating how the subsequent mural monuments at Crail were situated. The majority of the monuments noted in Figure 2.4 are built into the west wall, forming an aisle up which a potential viewer would have to walk in order to view the Lumsden monument. Most of the monuments on the south wall are late – from the beginning of the eighteenth century – and likely reflect a lack of available space on the western wall. In effect, the Lumsden monument created a circulation of people through space – it performed people, to borrow Turnbull’s phrase – in such a way that it encouraged the builders of subsequent monuments to group their works along an already extant axis, reinforcing and developing the spaces which it had brought into being. In short, then, everything that can be identified about the location of this monument in its immediate space is productive of meaning. It is situated so as to project specific religious and cultural meanings and it has shaped the nearby space in such a way that subsequent monuments have been built in line with the spatial circulation it created. Let us now cast our eyes further afield and think about the stone, not just in its space, but in its landscape. Tim Ingold, in his 2000 collection of essays, The Perception of the Environment, writes at length about what he calls “the practice of wayfinding”.43 In Ingold’s conception, wayfinding is the activity of the native, the local, while navigation is the activity of the outsider. He argues that: While dwelling in the world entails movement, this movement is not between locations in space but between places in a network of coming and going that I call a region. To know one’s whereabouts is thus to be able to connect one’s latest movements to narratives of journeys previously made, by oneself and others … places do not have locations but histories. Bound together by the itineraries of their inhabitants, places exist not in space but as nodes in a matrix of movement … a “region”.44

If we entertain, for a moment, this conception of movement, we can proceed to consider how a carved stone or monument could work within these “regions”. The stone itself is a waymarker, a point which has gravity in the landscape and which accretes to itself histories and narratives – that seems clear enough – but it also exists within a larger matrix that may contain other stones or other objects or spaces that act as analogies for stones. This leads us back to the layers of meaning discussed earlier. A viewer existing in the wayfinding space conceptualised by Ingold would not only be able to extract visual, symbolic, and/or textual meanings from a stone, they would also be able to extract inter-visual, inter-symbolic, and/or inter-textual meanings based on their experience of other stones or analogous forms elsewhere in the region within which they existed. Put more simply, the presence of a region – which may be geographical or may also be intellectual, in either case a horizon of knowledge –will condition the 43 44

Ingold 2000, 153. Ingold 2000, 155.

2.  Towards a theoretical model of the epigraphic landscape

31

ways in which any individual person reads a given stone; no stone fully exists in a vacuum, not even if the subsequent vagaries of time have ripped it completely from its original spatial context.

The stone in its landscape How might this work for the present case study? Beginning with the inter-visual, Figure 2.5 represents a monument in the kirkyard of Kilrenny, a small village a few miles to the southwest of Crail. One can immediately recognise the similarities in pillars, in the entablature, in the central heraldic panel, and, indeed, throughout the monument. A viewer familiar with the Crail monument would recognise its twin in Kilrenny, but how are the two related? In this case, as with the location of subsequent mural monuments in Crail, the Lumsden monument has been the model rather than the copy. James Lumsden of Airdrie’s nephew, also named James, followed a distinguished military career in the Swedish service – like so many other Scots during the seventeenth century – and retired to the small estate of Innergellie, just opposite the kirk of Kilrenny during the middle of the seventeenth century.45 The family continued to reside at Innergellie until the nineteenth century and at some point in or prior to 1823 the memorial in question was erected in Kilrenny kirkyard.46 It is striking, then, that it should have been so closely modelled on their kinsman’s tomb. Why, we might ask? The answer, it would seem, is a straightforward one of reflected glory. The family had lost both most of their fortune and the estate of Ardrie itself in the first decades of the seventeenth century and were subsequently nowhere near so prominent as they had been in 1598. The lavish reimagining of the Crail monument at Kilrenny was a way both of reconstructing and echoing the family’s past glories, creating a visual link between their forbearer at Crail and themselves. These sorts of analogies could be easily multiplied for almost any stone and in each instance a different set of echoes and additional meanings would be revealed. To remain close to the case study, however, let us turn to inter-symbolic codes. Both the ubiquity of the death’s head motif and the repetition of James Lumsden’s monogram on his house at Airdrie have already been mentioned. The extent to which any sort of recognisable symbolic system – monograms, trade symbols, emblems of mortality, etc. – could exist within a larger region- or landscape-wide matrix of meaning should also be emphasised. In this instance, heraldry offers an excellent example. The two heraldic panels on the Crail and Kilrenny monuments are immediately recognisable

Wood 1887, 383–4. The only inscriptions on the Kilrenny monument are “LUMSDAINE. 1823 INNERGELLIE.”, which has led to a nineteenth-century date generally being ascribed to the monument as a whole, but it should be noted that the inscriptions appear considerably fresher and less weathered than other sections of the tomb, suggesting they may have been added at a later date. 45 46

32

Kelsey Jackson Williams

Fig. 2.5: The Lumsdaine of Innergellie monument in Kilrenny. Photo © Kelsey Jackson Williams.

2.  Towards a theoretical model of the epigraphic landscape

33

as possessing the same heraldic device as, for example, a painted manuscript version of the same arms from an armorial dating to the period of the Crail monument.47 In the case of the Lumsden monument it is also important to think particularly in terms of inter-textual meanings present in its inscriptions; what one might think of as the most erudite or esoteric level of meaning contained. The presence of the Latin verses on the central frieze in Johnston’s Heroes ex omni historia Scotica lectissimi has already been discussed in its religious context, but one could go further in following the implications of this intertextuality. Johnston’s collection of poetical epitaphs included a host of famous Scots, beginning with the mythical King Ferchard.48 Placing James Lumsden within this august company raised his and his family’s stock considerably and this would add, in turn, to the cultural capital generated by the monument in the mind of a sufficiently well-read viewer. The marmoreal and printed versions of the text are in dialogue with each other and each brings their own echoes and contexts with them in the reading of their twin. The burial aisle of Sir James Melville of Halhill at Collessie, also in Fife and not so far from Crail, provides an additional point of triangulation. It was constructed about 1609, a little over a decade after the Lumsden monument, and like the latter is notable for two parallel verse inscriptions in Scots on its exterior wall facing the road.49 The two inscriptions, rhyming ABACCDD (i.e., rhyme royal, a form common to fifteenth- and sixteenth-century Scots poetry), simultaneously remind the passing “pilgrim” of their mortality and the need to repent, while sternly warning against defiling “Chrysts kirk with your carrion” (as quoted above). What is striking here is this use of parallel verse inscriptions, something not very common elsewhere in Fife and a textual quirk that seems to connect the Collessie mausoleum with the monument in Crail. Even more striking, the poems at Collessie appear to have been written by Elizabeth Melville, Lady Culross, one of the most accomplished Scots religious poets of her generation, and a close friend of James Lumsden’s sister-in-law Isobel Cor.50 While proof is lacking, it is not implausible that the poems on the Crail monument may have been written by Melville herself or a member of her circle. Either way, a Fife viewer of the early seventeenth century would have recognised in them a resonance with the Collessie monument and a further confirmation of the picture we have gradually built here of a religious allegiance to the hardline Calvinist faction in the Jacobean church as well as some degree of participation in the Presbyterian intellectual circles of St Andrews and the East Neuk of Fife. From these examples it can be seen that an inscription’s meaning comes not just from the inscription itself, but from its immediate and its more regional surroundings. Its meaning is thus contingent. It can change depending on the subsequent Maxwell Findlater 2008, 304–5. Johnston 1603. 49 Spicer 2000. 50 Melville 2010, 69. 47 48

34

Kelsey Jackson Williams

development of the epigraphic landscape around it and it can in turn effect change in that same landscape by virtue of its presence. To fully read a carved stone we need to go well beyond the stone itself.

Conclusion: The meaning and efficacy of an epigraphic stone First in his 1992 essay “The Technology of Enchantment” and subsequently in his posthumous 1998 magnum opus Art and Agency, Alfred Gell set out to restructure the anthropology of art.51 He attempted to push past the aesthetic valuation of art objects in order to better understand their anthropological significance, a project which has major implications for how we engage with epigraphy. Gell’s point was that: The work of art is inherently social in a way in which the merely beautiful or mysterious object is not: it is a physical entity which mediates between two beings, and therefore creates a social relation between them, which in turn provides a channel for further social relations and influences. This is so when, for instance, the court sculptor, by means of his magical power over marble, provides a physical analogue for the less easily realized power wielded by the king, and thereby enhances the king’s authority.52

This might at first sight seem to be a rather reductive way of reading an object, as simply a metonym for some more abstract power, but Gell goes on to elaborate his basic concept with reference to the canoe boards of the Trobriand islanders. Specifically, he argues that we must jettison aesthetic valuations of the canoe boards in favour of an instrumentalist view which focuses on the psychological effects – in this case, intimidation – which the boards are meant to produce and the cultural context in which they are produced. The carver of such a board, Gell wrote, Must exercise a faculty of aesthetic judgement, one might suppose, but this is not actually how it appears to the artist in the Trobriands who carves within a cultural context in which originality is not valued for its own sake, and who is supposed by his audience, and himself, to follow an ideal template for a canoe-board, the most magically efficacious one.53

Allowing for differences in cultural context, this offers a useful framework in which to understand carved stones. The inherently social stone is carved with a purpose and because of that purpose it, to return to Gell’s description, “provides a channel for further social relations and influences”.54 In other words, carved stones are tools, they are “system[s] of action, intended to change the world rather than encode symbolic propositions about it”.55 Their ability to successfully do so can be seen in what has already been written about the Lumsden monument and its subsequent effects both on its proximate and more distant landscapes. Gell 1992; 1998. Gell 1992, 52. 53 Gell 1992, 54. 54 Gell 1992, 52. 55 Gell 1998, 6. 51 52

2.  Towards a theoretical model of the epigraphic landscape

35

I have proposed here a model for understanding epigraphic objects by reading along two axes: visual-symbolic-textual and stone-space-landscape, each of which influence the other and each of which are productive of new and entangled meanings. The example used here offers a case study for how this model might operate in an early modern context, but it is equally applicable elsewhere, at least where a stone’s original context is known. Even where a stone has been detached from its original context, fragments of these axes of meaning continue to cling to it and new meanings are generated in the space in which it has come to rest, be that museum or rubbish heap. This leads, however, to a final proposition which follows inevitably from a model such as this: a stone removed from its context, placed in a museum for example, can only ever be a fragment, having been shorn of the meanings which it would have produced in its original environment. If such a loss or transformation of meaning occurs in the environment of a museum, what must occur when a carved stone is shorn of its context and placed, instead, in the pages of an epigraphic corpus? We cannot happily perform a reading of any epigraphic object out of context and assume that that meaning would hold were it restored to its intended landscape. Where a context can be recovered, however, this model offers the possibility of recovering a far richer web of meaning than any abstract reading of an object’s text alone. Once both people and landscape are restored to a stone, it becomes alive as a multivalent social object, which both influences and is influenced by the ebbs and flows of its environment.

Bibliography

Bain, Joseph, et al., eds. 1898–1969. Calendar of State Papers Relating to Scotland and Mary, Queen of Scots, 1547–1603, 12 vols. Edinburgh: General Register House. Beveridge, Erskine. 1893. The Churchyard Memorial of Crail: Containing a Full Description of the Epitaphs Anterior to 1800: Together with Some Account of the Other Antiquities of the Burgh. Edinburgh: Privately Printed by T. and A. Constable. Camden, William. 1605. Remaines of a Greater Worke, Concerning Britaine. London: Printed by George Eld for Simon Waters. Chernoff, Graham T. 2012. Building the Reformed Kirk: The Cultural Use of Ecclesiastical Buildings in Scotland, 1560–1645. PhD thesis, University of Edinburgh. Cohen, Kathleen. 1973. Metamorphosis of a Death Symbol: The Transi Tomb in the Late Middle Ages and the Renaissance. Berkeley, Los Angeles: University of California Press. Cooley, Alison E. 2012. Cambridge Manual of Latin Epigraphy. Cambridge: Cambridge University Press. Eisenstein, Elizabeth L. 1979. The Printing Press as an Agent of Change, 2 vols. Cambridge: Cambridge University Press. Febvre, Lucien and Henri-Jean Martin. 1958. L’apparition du livre. Paris: Albin Michel. Gell, Alfred. 1992. “The Technology of Enchantment and the Enchantment of Technology.” In Anthropology, Art, and Aesthetics, edited by Jeremy Coote and Anthony Shelton, 40–63. Oxford: Clarendon Press. Gell, Alfred. 1998. Art and Agency: An Anthropological Theory. Oxford: Clarendon Press. Graham, Angus. 1960. “Graveyard Monuments in East Lothian.” Proceedings of the Society of Antiquaries of Scotland 94: 211–71. Gruter, Jan. 1603. Inscriptiones antiquae totius orbis Romani. Heidelberg: ex officina Comeliniana.

36

Kelsey Jackson Williams

Gruter, Jan. 1707. Inscriptiones antiquae totius orbis Romani, edited by Johann Georg Graevius et al. Amsterdam: excudit Franciscus Halma. Ingold, Tim. 2000. The Perception of the Environment: Essays on Livelihood, Dwelling and Skill. London: Routledge. Jackson Williams, Kelsey. 2017. “Antiquarianism: A Reinterpretation.” Erudition and the Republic of Letters 2: 56–96. Johnston, John. 1603. Heroes ex omni historia Scotica lectissimi. Leiden: Excudebat Christophorus Guyotius, sumtibus Andreæ Hartii Bibliopolæ Edinburgensis. Kirk of Scotland. 1839–1845. Acts and Proceedings of the General Assemblies of the Kirk of Scotland, From the Year M.D.LX., 3 vols. Edinburgh: Bannatyne Club. Lahtinen, Anu and Mia Korpiola, eds. 2018. Dying Prepared in Medieval and Early Modern Northern Europe. Leiden: Brill. MacCoinnich, Aonghas. 2015. Plantation and Civility in the North Atlantic World: The Case of the Northern Hebrides, 1570–1639. Leiden: Brill. M’Crie, Thomas. 1824. Life of Andrew Melville, 2 vols. Edinburgh: William Blackwood. Macpherson, Robin G. 1998. Francis Stewart, 5th Earl of Bothwell, c1562–1612: Lordship and Politics in Jacobean Scotland. PhD thesis, University of Edinburgh. Mason, Roger and Steven Reid, eds. 2014. Andrew Melville (1545–1622): Writings, Reception, and Reputation. Farnham: Ashgate. Maxwell Findlater, Alex, ed. 2008. Lord Crawford’s Armorial. Bristol: The Heraldry Society of Scotland. Melville, Elizabeth. 2010. Poems of Elizabeth Melville, Lady Culross: Unpublished Work from Manuscript and ‘Ane Godlie Dreame’, 2nd ed., edited by Jamie Reid Baxter. Edinburgh: Solsequium. RCAHMS. 1933. The Royal Commission on Ancient and Historical Monuments & Constructions of Scotland Eleventh Report with Inventory of Monuments and Constructions in the Counties of Fife, Kinross, and Clackmannan. Edinburgh: Published by His Majesty’s Stationery Office. Record Commission. 1811–1816. Inquisitionum ad capellam domini regis retornatarum: quae in publicis archivis Scotiae adhuc servantur abbreviatio, 3 vols. London: s.n. Reid Baxter, Jamie. 2017. “Elizabeth Melville, Lady Culross: New Light from Fife.” Innes Review 68: 38–77. Salisbury, Marquess of. 1883–1976. Calendar of the Manuscripts of the Marquis of Salisbury, Preserved at Hatfield House, 24 vols. London: HMSO. Spicer, Andrew. 2000. “Defyle not Christ’s Kirk with your carrion: the development of burial aisles in Post-Reformation Scotland.” In The Place of the dead: Death and Remembrance in Late Medieval and Early Modern Europe, edited by B. Gordon and P. Marshall, 149–69. Cambridge: Cambridge University Press. Stair, James Dalrymple, Viscount. 1759. The Institutions of the Laws of Scotland, 3rd ed. Edinburgh: Printed for G. Hamilton, and J. Balfour. Stenhouse, William. 2005. Reading Inscriptions and Writing Ancient History: Historical Scholarship in the Late Renaissance. London: Institute of Classical Studies, University of London School of Advanced Study. Thiry, Steven. 2014. Matter(s) of State: Heraldic Display and Discourse in the Early Modern Monarchy (c.1480–1650). PhD thesis, University of Antwerp. Thomson, John Maitland, et al., eds. 1882–1914. Registrum magni sigilii regum Scotorum, 11 vols. Edinburgh: H.M. General Register House. Tilley, Christopher. 2004. The Materiality of Stone: Explorations in Landscape Phenomenology 1. Oxford and New York: Berg. Tilley, Christopher. 2008. Body and Image: Explorations in Landscape Phenomenology 2. Walnut Creek, Cal.: Left Coast Press.

2.  Towards a theoretical model of the epigraphic landscape

37

Turnbull, David. 2002. “Performance and Narrative, Bodies and Movement in the Construction of Places and Objects, Spaces and Knowledges: The Case of the Maltese Megaliths.” Theory, Culture & Society 19: 125–43. Willsher, Betty. 2005. Understanding Scottish Graveyards, 3rd ed. Edinburgh: Council for Scottish Archaeology and NMSE Publishing. Wood, Walter. 1887. The East Neuk of Fife: Its History and Antiquities, 2nd ed., edited by J. Wood Brown. Edinburgh: David Douglas. Wormald, Jenny. 1991. Court, Kirk, and Community: Scotland, 1470–1625. Edinburgh: Edinburgh University Press.

Chapter 3 Materializing epigraphy: archaeological and sociolinguistic approaches to Roman inscribed spindle whorls Alex Mullen*

Epigraphic perspectives This chapter explores sociolinguistic and archaeological approaches to epigraphy and demonstrates how they might work through a detailed analysis of the enigmatic corpus of Roman inscribed spindle whorls. Epigraphists might argue that they are already archaeologists and, indeed, within the textual realm of ancient world studies, they are some of the most field-based and object-oriented practitioners. Autopsy is such an important part of epigraphic analysis that many end up being intimately engaged with the inscribed objects, for example through making squeezes or more modern imaging and recording techniques such as Reflectance Transformation Imaging (RTI).1 When the objects are in situ, epigraphists will often expend effort reaching them, experiencing the topography, sight-lines, proximity to urban centres and light sources, all of which help to inform interpretations even if the details may not appear in the epigraphic corpora. Autopsy is no new thing: a drawing of one of the fathers of modern epigraphy, Theodor Mommsen, at work shows him perched partly on a donkey, partly on a ladder, both positioned in water, heading up to look This output received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 715626 (“LatinNow”). I am grateful to Monique Dondin-Payre, Mary Harlow, Marie-Thérèse Raepsaet-Charlier, Lacey Wallace, George Watson, and members of the LatinNow team, especially Pieter Houten, Noemí Moncunill and Simona Stoyanova, for their assistance. My thanks go especially to Eleri Cousins for thought-provoking discussions and encouragement. 1 A number of epigraphists have worked extensively with scientists to optimize imaging techniques such as multispectral analysis and RTI for ink-written/painted texts and incised texts respectively and to use computers to aid in character identification. Key early work was undertaken by Alan Bowman and Melissa Terras, see Bowman and Brady 2005 and Terras 2006. These imaging techniques now are arguably as much an epigraphical tool as an archaeological one. *

40

Alex Mullen

at an inscription on a bridge,2 and, a couple of millennia earlier, both Craterus of Macedon and Pausanias collected inscriptions and used them in their works, in some cases having perused them in context.3 It is fair to admit, however, that often objects and context have been of secondary (or lesser) concern and “epigraphists have often been viewed as narrow technicians whose conceptual myopia prevents them from seeing beyond the edges of their stones”.4 Epigraphic corpora have been designed primarily to present texts, through transcription, edition, and commentary. This can even be the case for the corpora produced by those with archaeological experience: the norms set by the discipline are followed. So even if deep contextual knowledge has informed the interpretation, users of these corpora may not appreciate the details, value, and role of that context. Texts can take on lives of their own in paper/digital form and become de-materialized. Some epigraphic corpora are so focused on a narrowly linguistic, rather than a sociolinguistic, perspective that scholars have even published the two versions of bi-version bilingual inscriptions in separate corpora, i.e., splitting them along language lines, severing the texts not just from the object but also from one another: CIL VIII and The Roman Inscriptions of Tripolitania (1952) give the Latin inscriptions but not the parallel Punic versions.5 Such divorcing from the social-cultural context in which the linguistic expressions were created does not help our historical and sociolinguistic analyses.

Sociolinguistic and archaeological epigraphy A sociolinguistic and archaeological approach to epigraphy puts people at the centre of the analysis. It entails integration of the analysis of macro and micro sociolinguistic features of epigraphic evidence and of archaeological approaches – for example, appreciation of materiality, context, and phenomenology – in order to understand social interactions and identities. Where possible, consideration of context, at all scales, is important. This means not only a detailed appreciation of the immediate context of the inscribed object itself including the uninscribed objects with which it was found, but also its broader site, region, provincial, even imperial, context, and its relation to other objects, inscriptions and society, language and culture.6 Materiality, namely the focus on the object and Bodel 2001, xvi. For Craterus, see FGrH 342 and Higbie 1999; for Pausanius, see Habicht 1984. 4 Bodel 2001, 1. The move towards more archaeological epigraphic corpora can be seen in the new project to re-edit the Gaulish inscriptions, RIIG (Recueil informatisé des inscriptions gauloises) https://riig. huma-num.fr/ (last accessed 11.5.2020). For some of the issues faced in creating digital editions which adequately present textual, material, and visual aspects of epigraphy in an encoding schema compliant with the EpiDoc guidelines, see Morlock and Santin 2014. 5 See Millar 1968, 131. 6 Modern sociolinguists have created a sub-field, Linguistic Landscapes, over the past two decades which is concerned with written language in urban contexts. Their approaches tend to be relatively ahistorical 2 3

3.  Materializing epigraphy

41

its relations to human practice, must also be included in a rigorously sociolinguistic and archaeological approach to epigraphy.7 This can be considered in situations when the object is known but its context is not, or only partially.8 Another ingredient in the mixture is phenomenology.9 This puts human experiences at the centre of the reconstruction of the ancient world and interrogates the realities of creating, displaying, viewing, touching,10 using the inscribed objects, thinking through how pleasurable, difficult or unusual these might have been in terms of the individual’s or community’s experience, and what sorts of cognitive and linguistic processes might have been informed by those and similar experiences. A phenomenological approach to an inscription carved into the rock face in the Cerdagne region (eastern Pyrenees), to take a random example, would require, amongst other things, analysis of its location (including sightlines and light sources), routes to the rock face for carving, best positions for reading/viewing, conditions at different times of the year and so on.11 Ancient sociolinguistics has gained significant traction in recent years. It is concerned with language use and change related to all aspects of society, and has deployed epigraphy as one of its key sources of evidence. It may take a range of different forms, including macro sociolinguistic analysis – addressing questions such as which language is used, when and where, for example – and micro sociolinguistic analysis, which may entail the collection of non-standard and standard linguistic features from epigraphic remains and making in-depth quantitative and qualitative analyses supported by data on social factors. There are numerous routes for exploration. I previously argued that it might be possible, using detailed cross-cultural knowledge of communities and individuals exhibiting language contact phenomena, to diagnose which ancient communities might have produced which types of bilingual texts and and they have not as yet linked up well with other fields that have had similar concerns for some time, see Pavlenko and Mullen 2015. Tools used extensively by archaeologists such as Geographic Information Systems can be usefully harnessed to plot epigraphic landscapes and to coordinate a range of nonepigraphic data to support detailed contextual analysis; this is part of the work in the LatinNow project, see https://latinnow.eu/ (last accessed 11.5.2020). 7 For a book-length treatment of materiality and texts, see Piquette and Whitehouse 2013. 8 A range of tools employed by the archaeological community, such as petrological or metallurgic analysis, can be used by epigraphists, for example to understand better the origins and therefore possible costs and effort required to obtain the material for inscription. J. Prag’s ERC-funded “Crossreads” project on the multilingual epigraphies of Sicily has petrological analysis as a research strand. See also d’Encarnação 1984. Often this information, if included in the printed corpora, does not get transferred into the online digital corpora. 9 For seminal works on phenomenology in archaeology, see Tilley 1994; Hamilton and Whitehouse 2006. 10 For work on touching writing materials, see Hoskin and New 2017 on fingerprints on medieval seals. 11 For the most recently published rock-cut inscription in Latin, with references to the other publications, see Ferrer i Jané et al. 2020. Viewshed analysis (for thinking about which features of the human-made or natural landscape can be seen from inscriptions and vice versa) might well work for these in situ inscriptions, whose rural context has remained similar over centuries. To my knowledge this has not been applied to contexts involving monumental Roman epigraphy, presumably partly because precise knowledge of original display contexts and the surrounding built environment can be elusive.

42

Alex Mullen

features.12 The matrix used the number of languages present in the community, the type of external links and levels of ethnolinguistic vitality as the main variables.13 The argument was that in circumstances where only bilingual epigraphic texts remain, very tentative assumptions could be made about the possible nature and contacts of the communities. However, such models of community dynamics and epigraphic remains, whilst useful in visualising how factors might interrelate, are necessarily reductive and tend not to cope adequately with the messiness and complexity of human linguistic relations which ancient sociolinguists are so keen to explore.14 The model is therefore just one element in the delicate balancing act of understanding partial evidence: applying the most effective sociolinguistic approach to ancient materials requires assembling as many tools as possible, operating at different scales of analysis as appropriate and carefully coordinating the results. In this chapter I argue for the utility of a new concept, translingualism, currently used in modern sociolinguistics, for our ancient world investigations. Translingualism puts a focus on the fluidity and complexity of linguistic repertoires and encourages us to think beyond bounded linguistic entities such as standard languages and the stock interpretations and concepts of bilingualism studies currently used in Classics (see below, pp. 57–59). These archaeological and sociolinguistic elements are closely intertwined and should, wherever possible, be used together. There is still plenty of scope for crossdisciplinary collaboration in epigraphy: understanding each other’s disciplines and learning from one another is essential. Archaeologists, partly because of the legacy of post-processualism (especially in the UK/US), the nature of the evidence, and the time and methods required in the recovery of that evidence, are acutely aware of the subjectivity and difficulty of interpretation. Epigraphists do not come from a disciplinary environment where such concerns are so pervasive. It might be argued that epigraphic subjectivity is generally more readily recognised at the level of readings and less so when it comes to reconstructing meaning, functions, and significance in broader context. Perhaps the most valuable aspect of a more self-consciously archaeological epigraphy is the constant questioning of assumptions and weighing up of possible interpretations. Many epigraphists already do this, but perhaps not with the doggedness of those trying to make material culture “speak”. Texts can make us think they are telling us what we need to know, and we ought to question that every time.

The inscribed spindle whorls of eastern Gaul The Roman inscribed spindle whorls from eastern Gaul will serve as a case study in which sociolinguistic and archaeological approaches can be combined, by applying Mullen 2012; 2013a. Ethnolinguistic vitality was introduced from modern sociolinguistic studies to Classics in Mullen 2012. The vitality of an ethnolinguistic group is “that which makes a group likely to behave as a distinctive and collective entity within the intergroup setting” (Giles et al. 1977, 308). 14 For ancient sociolinguistics, see Clackson 2015; Mullen 2016. 12 13

3.  Materializing epigraphy

43

a phenomenological perspective and an awareness of assumptions and uncertainties. In particular I shall consider whether assumptions about gendered interactions, and even language itself, have hindered the analysis of this corpus. Spindle whorls (French pesons de fuseau/fusaïoles, German Spinnwirteln) are common finds in Roman (and both later and earlier) contexts.15 They are weights placed on the spindle to increase the torsion of the twist and to allow the spinner to use one hand to draw out the thread and maintain the spin (Fig. 3.1).16 Since wool is spun before it is woven, vast numbers of these items must have been used across the Roman world over several centuries.17 They are made from a range of materials, including ceramic, bone, metal, and stone. A few examples from the Iron Age, Republican, and post-Roman periods inscribed in various non-Latin languages are known and yet, despite the Roman world’s obsession for writing on things,18 imperial-period whorls do not appear to have been inscribed, with the exception of an unusual corpus of two dozen from eastern Gaul.19 Fig. 3.1: Spinning with a distaff, Table 3.1 assembles the corpus of imperial-period drop spindle, and whorl. Drawing: inscribed spindle whorls, split into two parts, the first Jane Masséglia, LatinNow. listing 11 examples with find-spots in Autun (Fig. 3.2) and the second roughly the same number of examples that have been found beyond that settlement (Fig. 3.3). It seems likely that most of these items were made at Autun: some of the material has been scientifically analysed and the bituminous schist that has been identified can be traced to the quarries of Autun.20 Experienced workers of schist at Autun produced a range of materials, for example, wall and floor decoration, dice, game counters, jewellery, sometimes with inscriptions.21 Skill and planning would have been needed to cut the decoration and For textile manufacture in the western provinces, see Wild 1970; 2002; 2003; Allen et al. 2017, 221–30. Harlow forthcoming. See Barber 1992, 39–78 for practical details of spinning in prehistory. 17 The Rural Settlement in Roman Britain project found that they were “among the most common types of object recovered at sites across the province” (Allen et al. 2017, 226). 18 For inscribed examples in Celtiberian and Iberian language from the Iberian peninsula, see Castro Curel 1980; Ferrer i Jané 2008; Beltrán Lloris et al. 2021. 19 Héron de Villefosse 1914 collects a first corpus of this unusual group, correctly identifying the object type and linking them to inscriptions on drinking vessels. 20 For details of the petrological investigations, see Dondin-Payre 2006, 145 n. 10; Maggetti et al. 2009. 21 For the use of schist at Autun, see Rebourg 1996. 15 16

44

Alex Mullen Table 3.1: Corpus of imperial-period inscribed spindle whorls.

No.

Inscription

Reference

Notes

Find-spot: Autun (France) 1

ACCEDE / VRBANA

ILTG 523; Rebourg 1996, n°119; Dondin-Payre 2004, n°11

Collection Bulliot

2

SALVE / DOMINA

Rebourg 1996, n°126 bis22; Dondin-Payre 2004, n°3

3

AVE VALE / BELLA TV

CIL XIII 2697 and 10019.18; Rebourg 1996, n°127; Dondin-Payre 2004, n°1

4

GENETA / VISCARA

ILTG 526; Rebourg 1996, n°121; Dondin-Payre 2004, n°4; RIG II.2 L-114

5

LAVTA / LAVTA

Rebourg 1996, n°126; Dondin-Payre 2004, n°5

6

MARCOSIOR / MATERNIA

ILTG 527; Rebourg 1996, n°123; Dondin-Payre 2004, n°6; RIG II.2 L-117

Collection Bulliot (Figs 3.2, 3.4)

7

MATTA DAGOMOTA / BALINE E NATA

ILTG 528; Rebourg 1996, n°120; Dondin-Payre 2004, n°7; RIG II.2 L-115

Collection Bulliot

8

NATA VIMPI / CVRMI ILTG 529; Rebourg 1996, n°122; Dondin-Payre DA 2004, n°8; RIG II.2 L-112;

Collection Bulliot (Fig. 3.4)

9

NATA VIMPI B(ene) S(alve) V(ale) / TOTVNVCI

Chardron-Picault and Dondin-Payre 2000; Dondin-Payre 2004, n°12; RIG II.2 L-118

10

VEADIA TVA / [T] ENET

ILTG 531; Rebourg 1996, n°125; Dondin-Payre 2004, n°10; RIG II.2 L-116

Collection Bulliot

11

AVE DOMINA / SITIIO ILTG 524; Rebourg 1996, n°128; Dondin-Payre 2004, n°2

Collection Bulliot; unpierced and hemispherical in form

Collection Bulliot

Find spot: various 12

TAVRINA / VIMPI

ILTG 530; Dondin-Payre 2004, n°9; RIG II.2 L-113

Sennecey-le-Grand (France); Collection Bulliot23

13

NATA VIMPI / VI(nu?)M POTA

CIL XIII 10019.20; Dondin-Payre 2004, n°18; RIG II.2 L-121

Auxerre (Faubourg Saint-Martin) (France) (Fig. 3.4)

14

TIONO VIMPI / MORVCIN

CIL XIII 1324; Dondin-Payre 2004, n°20; RIG II.2 L-111

Gièvres (France) (Continued)

22 “Un autre peson aurait été découvert lors des mêmes travaux, mais il est dans une collection particulière” (Rebourg 1996, 109) (“Another spindle whorl [number 2] was apparently discovered during the same works [Plan d’eau du Vallon, 1976], but it is in a private collection”). 23 Erroneously assigned to the Autun set by various scholars, see RIG II.2 page 324 for information on the find-spot.

45

3.  Materializing epigraphy Table 3.1: (Continued) No.

Inscription

Reference

Notes

15

DA MI

CIL XIII 10019.21; Dondin-Payre 2004, n°15

Langres (France)

16

SALVE TV / PVELLA

CIL XIII 5885, 10019.19; Dondin-Payre 2004, n°19

Langres (France)

17

EMEME / FELIX

Barthèlemy 1976, p. 7 and pl. II; Dondin-Payre 2006

Mâcon, Flacé (France)

18

PACTVS / ITALIA

RIG II.2 p.319 i

Suin (France)

19

AVE / VIMPI

Dondin-Payre 2004, n°21; 2006; RIG II.2 L-122

Nyon (Switzerland)

20

MONI GNATHA GABI / BUÐÐVTON IMON

CIL XIII 2827; Desforges 1924; Dondin-Payre 2004, n°17; RIG II.2 L-119

Saint-Révérien, environs (France)

21

GENETTA IMI / DAGA VIMPI

ILTG 525; Héron de Villefosse 1914; Dondin-Payre Sens, or environs 2004, n°16; RIG II.2 L-120 (France)

22

IMPLE ME / SIC VERSA ME

CIL XIII 10019.17; Dondin-Payre 2004, n°14

Löwenbrücken, close to Trier (Germany); black ceramic

23

SALVE / SOROR

CIL XII 5688 19; Dondin-Payre 2004, n°13

Vienne, or environs (France) (ancienne Collection Girard); grey ceramic

24

CARA VIMPI / TO CARANTO

Binet and Dondin-Payre 2002

Amiens (France)

lettering into the small surface area of the whorl,24 which is almost always divided into two sections with roughly half of the text on each (Figs 3.2, 3.4). The similarities in the lettering suggest that some may have been inscribed by a community sharing epigraphic practices. The only two that are certainly not made of schist, numbers 22 and 23, are ceramic.25 Only half a dozen of the two dozen published to date have any indication of archaeological context and the dating of the corpus cannot be confidently offered except in loose terms.26 Dondin-Payre argues that the whorls should be dated to the first to third centuries AD, and that any more precise dates are “arbitraires pour la plupart, car elles sont fondées sur des préjugés culturels et non sur des critères The schist whorls are roughly 1.5 cm high × 2.5 cm diameter. Number 14, now lost, was described as being made of “serpentine noire” (RIG II.2 p. 320), though it could well have been schist. Number 19 was thought to be ceramic before petrological analysis (DondinPayre 2006). 26 The editor of the first corpus noted that none was found with others: “[t]ous ces petits monuments paraissent avoir été recueillis à l’état isolé; du moins on n’a jamais signalé de trouvaille en comprenant deux ou plusieurs” (Héron de Villefosse 1914, 226) (“all these small items seem to have been found in isolation; at least, no finds comprising two or more have ever been reported”). Numbers 9, 12, 17, 19, 21, 24 have some associated archaeological information. 24 25

46

Alex Mullen

Fig. 3.2: Inscribed spindle whorl from Autun (number 6) (Mullen and Darasse 2018, figs 30–31).

objectifs”.27 Loth, Meid, and Adams all state that they date to the third or fourth century AD, but without detailed supporting evidence.28 It is difficult to do much more than assign an imperial-period dating, though number 24 comes from a structure that was constructed in AD 90 and destroyed by fire in 125/130,29 9 was found in a context dated to the second half of the second century AD, 12 was found during excavations of a villa in 1858 with a coin of Domitian (c. AD 88–90) and 19 has been tentatively dated to the second century AD.30 This evidence, combined with Rebourg’s view that schist from Autun was most commonly worked in the second and third centuries, with a concentration in the Severan period, might suggest a date range of c. AD 90–235 for the whorls.31 If the similarities of practice can be attributed to a localised phenomenon, which may have been linked to the period of operation of a small number of carvers at one or more whorl-producing workshops/households, then the period of activity may have been shorter. I shall consider possible contexts for the creation of these inscribed items below (see below, pp. 48–55). Several of the whorls have texts composed in Latin, for instance numbers 1–3, 5, 11, 15–18, 22–3 in Table 3.1, and others in Gaulish, the Celtic language spoken in Gaul,32 Dondin-Payre 2005, 136, “for the most part arbitrary, since they are based on cultural prejudices and not on objective criteria”. 28 Loth 1916, 169; Meid 1983, 1030; Adams 2003, 196. 29 Binet and Dondin-Payre 2002, 133. 30 Dondin-Payre 2006, 153. 31 Rebourg also suggests that the evidence of inscribed instrumentum domesticum points to a Severan dating for the inscribed schist spindle whorls (1996, 15). There are not enough distinctive features in the texts on the whorls to support this narrower dating and number 24 cannot be Severan. 32 For recent introductions to Gaulish, see Lambert 2018; Mullen and Darasse 2020. 27

3.  Materializing epigraphy

47

Fig. 3.3: Find-spots of imperial-period inscribed spindle whorls. Map: Pieter Houten, LatinNow.

for example numbers 7, 14, 20, 21. A high proportion seem to be not obviously in a single language (4, 6, 8–10, 12–13, 15, 19, 24), which has excited linguists who, working with notions of languages as bounded linguistic resources, have deconstructed the utterances into two languages and used the concepts and terminology of bilingualism in their analysis (see below, pp. 56–57). The texts can be ascribed the function “speaking objects”, relaying direct speech or speaking themselves. They apparently address a female in several cases and some seem to have amatory/erotic content.33 For example, number 20: MONI GNATHA GABI Roman loom weights are more commonly inscribed and most carry texts relating to their production. One, from Zaragoza, dating to the early first century AD, however, seems similar to the Autun spindle whorl inscriptions, with direct speech and a reference to amatory relations: multas telas texat, bonum uirum inueniat. ama lateres! facimus fausta felicia (“Let her weave many threads, let her find a good husband. Love 33

48

Alex Mullen

/ BUÐÐVTON IMON a Gaulish utterance, which can be translated, using knowledge of Indo-European linguistics and the Celtic languages, as “Come girl, take my little kiss/ cock”.34 Scholars have assumed that these are gifts from men to women – “les galants qui offrent l’objet à la femme”35 – and there has, until now, been little questioning of this core assumption.

Uncertainties of context and interpretation The leading modern commentator on the inscribed spindle whorls has reminded us of “la nécessité de considérer ensemble l’implantation géographique, l’aspect, le matériau et les inscriptions pour évaluer la spécificité, donc l’importance historique des documents”.36 Unfortunately one striking thing about this set of inscribed material is the lack of archaeological context for most of the examples. As can be seen in Table 3.1, most appear in CIL XIII, indicating the early date of their discovery, and 7 of 11 from Autun were part of the Collection Bulliot, which provided no details of their original find context. Thus commentators have tended to create their own visions of by whom and where these objects were used. It has been suggested that the whorls are from “luxueux” domestic settings, not from “un environnement artisanal anonyme”.37 The motivation for this seems to be the context of number 24, a huge building covering originally at least 2,500 m2, with areas of habitation, plus commercial and storage facilities. Whilst the structures have been attributed to the “élite amiénoise”,38 given that the spindle whorl was found in a destruction level, it is impossible to say whether it was used by a Lucretia-style elite matrona spinning and then weaving in a comfortable atrium,39 or the result of a completely different

loom-weights! We make lucky and happy things”) (Beltrán and Beltrán 2012). The non-Latin inscriptions on spindle whorls from the Iberian Peninsula, dating to the second and first centuries BC, have been split into four types in Beltrán Lloris et al. forthcoming: short inscriptions of one to three signs (the majority); texts of pseudo writing or alphabets; texts containing names; a small group of longer texts that have been previously interpreted as amatory/erotic. The authors caution that the interpretation of this latter group is not at all secure, and one might suspect it may have been inspired by the texts in the Autun collection (the texts are unrelated, the only link being the choice of object). For a small number of possible earlier examples of inscribed spindle whorls from the Mediterranean world, see Tsori 1959 for Judaea; Bagnasco Gianni 1999 for Etruria; Sauvage and Hawley 2013 for Ugarit. The example from Çatal Höyuk published by Gevirtz in 1969 has been deemed a forgery (Levenson 1973). 34 See Eska 1998 for the tau gallicum, in this inscription represented by a double-barred D. 35 Héron de Villefosse 1914, 229, “the admirers who offer the object to the woman”. 36 Dondin-Payre 2005, 143, “the need to consider geographical location, appearance, material and inscriptions together to assess the specificity, and therefore the historical significance, of the documents”. 37 Binet and Dondin-Payre 2002, 137, “an ordinary artisanal environment”. 38 Binet and Dondin-Payre 2002, 133, “the elite from Amiens [Samarobriva]”. 39 In the Roman context textile work seems to have been deemed a worthy feminine pursuit for all sections of society. Apparently, Augustus wore clothes made by his sister, wife, daughter, and granddaughters (Suet., Aug. 73).

3.  Materializing epigraphy

49

scenario, involving non-elite members of the extended familia, for example, producing textiles in one part of the vast complex.40 There are only four archaeological descriptions on which to draw for possibly useful information concerning the context of the primary use of these objects in the textile-making process: numbers 9, 17, 19, and 24. Of these, number 17 is relatively vague – an urban domestic setting. Number 9, found in 1992, is described as having been found in a “zone artisanale”, dated to the second half of the second century, found along with four loom weights in a series of rooms. Number 19 was found in a modest urban domestic setting, in a small building with everyday objects, including a small number of higher quality. Number 24 was an isolated find with no precise context within the aforementioned vast “maison”, with commercial areas and storage to the west of Samarobriva, destroyed by fire in AD 125/130. Since the little archaeological information may point in part towards artisanal/lower-status domestic environments, we should be cautious in focusing exclusively on a higher-status domestic interpretation. Likewise, the circumstances of their production are uncertain. As mentioned previously, the material from which most of them are carved comes from Autun. Dondin-Payre argues that the concentration of texts on the whorls made in Autun is not a distribution created by preservation biases and modern practices but rather “[l]a raison réside dans la combinaison entre plusieurs facteurs: l’exploitation d’un support, ce schiste spécial et celle d’une compétence technique, le savoir-faire des artisans locaux conjugué avec une compétence linguistique et graphique imputable au niveau culturel élevé de la capitale des Éduens, où l’écrit est familier”.41 Whilst we might be nervous of making such generalisations about cultural levels and literacy, it is certainly the case that the inscribers of at least some of these texts had knowledge of lapidary epigraphy, since some of the features ape Roman epigraphic practices: the use of capitals, abbreviations, interpuncts, and, most strikingly, the ansate frame motif and hedera of number 9.42 So since the texts share numerous features, should we imagine a workshop producing (most of) these items, in which there were a small number of literate craftsmen who could “personalise” these objects with a message? In this scenario, many of the whorls then did not move far from their place of origin but some travelled beyond Autun with the spinners. Alternatively, the Autun craftsmen who produced these attractive whorls may have moved around to sell their items and 40 Slaves must have been employed extensively in textile production, see Harper 2011, 128–35. Slaves skilled at spinning were known as quasillariae. 41 Dondin-Payre 2005, 136, “[t]he reason lies in the combination of several factors: the exploitation of a medium, this special schist and the technical skill, the expertise of local craftsmen combined with linguistic and graphic skills ascribable to the high level of culture of the capital of the Aedui, where writing was common”. 42 This need not be exclusively a result of direct interaction with Latin epigraphy, of course, since GalloLatin lapidary examples, though significantly less common, also use epigraphic features derived from Latin models. The Gallo-Latin inscription from Alise-Sainte-Reine (RIG II.1 L-13), for instance, is inscribed within an ansate frame motif and includes hederae, interpuncts, and ligatures.

50

Alex Mullen

may have taken commissions for texts at markets or by the roadside. In both contexts spinners, or those buying the whorls for them, might have passed on messages orally to the engraver who placed them on the whorls. A completely different reconstruction in which people independently compose and write similar looking and sounding messages on items that are not usually inscribed in the Roman world, seems a less likely scenario. Carving these texts into the small schist whorls would have required some skill, and one possessed by those who worked regularly with the material. With both the context of production and the context of use uncertain, the objects themselves and the combination of their physical and textual characteristics become the principal portal into their social function. Commentators have been interested by the amatory/erotic nature of some of these texts and have tended to extrapolate from the small number that may be “suggestive” and have seen the greetings and exhortations to drink in the same light. The agents behind the speaking objects have, following gender stereotypes, been taken to be men.43 Meid notes in support of this perspective “[d]ass die Sprecher Männer sind, kann man aufgrund unserer Weltkenntnis vermuten. Die Äusserungen sind Ausdruck einer Art von Anbandelei, gehören also zum Ritual des amourösen Spiels”.44 Again we should question this assumption and consider other visions of interpersonal dynamics. One proposition to contemplate is that the agency behind the texts may not be, or may not only be, male. If some of these texts are used by women working in groups in workshops (indeed this reconstruction might be supported by the context of number 9), we might wonder whether some of these messages may be created by women for other members of the group, or for themselves, to enjoy.45 Here, a phenomenological perspective, focusing on the material capacities of the whorls, helps to draw out their 43 Beltrán and Beltrán 2012, 139 state in passing the view held by many that the texts are “seguramente realizados en los talleres y vendidos a varones para que, a su vez, los regalaran a muchachas” (“surely made in the workshops and sold to men so that they could, in turn, give them as gifts to girls”). The assumption that the authors are men also fits neatly with the view that levels of literacy amongst women were extremely low – an established “fact” that has not in fact been properly established, see now the commentary in Eckardt 2017, 154–75. There is much scope for detailed work on the social dimensions of literacy in the provinces, see Mullen 2021. 44 Meid 1983, 1030, “that the speakers are men can be assumed from our knowledge of the world. The utterances are expressions of a kind of flirtation, that is, they are part of the ritual of amorous play”. 45 There is no reason to believe that all the workers must have been women, but many, if not all, probably were, given the bulk of the literary, burial, and iconographic evidence. However, some male burials contain spindle whorls (though these are usually regarded as evidence of men owning workshops) (Rafel 2007) and there are depictions of males spinning, for example the depiction of the male thigh spinner from a sarcophagus now in the Terme Museum, Rome. This evidence and the comments that flax spinning was suitable for men in Pliny the Elder (HN 19.3.18) suggest that Lovén’s argument that it was “impossible for a man at any social level to be associated with wool work and, in particular, spinning, since it so distinctly represented femininity” (Larsson Lovén 2007, 233, see also Larsson Lovén 2013 for gender and textile work in Roman Italy) might need to be qualified. As Harlow remarks we need “to beware of taking an over-simplistic view of normative statements” (forthcoming). For gender and textile production in pre-history more generally, see Costin 2013. For the issue of assigning gendered use to small finds, see Allason-Jones 1995.

3.  Materializing epigraphy

51

potential for multi-sensory appeal. The dark-coloured whorls with white lettering would have created a striking visual effect, repeatedly spinning so that the object becomes a blur and then slowing to reveal the message. Co-workers in close quarters engaged in relatively monotonous tasks will often create distractions for themselves, for example work songs and in-group stories, language, and humour.46 The texts are short but two of the longer ones, numbers 8 and 13, may even have a rhythmic quality to them.47 Perhaps they tap into an in-group language to which we now have very limited access. We should hesitate before assuming that sexual “banter” is the preserve of men. Though there is on-going debate about the extent to which personal names feature in these texts, most linguists agree that there are perhaps only two or three names of addressees (?Maternia (number 6); Taurina (10); ?Italia (18)) and none of the addresser.48 We might wonder, therefore, whether that makes it less likely that these were gifts, since part of sending such amatory messages is often to inscribe the names of the people involved. Instead the references on the whorls are to (using nouns) domina “mistress”, geneta/genetta “girl”, puella “girl”, gnatha “girl”, soror “sister”; (noun and adjective) nata vimpi “pretty girl”, vimpi morucin “pretty girl”, ?cara vimpi “dear girl”; (adjectives) urbana “refined”, bella “pretty”, felix “lucky”. Whilst these could all be the outputs of male admirers, the possibility of women composing messages for themselves or others in the workshop should not be excluded. SALVE DOMINA, for example, might be a reference to the leader of the working group (it has a wide semantic range, spanning from a generic “Mrs” to sexual content)49 and SALVE SOROR is arguably just as likely to be the utterance of a woman than a man: soror is used by unrelated female friends from the first century BC (as with frater), 46 Examples of Scottish work songs (used to accompany numerous forms of repetitive work, such as spinning wool, but also fulling cloth, milking cows, churning butter etc.) can be found here: https:// blog.europeana.eu/2016/08/no-bees-no-honey-no-work-no-money-an-introduction-to-scottish-worksongs/ (last accessed: 1.8.2020). 47 We know of weaving/spinning/grinding work songs in the ancient world, e.g., in Catull. 64 and Plut., Conv. sep. sap. 14. Whatmough 1949, 389 notes that pretty much any text of more than a couple of words long can be scanned “after a fashion”, and warns against the tendency of some scholars to hunt for verse everywhere. For references to singing by female Roman textile workers, see Harper 2011, 135. 48 Dondin-Payre (2001, 318–27, 333–41; 2005) argues that certain words can be read both as names or as the lexemes that form them: Adiatu, Damus, Matta, Totunuca (analysed as Celtic); Cara (Celtic/Latin); Bella, Geneta, Lauta, Maternia, Taurina, Vimpus (‘noms latins à fréquence celtique’); Domina, Italia, Nata, Puella, Viscara, Vrbana (Latin). The names designated as Celtic are not secure and many of the other names are barely attested. For Bella, Geneta, Lauta, Vimpus, Domina, Nata, Puella an onomastic interpretation seems less likely. The names are not commonly used (a search across the LatinNow epigraphic dataset from the north-western provinces returns only two, possibly three examples of puella used as a name out of a total of around seventy) and vimpi, for example, is found as an adjective or an adjectival substantive in numerous cases, including in the phrase AVE VIMPI found on moulded brooches (RIB II.3, 2421.41 (Colchester) and Feugère and Lambert 2011 (Laon)). For the meaning “pretty” for vimpi, see Lejeune 1976, 96–104, which is preferable to the suggestion of Whatmough 1949 of an imperative “spin” (the moulded brooches now further undermine his case). 49 For details, see Dickey 2002, 77–109.

52

Alex Mullen

though it is also, more rarely, used with sexual reference.50 Number 21, found in 1913 in the tomb of a woman alongside four ceramic vessels, might be most eloquent for the question of agency: GENETTA IMI / DAGA VIMPI. This can be analysed as “girl, I am/my, good, pretty”, meaning either “I am a good, pretty girl” or “my good, pretty girl”. The uncertainty in the translation lies in the word IM(M)I in Gaulish, which may be either a verb or a possessive adjective.51 Since IMMI occurs on a bowl from Les Pennes-Mirabeau (Bouches-du-Rhône) (RIG I G-13) where it probably means “I am”, it seems on balance the more likely interpretation for the whorl, the suggestion of a possessive adjective perhaps being motivated by the assumption that these must be texts by men. As with many of the other texts, either reading of number 21 can be endowed (or not) with amatory content, depending on the preconceptions we bring to our interpretations. These erotic/amatory readings, however, need not erase female agency or female involvement in the creation of the whorls: indeed, when combined with a focus on aspects of the materiality and phenomenology of these objects, this sort of female-centred reading has the potential to undermine our normative models of the interplay between spinning and femininity. Those models have been shaped by the literary topos of the Roman matron sitting dutifully at her loom, and archaeologists have argued that chastity and moral standards are symbolised by spinning tools being used in the deductio ceremony from the early Roman period (part of the marriage ritual) and their appearance on tombstones (especially common in the eastern empire) and in burials.52 Spindle whorls, however, also exhibit a close connection to the female body. They were in close bodily contact with the spinner, who may have rolled the whorl down the thigh to begin the spinning process.53 This action may, but need not, support the interpretation of at least some of these textual messages as erotic.54 Commentators have also identified the action of inserting the

50 See Dickey 2002, 125–6. Dondin-Payre 2005, 138 notes that it could also be the object addressing the spinner or other parts of the spinning equipment. 51 For the term IM(M)I, see Lejeune 1976, 96–104. 52 See, for example, Cottica 2007 and Larsson Lovén 2007. Neither of these authors mentions the inscribed spindle whorls, which seem strangely absent from the non-linguistic/epigraphic scholarship. For deductio, see Torelli 1984. 53 The attractive little whorls may also have been hung around the neck when not in use. Indeed number 14 was initially thought to be an amulet (RIG II.2 p. 321). All commentators have assumed that they were used in spinning, rather than being made as replicas. Of the inscribed whorls for which I could find information, six are between 10–12 g, two are c. 21 g and one is very light, at just 6 g. These are towards the lighter end of Roman spindle whorl weights and would probably have been used to spin fine yarn. We cannot exclude the possibility that these were not used at all but were replicas, trinkets, or similar. Example 11 from Autun (but with no further contextual information) is interesting in this regard as it has not been pierced and is of hemispherical form. Whether it was intended to be a spindle whorl but had not yet had its centre bored through is unclear. Dice and games counters are also made in schist from Autun (Rebourg 1996). 54 The link between the act of weaving and physical union is discussed by Scheid and Svenbro 1994.

3.  Materializing epigraphy

53

material into the whorl as being symbolic of sexual relations.55 Further support for the erotic/amatory interpretation derives from the links between these and similar texts on drinking vessels, as Loth remarks: “Vénus fait une redoutable concurrence à Bacchus sur les vasa potoria de la Gaule”56 Indeed several of the whorl texts mention drinking specifically. So the whorls could potentially reflect the female enjoyment of sexuality, drink, and communal work, rather than the values of the ideal matrona. The texts, whatever their precise means of creation or manner of use, add to the evidence for a highly textual Roman world. But a world of texts does not necessarily mean a population of literates, and levels of literacy were never high.57 Certain occupations did encourage functional literacy, however, as demonstrated in the large pottery workshops such as La Graufesenque and schist carvers may have found writing for account-keeping and for inscribing their objects financially beneficial.58 The texts also suggest that perhaps some of the end-users could read them or that their acquaintances could, and a passive form of literacy, at different levels of competence, may have been relatively widespread, including amongst women. Who these “end-users” may have been is not clear from the texts, but Dondin-Payre pertinently remarks that just because names in these short texts are single names (linguists read just one or two single names, Dondin-Payre up to 17) does not mean that the women were not citizens,59 indeed after AD 212 they are perhaps quite likely to have been. She rightly reminds us that the functional nature of the object and the innuendo of some of the messages should not force us to assume participants of lower social status. But she goes even further in arguing that “ceux qui achètent, font graver, reçoivent, ou offrent des fusaïoles inscrites, ne sont pas des indigènes arriérés, incapables d’apprendre un latin correct, non romanisés et non concernés par une promotion civique”.60 Her view that the subtlety of the language points to educated and well-off clients will be reconsidered when we turn to the linguistic resources in play (see below, pp. 55–59). From other evidence we may know the names, or at least see the faces, of two possible spinners from Autun, whose funerary stelae were found with a group of around 200 uncovered in 2004 in excavations of the cemetery Pont-l’Évêque on the outskirts of Augustodunum. Several of the stelae present what appear to be tools of trade and all have been dated to the first half of the second century AD. One is around 1 m high and presents a woman in a rounded niche taking up the top half of the stone. In her right hand she holds a goblet (common in the iconography of the Meid 1983, 1043. Loth 1916, 178, “Venus is a formidable rival of Bacchus on the vasa potoria of Gaul”. For a categorization of erotic texts on instrumentum domesticum, see Thüry 2008. 57 For recent work on inscribed small finds and literacy, see references in Mullen 2021. 58 For the La Graufesenque graffiti, see Marichal 1988; Adams 2003, 687–724; Blom 2010–2012; Mullen forthcoming a. For schist carving at Autun, see Rebourg 1996. 59 Dondin-Payre 2005, 141. 60 Dondin-Payre 2005, 141, “those who buy, have engraved, receive, or offer inscribed spindle whorls, are not backward locals, unable to learn correct Latin, unromanised and unconcerned by civic advancement”. 55 56

54

Alex Mullen

stelae from this cemetery) and in her left a spindle and distaff are held to her chest (Fig. 3.4). The stele is briefly presented as number 22 in Venault et al. 2009 and the text edited as [Hila(?)]ricla | D(iis) M(anibus), with the name above the head and the abbreviated formula on either side. The stone is cut relatively roughly and the editors state that their suggestion for the name was arrived at “par désespoir”.61 The identification of the spindle and distaff is clear though. The other stele, bigger at 1.5 m, although damaged at the bottom, has the text D(iis) — Trita — M(anibus) (number 45), beneath the figure of a woman inside a rectangular niche. This woman carries a jug with a wide rim in her left hand and an object in her right hand, with thumb and index finger extended, which has been interpreted as “without a doubt” a distaff.62 What makes these Fig. 3.4: Image of funerary stele depicting a two stelae, found in the same city that woman holding a goblet, spindle, and distaff produced the unusual inscribed spindle from cemetery Pont-l’Évêque, Autun (Venault et whorls, particularly interesting, is that al. 2009, number 22). © Loïc de Cargouët, Inrap spinning representations in funerary (CC BY-NC-ND 4.0). contexts are not very common in the western provinces (though they are more so in the east). Whilst it is possible that the imagery might make reference to the deductio ritual where the bride would carry a spindle and distaff, it seems more likely, given the presence of professional tools in several of the other reliefs from the same cemetery (e.g., the hammer and tongs of metal working (number 5)), that these can be related to the occupation of the women depicted. Unfortunately the textual information on the stelae is not especially illuminating – arguably the single name may suggest peregrine status (unlike on the whorls, there is clearly space for further names), but it is hard to say more. The name in the first has not been transmitted with any certainty. The name of the second may well be Celtic. In the presentation of the stelae the editors note that Kajanto argues that Tritus -a is an Illyrian name but that this evidence indicates that, though rare, it is in fact a Latin name from the past

61 62

Venault et al. 2009, 155, “out of desperation”. Venault et al. 2009, 167.

3.  Materializing epigraphy

55

participle of Latin tero.63 It is more likely that this is simply the commonly attested Celtic personal name meaning “third” (equivalent to Latin Tertia).64 The clothing depicted in both reliefs matches local styles, the wide sleeved robe in number 45 being similar to that in the set found in the archaeological assemblage from a secondcentury AD burial at Les Martres-de-Veyre.65 The combination of evidence perhaps makes it likely that this woman may have had a local (possibly, but not necessarily, Celtic-speaking) background. Sadly, though it is an enticing link to make, we have no idea whether these women had anything to do with people involved with the inscribed whorls from Autun. But nevertheless the stelae show us the faces, just as the whorls may offer us some words, of those ubiquitous spinning women who are not usually seen or heard.

Translingualism: flexibility of linguistic resources The language used on the whorls may help us to think further about the composers of the texts and their relationships with the linguistic context in which they were writing. The issue of the heterogeneity of the linguistic composition of the texts on whorls has attracted interest from scholars exploring bilingualism in the Roman world. Implicitly following the conception of the linguistic repertoire being split into bounded entities called “languages”, commentators have spent time analysing the texts in terms of whether they contain what is called “Latin” and “Gaulish”. Particularly intriguing have been the texts that do not fit neatly into either category, but instead show elements of both. Take, for example, numbers 8 and 13 (Fig. 3.5): NATA VIMPI / CVRMI DA “pretty girl, give me beer” (Autun) NATA VIMPI / VI(nu?)M POTA “pretty girl, drink ?wine” (Auxerre) (g)nata, “girl”, which also occurs as nata in number 9 and as gnatha in 20, is a noun in Latin and Gaulish from their shared Indo-European inheritance. Adams tentatively suggests that “the similarity of natus, -a to Gaulish gnatus, -a gave it some currency in the Latin of Gaul alongside the more usual terms filius and filia, and by extension puer and puella, particularly in the feminine”.66 This would be a clever choice of appellation if one wanted to communicate simultaneously to both Latin and Gaulish speakers. vimpi, here used in the vocative, means “pretty” in Gaulish, and is commonly attested in these spindle whorls and on other instrumentum such as brooches.67 The origin of the word is unclear but it is likely to be related to Welsh gwymp. Since it appears in Venault et al. 2009, 167, citing Kajanto 1965, 356–7. For examples see Delamarre 2007, 185. 65 See van Driel-Murray 1999. 66 Adams 2007, 303. 67 For vimpi, see n. 57 and Meid 1983, 1032–3; RIG II.2 pp. 321–2. 63 64

56

Alex Mullen

repeated phrases such as AVE VIMPI it might also have been current in a regional form of Latin (which we might loosely term “Gallic Latin”), and hence may also have worked bilingually. The second half of the example from Autun follows the same pattern: the first word, curmi “beer”, is Gaulish (seen also in the personal name Curmisagios “beer seeker” and Old Irish cuirm, Welsh cwrw “beer”),68 Fig. 3.5: Replicas of spindle whorls numbers 6, 8, 13, but it is likely to have been borrowed made for LatinNow by Potted History. Photo: Pieter into the Latin of the area. The drink Houten, LatinNow. was a staple of western provincial life and was produced locally and of 69 pre-Roman heritage. Terms for it seem to have been borrowed from local languages into regional varieties of Latin.70 da is the imperative of the verb “to give” and, thanks to shared Indo-European origins, exists in both Latin and Gaulish.71 Following this analysis all four words could be understood as entirely Gaulish, entirely Gallic Latin or both. The second half of the example from Auxerre is more difficult to interpret, due to the uncertainties over the interpretation of VIM. This has been taken in unabbreviated form as Latin vim, meaning literally “force”, and here perhaps having sexual reference,72 plus pota “drink”, or Gaulish vimpota (a hypothetical form based on vimpo-, meaning unclear) or as an abbreviation of Latin vinum “wine”, plus pota “drink”, or potavim(us) (either “we have drunk” or (for potabimus) “we shall drink”).73 Trying to interpret this message reminds us of the importance of not “fixing” the text in print: the text is written around the curved exterior of the whorl with no obvious starting point, meaning the words could be read VIM POTA or POTA VIM. The most For Curmisagios, see Delamarre 2007, 80. For alcoholic drinks in Gaul, see Laubenheimer 2015. 70 On cervesa as “wheat beer” and curmi as “barley beer”, see Nelson 2003. Cervesa is attested c. AD 100 at Vindolanda (Tab. Vindol. 628): cervesam commilitones non habunt quam rogó iubeas mitti “my fellow-soldiers have no beer: please order some to be sent”, and in a number of provincial inscriptions, including one on a third–fifth-century AD ceramic cup from Vannes (Morbihan): […] BIBIS C[ER]VESA GRATIS “you drink beer for free” (Simon 2001, 29) and another on a large, fourth-century AD vessel from Mainz (Germany): IMPLE OSPITA OLA DE CERVESA DA “Hostess, fill the vessel with good beer (?)” (Année épigraphique 1992, no. 1287). Marcellus of Bordeaux mentions curmi and cervesa as ingredients to put into a cough mixture: in potionem cervesae aut curmi mittat (XVI 33). 71 RIG II.2 p. 323. Meid 1983, 1034 urges caution on the assignment of Latin/Celtic labels to this form, but then opts in preference for Latin. 72 I have not, however, found the phrase vim potare with sexual reference elsewhere and it does not occur in Adams 1982. 73 See RIG II.2 p. 334 for these options. 68 69

3.  Materializing epigraphy

57

likely interpretation, “drink wine”, would take the first half as Latin/Gaulish/both and the second as Latin. The inscribed whorls were described by Meid as being in a mixed jargon,74 a “typisches Kompromißprodukt”75, which made communication easier in a bilingual environment. Adams, in his ground-breaking work Bilingualism and the Latin language,76 rejected this description, stating that “[t]here are certainly no grounds for setting up a mixed language, neither fully Latin nor fully Gaulish, which might have become established at a transitional stage in the process of Romanisation”.77 Instead he interpreted these short texts as showing “code-switching”, the switch from one language to another within or between sentences. He surmises that “[a]t a time of advanced Romanisation, when Gaulish was fading from use, code-switching into Gaulish or the use of simple Gaulish phrases might have offered a sort of language of intimacy, a language which has become fossilised in semi-public form in the banter of the spindle whorls”.78 Code-switching has been a popular topic amongst classicists in recent years and has led to insights into the use of languages, cultural interactions, and identities in the Roman world.79 Used beyond the narrowly linguistic, code-switching is a way to approach identities that does not assume one or the other identity (for example Roman or indigenous) or even hybridity. Wallace-Hadrill supported bilingualism, and specifically code-switching, as a model for understanding cultural interaction in the Roman world because in his view an individual did not need to be Greek or Roman or native, nor a fusion, but could be all three at the same time.80 Other models, including even hybridization, assume a replacement of old identities with new, whereas the model of bi/multilingualism “points the way to other possibilities: of populations that can sustain simultaneously diverse culture-systems, in full awareness of their difference, and code-switch between them”.81 Code-switching reflects “the power of multiple identities” and “their strategic deployment in diverse contexts”.82 Meid 1983, 1030. Meid 1983, 1034. 76 Studying Classics has always entailed an appreciation of bilingualism and biculturalism, of course, but it is only in the last two decades that full engagement with modern bi- and multilingualism theory and practice has begun, following pioneering work by Adams. For studies using evidence other than the literary see, for example, Adams et al. 2002; Adams 2003; Biville et al. 2008; Cotton et al. 2009; Mullen and James 2012; Mullen 2013a. 77 Adams 2003, 197. 78 Adams 2003, 197. The perceived linguistic context – a decline of Gaulish – seems to have motivated the late dating by linguists. This context, however, is not necessarily indicated by the linguistic content. Texts in more than one language do not necessarily reflect a lack of competence and a decline in the vitality of languages. 79 See Elder and Mullen 2019 for a detailed study of code-switching in Roman letters and its value for understanding individuals, politics, culture, and society and for extensive references to earlier secondary literature. 80 Wallace-Hadrill 2008, 3–7. 81 Wallace-Hadrill 2008, 27–8. 82 Wallace-Hadrill 2008, 85. 74 75

58

Alex Mullen

One issue with the code-switching analysis, however, is that it generally assumes the interplay of bounded linguistic entities which we call “languages” and a link from these to specific identities.83 Discussions of code-switching tend to have reductive tendencies: the switch is between different cultural identities encoded through language, Latin for Roman and other, Gaulish, in this case, for local/indigenous. Both concepts – identities and languages – are not always so straightforward to capture. Identities are complex, fluid, and overlapping: in short, hard to identify. In a period before nation states and without systematic, universal education, evidence on the ground amongst the provincial population sometimes suggests that language was not carved up into linguistic entities in the way that we, or some high-status Romans, were trained to recognize, and that speech was a more flexible linguistic resource for its users than we sometimes assume armed with our Indo-European lexica and grammars of Latin. Some of these texts on whorls seem to resist code-switching analysis: they can be simultaneously read in either, or both, languages, rather than alternately one language then the other. Rather than trying to force the bilingual texts into a code-switching scheme we might instead appreciate the flexibility of linguistic resources at play. In other linguistic terms these texts show “bilingual homonymy” and examples of “lexical ambiguity”. Some modern linguists explore the cognitive processes that underlie lexical ambiguity resolution/lexical disambiguation and might see the “ambiguous” words in our texts almost as a problem to be resolved. Instead, these might be skilful ways to address various linguistic competences. Here language cannot be attached to one language at all: the polysemy is deliberate. The linguistic resources, from the perspective of those using them at least, might not be seen as strictly composed of two languages, but rather as a continuum of repertoire that could be used flexibly, providing windows into culture and identities that sometimes overlap, sometimes merge, and sometimes stay distinct. We could argue that the output may be a way of showing awareness of, and ability to negotiate, multiple identities, but this may be an overly academic commentary: the output may be playful. This flexibility of linguistic practices is seen in multilingual contexts across time and space: it does not necessarily involve creating stable mixed languages or switching between two separate languages as in the well-documented process of code-switching, but encompasses a wider range of subtle and fluid, sometimes ephemeral, linguistic practices. The multilingual skills on display in the spindle whorls are by no means necessarily the preserve of the highly educated and well-to-do. Indeed, the very well trained might arguably be less likely to accept “non-normative” language in writing. Modern sociolinguists might employ the term translingualism to describe this multilingual linguistic fluidity. This term will serve as a useful addition to our conceptual toolkit when dealing with multilingual inscriptions such as those on the whorls which do not neatly fit into our existing terminologies and helpfully reminds

83

For issues with using the model of bilingualism for cultural contacts more broadly, see Mullen 2013b.

3.  Materializing epigraphy

59

us that the languages carved up, described, and labelled by linguists may not map onto the linguistic experiences of those that use them. Translingualism (linked to the field of translanguaging, rather than to the earlier literary translingualism)84 refers to the notion of “going-beyond” Languages (with a capital L), namely “the deployment of a speaker’s full linguistic repertoire without regard for watchful adherence to the socially and politically defined boundaries of named (and usually national and state) languages”.85 Whilst many support the social justice focus of the advocates of translanguaging, there have been criticisms.86 Most relevant for us, it seems that in the concerted attempt to move entirely away from “Language”, some scholars have effectively denied the existence of languages altogether. But this is an untenable stance: clearly Languages do have meaning in some contexts, for example, they have linguistic reality for linguists and as the standard languages of nation states. These perspectives can be relevant when analysing the intricacies of linguistic resources, practice, and interaction. The concept of translingualism, used to refer to the complexities of linguistic realities in bi- and multilingual situations (both the outputs and the mindsets), but not pushed so far that “Languages” no longer exist, is relevant for some modern multilingual environments and for thinking about how some individuals and communities in the Roman provinces used, and may have viewed, their language.87

A new spin on old material The implements of the industry may be lost forever, like the “songs of the weaving women” that lilted through the streets of a late ancient city, but it is the historian’s task to sense the vanished artefact and to hear the “rhythms” of those whose labor was taken in the endless cycles of the spindle and loom.88

Thus an historian of slavery encourages us to reanimate the long silent spinning women of the Roman world. Strangely the words on the inscribed whorls from Autun have been absent from the work of those who have otherwise done so much to shine a light into the often-overlooked work of millions of women across the Roman world. The texts have been known for over a century and the corpus now totals two For literary translingualism and the Graeco-Roman world, see Bozia and Mullen 2021. Otheguy et al. 2015, 281. The main drivers behind this concept have been pedagogical, with adherents arguing that monolingual teaching environments that do not appreciate the complexity of linguistic resources of bilingual children are poor contexts for their learning. 86 Note, for example the lively exchange, between MacSwan 2017 and Otheguy et al. 2019. One key criticism has been that it is not all as new as they would have us believe: the notion that users of more than one language do not necessarily view their linguistic resources as the bounded entities that linguists describe and that psycholinguistically they are not two (or more) monolinguals in one person are arguments that pre-date the recent interest in translanguaging (e.g., Grosjean 1989). 87 See further, Mullen forthcoming b. 88 Harper 2011, 135. 84 85

60

Alex Mullen

dozen. They have intrigued epigraphers and particularly those linguists interested in investigating Gaulish and bilingualism in Gaul. But the disciplinary boundaries that, despite the rhetoric, are still strong, have prevented their incorporation into the extensive scholarship on ancient textiles and the societies and individuals involved with them. This chapter has brought the inscriptions and their contexts closer together and has presented a new spin which must be considered in discussions of the topos of the virtuous woolworker. An archaeological and sociolinguistic lens has enabled us to be explicit about the limits of our knowledge and the role our assumptions play in constructing our interpretations. Our ability to reconstruct ancient social realities is restricted and texts have meanings that are not “set in stone”. Nevertheless, we have exploited the epigraphic remains to the fullest by deploying an interdisciplinary epigraphy, combining archaeological and sociolinguistic perspectives, and bringing in evidence not previously considered with these materials, for example, the stelae from Autun. This has led to new commentary on the dates, social backgrounds of the spinners, and possible contexts for the creation of the whorls. Reconstructions of possible realities have been offered through a phenomenological approach based on clues from the material itself. Another significant step, which will be relevant for other textual materials from the ancient world, is the introduction of the term translingualism as a way to describe the fluidity of multilingual resources deployed in the texts from Autun, and the possible linguistic perspectives of their users. Our existing terminology of bilingualism is not sufficient to cope with these enigmatic offerings. The argument is not that code-switching was not a feature of bilingual communities and individuals in the ancient world – it clearly was89 – but that we should not attempt to force these multilingual texts to fit into a code-switching analysis. Language itself is a social construct (with “Language-focused” meaning for many, but not necessarily all users) and identities may not be easily directly identified with languages. The particular ethno-national linguistic perspective of modern nation states, which sees languages as reified and linked directly to territories and ethnic/national identities, should not be automatically assumed for the whorl carvers and spinners of Autun.

Bibliography

Adams, James N. 1982. The Latin Sexual Vocabulary. London: Duckworth. Adams, James N. 2003. Bilingualism and the Latin Language. Cambridge: Cambridge University Press. Adams, James N. 2007. The Regional Diversification of Latin 200 BC – AD 600. Cambridge: Cambridge University Press. Adams, James N., Mark Janse, and Simon Swain. 2002. Bilingualism in Ancient Society. Language Contact and the Written Text. Oxford: Oxford University Press.

89

See Adams 2003; Elder and Mullen 2019.

3.  Materializing epigraphy

61

Allason-Jones, Lindsay. 1995. “‘Sexing’ Small Finds.” In Theoretical Roman Archaeology: Second Conference Proceedings, edited by Peter Rush, 22–32. Aldershot: Avebury. Allen, Martyn, Lisa Lodwick, Tom Brindle, Michael Fulford, and Alexander T. Smith. 2017. New Visions of the Countryside of Roman Britain: Volume 2: The Rural Economy of Roman Britain. London: Society for the Promotion of Roman Studies. Bagnasco Gianna, Giovanna. 1999. “L’acquisizione della scrittura in Etruria: materiali a confront per la ricostruzione del quadro storico e culturale.” In Scritture Mediterranee tra il IX e il VII secolo a. C., edited by Giovanna Bagnasco Gianna and Federica Cordano, 80–105. Milan: Ed. ET. Barthèlemy, A. 1976. “Découvertes archéologiques sur le chantier du nouvel hôpital à Flacé.” Annales de l’Académie de Mâcon 53: 1–14. Beltrán Lloris, Francisco and Miguel Beltrán Lloris. 2012. “Ama lateres! Sobre una pesa de telar relativa al lanificium.” Sylloge Epigraphica Barcinonensis (SEBarc) 10: 127–48. Beltrán Lloris, Francisco, Carlos Jordán Cólera, and Ignacio Simón Cornago. 2021. “The Inscriptions on Pottery of Hispania Citerior (2nd-1st Century BCE): A Case Study of the Inscriptions on Spindle Whorls.” In Epigraphy on Ceramics. The Instrumentum Domesticum, Its Textual Genres and Its Functions in Ancient Societies, edited by Wim Broekaert, A. Delattre, Emmanuel Dupraz, and María José Estarán Tolosa, 139–161. Geneva: Droz. Binet, Éric and Monique Dondin-Payre. 2002. “La première fusaïole inscrite trouvée à Amiens.” Revue Du Nord 84: 133–7. Biville, Frédérique, Jean-Claude Decourt, and Georges Rougemont, eds. 2008. Bilinguisme gréco-latin et épigraphie. Lyon: Maison de l’Orient méditerranéen. Blom, Alderik H. 2010–2012. “Language at La Graufesenque: The Evidence of the Potters’ Graffiti.” Keltische Forschungen 5: 13–53. Bodel, John. 2001. “Epigraphy and the Ancient Historian.” In Epigraphic Evidence: Ancient History from Inscriptions, edited by John Bodel, 1–56. London: Taylor and Francis. Bowman, Alan K., and Michael Brady, eds. 2005. Images and Artefacts of the Ancient World. Oxford, New York: Oxford University Press. Bozia, Eleni and Alex Mullen. 2021. “Literary Translingualism in the Greek and Roman Worlds.” In The Routledge Handbook of Literary Translingualism, edited by Steven G. Kellman and Natasha Lvovich, 45–59. New York, Abingdon: Routledge. Castro Curel, Zaida. 1980. “Fusayolas Ibéricas, antecedentes y empleo.” Cypsela 3: 127–46. Chardron-Picault, Pascale and Monique Dondin-Payre. 2000 [2002]. “Une nouvelle fusaïole inscrite à Autun.” Antiquités Nationales 32: 209–17. Clackson, James. 2015. Language and Society in the Greek and Roman Worlds. Cambridge: Cambridge University Press. Costin, C. L. 2013. “Gender and Textile Production in Prehistory.” In A Companion to Gender Prehistory, edited by Diane Bolger, 206–27. London, Wiley-Blackwell. Cottica, Daniela. 2007. “Spinning in the Roman World: From Everyday Craft to Metaphor of Destiny.” In Ancient Textiles: Production, Crafts and Society, edited by Carole Gillis and Marie-Louise Nosch, 220–8. Oxford: Oxbow Books. Cotton, Hannah M., Robert G. Hoyland, Jonathan J. Price, and David J. Wasserstein. 2009. From Hellenism to Islam. Cultural and Linguistic Change in the Roman Near East. Cambridge: Cambridge University Press. d’Encarnação, José. 1984. Inscrições Romanas do Conventus Pacensis. Coimbra: Coimbra Inst. de arqueologia. Delamarre, Xavier. 2007. Noms de personnes celtiques dans l’épigraphie classique. Paris: Errance. Desforges, A. 1924. “Fusaïole à inscription gauloise.” Bulletin de La Société Préhistorique de France 21: 260–1.

62

Alex Mullen

Dondin-Payre, Monique. 2001. “L’onomastique dans les cités de Gaule Centrale (Bituriges Cubes, Eduens, Senons, Carnutes, Turons, Parisii).” In Noms, identités culturelles et Romanisation sous le Haut-Empire, edited by Monique Dondin-Payre and Maria-Theresa Raepsaet-Charlier, 193–341. Brussels: Le livre Timperman. Dondin-Payre, Monique. 2004. “Les pesons de fuseaux inscrits en Gaule Romaine.” Bulletin de la Société Nationale des Antiquaires de France 2000: 198–205. Dondin-Payre, Monique. 2005. “Épigraphie et acculturation:  l’apport des fusaïoles écrites.” In Le monde Romain à travers l’épigraphie : méthodes et pratiques, edited by Janine Desmulliez and Christine Hoët-Van Cauwenberghe, 133–46. Lille: Université Charles-de-Gaulle – Lille III. Dondin-Payre, Monique. 2006. “Du nouveau dans un dossier original: la fusaïole inscrite de Mâcon (Saône et Loire) et la fusaïole inscrite de Nyon (Confédération Helvétique).” L’antiquité Classique 75: 143–54. Elder, Olivia and Alex Mullen. 2019. The Language of Roman Letters: Bilingual Epistolography from Cicero to Fronto. Cambridge: Cambridge University Press. Eska, Joseph. 1998. “Tau Gallicum.” Studia Celtica 32: 115–27. Ferrer i Jané, Joan. 2008. “Ibèric Kastaun : un element característic del lèxic sobre torteres.” Cypsela 17: 253–71. Ferrer i Jané, Joan, Javier Velaza, and Noemí Moncunill. 2020. “Une nouvelle inscription rupestre latine d’Err (Pyrénées-Orientales).” Latomus 79: 78–88. Feugère, Michel and Pierre-Yves Lambert. 2011. “Une belle gauloise... A propos d’une fibule inscrite de Laon.” Etudes Celtiques 37: 147–52. Gevirtz, S. 1967. “A spindle Whorl with Phoenician Inscription.” Journal of Near Eastern Studies 26: 13–16. Giles, Howard, Richard Bourhis, and D. Taylor. 1977. “Towards a Theory of Language in Ethnic Group Relations.” In Language, Ethnicity and Intergroup Relations, edited by Howard Giles, 307–48. London, New York: Academic Press. Grosjean, F. 1989. “Neurolinguists, Beware! The Bilingual Is Not Two Monolinguals in One Person.” Brain and Language 36: 3–15. Habicht, Christian. 1984. “Pausanias and the Evidence of Inscriptions.” Classical Antiquity 3.1: 40–56. Hamilton, Sue and Ruth Whitehouse. 2006. “Phenomenology in Practice: Towards a Methodology for a ‘subjective’ Approach.” European Journal of Archaeology 9.1: 31–71. Harlow, Mary. Forthcoming. “Spinning – the Invisible Profession.” In The Value of Making: Theory and Practice in Ancient Craft Production, edited by H. Hochscheid and B. Russell. Turnhout: Brepols. Harper, Kyle. 2011. Slavery in the Late Roman World, AD 275–425. Cambridge, New York: Cambridge University Press. Héron de Villefosse, H. 1914. “Un peson de fuseau portant une inscription latine incisée, trouvé à Sens.” Bulletin Archéologique du Comité des Travaux Historiques et Scientifiques, 213–30. Higbie, Carolyn. 1999. “Craterus and the Use of Inscriptions in Ancient Scholarship.” Transactions of the American Philological Association 129: 43–83. Hoskin, Philippa and Elizabeth New. 2017. “Imprint: A Forensic and Historical Investigation of Fingerprints on Medieval Seals.” Medieval Prosopography 32: 249–52. Kajanto, Iiro. 1965. The Latin Cognomina. Helsinki: Societas scientiarum Fennica. Lambert, Pierre-Yves. 2018. La langue gauloise. 3rd ed. Paris: Errance. Larsson Lovén, Lena. 2007. “Wool Work as a Gender Symbol in Ancient Rome. Roman Textiles and Ancient Sources.” In Ancient Textiles: Production, Crafts and Society, edited by Carole Gillis and Marie-Louis Nosch, 229–36. Oxford: Oxbow Books. Larsson Lovén, Lena. 2013. “Female Work and Identity in Roman Textile Production and Trade: A Methodological Discussion.” In Making Textiles in Pre-Roman and Roman Times. People, Places Identities, edited by Margarita Gleba and Judit Pásztókai-Szeőke, 109–25. Oxford: Oxbow Books.

3.  Materializing epigraphy

63

Laubenheimer, Fanette. 2015. Boire En Gaule. Paris: CNRS. Lejeune, Michel. 1976–1978. “Notes d’étymologie gauloise.” Études Celtiques 15: 95–104. Levenson, Jon D. 1973. “The Spindle Whorl from Chatal Hüyük: A Forgery.” Bulletin of the American Schools of Oriental Research 209: 37–40. Loth, Joseph. 1916. “Remarques aux inscriptions latines sur pesons de fuseau trouvés en territoire gaulois et, en particulier, à l’inscription celtique de Saint-Révérien (Nièvre).” Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres, 168–86. MacSwan, Jeff. 2017. “A Multilingual Perspective on Translanguaging.” American Educational Research Journal 54, no. 1: 167–201. Maggetti, Marino, Albert Jornet, Monique Dondin-Payre, and Véronique Rey-Vodoz. 2009. “La fusaïole Gallo-Romaine inscrite de Nyon: provenance et technique.” Annuaire d’Archéologie Suisse 92: 252–5. Marichal, Robert. 1988. Les graffites de La Graufesenque. Paris: CNRS. Meid, Wolfgang. 1983. “Gallisch oder Lateinisch? Soziolinguistische und andere Bemerkungen zu populären Gallo-Lateinischen Inschriften.” Aufstieg und Niedergang der römischen Welt II.29.2: 1019–44. Millar, Fergus. 1968. “Local Cultures in the Roman Empire: Libyan, Punic and Latin in Roman Africa.” Journal of Roman Studies 58: 126–34. Morlock, Emmanuelle and Eleanor Santin. 2014. “The Inscription between Text and Object: The Deconstruction of a Multifaceted Notion with a View of a Flexible Digital Representation.” In First EAGLE International Conference on Information Technologies for Epigraphy and Cultural Heritage, The Europeana Network of Ancient Greek and Latin Epigraphy, edited by Silvia Orlandi, Raffaella Santucci, Vittore Casarosa, and Pietro Maria Liuzzo, 325–50. Rome: Sapienza università editrice. Mullen, Alex. 2012. “Introduction. Multiple Languages, Multiple Identities.” In Multilingualism in the Graeco-Roman Worlds, edited by Alex Mullen and Patrick James, 1–35. Cambridge: Cambridge University Press. Mullen, Alex. 2013a. Southern Gaul and the Mediterranean. Multilingualism and Multiple Identities in the Iron Age and Roman Periods. Cambridge: Cambridge University Press. Mullen, Alex. 2013b. “The Bilingualism of Material Culture? Thoughts from a Linguistic Perspective.” HEROM 2: 21–43. Mullen, Alex. 2016. “Sociolinguistics.” In The Oxford Handbook to Roman Britain, edited by Martin Millett, Louise Revell, and Alison Moore, 573–98. Oxford: Oxford University Press. Mullen, Alex. 2021. “Socio-Literacy: An Interdisciplinary Approach to Understanding Literacy in the Roman North-West.” In Aprender la escritura, olvidar la escritura: nuevas Perspectivas sobre la historia de la escritura en el Occidente romano, edited by Noemí Moncunill and Manuel Ramírez-Sánchez, 349–372. Vitoria: Universidad de Pais Vasco. Mullen, Alex. Forthcoming a. “Transformations de la Gaule sous l’Empire Romain: une culture de l’écrit au service de la production de masse.” In La culture de l’écrit en Méditerranée occidentale à travers les pratiques épigraphiques (Gaule, Ibérie, Afrique du Nord), edited by Michèle ColtelloniTrannoy and Noemí Moncunill, 131–62. Leuven: Peeters. Mullen, Alex. Forthcoming b. “Language shift, attitudes and practices in the Roman West.” In Multilingual Practices Through the Ages: From Antiquity to the Present Day, edited by Aneta Pavlenko, Pia Lane and Alexandre Dûchene. Cambridge: Cambridge University Press. Mullen, Alex and Coline Ruiz Darasse. 2018. Gaulish. Language, Writing, Epigraphy. Zaragoza: Universidad de Zaragoza. Mullen, Alex and Coline Ruiz Darasse. 2020. “Gaulish.” Palaeohispanica 20: 749–83. Mullen, Alex and Patrick James. 2012. Multilingualism in the Graeco-Roman Worlds. Cambridge: Cambridge University Press. Nelson, Max. 2003. “On a Beautiful Girl and Some Good Barley Beer.” Etudes Celtiques 35: 257–9.

64

Alex Mullen

Otheguy, Ricardo, Ofelia García, and Wallis Reid. 2015. “Clarifying Translanguaging and Deconstructing Named Languages: A Perspective from Linguistics.” Applied Linguistics Review 6, no. 3: 281–307. Otheguy, Ricardo, Ofelia García, and Wallis Reid. 2019. “A Translanguaging View of the Linguistic System of Bilinguals.” Applied Linguistics Review 10, no. 4: 625–51. Pavlenko, Aneta and Alex Mullen. 2015. “Why Diachronicity Matters in the Study of Linguistic Landscapes.” Linguistic Landscapes 1: 114–132. Piquette, Kathryn E. and Ruth D. Whitehouse, eds. 2013. Writing as Material Practice: Substance, Surface and Medium. London: Ubiquity Press. Rafel Fontanals, Núria. 2007. “El textil como indicador de género en el registro funerario Ibérico.” Treballs d’Arqueologia 13: 113–44. Rebourg, Alain. 1996. L’oeuvre au noir: l’emploi du schiste à Augustodunum. Autun: Musée Rolin. Sauvage, Caroline and Robert Hawley. 2013. “Une fusaïole inscrite en ougaritique au Musée d’Archéologie Nationale de Saint-Germain-En-Laye.” In Études Ougaritiques III, edited by Valérie Matoïan and Michel Al-Maqdissi, 365–94. Leuven/Paris/Walpole MA: Peeters. Scheid, John and Jesper Svenbro. 1994. Le métier de Zeus: mythe du tissage et du tissu dans le monde Gréco-Romain. Paris: La Découverte. Simon, Laure. 2001. Vannes: Le sanctuaire gallo-Romain de Bilaire. Volume 2: les mobiliers. DRAC Service Regional de l’Archéologie de Bretagne. Terras, Melissa. 2006. Image to Interpretation: An Intelligent System to Aid Historians in Reading the Vindolanda Texts. Oxford: Oxford University Press. Thüry, Günther E. 2008. “Die erotischen Inschriften des instrumentum domesticum: ein Überblick.” In Instrumenta Inscripta Latina. 2, Akten des 2. Internationalen Kolloquiums, Klagenfurt, 5.–8. Mai 2005, edited by Manfred Hainzmann and Reinhold Wedenig, 295–304. Klagenfurt: Verlag des Geschichtsverein für Kärnten. Tilley, Christopher. 1994. A Phenomenology of Landscape. Places, Paths and Monuments. Oxford: Berg. Torelli, Mario. 1984. Lavinio e Roma: riti iniziatici e matrimonio tra archeologia e storia. Rome: Quasar. Tsori, N. 1959. “A Spindle Whorl with Hebrew Inscription.” Israel Exploration Journal 9, no. 3: 191–2. Van Driel-Murray, Carol. 1999. “A Set of Roman Clothing from Les Martres-de-Veyre, France.” Archaeological Textiles Newsletter 28: 11–15. Venault, Stéphane, Simone Deyts, Yann Le Bohec, and Yannick Labaune. 2009. “Les stèles funéraires de La nécropole de Pont-l’Évêque. Contextes de découverte et étude du corpus.” Bulletin Archéologique 35: 129–204. Wallace-Hadrill, Andrew. 2008. Rome’s Cultural Revolution. Cambridge: Cambridge University Press. Whatmough, J. 1949. “GAULISH uimpi.” Language 25, no. 4: 388–91. Wild, John Peter. 1970. Textile Manufacture in the Northern Roman Provinces. Cambridge: Cambridge University Press. Wild, John Peter. 2002. “The Textile Industries of Roman Britain.” Britannia 33: 1–42. Wild, John Peter. 2003. “The Romans in the West, 600 BC–AD 400.” In The Cambridge History of Western Textiles, edited by David T. Jenkins, Vol. 1, 77–93. Cambridge: Cambridge University Press.

Chapter 4 Written to be (un)read, written to be seen: beyond Latin codes in Latin epigraphy M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

Introduction The aim of this contribution is to analyse the relationship between images and the transmission of meaning in epigraphic Roman tombstones, resulting in a language that is neither pure Latin, nor pure image. Our main purpose is to delve deeper into the inscriptions’ code, underlining the fact that an inscription is not always read but rather interpreted. To a large extent, inscriptions were conceived to be seen rather than to be read, establishing a codified relationship between author and user. For this initial study, we will focus on funerary epigraphy, for a number of reasons. Epitaphs are a type of record that show a multiplicity of interactions between text and image. Furthermore, they involve a certain level of personal choice. Funerary monuments thus provide us with an ideal record to evaluate the interaction between text and image. In fact, in some provincial contexts there are records of certain types of sculpture where inscriptions were retrospectively superimposed in a later period under the influence of Roman culture.1 In these cases, the starting point is therefore a previous iconographic tradition, to which a text is introduced later. Inserting the text in Latin – the language of the political and economic power – was a means of conferring prestige upon the funerary customs. Finally, yet importantly, funerary inscriptions made minorities visible, especially women and liberti. Thus, the visible spectrum of human society became wider and allowed for different messages intended to be read by almost everyone. This would be one reason why they appealed to passersby. Accordingly, in this chapter we assume that funerary epigraphy can provide a different and underexploited vision of both epigraphic messages and their readers that is not necessarily accessible in other sorts of texts and evidence. In what follows, 1

Rose 2003; Rodríguez Corral 2019.

66

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

we will employ diverse levels of interpretation in a series of case studies in order to better understand not only the producers of funerary epigraphy, but also its readers. Any ancient observer of a funerary inscription could perceive different kinds of information at different levels. These would be fully understood only by members of the same culture and society within which the epigraph was created. Firstly, a monument shows status and a personal way of respecting death, which derives from widespread shared rituals and funerary customs. Secondly, there is a dialogue between image and text, even when the observer was not able to read it. Images were not coincidental, but rather they complemented the text. Finally, if the beholder could read a modicum of Latin, specific information about the person who was buried could be comprehended. Not all individuals had the same level of literacy and iconographic comprehension, but they would have been aware of all of these elements to varying degrees. Some could not reach all aspects of interpretation and might, for instance, have required other people to read the text out loud for them. To this end, in this chapter inscriptions are not studied following a solely philological methodology, since code and context are more important than they might initially seem, and so we want to tackle the perception of different kinds of ancient spectators. Finally, we should add that Latin had dialectal variations in its spoken form that were transferred into written inscriptions. This fact shows us that local people understood their inscriptions slightly differently from outsiders, even if both were able to communicate. Two years ago, a professor of Latin Philology mentioned to us during a casual conversation that she was amazed at how specialists in Latin epigraphy could understand the content of inscriptions; the concoction of letters and symbols with hardly any complete words seemed to her anything but Latin. This professor was used to reading Latin every day, but only texts belonging to written sources in literary Latin, preserved through the manuscript tradition. Though this was little more than a typical academic joke, the comment exposed an issue that we had already considered from our respective fields of work in relation to epigraphy. Specialized work with Latin epigraphs has in some sense deformed our knowledge of Latin, since at the end of the day we are more familiar with epigraphic Latin, so to speak, than with the literary one. Our Latin was not “standard”, and it changed according to local peculiarities. In these local differences, there are dialectal language uses, accents and Latin deformations, which can be studied with different purposes to that of just reading an inscription. From this Latin we can trace the state of the language, as well as conventions that were sometimes simply local. This information has been already used in order to identify inscriptions without context, but can also be employed for other interpretative ends, as we will discuss. Some months after the above-mentioned conversation, we realized that Latin epigraphy has been used throughout its history as a mixture of code and image, even when Latin was not a common language anymore. In fact, if we radically separate image from text, we miss an important part of information. When people employed a language with which they were not familiar, this could result both in mixes and mix-

4.  Written to be (un)read, written to be seen

67

ups. This happens even today. Some years ago in the Spanish city of Cádiz, an official monument was set up, promoted by the Arts and Crafts Academy and the city mayor. They wanted to commemorate the foundation of the first photography studio in the city. When writing “1996” with Roman numerals, they carved MDCCC96,2 showing that Roman numbers could be used in a public inscription but not necessarily correctly. The conventional use of Latin in monumental and official inscriptions for centuries has two consequences: Latin is a language of prestige and some modern inscriptions still use this language for certain elements. Roman numerals, notably more difficult to use than Arabic ones, are still commonly employed to indicate dates in epigraphy, even if they are not used on a daily basis. The person who ordered the epigraph in Cádiz wanted this section in Roman numerals, but the designer and the stonecutter were not able to realize that they were mixing codes. This anecdote, which anyone can still personally verify in Cádiz, shows us that, even nowadays with the abundance of sources available on the internet and through academic research, the writing of an epigraph in Latin demands knowledge of the code. This made us consider a certain line of analysis: epigraphy is a science close to both philology and history, so it must be approached from a different perspective. Ever since its origins as a specialist sub-discipline, epigraphy has been marked by a philological bias in which an epigraph is translated, and reduced, as a text written on a paper, since this is understood as its main component. Initially, only the inscribed text was studied, with images largely ignored. However, an epigraph is a monument consisting of different parts: context, shape, size, material, text, encoding of text and image. Taken together, all the parts create a common message. Thus, epigraphy has traditionally established an order of priority in which “who reads what” has been predominant over “how is it read or seen and within what frame”. This mode of thinking leads to a reduced perception of epigraphic communicative channels since, as we will try to demonstrate, the reading of a text in antiquity, if it happened at all, would be the last stage in a series of approaches to the implicit content of an epigraph. Knowing and reading Latin as a functional language is not equivalent to reading and understanding Latin epigraphy, however contradictory that may seem, just as it is not the same to master a living language such as English or Spanish and to be able to read and write them in the abbreviated kind of messages employed on social networks, where it is common to mix abbreviations that are created generation after generation. Of course, this comparison should always be understood bearing in mind each kind of message within its context: texts today are temporary and generally poorly elaborated despite the high literacy level of the society creating them. The conclusion we came to, and which we will propound and defend here, is Fotografía Reymundo/ se estableció en esta casa / en el año MDCCC96. / El exc(elentísi)mo Ayuntamiento de Cádiz / y la Real Academia Provincial / de Bellas Artes / conmemoran el centenario. Marble plaque. 1996. Plaza de Mina, 2, Cádiz (Spain). 2

68

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

that epigraphy itself implies a set of codes for expression and comprehension that only partially matches those of language. In order to understand all the implications of the message in an inscription, knowing the supporting language is not enough. Code is a quite abstract concept that varies throughout human history depending on social circumstances and education. Scholars nowadays take account not only of the amount of writing but also of its quality, in order to determine how many people were able to understand it at different levels of literacy.3 We will focus on specific examples in order to propose a fundamental theory: as epigraphists, we have traditionally approached inscriptions in the opposite direction from the experience of ancient observers when decoding inscriptions. With this aim, this work is structured following the traditional approach to epigraphs: from textual reading, through the piece towards the context. Or final goal is to show that we should consider the opposite order as the correct one, so that our questions and approach to the epigraphic record can take into consideration the perception of ancient observers as referents instead of our perspective as contemporary viewers and researchers. We provide different examples of subtext problems that have been wrongly interpreted through centuries, as well as instances where not taking into account all the elements that are part of an inscription has resulted in wrong interpretations.

Latin origins Latin’s first appearance in a publicly written form takes place through epigraphy, which must be understood as a prestigious and political medium, since it works as sort of a letter of introduction for dedicators and their families and entails an economic outlay.4 Stone epigraphy may have not been the first kind of text written in Latin, especially in the provinces, but it is the oldest one that we have and can interpret. When an epigraph is found, we should assume that some people were able to interpret it or, at least, that its concept could be understood by a certain group of people. A public epigraph written in Latin was a symbol of prestige because it was related to power. At a more or less effective level of interpretation, commoners should have been able to, at least, appreciate it. Nevertheless, to appreciate it does not necessarily mean to fully understand it. On the other hand, the first epigraphs known are a very valuable source for scholars for different reasons. An epigraph is a primary source which has not been copied time after time, nor has it accumulated reading mistakes or copying ones. Latin manuscripts, conserved in monasteries, allow us to study texts by orators, philosophers, or artists, but they must be approached from a different point of view. It is obvious that when we have two manuscripts, they do not have exactly the same information. The texts of the plays of Euripides, for example, present small variations 3 4

Corbier 2013, 13. Beltrán Ortega 2015, 52–4.

4.  Written to be (un)read, written to be seen

69

as the result of a complex manuscript tradition. As a result, textual criticism is of course one of the foundation stones of philology, even if it still has certain problems.5 It has, however, also created a misunderstanding within epigraphy, since scholars often try to use the same methodology for reconstructing or reading texts written on stone. There are no mistakes made by second hands in epigraphy. A funerary inscription is a carefully thought-through document, displayed in a public spot to be read, mostly aloud, and today we can interpret it in its original format. Consequently, our study must not be the same as the traditional analysis of the Latin of literary works preserved through the manuscript tradition. The production of an inscription seeks the best acceptable result, especially in a long-lasting format with a planned and highly symbolic message. If the language is therefore different from “regular Latin”, we should search for another explanation than simply “getting it wrong”. With the exception of lost epigraphs, recorded only through academic publications, inscriptions offer us direct access to an original Latin text. Unlike with codices, we can read it and try to reconstruct it on its original monument. By overcoming the philological epigraphic paradigm, we can approach features such as “mistakes” in inscriptions in a different way from any other written source. The philological methodology has been perfectly developed in its own field but cannot be used unchanged for epigraphic interpretations.

Who reads? The imposition of Latin as the official language of Rome in the western provinces was not a controlled process from the metropolis; for example, there was no systematic education or systematic promotion of privileges for Latin speakers. On the other hand, there was no “patriotic” exaltation of local languages; rather, these were progressively forgotten as the conquerors’ language was adopted.6 But when did the process finish? And what was the Latin spoken on the streets before and after that? One of the main aspects in this issue is determining who was able to read. An epigraph is usually placed in a public space so that its content can reach people passing by. With all due reservations, Harris considered only around 10% of population were literate,7 with the possible exception of areas such as Campania, where it has been argued that everyone could recognize letters from the time of the early Empire.8 We assume that the population able to read a text – always aloud – was numerically Tov 1982, 439–45. Although Latin grew as a spoken language, “the interaction with Latin epigraphy in the Republican period seems to have stimulated the elaboration and expansion of epigraphies in local languages” (Mullen 2019, 3). For further information about this process and its decline at the beginning of the principate, see the works of the LatinNow Project accessible at https://latinnow.eu/publications-and-online-resources/. 7 Harris 1989, 259–73. 8 Todisco 2013, 295–301. It has even been argued that women in Campania, while not at men’s level, were relatively literate (Mano 2008, 205). 5 6

70

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

limited and associated with a high social status, with the exception of slaves trained to help in business. The highbrow literary culture would only concern the political and administrative elite, who would employ written Latin on a daily basis. We do not refer to the highest social class, who could read ancient Greek fluently as a cultivated language,9 since appealing only to this social class would be pointless. Knowing Latin exclusively for business reasons was associated with lower classes. We must remember the criticism towards freedmen in Petronius’ Satyricon,10 where the freedman Trimalchio discusses numerous issues, about which he shows a great lack of knowledge, as a clear attempt to criticize the nouveau riche. Thus, first- and second-class citizens were socially differentiated according to their access to basic Latin. We can consider letters to be regarded as a sort of ignoble science when employed only for businesses. Some masters would even prefer their slaves to be instructed. In sum, writing was indispensable but being educated was not. Not every child could be fully educated; rather, they just went through an initial literacy stage, as the characters in the Satyricon show.11 Trimalchio and his fellow freedmen were content with being able to read a text on stone (in capital letters), but one of Trimalchio’s friends was also able to speak with specific vocabulary relating to the funerary world. He is quite specific about the measures of his tomb, information only known by people who could correctly interpret an epitaph12 and he clearly shows off his knowledge. This example from a literary source is clear evidence that not all people could read at the same level, and that epigraphs were interpreted more than read. Having full knowledge of all the content written in an inscription was shown in public with pride. Evidence also comes from certain trading documents, in which it is specified that the author of the transaction was not able to write and required assistance. In a tabula cerata13 from Pompeii, M. Barbatus has to be represented by Q. Aelius Romanus Youtie 1975, 101–8. Horsfall 1989, 74–89, 194–209. 11 Petron., Sat. 58.7–8: Non didici geometrias, critica et alogias nenias, sed lapidarias litteras scio, partes centum dico ad aes, ad pondus, ad nummum. Ad summam, si quid uis, ego et tu sponsiunculam: exi, defero lamnam. Iam scies patrem tuum mercedes perdidisse, quamuis et rhetoricam scis. Ecce “Qui de nobis longe uenio, late uenio? Solue me.” (“I didn’t learn geometry, critical theory and meaningless nonsense, but I know letters on stone, I do percentages in measures, weights, and money. In short, if you like, I and you’ll have a little bet: come on, I’m putting my money down. You’ll soon realize that your father wasted his education fees, even if you know rhetoric. Look here at the riddle: ‘I’m someone of us, I come long and wide. Solve me.” (After the Loeb translation.)) 12 Keegan 2013, 53. 13 AE 1984, 227 (= HD003034): Ἐπὶ ὑπάτων Μάρκου Ἀκύλα Ἰουλι/άνου καὶ Ποπλίου Νωνίου Ἀσ/πρήνα πρὸ τριῶν εἰδῶν / Ἀπριλίων ἐν Δικαρχήα. Μενέλαος Εἰρηναίου Κερα/μιήτης ἔγραψα ἄπεχιν μοι / παρὰ Πρίμου Ποπλίου Ἀττίου Σεβή/ρου δούλου δηνάρια χίλια / ἐκ ναυλωτικῆς ἐκσφραγισμένης, / ἅ καὶ ἀποδώσω ἀποκολούθως / τῇ ναυλωτικῇ ἥ πεποίημαι πρὸς αὐτόν· κατέστησα δὲ ἔνγυον / ἐις ἔκτισιν τῶν προγεγραμμένων / δηναρίων χιλίων Μάρκον Βαρ/βάτιον Κέλερα. // Q(uintus) Aelius Romanus scripsi rogatu et / mandatu M(arci) Barbati Celeris coram / ipso quod is litteras nesciret eum / sua fide iubere eos (denarios) (mille) q(ui) s(upra) s(cripti) sunt / Primo P(ubli) Atti Seueri ser(uo) pro Menela/uo Irenaei f(ilio) Ceramietae ita / uti supra scriptum es[t]. Tabula cerata. 11 April 38 AD. Villa Murécine (Pompeii, Italy). 9

10

4.  Written to be (un)read, written to be seen

71

in the transaction, because Barbatus is not able to recognize the letters: ipso quod is litteras nesciret eum. This tabula provides another example of the lack of literacy among certain members of Roman society, and the use of assistants in order to interpret written documents. Some scholars consider that Roman society was mostly able to read and write. The main argument for this is the multitude of written objects, graffiti and property marks known from the Roman world.14 Regardless, funerary inscriptions are often comprised largely of formulaic phrases and abbreviations, to the extent that their texts can be confidently restored when only some parts of the inscription remain.15 Therefore, it was not necessary to know how to read and write to interpret an inscription whose content was, more or less, similar to all the funerary inscriptions that remained in a necropolis. Complete literacy was not needed, since context and code could be clearly understood. Cavallo proposes adjusting the limits between literate and illiterate population.16 He suggests that a large proportion of people would have only been able to recognize capital letters. It is likely that not all could even do that, since they had not received a formal education. However, some of them would be sporadically taught and under the constant stimulus of inscriptions and graffiti. These people would recognize certain sequences of letters and could perhaps even reproduce them, especially when linked to a specific context, but could not necessarily ascribe meaning to those sequences. They should therefore not be considered literate, though they would not be totally ignorant of the writing code. This idea shows that some members of the population could potentially recognize certain letters but were not able to read them. Such a group would not be sustainable without an intermediate level, that of semi-illiterates, i.e., people perhaps able to read and write their name and recognize certain formulae, but little more. Even if they were not able to read a full text, this group could offer readings of specific elements, or generalized interpretations, to illiterate companions. Within this semi-illiterate group, orality and memory would play a very important role in their mediation of text. In any provincial context, then, the presence of individuals able to read inscriptions aloud would be essential to the functioning of epigraphic culture. These people would not be able to read literary texts, but the common code and the use of repetitive formulae would make the message of inscriptions accessible to them, and through them, to others. Thus, orality would create the channel of mediation between the bulk of viewers and inscriptions. In conclusion, considering that very few people could read the full text and that only some others could interpret epigraphic formulae, the message would reach the majority of the population through the habit of reading out loud and the development of memory skills in ancient societies. Mayer i Olivé 2019, 15–25. Fernández-Martínez 2019, 63–5. 16 Cavallo 1978, 466–87. 14 15

72

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

If we take into consideration this hypothesis, literacy levels must have affected local epigraphic norms, and so local mistakes become the key to localized Latin pronunciation and formulae meanings. We usually talk about typos in inscriptions that show a substitution of c by k or double addition of sounds, just to mention some examples from the western provinces. These mistakes are quite common in certain areas and could be a result of their inhabitants’ spoken Latin. We do not know whether inscriptions were commonly commissioned orally or in writing. If the former, it would be very easy to make this kind of mistake. Another stage in which they could appear was when a stonecutter with a limited knowledge of Latin wrote an inscription from memory. Alien as a stone monument may seem to us, we must never forget that it required a significant economic outlay, and therefore we should be wary of interpreting these variations as typos. Everyone writes at his or her best, especially in a long-lasting format with a planned and highly symbolic message. We must understand that epigraphy is not a literary form of writing, with the exception of carmina.17 Local inhabitants, with a more or less limited knowledge of Latin, then interpreted a series of familiar formulae. However, these observers would not see typos as we conceive them today. Not only would locals be familiar with the same kind of typo in other inscriptions – thus creating a local variant of language – but additionally, anyone encountering the inscription via a shallow reading, or by hearing it read aloud by a second individual, would not find any difference in comparison with the sound of another inscription properly written in grammatical terms. One of the most famous examples of variation in the epigraphic inscription of a word, the term uxor, offers an interesting case study in this respect. There are several examples of variations recorded: uxsor, ucsor, ucxor, uxxor. A preliminary analysis shows that uxsor is commonly found in Italy, Gallia Narbonense and Hispania, whereas ucsor and ucxor have a notable concentration in the territories of Hispania, northern Africa, the Black Sea coast and, of course, Italy. Finally, uxxor is the rarest variation except for in Christian inscriptions from Rome.18 These typos show a spelling that tends to keep the same phonetical outcome as in the original term, uxor, such as swapping C for K or the fusion of the X and S sounds. The connection of these variations with the way people spoke is obvious: the additional letters must reflect phonetic spelling. However, an individual predominantly familiar with standard Latin orthography may have had problems decoding the inscription, due to the unorthodox part of the code (i.e., confusion between certain sounds). This situation is not exclusive to ancient Rome. Code continues to manifest in writing, even as Latin keeps being written. The wish to imitate Roman epigraphy, at a time when the understanding of the Roman epigraphic code has been lost, leads to counterfeits such as the Caesar Cippus at Rimini. This object is an attempt Carmina are considered an exception. They are poems written ad hoc, unique texts without formulae or abbreviations. Most probably, they were difficult to read for people with incomplete knowledge of Latin. 18 The search was conducted on EDR, HD, EDB, HEpOl and EAGLE on 10 January 2020. 17

4.  Written to be (un)read, written to be seen

73

to create a commemorative inscription that was expected to be taken as real in its time, but is now easily identified as a forgery by current researchers – who are well aware of Latin epigraphy codes – and likewise would be recognizably “wrong” to a hypothetical ancient observer brought forward in time to read the epigraph.19 It is obviously correct Latin, properly declined, and claimed in the Renaissance to be an ancient epigraph in a bad state of conservation that was remade to be exposed in a public place.20 However, the content, with its lack of abbreviations, is totally alien to the Roman epigraphic code. But this mistake could only be recognized either by an actual ancient individual, who knew instinctively the system of abbreviation, or an epigraphist who has deeply studied and absorbed Latin epigraphic codes. For Renaissance viewers, it could be conceived as a historic epigraph, but it has no aura of authenticity for either an ancient Roman or a modern scholar. Finally, we can transfer this problem to written codes of today. In all our daily communication situations, channels and codes have evolved and changed within formal and informal contexts. In WhatsApp and Twitter, quickness, immediacy, and clarity are imperative for messages written in haste among many other daily activities. These kinds of texts obviously do not have the same permanency as monumental inscriptions – as we previously mentioned defining them as “momentary” – but they leave a trace that can be tracked so that individuals can judge and be judged through them. In this case we can see how people who share the same language cannot necessarily understand each other, since they do not share the same code. One example of this would be the case of a hypothetical teenager, used to employing the typical abbreviations of his or her generation, who cannot communicate with his or her mother. Again, it is orality, the common channel of communication, which finds obstacles in a written format. In another example, even if a teenager uses abbreviations in social media, it will still be inappropriate to use them in an academic work. A similar situation arises if abbreviations are used in an official advertisement, for example, in newspapers. In short, codes and abbreviations depend on context and format even today, and not everybody is able to understand all of them. What arises from these examples is that the recipient is a key factor, and we can assume that the same holds true with an epigraph. In antiquity, the way of reading out loud, coupled with the standardized code of inscriptions, permitted the content of an epigraph to be spread, even if sometimes the way of writing this information was shaped by local formulae and other “mistakes”, none of which would have been considered such by locals. Let us focus now on what is read in a funerary epigraph.

CIL XI, *34: C(aius) Caesar / dict(ator), / Rubicone / superato, / ciuili bel(lo) / commilit(ones) / suos hic / in foro Ar(imini) / adlocutut(us est). Gravestone cippus. Piazza Tre Martiri a Rimini (Italy). 20 Ravara 2006, 68–79. 19

74

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

Fig. 4.1: Epitaph of Atellia Cleunica. Museo Arqueológico Municipal de Cartagena (Spain). Photo © M. Cristina de la Escosura.

What is read? After determining who reads, the traditional epigraphic analysis – philologically biased – turns its attention to what is read. As mentioned above, what an ancient observer scrutinizes in an epigraph has little, or nothing, to do with literary Latin, except perhaps in the case of carmina, which follow rhythmic structures. The majority of Roman inscriptions are characterized by an abundance of abbreviations and formulae. The former allow a large number of people to identify the sense of the text within the epigraph. Observers would have been used to seeing the same abbreviations with the very same uses. However, despite Latin epigraphy manuals containing long lists of abbreviations, this tool is not homogeneous throughout time and space. Abbreviations reflect epigraphic trends or even lines of evolution in Latin.21 We can pick two inscriptions from Carthago Noua (Cartagena, Spain), in Hispania Citerior, as examples. This city was inhabited by a great many Italic immigrants, many of whom were strongly familiar with Roman high culture,22 and Latin archaisms can be witnessed in several of its gravestones. Such is the case of the epitaph of Atellia Cleunica, dated to the republican period, namely 70–30 BC (Fig. 4.1).23 It is a local stone decorated as a tabula ansata. The text is very brief and formulaic, with scant information: the deceased’s name, Atellia Cn. l. Cleunica, and the funerary formula, Pocetti 2017, 8. De la Escosura Balbás 2021, 38–48. 23 CIL I2 2272 (= CIL II 3451 = HD037681 = HEpOl 9538): Atellia / Cn(aei) l(iberta) Cleunica / heic sitast. 70–30 BC. 21 22

4.  Written to be (un)read, written to be seen

75

Fig. 4.2: Epitaph of Cn. Atellius Toloco. Museo Arqueológico Municipal de Cartagena (Spain). Photo © M. Cristina de la Escosura.

heic sitast. From the same city comes the epitaph of the freeman Cn. Atellius Toloco (Fig. 4.2),24 whose text, from the first half of the first century AD, has the same concept abbreviated: h(ic) s(itus) e(st). This example is especially interesting because both came from the same gens but from different periods. Therefore, we can appreciate the development of abbreviations within the same group.25 However, the most common sequences of abbreviations in Roman epitaphs are even more complex, making the knowledge of the epigraphic code even more necessary. How else could a sequence such as ANNXXHSESTTL, ann(orum) XX h(ic) s(itus) e(st) s(it) t(ibi) t(erra) l(euis), be decoded? The abbreviation comes from the funerary stela of Pompeia Nereis, also from Carthago Noua (Fig. 4.3).26 The sequence 24 CIL II 3450 (= HD037605 = HEpOl 09537): Cn(aeus) Atellius Cn(aei) (libertus) / Toloco h(ic) s(itus) e(st). First half of the first century AD. 25 There was onomastic transmission in the gens Atellia of Carthago Noua (De la Escosura Balbás 2021, 95–6). Koch 1988, 403 defends the idea that the nine Atellii were members of the same family, but that is unlikely. 26 ILER 6250 (= HD037787 = HEpOl 14115): Pompeia Ne/ṛeis an(nnorum) XX / h(ic) ṣ(ita) e(st) s(it) t(ibi) t(erra) l(euis). Second half of the first century AD.

76

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

Fig. 4.3: Stela of Pompeia Nereis. Museo Arqueológico Municipal de Cartagena (Spain). Photo © M. Cristina de la Escosura.

HSESTTL and/or the abbreviation ann(orum) – instead of ann(is) – can be seen in several pieces from almost every western Roman province, but it is especially employed by hispanii (whether inside Hispania or when they die abroad).27 This is reflected on dispersion maps of the material in the Epigraphic Database Heidelberg.28 With respect to Hispania, we find this sequence, for instance, in the marble stela of Peregrina, from Hispalis, Seville, in the province of Baetica.29 It also appears in the case of the stela of L. Marcius Marcianus, a legionary from Tucci (Martos, Jaén, Baetica) who died in Carnuntum (Pannonia Superior).30 However, away from Hispania, we can also see it on the tombstones of the Papirii, a couple from Rome,31 or of the legionary Sex. Pilonius Modestus, from Beneuentum (Italy), who died in Dacia (Sarmizegetusa, Romania).32 The more complex the lists of capital letters forming the abbreviations become, the more essential it is to know the code. Such is the case of the funerary inscription of Alexander, son of Q. Canuleius Alexander and Clarina, from Rome, dated between AD 31 and 150 (Fig. 4.4).33 Here we

De la Escosura Balbás 2019. Text–DB: Map View, Epigraphic Database Heidelberg, https://edh-www.adw.uni-heidelberg.de/ inschrift/karte?hd_nr=&land=&fo_antik=&fo_modern_fundstelle=&literatur=&dat_jahr_a=&dat_jahr_ e=&hist_periode=&atext1=sit+tibi+terra+levis&bool=AND&atext2=&sort=hd_nr&anzahl=20 (last accessed 10.1.2020). 29 CILA II.1 78 (= HEpOl 4592): Peregrina / h(ic) s(ita) e(st) s(it) t(ibi) t(erra) l(euis). 30 BC–AD 30. 30 AE 1929, 189 (= HD024613): L(ucius) Marcius L(uci) / Ser(gia) Marcian/us Tucc(is) mil(es) leg(ionis) X ge(minae) / (centuria) Iusti ann(orum) XXXV aer(um) / XI hic s(itus) e(st) s(it) t(ibi) t(erra) l(euis) / frater ex uolun/tate sua f(aciendum) c(urauit). AD 63–68. 31 AE 2014, 190 (= EDR145041): D(is) M(anibus) / Papiria Eutychia / Cn(aeo) Papirio Isochryso / marito carissimo sibi / et dulcissimo fecit / H(ic) s(itus) e(st), s(it) t(ibi) t(erra) l(euis). Marble tabula. Second century AD. 32 CIL III 1480 (= ILS 2654 = HD047677): Sex(tus) Pilonius / Sex(ti) f(ilius) Ste(llatina) Mode/stus Beneuento / (centurio) leg(ionis) IIII F(lauiae) F(elicis) III hast(atus) / post(erior) ann(orum) XXXVII or/dine(m) accepit ex / equite Romano / militauit in leg(ione) / VII C(laudia) p(ia) f(ideli) et VIII Aug(usta) / XI C(laudia) p(ia) f(ideli) I Miner(uia) p(ia) f(ideli) / stipendi(i)s centurio/nicis XVIIII / h(ic) s(itus) e(st) s(it) t(ibi) t(erra) l(euis). Local stone tabula. AD 107–200. 33 AE 1990, 87 (= EDR081618 = HD015876): D(is) M(anibus). / Alexander uixit annis I̅I̅I̅, / mensibus IV, diebus XIIX; / Q(uintus) Canulêius Alexander / pater et Clarina / mater fil̂io carissimo / pientissimo bene m(erenti) fecer(unt) / H(ic) c(onditus) e(st); t(e) r(ogo) p(raeteriens) d(icas): s(it) t(ibi) t(erra) l(euis). Marble. Tabula mensa podiale. AD 31–150. 27

28

4.  Written to be (un)read, written to be seen

77

Fig. 4.4: Tabula mensa podiale of Alexander. Musei Capitolini (Rome, Italy). Photo © M. Cristina de la Escosura.

can read HCETRPDSTTL: h(ic) c(onditus) e(st); t(e) r(ogo) p(raeteriens) d(icas): s(it) t(ibi) t(erra) l(euis). Abbreviations are developed according to the writing system’s conventions, changes in the writing medium, or the means of communication employed. When the contexts and functions of written communication change, so do abbreviations or their meaning. This takes place because the existence of these codes presupposes not only a reliance on basic linguistic competences, but also belonging to a circle of users with a common baseline of communication.34 Indeed, abbreviations go beyond even this because they mix codes and sub-codes. For example, Classical Latin shows certain abbreviations which are still present in the Middle Ages and Renaissance. In some circumstances, the code remains unchanged, as in the case of PONT.MAX, generally accompanying popes’ names. An ancient Roman would be able to read the abbreviation, pont(ifex) max(imus), since he or she would be familiar with it from his or her own time. However, the sub-code would have changed, and consequently the message transmitted. We find the abbreviation in the tomb of Pope Gregory XIII (1572–1585), responsible for the famous calendar 34

Pocetti 2017, 11.

78

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

change, at the Basilica of St Peter in the Vatican. His tomb was decorated in 1723 by Camillo Rusconi, and it clearly shows his name followed by PONT.MAX, a very similar sequence to that in Roman imperial titles. One of the hundreds of examples is a Hadrianic epistyle from Cingulum (Cingolo, Macerata, Italy).35 However, the pontifex maximus is no longer the same religious Roman magistracy, since this use was abandoned by the emperor Gratian in AD 381. It would eventually be adopted by bishops of Rome in the fifteenth century.36 The lack of knowledge about the code on the part of people who know the language has led to the discovery of some important attempts at forgery, such as the graffiti on ceramics from the site of Iruña–Veleia, in the north of Spain. The Latin ostraca37 we refer to were initially dated to the mid-third century AD, but declared a forgery in 2008 by a committee of specialists, who determined that the materials were ancient, but not the graffiti.38 One of the ostraca shows a scene of Calvary, with the inscription r(equiescat) i(n) p(acem) on Christ’s cross. Though this expression is employed on Christian tombstones from the fourth century AD,39 its use is not widespread until the eighteenth century. However, under no circumstances does it make sense on Christ’s cross, where I(esus) N(azarenus) R(ex) I(udaeorum) is to be expected, after the Gospel of St John,40 a phrase which also makes more sense theologically, since, according to the Christian faith, Christ was resurrected and did not “rest in peace”.41 The forger shows a knowledge of the language and the Latin codes of his or her time, but is not aware of the Latin codes of the third century AD, leading to a wrong use of abbreviations. The forgery is thus quite obvious due to the ignorance of the written code, as well as the chronology associated with the image.42 To conclude, when Romans read inscriptions, they needed to decode a great deal of abbreviations and formulae to be able to understand them. However, these sequences of capital letters changed their meaning with the passage of time. Knowing the code also means putting it in the right chronological context, that is, to comprehend the sub-code. What is read is deeply linked to what is understood. Even if the Latin 35 CIL IX 05681 (= AE 1998, 420 = EDR015004 = HD032620): [I]m[p(erator)] Caesar d[iui Traiani Parth(ici) f(ilius)] diui Nervae nepos Tr[aianus Hadrianus Aug(ustus)], pont(ifex) max(imus), trib(unicia) pọ[t(estate) ---]I̲ c̲o̲(n)s̲(ul) [---], aquae ductum ue[tust]a̲t̲e̲ c̲o̲n̲l̲a̲[psum] / pecunia su[a] C̲i̲n̲g̲u̲l̲a̲[nis] rest[ituit]. AD 117–138. 36 Dijkstra and Van Espelo 2017, 319–25. 37 Even though the press and the academy had been calling it “graffiti”, it is more accurate to talk of ostraca, as pointed out by Gorrochategui 2011, 251. They are ceramic fragments reused to make inscriptions instead of ceramic vases inscribed when the piece was complete. Ostraca are extremely rare in the Roman West, unlike graffiti. 38 Gorrochategi 2011, 242–5. 39 E.g., ICVR II 5957 (= EDB17244): hic es⌜t⌝ posita / Paula qu⌜ae⌝ ui/xit ann(is) III / requiescet / in pacem. Marble tabula. Fourth century AD. Rome (Italy). See more in Gorrochategui 2012, 54–5. 40 John 19:19. 41 Gorrochategui 2012, 53. 42 The first representations are from the fifth century AD. Such is the case of the Ivory Casket from Rome (Italy), but preserved at the British Museum (Inv. no. 1856,0623.5). It is dated between AD 420 and 430.

4.  Written to be (un)read, written to be seen

79

abbreviation is the same, the reader needs to be familiar with the code. That is connected with how it was read.

How is it read? This question suits our approach better if we reformulate it to ask “what is seen?”. Abbreviations and formulae help the text to become images. We, as modern native speakers of a language that uses the Latin alphabet, observe a Latin inscription and can see letters in it. We are able to do so not only because we share an alphabet with the Romans; we are also part of a literate society, able not only to identify letters but also to read them even if we do not understand the meaning of the words they compose. As we saw above, this scenario did not totally apply for ancient Romans. For them, the carved letters of inscriptions were also images, or at least could be, depending on their ability to read. Whether they were able to identify concepts, read letters, or understand words, depended on their literacy. The best example is the most common Roman funerary formula between the mid-first and the third century AD, namely the imprecation of the Manes.43 We Fig. 4.5: Stela of Boderus. Museo can trace how its abbreviation, DM (Diis Manibus) or de León (Spain). Photo © David Serrano Lozano. DMS (Diis Manibus Sacrum), evolved from being part of the epigraphic text to become part of decoration. This does not imply a linear process through which a non-abbreviated expression is always chronologically followed by a text abbreviation. Since DM is such a widespread formula throughout the western Empire, each region shows its own features and timings. This does not affect the fact that DM in every province is better perceived as an image than as a text. There are hundreds of examples of this. However, we choose to illustrate our point with the epitaph of the slave Naidi, where the formula is entirely written out, without abbreviations: Dis Manibus / Naidi Phainus / contubernali / bene meritae / fecit.44 The ordinatio of the inscription is symmetric and the capital letters are carved to occupy all of the epigraphic space. The mention of the Manes seems to be like the rest of the words. It is a dis manibus formula working as a text. Another Naidi from Rome, also dated in the second half of the second century AD,45 presents in Tantimonaco 2017, 82–7. AE 1985, 134 (= EDR079615 = HD002364). Second half of the first century AD. Marble tabula. Rome (Italy). 45 CIL VI 10141 (= ILS 5261 = EDR107497): Dis Man(ibus) Naidi Caesaris uernae, / ex numero Pyrriĉhe u(ixit) a(nnis) XXV, m(ensibus) II, d(ie) uno. / Onesimus Caeseris n(ostri) coniugi b(ene) m(erenti) / fecit et sibi et suis posterisq(ue) eorum. Marble tabula. Second half of the first century AD. Rome (Italy). 43 44

80

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

her epitaph a slightly abbreviated formula, Dis Man(ibus), although there is no need of abbreviation, since almost half of the epigraphic area is empty. A third Naidi is recorded in Rome.46 In this case, the formula is fully abbreviated, D(is) M(anibus). The two letters are central and symmetric in the piece’s ordinatio, carved almost as a heading, but inside the epigraphic space. We can find this kind of example in every province of the western empire, on tombstones with a wide range of onomastics and juridical status. In the second century AD, it becomes more frequent to use the abbreviation DM as part of the decoration, that is, outside the epigraphic area. The funerary stela made by Flavia Euhodia to her husband, M. Tessius Lucrio, depicts a man leaning back in a lectus triclinaris. Between the frontal legs are carved the D, near the left leg, and the M, near the right one. Thus, the mention of the Manes is integrated into the decoration. Fig. 4.6: Stela of M. Caecilius Flavinus and M. The DM as the main iconographic element of Caecilius Flavus. Museo de Navarra (Pamplona, inscriptions is a common phenomenon. In Spain). Photo © M. Cristina de la Escosura. the stela of Boderus (Fig. 4.5), for example, from north-western Spain, the DM is inside a torque, surrounded by a simplified tree in each side.47 The text is inscribed below, independent of the iconography. In the double epitaph of M. Caecilius Flauinus and M. Caecilius Flauus (Fig. 4.6),48 the text has its own epigraphic area delimited by a tabula ansata carved in the middle of the local stone stela. In the space above the tabula there is a DM with a crescent moon between both characters. The same iconographic topic is repeated below the text. Even more complex is the iconography in Annia Buturra’s epitaph (Fig. 4.7).49 The epigraphic area occupies only a small portion in the middle of the stone, whereas the DM is inserted in the middle of a profuse decoration, on top of 46 EDR005356: D(is) M(anibus). / Naidi fil(iae) / Lollius Florus / pater / u(ixit) m(ensibus) VIII. Marble stela. Late first or early second century AD. Rome (Italy). 47 HEpOl 12031: D(is) M(anibus) / Bodero / Bodiues(cum) / Doideri f(ilio) / a(nnorum) XXV / filio suo / m(ater) p(osuit) h(ic) s(itus) / est. Local stone stela. Second half of the second century AD. Cistierna (León, Spain). 48 IRMN 53 (= HEpOl 19182): D(is) M(anibus) / Ma(rco) Cae(cilio) Flaui/no an(norum) LX et / Ma(rco) Cae(cilio) Flauo / an(norum) XXXV Doite/na Ambati Cel/ti f(ilia) soc(ero) et ma/rito f(aciendum) c(urauit). Local stone stela. Marañón (Navarra, Spain). 49 CIL II 2970 (= IRMN 42 = HEpOl 8804): D(is) M(anibus) / An(n)i(a) Buturra / Viriati filia / an(norum) XXX h(ic) s(ita). Local stone stela. Gastiain (Navarra, Spain).

4.  Written to be (un)read, written to be seen

Fig. 4.7: Stela of Annia Buturra. Museo de Navarra (Pamplona, Spain). Photo © M. Cristina de la Escosura.

81

Fig. 4.8: Stela of Λ(ουκίῳ) Αἰλίῳ Μελιτίνῳ. Musei Capitolini (Rome, Italy). Photo © M. Cristina de la Escosura.

the encircled female figure.50 We can determine in these four cases that DM is working as an image. Even when the written concept is translated into Greek inscriptions, the characters making reference to the Manes become part of the decoration.51 Such is the case in the marble stela of Λ. Αἰλίος Μελιτίνος (Fig. 4.8), from Rome, where the DM concept was literally interpreted to become Θ(εοῖς) Δ(αίμοσιν).52 Centuries later, we witness the same phenomenon of text turned into images in Arabic carvings on Christian medieval monuments. During the Middle Ages in the Iberian Peninsula, Christian kingdoms and Muslim realms struggled for territorial control. Nevertheless, beyond the fighting there were also chances for peaceable interactions and co-existence (though not without problems) between Jews, Christians, and Muslims, regardless of the realm’s official religion. Muslims who See Marco Simón 1979 for a detailed explanation of the decoration. For the use of the formula in Greek texts, see Tantimonaco 2017, 79–82. 52 CIL VI 10736 (= IG XIV 1337 = IGUR II 291 = EDR110969 = PacHum 187925): Θ(εοῖς) Δ(αίμοσιν). / Λ(ουκίῳ) Αἰλίῳ Μελιτίνῳ / τέκνῳ γλυκυτάτῳ / Φηλῖκλα μήτηρ καὶ Μύ/ρων πατὴρ ἀτυχέστα/τοι ἐπόησαν. Ἔζησεν / μησὶν δέκα τρισίν, / ἡμέραις ἐννέα. Μὴ / ἐνοχλήσῃς τῷ τάφῳ / μὴ τοιαῦτα πάθῃς / περὶ τέκνων. Ne sis / molestus, ne patiaris hoc /et ollas inclusas caue. Marble stela. Rome (Italy). 50 51

82

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

lived in Christian cities, the so-called Mudejars,53 were celebrated as elite builders. Therefore, Christian kings and cities did not hesitate to make use of their skills in their most prestigious civil and religious buildings. This led to paradoxes that can be observed in a number of religious monuments in Castile, in the central Iberian Peninsula.54 Such is the case, for instance, at the Abbey of Santa Maria La Real of Las Huelgas (Burgos, Spain). This important place holds the remains of dozens of Christian kings and was the site of both coronations and ennoblements during most of the Middle Ages. Alfonso VIII, king of Castile, and Leonor Plantagenet founded it in 1087 as a monastery for Cistercian nuns.55 Destined to be the burial place for royal families, the abbey was granted many privileges. Among these was the service of some Muslim craftspeople who contributed to its construction along with other Muslims from the nearby city of Burgos.56 The royal abbey thus has numerous Arabic inscriptions in Kufic in its cloister and small external chapels, praising Allah in different ways. This was possible because Christian individuals looking at these inscriptions could only see decoration in them: they did not know the linguistic code necessary for distinguishing the Muslim verses inside the Christian temple. Although there were cases in which Christian verses were introduced in this kind of Arabic iconography, they were very rare in the Iberian Peninsula, while Arabic inscriptions were common in Christian buildings. Exactly the same phenomenon is experienced by present-day visitors to Muslim buildings such as the Alhambra, where the great display of Arabic epigraphy on its walls, both religious and poetic, will be seen by most simply as decoration.57 Nowadays, our own version of text as image is increasingly present in our daily life: emoticons. Although many users are no longer aware of it, several of the images we currently use in social networks or texting were originally combinations of already existing letters and punctuation marks. This was a primary method for abbreviating textual length in digital communication, before the present availability of images, gifs, and emojis. Each language in fact had its own variations, while the use of images has unified them nowadays. However, the use of xD / XD58 is especially interesting since it is one of the very few original emoticons which is still employed both as letters and as an image. The concept is also used as a visual joke: “Why are 490 Romans funny? Because XD.” In Spanish, speakers even go beyond this and today write the name of the two letters, equisdé (ex dee), to express the concept. It is used in comedy radio programmes, YouTube channels, restaurant names, memes, and elsewhere. It is the 53 N.B. the use of “Mudejar” as an artistic style is currently being debated since it is the result of a nineteenth-century nationalist trend. See García Nistal 2009. 54 A detailed list of examples with some photos can be found in Pavón Maldonado 2009, 512–16. It also contains the rare examples of Christian words in Arabic inscriptions 55 Carrero Santamaría 2004. 56 See Villanueva Zubizarreta and Araus Ballesteros 2016. 57 Puerta Vílchez 2015. 58 “Grinning Squinting Face,” Emojipedia, https://emojipedia.org/grinning-squinting-face/ (last accessed 10.1.2020).

4.  Written to be (un)read, written to be seen

83

reconversion of an image into text, but again only Spanish speakers have the code to understand it, while the XD could be identified by everyone in any language. In sum, Romans did not need to be literate to understand the capital information contained in an epigraph. When they looked at funerary inscriptions, they saw images with the form of letters. Abbreviations and formulae help the text to become images. This is the case for the Manes’ imprecation, the praenomen, the filiation, the identification of the inscription as an epitaph (HSE), the age of the deceased, the measures, and so on. In the same way in which identifying letters does not mean understanding their significance, comprehending images does not indicate that you are able to individualize the letters that form them. With this established, we move on to the next question. To perceive every bit of information from an inscription, we need to consider not only what is seen but also where.

Where is it read? In his classic work, Ramsay MacMullen defined the background cause for the rise and fall of the epigraphic habit as “the sense of audience”.59 This term referred to people’s expectation that “their world still continuing in existence for a long time to come”. Nevertheless, it is notable how pioneering this point is in placing the driving force of the epigraphic habit on viewers.60 For a long time, epigraphy has been thoroughly analyzed for its (textual) content, philological elements, dispersion, or socio-political implications. However, the role of viewers, understood in a wide sense as the process of observing, understanding, and interpreting epigraphy in ancient times, still remains largely uncharted territory.61 Access to the content of inscriptions required both the ability to understand the linguistic code of Latin (i.e., the modern literary conception of reading) and for “the reader to supply a good deal of knowledge” 62 to understand its formulaic nature. Due to the written and monumental nature of epigraphy, 63 our approach to the original perception of inscriptions must be more complex than the reading of an array of texts. As a material record with a notable visual component, epigraphy conveyed a complex content in which “images and texts worked together to communicate with viewers”. 64 The specific weight of internal and external elements could obviously vary for each inscription. However, even in those pieces where the text was apparently the central part of the message (such as plaques, or honorific inscriptions), there is an additional layer of meaning which contributes to its configuration. From the largest scale (context of display, MacMullen 1982, 246. Keegan 2013, 55. 61 Esmonde Cleary 2013, 217. 62 Woolf 2009, 53. 63 Woolf 1996, 24. 64 Rose 2003, 155. 59 60

84

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

place of exhibition) to the smallest one (materials, format, decoration) there were non-random external components that contributed to the construction of each inscription’s performance65 within its environment.66 As Greg Woolf has written, “competence with sign systems involves complex skills and sensitivity to context”.67 Consequently, the total or partial lack of evidence about the original context of display for most inscriptions is undoubtedly one of the greatest obstacles in epigraphic research. However, this absence often leads us – unnecessarily – to forget the fact that they had an original context of display at all. In recent years, there has been much interesting work analysing certain types of inscriptions in their original locations, such as tombstones alongside roads at the entrance of cities or plaques in columbaria.68 However, we are still unaware of most of (if not all) the specifics about the visuality of, for example, inscriptions displayed in urban environments outside fora,69 rural epigraphy,70 non-funerary inscriptions next to roads,71 or the manifold possibilities of exhibition within different types of sanctuaries.72 Most original epigraphic contexts are virtually impossible to reconstruct. However, this must not be a hindrance to considering that they determined several cognitive and symbolic processes implicit in what we consider to be the first stages of the perception of any epigraph for an ancient viewer. This refers not only to the very physical context of display of a certain piece. It also concerns the epigraphic contextual environment defined by the different inscriptions produced in a region, area, or place, which follows a certain common cultural frame (fashion, imitation, ritual, etc.). This corresponds with the idea of the “epigraphic landscape”, as conceived by Giancarlo Susini, and employed for decades.73 The network of interrelationships among inscriptions in a particular epigraphic landscape is fundamental. This factor would delimit the capacity of a hypothetical ancient viewer to apprehend an initial set of information from a piece, before any possible readings even took place. A particular format, display position, or external feature could be crucial for an illiterate viewer to understand, for instance, the funerary content of an epitaph or the attribution of a dedication to a specific divinity.74 Possibly conditioned by having been born and raised in an industrial world, we tend to equalize and standardize via the use of categories such as “decorated epitaph”. Thus, we have a tendency to equate “functionally analogous” with “semantically Revell 2000, 1. Esmonde Cleary 2013, 230. 67 Woolf 2015, 37. 68 Esmonde Cleary 2013; Keegan 2013. 69 Corbier 2013. 70 Häussler 2014. 71 Marco Simón 2007. 72 McCullough 2018. 73 Susini 1982, 15, 47; Cooley 2000, 1; Ruíz Gutiérrez 2013. 74 Barrett 1993, 237; Keegan 2013, 49. 65 66

4.  Written to be (un)read, written to be seen

85

Fig. 4.9: Sample of four different funerary inscriptions with analogous textual structures. 1: HEp 1994.353 = HEpOl 13910 (Santa Comba, La Coruña, Spain; now in Museo Arqueológico de La Coruña, Spain). 2: HEp 1996.780 = HEpOl 13968 (Vigo, Pontevedra, Spain; now in Museo Arqueológico de Asturias, Spain). 3: AE 1975, 507 = HEp 1994.336 = HEpOl 13917 (Oza de los Ríos, Mazarelas, La Coruña, Spain; now in Museo Arqueológico de La Coruña, Spain). 4: CIL II 5705 (Acebedo, León, Spain; now in Museo Municipal de Vigo “Quiñones de León”, Spain). Photos © David Serrano.

equivalent”. The key to deconstructing this principle is to contemplate that different “decorated epitaphs” could entail markedly different meanings, depending on their differing uses of contextual and external elements. The text would merely be the central axis that defined specific content (such as onomastics), but the external elements, together with the geographical, chronological, contextual, or cultural components, would be spinning around it, providing the semiotic structure that shaped most of its message.75 Let us picture a certain set of four epitaphs (Fig. 4.9). They come from four different places within the Roman Empire, but all of which nevertheless lie within a not-tooextended geographic area, namely north-western Hispania (Fig. 4.10). Likewise, the four inscriptions show analogous text in terms of content, since they display the most typical components of information in epitaphs: formulae (e.g., DMS, HSE, STTL, FC), onomastics, origo, age, relatives, and/or dedicants. Nevertheless, each piece shows 75

Esmonde Cleary 2013, 227.

86

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

Fig. 4.10: Distribution map of the find spots of the four inscriptions in Fig. 4.9.

radically different external features, including their size, proportions, decorations, or composition of their texts. Let us imagine a hypothetical observer from the third century AD who had the chance to see these four inscriptions in situ without being familiar beforehand with their geographical and cultural contexts. The experience of this observer would entail conditions and features separating it from our perspective as observers of the archaeological record, namely covering distances to visit each one in turn, keeping only a personal memory of each piece (never a single common picture), maintaining a perception of the original context of display and of different states of conservation. It also goes without saying that the background of this individual would lead to a rather different relationship with the epigraphic habit from that of a researcher today. All this considered, to what extent could this viewer extract all the signifiers implicit in each piece, and the differences among them? The individual scale and position of this observer would provide additional (and partial) information to decode an inscription only in certain cases. The first inscription comes from a site some miles away from the capital city of a conuentus (Lucus Augusti), and its outstanding size (almost 3 m tall) and antiquity by the third century could transmit a perception rather different than the one coming from the current state of the record: a unicum in the whole north-western area.76 76

González García and López Barja 2011.

4.  Written to be (un)read, written to be seen

87

For their part, inscriptions number two and three would provide radically different contextual information. Number two belongs to the epigraphy of the Ciuitas Vadiniensis, a region in northern Spain characterized by a widely dispersed record in a completely rural region, where there is not a single site with a concentration of inscriptions and there is a notable consistency in the use of a distinctive aesthetic (horses, torques, palms) and material (large rolling stones).77 In contrast, number three is part of the epigraphic record from the Roman necropolis of Vigo (Pontevedra, Spain), where the assemblage of 30 pieces discovered in 1953 shows a remarkable consistency in the external elements of the epitaphs (mainly rosettes, arches, crescents, and human figures).78 So, each piece would be framed in radically different immediate contexts for an ancient observer. Number two would certainly show a peculiar and striking decoration, but its location in a place without spatial relationships with other inscriptions would force a viewer to confront a lack of comparison or context within its epigraphic landscape. By contrast, the piece from Vigo could have been interpreted via the context provided by a set of inscriptions with varied but consistent typologies, so that our observer could have understood it as belonging to a kind of epitaph typical of the city. Finally, inscription number four would offer the most challenging contextual conditions. It belongs to a group of pieces identified as a rural group in the area of Santa Comba (Galicia, Spain), characterized by wide dispersion and typological variety, with the only common element being prominent top parts in funerary inscriptions.79 So, a mid-size thin piece, with a rather unique design and located in a rural place with scant or no relationships with other inscriptions in the immediate area would have strongly depended upon its text to be apprehended by a foreign observer. The contrast among these four hypothetical situations shows how the first set of signifiers of many inscriptions would not be placed in their texts, but in all the non-textual components surrounding and framing them.80 Independent of an observer’s efficiency in reading the textual content, familiarity with the symbolic codes associated with the context of display would be necessary for the reader to “make sense of that inscription by relating it to what they and others know”, since “different expectations will give the inscription different significances”.81 In the present day, we are in an analogous, or perhaps even more complex situation. Due to the historical distance and lack of knowledge about the original symbolic and visual code, we are another observer with different expectations (academic and so mostly analytical) from the ones of an ancient audience. How often is such a facet considered when analyzing a certain inscription or an ensemble of inscriptions? Of course, the location of inscriptions in secondary Martino García 2012. González García 2010. 79 Serrano Lozano 2017, 164–6. 80 Laurence and Sears 2013, 1–2. 81 Barrett 1993, 237. 77 78

88

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

contexts make this kind of consideration notably difficult, if not virtually impossible. Nevertheless, every epigraphic analysis should be aware of the potential loss of information implied in the lack of context. Otherwise, the tendency is to assume a tacit but untrue sense of having attained a definitive interpretation of a piece or a set of pieces based on its text and external elements. Likewise, how often is an epigraph analyzed, mainly or exclusively, based on its text? Furthermore, how often are external elements taken into account only when they are particularly remarkable or show an evident relationship with the written text? If we work in this way, we are likely to be proceeding in exactly the opposite direction to the ancient perception of epigraphy. Cultural and spatial contexts would have been the semantic frames within which the specific content of epigraphs would be placed in the past, both individually and as sets. This was even more so in the numerous cases in which texts did not provide an especially profuse amount of information,82 both due to the use of formulae, as we have seen, and to the scarcity of individualizing elements (e.g., onomastics, origines, theonyms). Moving from the surrounding context to the very pieces themselves, there is an obvious second level of non-textual reading that comes from external elements to the text, such as formats, materials, or decorations. These are to be considered as nonrandom self-speaking components. Every external element (size, format, proportions, image–text ratio) would belong to a web of intertwined pieces of information composing a resultant complex message for each inscription, or inscriptions, in any epigraphic landscape. In consequence, the semantics of external epigraphic elements merit more consideration than their mere inclusion as part of a systematic typological record. This notion must be brought into consideration if we are to really approach the processes of original epigraphic perception, such as visibility, spaces of display, or interrelationships among inscriptions in the eyes of viewers. As a discipline, epigraphy has progressed to some extent in the analysis and comprehension of the implicit and explicit meanings in the external elements of inscriptions. Nonetheless, the dominant tendency is still to set fixed rules rather than to regard the textual and non-textual information as two sets of mutually interacting meanings. Thus, a tacit univocal framework is developed, which tends to link certain typological features with specific kinds of text, so that even illegible texts can be categorized according to the shape or form of the inscription itself. The risk of this approach is that we consider it as an immutable quality of the categorization of inscriptions, so that when pieces do not reflect this pattern, we can even fail to identify any variation at all. Such could be the case, for instance, with the phenomenon of epitaphs carved in forms that reflect the typical altar shape, or funerary altars. This kind of piece has never been specifically studied, as far as we are aware, and so perhaps we are missing aspects such as its possible contrast with regular epitaphs or its potential chronological connotations.83 82 83

Woolf 2009, 53. Sadurska 1959, 73; Laubry 2009, 137.

4.  Written to be (un)read, written to be seen

89

Fig. 4.11: Monument of Piedra Escrita (Cenicientos, Madrid, Spain). Sources: https://es.wikiloc.com/; Canto 1994, 277.

An illustrative example of the potential risks of ignoring the interpretative weight of epigraphic external elements can be found in the so-called monument of Piedra Escrita (Spanish for “Scripted Stone”).84 Located 65 km south-west of Madrid, this is a 7 m high and 9 m in diameter granite monolith. In the centre of the rock, the image of three human figures wearing togas and disposed around an altar are carved inside an alcove, together with an ox and, possibly, a goat below. Next to this group, there are two remaining lines of text corresponding to an individual consecration to Diana in the context of a rural sanctuary (Fig. 4.11). At some point in the medieval or modern period, the three figures were reinterpreted as three women and, consequently, the inscription was re-carved to read Las tres Marías (Spanish for “The Three Marys”). Ignorance about the meaning of the external elements led to this specific re-interpretation, understandable in a strongly Catholic cultural context.85 Apart from the obvious chronological and cultural gap, there is no procedural difference between the misconceptions carried out during the re-carving of the monument and the ones that we, as researchers, might currently perform in epigraphic analysis. The assumption of predominant templates and (alleged) standards in epigraphic external elements can make us miss the potential content of possible variations in the use of these external components as layers of meaning. Local uses, provincial understandings, and regional standards or re-interpretations could stay off our radar if we are to univocally associate specific external elements (formats, icons, scenes) with specific – and exclusive – contents or uses. Again, this procedure would not differ markedly from the fixed standard by HEp 1996.642 (= HEpOl 16199): A(nimo) l(ibens) s(oluit) Sisc(inius?) Q(---) / Dianae. Granitic monolith. Cenicientos (Madrid, Spain). 85 Canto 1994, 272, 276–9. 84

90

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

which it was thought that three human figures carved on stone could only represent the Three Marys. Another contemporary example can further illustrate our point here. Even in such a textual and hyper-textual world as ours, we often (un)consciously employ contextual and external sets of implicit meanings in the perception of inscriptions. Let us consider a well-known set of inscriptions like the Hollywood Walk of Fame. A simple gathering of (artistic) names and surnames, all carved with the same format and dimensions, with a monumental nature understandable by its size and location. All the complementary information involved is never explicitly displayed by the texts. It is the audience who provides the rest of this defining information which gives sense to the “monument”, by being aware not only of its intention and meaning, but also by being familiar with the person behind specific star(s). We could imagine, as a brainstorming exercise, the obstacles faced by a hypothetical researcher 2,000 years in the future, trying to decode the decontextualized remains of the Walk of Fame. The possibilities of establishing dominant interpretative templates would be manifold: e.g., similar external appearance imply similar status of the individuals, similar sizes entail similar economic levels. An analogous parallel could be identified in bench memorial plaques. In this case, textual information is variable and external components are complex and set up at two scales (plaque and bench), all of it totally framed by the context of display (local, professional, geographical). So, a hypothetical future record consisting exclusively of a gathering of plaques corresponding to a certain area or site would be equally prone to dominant premises when considering external elements: small plaques size points to limited visibility, or similar plaques suggest similar individual status. Returning to our four inscriptions from north-western Hispania, could we consider that these four pieces originally transmitted the same meaning because they share very similar texts? Of course, regional, cultural, or typological slants could be proposed for each of them, but it is not unreasonable to suggest that the message perceived by the audience in each case would not be totally equivalent. As Greg Woolf has stated, “many tombstones, for example, proclaim their subject before one word is read”.86 A broad majority of funerary inscriptions – the most numerous form of inscription in the epigraphic record87 – actually provide quite similar and, to some extent, succinct pieces of information. Consequently, it must be a necessary analytical condition that a great deal of information transmitted by these pieces to their original observers was channelled through their non-textual features. Otherwise, we would be assuming that thousands of pieces along centuries and three continents were essentially transmitting an equivalent or analogous message, but for specific individual data, despite the vast typological mosaic recorded. In consequence, epigraphic analysis should tend to follow an approach going from a semantic structure defined by a general context, context of display, format, and 86 87

Woolf 2009, 57. Chioffi 2015, 627.

4.  Written to be (un)read, written to be seen

91

external features, converging on a textual specific content. This way, perhaps we could start approaching inscriptions in a manner more similar to that of an ancient population that was mostly illiterate or functionally illiterate.

Conclusion Since the pioneering works of Morcelli, Mommsen, or Hübner, the way epigraphy, as a discipline, has approached inscriptions has been strongly influenced by a philological principle by which the text was regarded as the first and main source of historical information.88 This synecdochical conception, which understood the whole entity of an epigraph through its textual content, responds to a post-industrial literate perception rather than to an ancient approach towards exposed monumental inscriptions. This traditional vision, in which what is read is the focal point of analysis and any other aspect is subordinated, has – to a large extent – conditioned the recording, querying, and analyzing of the epigraphic record for roughly a century. As we have tried to demonstrate in this paper, the very idea of epigraphic texts being generally read in ancient times, as we currently understand the action of reading, is questionable.89 In a mostly illiterate society where both the social perception and performing technique of reading were rather different than ours, formulistic reading, the concept of text as image and the importance of implicit non-textual contents in epigraphy are aspects that must be put to the forefront. Even when inscriptions were fully read following our contemporary concept of this action, this very act would not have been ruled by the same patterns of writing, grammar, and display of ancient literary texts. Indeed, the text in an epigraph would be but one piece of information framed by a structure of unwritten signifiers. These would place the text within a specific context and meaning, culminating in the written information as the last – not the first – communicative step. Since written language would not have been its only channel for transmitting information, epigraphy was not dependent on orthodox forms of language in order to be comprehensible. Thus, epigraphic writing developed its own tools for re-elaborating the written text, though in relationship with the rest of written expressions of the same language, such as literature or cursive script.90 Therefore, we propose that, in order to attempt an approach to ancient inscriptions that mirrors the perceptive experience of an average observer in the ancient world, we should consider the opposite direction to the traditional one. That is, we should consider the text as the last stage in approaching the inscriptions, from the widest to the most specific scale. This implies going from the implicit non-written information to the textual content as the final step, not necessarily perceived under the contemporary patterns of reading. In so doing, we could begin to harmonize some Buonocore 2015. Williamson 1987, 166–70; Bodel 2015. 90 Woolf 2009, 61; Poccetti 2016, 11, 28–9. 88 89

92

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

apparently irreconcilable issues such as the great presence of epigraphy in a mostly illiterate society, the presence of less orthodox epigraphic uses, or the recognition of epigraphs as a speaking, performative material on its own.

Bibliography

Barrett, John C. 1993. “Chronologies of Remembrance: The Interpretation of Some Roman Inscriptions.” World Archaeology 25, no. 2 (October), Conceptions of Time and Ancient Society: 236–47. Beltrán Ortega, Alejandro. 2015. Epigrafía y Territorio: La Civitates de La Asturia Meridional y La Lusitania Nororiental. PhD thesis, Universidad Complutense de Madrid. Bodel, John. 2015. “Inscriptions and Literacy.” In The Oxford Handbook of Roman Epigraphy, edited by Christer Bruun and Jonathan Edmondson, 745–63. Oxford: Oxford University Press. Buonocore, Marco. 2015. “Epigraphic Research from its Inception: The Contribution of Manuscripts.” In The Oxford Handbook of Roman Epigraphy, edited by Christer Bruun and Jonathan Edmondson, 21–41. Oxford: Oxford University Press. Canto, Alicia. 1994. “La piedra escrita de Diana, en Cenicientos (Madrid), y la frontera oriental de Lusitania.” Cuadernos de Prehistoria y Arqueología 21: 271–96. Carrero Santamaría, Eduardo. 2004. “Observaciones sobre la topografía sacra y cementerial de Santa María la Real de Las Huelgas, en Burgos, y su materialización arquitectónica.” In La clausura femenina en España. Actas del simposium 14-IX-2004, vol. 2, edited by Francisco Javier Campos y Fernández de Sevilla, 695–716. San Lorenzo del Escorial: Ediciones Escurialenses (EDES). Castillo, Carmen, Joaquín Gómez–Pantoja and María Dolores Mauleón. 1981. Inscripciones romanas del Museo de Navarra. Pamplona: Diputación Foral de Navarra. Cavallo, Guglielmo. 1978. “Dal Segno Incompiuto Al Segno Negato.” Alfabetismo e cultura scritta 13, no. 38 (2): 466–87. Chioffi, Laura. 2015. “Death and Burial.” In The Oxford Handbook of Roman Epigraphy, edited by Christer Bruun and Jonathan Edmondson, 627–48. Oxford: Oxford University Press. Cooley, Alison. 2000. The Epigraphic Landscape of Roman Italy. London: Institute of Classical Studies, School of Advanced Study, University of London. Corbier, Mireille. 2013. “Writing in Roman Public Space.” In Written Space in the Latin West, 200 BC to 300 AD, edited by Gareth Sears, Peter Keegan and Ray Laurence, 13–47. London: Bloomsbury Academic. Dijkstra, Roald and Dorine Van Espelo. 2017. “Anchoring Pontifical Authority: A Reconsideration of the Papal Employment of the Title Pontifex Maximus.” Journal of Religious History 41 (3): 312–25. De la Escosura Balbás, María Cristina. 2019. “Documenting Hispanic Immigrants in Italia, Gallia, and Britannia.” In From Document to History: Epigraphic Insights into the Greco-Roman World, edited by Carlos F. Noreña and Nikolaos Papazarkadas, 326–55. Boston, Leiden: Brill. De la Escosura Balbás, María Cristina. 2021. La población de Carthago Noua de la conquista al Principado. Epigrafía y onomástica. Barcelona: Universitat de Barcelona Edicions. Esmonde Cleary, Simon. 2013. “The City as Preferred Written Space: The Case of Aquitania.” In Written Space in the Latin West, 200 BC to 300 AD, edited by Gareth Sears, Peter Keegan and Ray Laurence, 217–30. London: Bloomsbury Academic. Fernández–Martínez, Concepción. 2019. “¿Cómo se reconstruyen las inscripciones deterioradas?” In Siste viator, la epigrafía en la Antigua Roma, edited by Alvar Ezquerra, Antonio, 63–71. Alcalá de Henares: Universidad de Alcalá, Servicio de Publicaciones. García Nistal, Joaquín. 2009. “El mudéjar o la formulación romántica de un estilo medieval hispánico.” In Perspectives contemporaines sur le monde médiéval, edited by Laura Bădescu, 242–7. Pitesti: Editura TIPARG.

4.  Written to be (un)read, written to be seen

93

González Fernández, Julián. 1991. Corpus de inscripciones latinas de Andalucía. Volumen II Sevilla. Tomo I: La Vega (Hispalis). Sevilla: Consejería de Cultura y Medio Ambiente de la Junta de Andalucía. González García, Francisco Javier. 2010. “Hábito epigráfico, decoración plástica e interacción cultural en el noroeste hispano en época romana. Análisis de las estelas funerarias de Vigo (Pontevedra).” Madrider Mitteilungen 51: 397–418. González García, Francisco Javier and Pedro Manuel López Barja. 2011. “La estela de Crecedente: reflexiones sobre el proceso romanizador en la Galicia antigua.” In Arqueología, sociedad, territorio y paisaje: estudios sobre Prehistoria reciente, Protohistoria y transición al mundo romano en homenaje a Mª Dolores Fernández Posse, edited by Primitiva Bueno, Antonio Gilman, Concha Martín Morales, and F.-Javier Sánchez-Palencia, 349–60. Madrid: Consejo Superior de Investigaciones Científicas, Instituto de Historia. Gorrochategui Churruca, Joaquín. 2011. “Hic et nunc. Falsificaciones contemporáneas. El caso de Iruña–Veleia.” In El Monumento epigráfico en contextos secundarios: procesos de reutilización, interpretación y falsificación, edited by Joan Carbonell Manils, Helena Gimeno Pascual and José Luis Moralejo Álvarez, 241–61. Barcelona: Universitat Autònoma de Barcelona. Gorrochategui Churruca, Joaquín. 2012. “Las Armas de la Filología: la cuestión del Vasco Antiguo y los hallazgos de Iruña–Veleia.” In 2nd Conference of the Luis Michelena Chair = Koldo Mitxelena Katedraren II. Biltzarra = II. Congreso de la Cátedra Luis Michelena, edited by Joseba A. Lakarra, Joaquín Gorrochategui and Blanca Urgell, 41–70. Vitoria–Gasteiz: Universidad del País Vasco, Servicio Editorial. Harris, Willam V. 1989. Ancient Literacy. London, Cambridge, MA: Harvard University Press. Häussler, Ralph. 2014. “Differences in the epigraphic habit in the rural landscapes of Gallia Narbonensis.” In Öffentlichkeit – Monument – Text. XIV Congressus Internationalis Epigraphiae Graeca et Latinae, 27–31 Augusti MMXII. Akten (Corpus Inscriptionum Latinarum, Auctarium, series nova, XIV), edited by Werber Eck and Peter Funke, 323–45. Berlin: De Gruyter. Horsfall, Nicholas. 1989. “The Uses of Literacy and the Cena Trimalchionis: I.” Greece & Rome 36, no. 1 (April): 74–89. Keegan, Peter. 2013. “Reading epigraphic culture, writing funerary space in the Roman city.” In Written Space in the Latin West, 200 BC to 300 AD, edited by Gareth Sears, Peter Keegan and Ray Laurence, 49–64. London: Bloomsbury Academic. Laubry, Nicholas. 2009. “Les inscriptions fun´eraires des monuments lyonnais”. In Rites funéraires à Lugdunum, edited by Christian Godineau, 135–53. Paris: Errance. Laurence, Ray and Gareth Sears. 2013. “Written Space.” In Written Space in the Latin West, 200 BC to 300 AD, edited by Gareth Sears, Peter Keegan and Ray Laurence, 1–9. London: Bloomsbury Academic. MacMullen, Ramsay. 1982. “The Epigraphic Habit in the Roman Empire.” American Journal of Philology 103, no. 3 (Autumn): 233–46. Marco Simón, Francisco. 1979. “Las estelas decoradas de época romana en Navarra.” Trabajos de arqueología Navarra 1: 205–50. Marco Simón, Francisco. 2007. “Within the Confines of the Romano-Celtic world: The Gods of the Roads.” In Auf den Spuren keltischer Götterverehrung. Akten des 5. FERCAN. 5. –Workshop, Graz, 9–12 Oktober 2003, edited by Manfred Heinzmann, 197–205. Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Martino García, David. 2012. “Acerca de la cronología de la epigrafía Vadiniense. Revisión historiográfica y nuevas propuestas.” Espacio, Tiempo y Forma, Serie II, Historia Antigua 25: 305–26. Mayer i Olivé, Marc. 2019. “La cultura epigráfica como reflejo de un mundo alfabetizado.” In Siste viator, la epigrafía en la Antigua Roma, edited by Antonio Alvar Ezquerra, 15–25. Alcalá de Henares: Universidad de Alcalá, Servicio de Publicaciones.

94

M. Cristina de la Escosura Balbás, Elena Duce Pastor, and David Serrano Lozano

McCullough, Kimberly. 2018. Ritual Religion at Rock Sanctuaries in Roman Spain: An Alternative Model. PhD thesis, University of British Columbia. Mullen, Alex. 2019. “Was Latin epigraphy a killer?” Centre for the Study of Ancient Documents 23: 2–5. Pavón Maldonado, Basilio. 2009. Tratado de arquitectura hispanomusulmana. IV. Mezquitas. Madrid: Consejo Superior de Investigaciones Científicas. Poccetti, Paolo. 2016. “Abbreviare la Pietra. Prassi e percorsi nell’epigrafia antica. Tra lingua e scritura.” In Scriver veloce. Sistemi tachigrafici dall’antichità a Twitter. Atti del convegno, Rovereto, 22–24 maggio 2014, edited by Alessandro Tedesco, 7–39. Florence: Leo S. Olschki. Puerta Vílchez, José Miguel. 2015. Leer la Alhambra. Guía visual del Monumento a través de sus inscripciones. Granada: Patronato de la Alhambra y Generalife Edilux. Ravara, Cristina. 2006. “Il Cippo Riminese Di Giulio Cesare. Omaggio Ad Augusto Campana.” L’arco: quadrimestrale di attualità e cultura della Fondazione Cassa di risparmio di Rimini 1: 68–79. Revell, Louise. 2000. “The Creation of Multiple Identities in Roman Italica.” In TRAC 99: Proceedings of the Ninth Annual Theoretical Roman Archaeology Conference, edited by Garrick Fincham, Geoff Harrison, René Holland and Louise Revell, 1–7. Durham, Oxford: Oxbow Books. Rodríguez Corral, Javier. 2019. “Arming Landscapes: Connectivity and Resistance in Northwestern Iberia in Late Prehistory.” Journal of Anthropological Archaeology 53: 288–303. Rose, Fiona A. 2003. “Text and Image in Celtiberia: The Adoption and Adaptation of Written Language in Indigenous Visual Vocabulary.” Oxford Journal of Archaeology 22 (2): 155–75. Ruíz Gutiérrez, Alicia. 2013. “El paisaje epigráfico de la ciudad romana: concepto y perspectivas de studio.” In Paisajes epigráficos de la Hispania romana: monumentos, contextos, topografías, edited by José Manuel Iglesias Gil and Alicia Ruíz Gutiérrez, 13–27. Rome: L’Erma di Bretschneider. Sadurska, Anna. 1959. “Quelques remarques sur la datation des épitaphes romaines fondée sur la décoration en relief.” In Atti del terzo Congresso Internazionale di Epigrafia Greca e Latina, 71–76. Rome: L’Erma di Bretschneider. Serrano Lozano, David. 2017. “Sobre algunos conjuntos epigráficos rurales del interior de la Gallaecia romana: una propuesta de interpretación.” Conimbriga 56: 158–99. Susini, Giancarlo. 1982. Epigrafia Romana. Rome: Jouvence. Tantimonaco, Silvia. 2017. Dis Manibus. Il culto degli dei Dei Mani attraverso la documentazione epigrafica. Il caso di studio della Regio X Venetia et Histria. Getafe: Arys. Todisco, Elisabetta. 2013. “Esempi Di Alfabetismo Nella Campagna Romana in Età Imperiale (Italia e Province Occidentali).” Zeitschrift für Papyrologie und Epigraphik 187: 295–301. Tov, Emanuel. 1982. “Criteria for Evaluating Textual Readings: The Limitations of Textual Rules.” Harvard Theological Review 75, no. 4 (October): 429–48. Villanueva Zubizarreta, Olatz and Luis Araus Ballesteros. 2016. “Espacios, identidades y relaciones de los musulmanes de la ciudad de Burgos durante su minoría mudéjar.” In De la alquería a la aljama, edited by Ana Echevarría Arsuaga and Adela Fábregas García, 167–94. Madrid: Universidad Nacional de Educación a Distancia (UNED). Williamson, Callie. 1987. “Monuments of Bronze: Roman Legal Documents on Bronze Tablets.” Classical Antiquity 6, no. 1 (April): 160–83. Woolf, Greg. 1996. “Monumental Writing and the Expansion of Roman Society in the Early Empire.” Journal of Roman Studies 86: 22–39. Woolf, Greg. 2009. “Literacy or literacies in Rome?” In Ancient Literacies. The culture of Reading in Greece and Rome, edited by William A. Johnson and Holt N. Parker, 46–68. Oxford: Oxford University Press. Woolf, Greg. 2015. “Ancient Illiteracy?” Bulletin of the Institute of Classical Studies 58, no. 2 (December): 31–42. Youtie, Herbert C. 1975. “Because They Do Not Know Letters.” Zeitschrift für Papyrologie und Epigraphik 19: 101–8.

Chapter 5 Epigraphic strategies of communication: the visual accusative of Roman Republican dedications of spoils Fabio Luci

Introduction When we look at an inscription combined with an artwork or monument, attention is usually paid to what information is provided by its text. In many cases, the inscription is considered as a “caption” for its related visual object. Considering inscriptions as “captions” can be useful for answering important, but general questions, for example, the identity of the honoured person and/or the dedicator(s) represented on a statue, the historical circumstances of such an honour or whatever other issues are addressed by the information included in the inscription. The major risk in considering inscriptions as captions is to regard the role of the inscription as subordinate to the whole composition, which drastically limits its potential. More recent works on art and text demonstrated that there is an increasing focus on cross-disciplinary analysis in the visual and textual culture of the ancient world.1 John Ma’s Statues and Cities provides an important contribution towards understanding not only how art and text interacted with their space and audience, but also the physical interplay between art and text themselves, warning about the risk in considering inscriptions as only captions of their visual objects. It is worth strongly emphasizing this last point: that In art history, the first word encountered when dealing with image and text is ekphrasis, literally “writing on art”. The approach developed in this field (already known by ancient authors such as Pliny, Vitruvius, Lucian, Philostratus, and Pausanias) offers today important studies on the intertextuality between art and text (Robillard and Jongeneel 1998; Webb 1999; Elsner 2002; 2007; 2010; Goldhill 2007; Squire 2015; Roby 2016). In another subfield of scholarship, inscriptions are at the centre of multidisciplinary analysis that focuses on the physicality of the inscriptions and their relationship with the space and the ancient viewers: Cooley 2008; Ma 2013; Sears et al. 2013; Petrovic et al. 2018. In these works, as in my own, inscriptions are also integral parts of their monuments, and they work to reshape their physical features as well as their political meanings.

1

96

Fabio Luci

an inscription is not only the caption for a composition. Rather it is a concrete and stand-alone object that is correlated with its artwork.2 This paper offers a different approach to monuments and inscriptions, focusing on compositions as a set produced by a single matrix. Inscriptions attached to monuments were more than simple carved words informing the reader about the object itself, the dedicator(s) or the recipient(s). Rather, in many cases, inscriptions had a central role in determining the way their readers approached them and the dedications themselves. It is worth restating the obvious: monuments, inscriptions, artworks, visual objects, complexes, dedications, etc. had remarkable communicative resonance with their audience; inasmuch as the audience could interact with them and understand their significance. It does not surprise that the relationship between the visual object and the text was strategic and important in shaping the monument’s overall meaning. The way inscriptions and artworks interacted to create such a consistent and effective message to their audience was through the construction of a syntax, capable of transforming the visual objects into almost textual elements of their inscriptions. The audience was the catalyst to trigger the transformation by actively interacting with the compositions. A series of inscriptions from the Roman Republic dedicated by Roman commanders to tell their audience about their victorious military campaigns and their political successes offers a perfect example for the proposed methodological framework. In this kind of inscription, the direct object is always textually omitted, yet it was present in a visual form. This strategy creates a tension for those who read the apparently incomplete sentence, leaving to the audience the task of completing its meaning by looking at the artwork as the (visual) accusative of the inscription. The benchmark case used in this work is the inscription of Claudius Marcellus in Rome in 211 BC (CIL I2 608), focusing on its syntax to show how the statue itself is a surrogate for the missing direct object of the inscription. Metaphorically speaking, the visual object became part of the syntax of the inscription, whose accusative was intentionally not expressed by words. This analysis has two aims: the first is to set a new, and broadly adaptable, methodological approach in understanding the way epigraphy and artworks were strategically used by dedicators to communicate with their audience. Through the syntax of the inscription, each element of the phrase – subject, verb, and then the direct object, metaphorically embodied by the visual artwork – are perfectly organized and dependent on each other. The strategy adopted by dedicators pushed the audience in integrating the visual object into the incomplete syntax of the epigraphy; by doing so they restored the message promoted by the whole composition. The second is to show how this methodology, in some measure, is capable of analyzing monuments by only looking at their inscriptions when the artworks are now lost. In this sense, text and visual object are part of the same syntax, and the inscriptions can be 2

Ma 2013, 15­­­–24.

5.  Epigraphic strategies of communication

97

considered as part of the missing visual artwork’s body, just as a torso or a head are to a fragmentary statue.

Words and images: a syntactical framework Inscriptions and monuments are produced from the same matrix: by those historical contingencies (events, actions, and consequences) that were considered worthy of being remembered and celebrated. In this light, inscriptions and monuments are equally vehicles of information, but they diverge concerning how they deliver the message to their audience: the first in linguistic form, the latter in visual form. In a single composition, for example, if we consider a statue placed over an inscribed base, both its artistic features and textual languages communicate a particular message to the audience. The main characteristic of this double channel of communication is how images and texts work together to convey clear messages to their audience. These two different vehicles of information contained a built-in multiplicity of meaning, such as civic and military values of the Roman Republic both in visual and textual form. What was crucial for the patrons of honorary dedications was to deliver clear messages that could benefit both from the strength of visuality and the conciseness of the text. This attitude is not too hard to imagine if we think about the importance of honorary practice in Republican Rome, where the competition among families and individual was played through the display of power, prestige, values, and ideals. Words and images function in different ways because of their contrasting textual and visual natures. Whereas words need to be read, images are appreciated for their visual features: theme, iconography, and style. Understanding the content of an inscription required the ability to read simple Latin.3 In addition, visual features could not speak by themselves unless the audience were familiar with local and foreign artistic traditions and capable of distinguishing the differences between one and the other. Connoisseurship in art was more easily accessible to the elite class. However, it is absurd to think that the consumption of compositions of artworks and inscriptions was restricted to the elites, not only considering the location chosen for honorary dedications, such as in the Forum, but also because they addressed a heterogeneous public, including a solid base of clients and supporters, to increase the prestige of the honoured individuals and their families. Inscriptions had a pivotal role in forming the message of the whole composition. Their words guided the audience in discerning the meaning, not only of the event that was being celebrated by the honorific dedication, but also in understanding the connection between the visual object – its content, style, and iconography – and the words expressed in its inscription. Through the repetition of formulaic epigraphic 3 In some cases, it was not even necessary to have a high level of literacy because the honorary inscriptions of the Republic were quite concise and tended to resort to well-known words (e.g., names with filiation, titles, offices) framed in formulaic sentences (Corbier 2013, 13–38; Revell 2013, 231–3). On the nature of inscriptions, see the thought-provoking essay by Panciera 2012.

98

Fabio Luci

sentences, the audience became accustomed to an epigraphic landscape that helped them to immediately grasp what the composition was about. The combination of political, social, and military ideal values (such as auctoritas, clementia, concordia, pietas, maiestas, virtus) was traditionally understood in the Roman culture via individual cases as exempla. One of the channels through which these ideal values were communicated was figurative art, by associating ideal values to the various stylistic forms drawn from Greek artistic tradition. The abstraction of ideal values became a link between subjects represented and forms of imagery, creating a versatile and adaptive system for Roman art. As Hölscher has pointed out, this “ideological system”, formed during the Republic and consistently applied during the following centuries, was used to exemplify intellectual ideals more than to record historical reality.4 Although this system is now well accepted, it is less clear how the process of abstraction worked and how the artistic forms were understood by the audience. Inscriptions can be much more helpful in understanding the conceptualization of artistic forms. Artistic styles and iconography could be understood as connected to particular ideal values, as argued by Hölscher and Zanker.5 However, each individual artwork was also the result of a historical event that generated, in the first place, the visual and textual compositions. In this respect, inscriptions informed the audience about which event was commemorated and why. This information contained abstract concepts: for example, an inscription can inform its viewers about the capture of a city, but it is not limited to the factual event. Inscriptions contain far more information, or they can imply something, such as the narrative of a successful military action performed by a commander and, thus, the ideal values connected to it: the virtus of the commander who achieved it, but also the euergetism of the commander who dedicated a beautiful artwork to the city or his pietas in dedicating it as a votive offering. Thus, inscriptions pointed more directly at specific ideal values from the Roman value system. By juxtaposing inscriptions with artworks, any ideal values connected with the inscriptions (from the previous example, virtus, euergetism, or pietas), were also attached also to the artistic style of the visual object. Only subsequently could artistic forms, whose meaning was translated and understood with the support of the language in the form of inscribed letters, be conceptualized into ideals and values. In other words, the ideal values produced by the significance of the event remembered by the words of an inscription were transferred to the style of the artwork. The major questions that stem from this set of assumptions are, first, how inscriptions helped viewers understand what the images represented and, second, how two different communicative systems could be integrated into a single narrative. The answers can only be found by looking at the interaction between language and image and at the processes that establish such an interaction. It is useful to borrow the notion of syntax 4 5

Hölscher 2004, 88–9. Hölscher 2004; Zanker 1988.

5.  Epigraphic strategies of communication

99

from linguistics exclusively for the purposes of this discussion and apply it as an effective solution for analyzing the interactions between language and image. In linguistics, “syntax” means a set of rules and tenets of a language that organize and arrange words, phrases, and clauses to fully form logical sentences. From this generic definition of syntax, it is possible to narrow down, for the benefit of this discussion, the grammatical function of the syntax to the way in which a specific word or element of a sentence can be arranged and defined in its function with other words or elements of the same sentence.6 In Latin, unlike in English, in order to accomplish the formation of sentences and their subsequent meanings, words, in our cases the words of an inscription, adopt specific grammatical cases that define what function they play within the sentence. The role of each word in a sentence is defined by its grammatical function, and the correlation between the words is determined by a specific syntax. The syntax, in organizing the words, creates logical and clear phrases. Roman dedicatory inscriptions are expressed through sentences that can also be analysed in terms of “grammar” and “syntax”. Each word of an inscription is related to the others to form a complete sentence. As is demonstrated in this section, not all the syntax contained in all the inscriptions of Roman dedications has been completed and, as a result, needs to find its connections and relationships “outside” of the text. Although Roman dedicatory inscriptions are expressed by words alone, their meanings only became complete and intelligible to their readers with the integration of non-textual elements related to the inscriptions, namely, the artwork itself and, in some circumstances, the space in which it was placed. In this study, the grammar and syntax of Roman dedicatory inscriptions play a central role not only in revealing the grammatical function of the words of the inscription, but also in distinguishing and emphasizing the correlation between the textual and non-textual elements and their respective and metaphorical roles. The verbal syntax of the inscriptions is integrated with the non-textual elements, such as the monument itself and the space in which it has been placed. In this way, the syntax is not only able to complete the meaning of the whole composition by considering all its elements at once, but it also cooperates in translating the artistic form into intellectual ideals. The final result is the presentation of a single consistent and intelligible narrative of the whole composition to its audience.

6 The defining contribution to linguistic semiotics was made by Ferdinand de Saussure’s Course in General Linguistics, published in 1916. In Saussure’s formulation, language is a system of signs, each sign consisting of the actual form the sign takes, that is, a signifier, and the concept it represents, that is, a signified. The relationship between the signifier and signified is conventional, which means that it is dependent on the social context and its particular cultural conventions.

100

Fabio Luci

Inscriptions of Roman Republican dedications of spoils The pattern of Roman Republican dedicatory inscriptions is quite simple. These inscriptions usually contain a subject (rendered in the nominative case), the typical filiation that grammatically corresponds to the apposition of the subject, a transitive verb of the sentence and an ablative of separation that indicates the location from where the item was taken (Leucadus, Enna, or another city). It is evident that in such a sentence there is something missing. Although the subject of the action, the action, and the location are there, there is no trace of the direct object, which, however, must be present due to the transitive verb. In many cases, the recipient of the dedications is also absent, although it is not compulsory for the sentence consistency. The missing elements must be searched for in non-textual elements and the artwork itself becomes the direct object of the sentence because it is related to the action expressed by the sentence verb (i.e., capere or dare). In this sense, the artwork assumes the syntactic role as the “visual accusative”; the physical space in which the artwork was placed might define a metaphorical dative of interest. The deliberate omission of words by the dedicators has the effect of creating a “tension” for those who read an apparently incomplete sentence, leaving to the audience the task of completing the meaning. In completing the message, the audience are pushed to look at non-textual elements as part of the inscription’s syntax in order to close the textual gap. Hence, the participation of the audience is fundamental and complementary to the metaphorical creation of a syntax that considers both textual and non-textual elements. A dedicatory inscription carved on the base of a statue by Lucius Quinctius Flamininus at Praeneste in 192 BC offers a specific example of how non-textual elements are metaphorically part of the syntax (Fig. 5.1): [L. Quinctius L(uci) f(ilius) Le]ucado cepit / [eidem conso]l dedit.7 Lucius Quinctius, son of Lucius, took from Leucadus, he gave as consul. The first observation is that this category of inscriptions is commonly translated with the inclusion of the direct object in the form of “this”, despite its absence from the text. For example, this inscription is usually translated in manuals and books as: “Lucius Quinctius, son of Lucius took this as booty from Leucas; he likewise made a gift of it when he was consul”;8 or “L. Quinctius, son of Lucius, took this from Leucadus. He gave it as consul”.9 The word “this” is absent from the Latin text and should not be translated whatsoever. Including the word “this” in the translation could, apparently, be correct to reconstruct the meaning of the sentence. However,

CIL I2 613. Warmington 1940, 77. 9 Riggsby 2006, 218. 7 8

5.  Epigraphic strategies of communication

101

Fig. 5.1: Dedication by L. Quinctius Flamininus in Praeneste (Center for Epigraphical and Palaeographical Studies, Ohio State University).

its inclusion downplays and undermines the significance of the artwork and its relationship with the inscription.10 Although the inclusion of “this” in translations is common practice and can be useful to provide a full understanding of the inscription’s texts, it nevertheless affects the meaning of the whole composition because it deprives the real object, the artwork, of its syntactic function. It is important, therefore, to approach the translation of the text of the inscription exclusively from what is actually present in order to understand the interconnection between object and text.11 Accordingly, my translation is: Lucius Quinctius, son of Lucius, took from Leucadus, he gave as consul. The second observation is that the missing elements that complete the meaning of the sentence and of the whole composition, are the accusative dependent on cepit (to take) and the dative of interest implied by dedit (to give). The accusative can be integrated with the artwork itself, which assumes the role of a “visual accusative”, acting as a surrogate for the textual accusative. Conversely, the textual absence of the dative can be reintegrated by the physical space in which the composition was placed, which points to its destinatory: for example, to the people of a city, if it was set in a public space, to a god, if it was set in a temple. In this syntactic system, both space and audience can assume the role of metaphorical datives. As datives, space and audience not only contribute to completing the meaning of the sentence, but they also strengthen the effectiveness of the dedication message. In this specific In fact, in La Trahison des images by Magritte, determining the relationship between the image and the demonstrative pronoun, “ceci”, can be particularly controversial. The Roman dedicator did not want such ambiguity. 11 The translations of all inscriptions are mine. In my translations the nominative might present the cognomen of the subject, for example, (Marcellus), (Nobilior) etc., to clarify for the reader who the individuals mentioned are. 10

102

Fabio Luci

case, the Temple of Fortuna Primigenia provides the identification of the indirect object of the composition – the goddess – although this is absent from the text of the inscription. The citizens of Praeneste are the audience looking at this dedication, and they take up the role of the dative of advantage because they indirectly benefit from the consumption of the artwork. The two metaphorical datives represented by space and audience, respectively, do not necessarily overlap. They assume two different functions: whereas the indirect object directly exalts the relationship of the Quinctii with Fortuna Primigenia,12 the dative of advantage shows the sphere of influence of the Quinctii in Praeneste.13 The significance of this model in reconstructing the connections between art and text based on a syntactic structure is not a mere intellectual exercise. Its applicability to the honorific dedications has several implications for our understanding of Roman culture and its artistic production. In the first instance, this model can be very helpful in highlighting the complexity and the novelty of the artistic and epigraphic output that the Roman political culture produced. Second, it negates the idea that republican artworks are completely lost, as scholarship tends to suggest. Inscriptions and artworks were conceived as a whole by their patrons and consumed as a whole by their viewers. This means that inscriptions should not be considered as any different from fragments that remain of a statue or monument. By considering inscriptions as fragments of otherwise lost artworks, it is possible to collect more information about the style and the iconography of a monument, no less than an arm, a torso, or a capital can provide. This syntactic model exploits the mechanics of how the text of an inscription has a syntactic link with its artwork. In most of the inscriptions related to spoils there are no textual accusatives to complete the significance of the verbs in the sentences, as shown in Table 5.1. The apparently missing direct object exists, but in a visual form. This intentional omission would avoid the redundancy of repeating the direct object textually, because it is present visually. The missing direct object pushes the audience to participate in order to complete the meaning of the sentence with the inclusion of the visual object as a figurative direct object of the sentence. The model presented can offer significant insight into the reception of the artworks and inscriptions. The interplay between text and artwork can help us understand how the audience understood and interacted with the honorific practice. When they approached a monument, a complex, or any brand new or refurbished building, they searched for the inscriptions. Once found, the viewers could fully appreciate and understand the significance of the whole composition. This is suggested not only by The relationship between the Quincti and Fortuna is attested also from a dedication carved on an altar: Fortuna[(e) - - - ] / sac[(rum) - - - ] / T. Quincti[ - - - de] / Senati sente[ntiae - - - ] (CIL I2 656). 13 Clientships and political connections between the Quinctii and the citizens of Praeneste began in 380 BC and were started by Titus Quinctius Cincinnatus Capitolinus. See Demma 2011 for a broader discussion of the Quinctii, esp. 40–1. 12

103

5.  Epigraphic strategies of communication

Table 5.1: Inscriptions of manubial dedications erected by Roman commanders in Rome and Italy (264–146 BC). Dedicators

Date (BC) Location

Inscription

Edition

M. Fulvius Flaccus

264

Rome

M(arcus) Folv[io(s) Q(uinti) f(ilius) co(n)s]ol [dede]d Volsi[niis] cap[tis]

CIL I2 2836a

M. Minucius Rufus

217

Rome

Hercolei / Sacrom / M(arcus) Minuci(us) C(ai) CIL I2 607; ILLRP 118 f(ilius) / dictator vov / it

L. Quinctius Flamininus

192

Praeneste

[L. Quinctius L(uci) f(ilius) Le]ucado cepit / [eidem conso]l dedit

CIL I2 613; ILLRP 321

M. Acilius Glabrio

191–190

Luni

M(anius)Acilius C(ai) f(ilius) / co(n)s(ul) / Heracelea cepi(t)

AE 1993, 643

M. Acilius Glabrio

191–190

Luni

M(anius) Acilius C(ai) f(ilius) / co(n)s(ul) / Scarp(h)ea cepi(t)

ILLRP 321a

M. Fulvius Nobilior

189

Rome

M(arcus) Folvius M(arci) f(ilius) / Ser(vi) CIL I2 607; n(epos) Nobilior / co(n)s(ul) Ambracia / cepit ILLRP 118

L. Mummius

146

Nursia

L(ucius) Mum(m)ius / co(n)s(ul) ded(it) N(ursinis)

CIL I2 628; ILLRP 329

L. Mummius

146

Parma

L(ucius) Mummius / co(n)s(ul) P(opulo) P(armensi)

CIL I2 629; ILLRP 330

L. Mummius

146

Cures

[L(ucius) Mummius]] co(n)s(ul) Achaea capta

CIL I2 631; ILLRP 328

the ubiquitous presence of inscriptions or graffiti that have been found in many kinds of contexts – public, private, sacred, and funerary – but also by the attention paid by ancient writers to reporting the presence (or absence) of inscriptions connected to specific monuments.14 A fundamental element to consider is that the chief aim of writing an inscription is that it should be read. The result of including the visual elements of the artwork in the verbal syntax of the inscriptions was to reinforce the reception of their combined message by its audience.

The visual accusative as a syntactic strategy: the case of Marcellus The inscription CIL I2 608 relating to Claudius Marcellus (end of the third century BC) offers an ideal test-case in which textual and non-textual elements can be analyzed in the proposed syntactic structure: namely, how the visual object becomes part of its syntax. The test-case shows the potential that my syntactic model can offer for analyzing epigraphic dedicatory examples, including a second coeval inscription relating to Marcellus (CIL I2 609). Indeed, these two texts offer the chance to observe the dynamism of the self-promotion strategies used by patrons in their dedications. E.g., the inscriptions of Anicius at Praeneste (Livy 23.19.17–18); or the absence of Metellus Macedonicus’ inscription on the two temples of his portico (Velleius Paterculus 1.11.4).

14

104

Fabio Luci

The inscriptions of Marcellus are not an isolated case, but fit into a customary pattern used by many other victorious commanders from the third and second centuries BC. The interplay between text and art was, therefore, a generally accepted system, used by patrons, and understood and consumed by viewers. The contact point between art and text is established by the role that the artwork acquires if considered within the inscription. By integrating the missing direct object of the inscription with its related artwork, the sentence’s syntax is completed by the “visual accusative”, which completes the message of the composition. The integration of the artwork with the inscription’s syntax, however, is triggered only by the participation of the audience in reading, looking at, and consuming the whole composition. Without the active participation of an audience familiar with the Latin language, eliciting the meaning conveyed by the artwork and inscriptions of these dedications would not be possible. The contemporary audience were surely accustomed to this process, and their participation was mandatory, as will be discussed later on. For us, the process may not be prompted naturally. In order to rebuild the original meaning of the composition it is necessary to approach the artworks and their inscriptions in the same fashion in which their original audience would have done, and with the same mechanics with which these compositions were conceived during their own time. Three elements of the inscription An epigraphic dedication belonging to a series of inscriptions from victorious commanders who conquered cities during the third and second centuries BC shows both the conciseness and the syntactic co-dependency of the textual elements and the object displayed. The inscription and dedication, now lost, but recorded in CIL, was found in Rome on the Esquiline. It was set up after the conquest of Enna in 214 BC by M. Claudius Marcellus, consul quinquies, conqueror of Syracuse: M(arcus) Claudius M(arci) f(ilius) / consol / Hinnad cepit.15 Like L. Quinctius Flamininus’ inscription in Praeneste, this inscription is commonly translated with the inclusion of the direct object in the form of “this”, despite its absence from the text. For example: “Marcus Claudius, consul, son of Marcus, took this as booty from Henna”;16 or “M. Claudius, son of Marcus, consul, took this from Hinna”.17 The word “this” is not present in Latin, and such an omission is respected in my translation in order to better understand the interconnection between the object dedicated by Marcellus and its inscriptions. A different translation is thus possible: Rome; CIL I2 608. Extensive bibliography in Cadario 2005. See also Edwards 2003 on the displays of spoils as art in Rome. 16 Warmington 1940, 77. 17 Riggsby 2006, 218. 15

5.  Epigraphic strategies of communication

105

Marcus Claudius (Marcellus), son of Marcus, consul, took from Enna. This inscription was carved on a base of a piece of artwork taken from Enna and displayed in Rome to celebrate the conquest of the city by Marcellus in 214 BC through the combination of the inscription and the artwork.18 The text of the inscription is composed of three lines and elements present in the text are grammatically incomplete: the dedicator is expressed in the nominative case followed by the filiation; the status of the dedicator as consul is in apposition to the subject; and the active perfect of capere together with the ablative of separation identifies the location from which the object dedicated was taken.19 Missing from this textual construction are at least two elements: the accusative direct object dependent on cepit; and the dative of interest, implied by the dedicatory nature of the inscription. The former is syntactically presented as the object dedicated, which plays the role of a visual accusative. The latter is metaphorically defined by the topographical position in an urban context, which, in turn, defines the kind of audience that the whole composition was addressing: the people of Rome. The analysis of this inscription is here divided into three parts, each of which corresponds to the three lines that constitute the text: the nominative and filiation in the first; the apposition of the subject in the second; and the verb with the ablative of separation in the third. First line: the nominative and filiation In the first line, the name of the dedicator, Marcus Claudius (Marcellus), along with his filiation, is indicated by the nominative case. The hierarchical order created by the syntax suggests the grammatical function of the nominative case as the subject of the action, with the name of Marcellus directly presenting him as the main actor of the whole composition. The filiation, which is rarely omitted in this category of inscriptions, subsequently follows Marcellus’ name, completing its semantics.20 In this concise inscription, the filiation assumes a meaningful role in strengthening the relationship between the Claudii Marcelli and the Urbs and its institution. The Claudii Marcelli family had already produced two consuls: Marcus Claudius Marcellus in 331 BC and his son Marcus Claudius Marcellus in 287 BC. The latter was also the grandfather of the conqueror of Syracuse. The filiation Marci filius works as reminder of See Livy 24.39 for the narrative of the conquest of Enna. A first approach to foreign names of cities conquered by Roman people happened during the pompa triumphalis. See Östenberg 2009b; Cadario 2011, 31–2; 2014, 86–7; Tarpin 2011, 683–4. 20 The few exceptions include Corneli(us) / Scipio /Cartha(gine) / capta (CIL I2 625); L. Mummius cos. / vico (CIL I2 627); L. Mumius / co(n)s(ul) ded(it) N(ursinis) (CIL I2 628); L. Mummius / co(n)s(ul) p(opulo) P(armensi) (CIL I2 629); L. Mummius / co(n)s(ul) Achea capta (CIL I2 631). These inscriptions were not located in Rome and they follow different self-promotion strategies. The first case adopts the Saturnian meter, typical of the Scipionic epitaphs and inscriptions, in which the filiation is not always canonically used. Through the chiastic position (ABBA) of its textual elements, the inscription emphasizes the role of Scipio in conquering Carthage. Mummius’ tituli will be addressed later in this section. See Graverini 2001. 18 19

106

Fabio Luci

the consulships obtained by Claudius Marcellus’ ancestors.21 The filiation of the Claudii Marcelli assumed more emphatic tones during the second century BC with M. Claudius Marcellus, consul in 166, 155, and 152 BC and grandson of the fivetimes consul. He erected three statues near the Temple of Honos and Virtus depicting himself, his father and his grandfather (the consul quinquies), and set up an inscription reported by Asconius that read “III MARCELLI NOVIES COSS”.22 In the case of the triple statue, the filiation is not mentioned in the text, but is highly emphasised by the visual juxtaposition of the statues of the three Marcelli that a Roman reader might reconstruct as M. Claudius Marcellus, Marci filius, Marci nepos. Thus, textual Fig. 5.2: Dedication by M. Fulvius Nobilior in Rome elements strengthen the continuation (Rome, Musei Capitolini, Centrale Montemartini. of a successful career among members Photo: Archivio Fotografico dei Musei Capitolini. © of this family, exalting their important Roma, Sovrintendenza Capitolina ai Beni Culturali). role in the Republic through the total of their consulships (honores). The filiation, filius and nepos, recalling illustrious predecessors, emphasises and confirms the prestigious socio-political status of the protagonist of the dedication. This strategy had been adopted in the inscription belonging to Marcus Fulvius Nobilior, consul in 189 BC, after his victory against the Aetolians and the capture of the Greek city of Ambracia (Fig. 5.2): M(arcus) Folvius M(arci) f(ilius) / Ser(vi) n(epos) Nobilior / co(n)s(ul) Ambracia cepit.23 Marcus Fulvius Nobilior, son of Marcus, grandson of Servius, consul, took from Ambracia. 21 The link between the filiation and the consulship could have also worked as a means to overshadow an embarrassing episode that happened to Claudius Marcellus, consul in 331, who was at the centre of a controversy during his appointment as dictator in 327 BC. In that occasion he was disapproved by the augurs instigating an interregnum. Livy 8.23.14–17. Cf. Oakley 2005, 84–5. 22 Asc., Pis. 12.15–19. See Sehlmeyer 1999, 163–5; Papini 2004, 401–2; Cadario 2005, 165–6; Lewis 2006, 24–5. 23 Rome, CIL I2 615; Ruck 2004, 485. It is possible that the base and its artwork were dedicated to the Muses in the Aedes Herculis Musarum, which was erected by the same M. Fulvius Nobilior. Cf. Cic., Arch. 11.17; Plin., HN 35.66. Cf. CIL I2 616, Marcus F(ulvius) M(arci) f(ilius)/Ser(vi) n(epos) co(n)s(ul)/Aetolia cepit: a base found in Tusculum’s theatre, probably a copy belonging to the Augustan age.

5.  Epigraphic strategies of communication

107

The grandfather of M. Fulvius Nobilior was Servius Fulvius Paetinus, consul in 255 BC, who celebrated a triumph for the victory over Cossyra and a naval triumph over the Carthaginians in 254 BC as proconsul.24 The long filiation bridged the triumph celebrated by Fulvius Paetinus with one of his grandsons, showing a continuity of the gens Fulvia in defeating powerful enemies of Rome. Furthermore, the parallelism between Nobilior and his grandfather emphasizes the consulships that both had held and, at the same time, overshadows the lesser prestige of his father, who did not obtain such an office. In Roman honorific inscriptions, filiation is usually expressed with the father’s initial only. In Nobilior’s case, however, the slightly less usual choice of including both father and grandfather suggests that the aim of the subtle strategy conveyed is to emphasize specific members and, therefore, their achievements. A similar strategy was also adopted by the Claudii Marcelli. The evidence is reported by Asconius with the statuary group inscription citing the III Marcelli novies coss. The strategy adopted seems to point to the numeric sum of the consulships held by the three members. In this sense, the presence of the illustrious grandfather Marcellus as well as of the father was fundamental in promoting the political weight of the Marcelli and in counterbalancing the unexceptional single consulship that Marcellus’ father held in 196 BC.25 In this case, the strategy adopted to promote and advertise the role of the Claudii Marcelli is even more evident because Marcellus’ consulship in 215 BC is included in the total: he was forced to resign after a bad omen.26 Similarly to Marcellus’ and Nobilior’s inscriptions, a substantial number of honorific inscriptions belonging to victorious generals follow this syntactic strategy in displaying the nominative and the filiation in the first line of the text.27 Thus, not only was the prestige of individuals celebrated through the main position of their names but, in addition, their achievements were framed as a reflection and consequence of being a member of illustrious gentes and familiae, which were evoked and brought into the action through filiation. In our inscription, then, Marcellus followed an established pattern that emphasized his noble lineage. Second line: the apposition In the second line of CIL I2 608, the apposition completes and identifies the subject as consul. The apposition following the name is almost ubiquitous in Republican honorific inscriptions, qualifying the status of the subject of the dedication with his office. In the case of M. Claudius Marcellus’ inscription, consul stands alone between Zonar. (Cass. Dio XI) 8.14; Polyb. 1.36.10; Livy 42.20.1; Fasti triumphales: Ser(vius) Fulvius M(arci) f(ilius) M(arci) n(epos) Paetinus a(nno) CDX[CIX]/Nobilior pro co(n)s(ule) de Cossurensibus/et Poeneis navalem egit XIII K(alendas) Febr(uarias). 25 Asc., Pis. 44; 12 C. See Cadario 2005, 165. Cf. Flower 2000, 46–7 for the consulship held by Marcellus’ descendants. 26 Livy 23.31.12–14; Plut., Marc. 12.1. See Flower 1996, 71–2. 27 E.g. CIL I2 48; CIL I2 608; CIL I2 623; CIL I2 615; CIL I2 616; CIL I2 2926; AE 1993, 643; CIL I2 613; CIL I2 626; CIL I2 622. 24

108

Fabio Luci

the first and third lines. Its position in the line helps to emphasize the meaning of the word, at the same time stressing the relevance of the office obtained by Claudius and members of his family. In other words, the function of the consulship displayed in such a way is not merely limited to emphasizing the prestigious office held, but also establishes the connection between his name and the action of the verb in the third line. It was Marcellus’ consulship that gave him the authority to lead Rome’s army and capture booty to dedicate.28 The consulship, expressed syntactically as the apposition of subject, was specifically used as a leitmotif by the Claudii Marcelli in promoting their achievements. For example, the textual strategy adopted by the inscription of the tres Marcelli novies consules focuses on the total of the consulates achieved by three members of the family. The adoption of this syntactic strategy persists and is emphasized over time in other inscriptions, connecting the conqueror of Syracuse (CIL I2 608 and CIL I2 609, Martei / M(arcus) Claudius M(arci) f(ilius) / consol dedit) to his descendants. For instance, the grandson M. Claudius Marcellus, who was consul in 155 BC, erected his statue in the Forum of Luna29 to celebrate his victory against the Apuani, a Ligurian community.30 M. Claudius M(arci) f(ilius) Marcelus / consol iterum.31 Marcus Claudius Marcellus, son of Marcus, consul for the second time. The legacy of the consulships held by the Claudii Marcelli is still present and evoked during the middle of the first century BC with a denarius minted by the triumvir monetalis P. Cornelius Lentulus Marcellinus in 50 BC that shows on the reverse the episode of the spolia opima obtained by M. Claudius Marcellus at Clastidium in 222 BC and dedicated to the Temple of Jupiter Feretrius with the inscription Marcellus cos quinq(uies).32 In both inscriptions of Marcellus and Nobilior, the consulship (consol and consul) are placed on the second line; yet, they have slightly different meanings. The second line of Nobilior’s inscription includes both the filiation and the consulship to link his office with his famous grandfather, Paetinus, who was victorious against the Carthaginians. This establishes a connection with his own deeds in capturing Ambracia, overshadowing the criticism as to how he conducted his war.33 Conversely, For the booty used by M. Claudius Marcellus as “cultural capital”, in the words of Bourdieu 1984, 53–4, see Holliday 2002, 196–7. Cf. Riggsby 2006, 196–7; Beck 2011, 77–96. 29 Livy 41.13.4–5 for the foundation of the colony of Luna in 177 BC. 30 Cf. Fasti Triumphales: [M(arcus) Claudius M(arci) f(ilius)] M(arci) n(epos) Marcellus II co[(n)s(ul)] II a(nno) DX[CIIX] [de ---]us et Apua[neis ---]; MRR I, 448; See also Angeli Bertinelli 1983; Sehlmeyer 1999, 112. 31 CIL I2 623. 32 RRC 439. See Coarelli 1996; Spannagel 1999, 149. For the identification of Gnaeus Cornelius Lentulus Marcellinus, consul in 56 BC, on the obverse of the coin, Lahusen 1985. For a possible influence of a painting on this image, Flower 2000, 47. For the debate on the identification of the portrait on the obverse, Papini 2004, 436–7. See also Cadario 2005, 170–3. 33 Polyb. 21.29–30; Livy 38.9.13; Plin., HN 35.36.66. See Östenberg 2009a, 212. 28

5.  Epigraphic strategies of communication

109

M. Claudius Marcellus’ consulship stands alone to emphasize the outstanding and well-known deeds of the general.34 The two examples demonstrate the versatility of the textual strategies used: the semantics of the apposition (consul) can, or cannot, be reinforced by other textual elements such as the filiation, as in the case of Nobilior. Third line: verb and ablative The third line of M. Claudius Marcellus’ inscription is composed of cepit, the perfect active verb of capere (take), and Hinnad as the ablative of separation that identifies the location in which the object dedicated was originally placed. The verb cepit (took) emphasizes the action performed by military commanders towards hostile cities and enemies. Syntactically speaking, the verb is at the very heart of the whole text, introducing and combining different elements of the inscription. The subject of the action is M(arcus) Claudius M(arci) f(ilius), who performed it through the imperium militiae suggested by the apposition consol. The ablative of separation Hinnad specifies and brings into the action the victim of the conquest. Therefore, the verb does not only emphasize the action performed by the subject, but also directly address the iter that produced the composition through the ablative of separation. Both subject and action act together to emphasize the existence of the whole composition itself as the result of their combination: individual (Marcellus) + legitimate authority (cos) + military victory (cepit) = benefit to city (artwork). The combination of cepit with the inscription’s other textual elements achieves the specific aim of celebrating both the military victory and the political dominance of Marcellus. The grammatical function of cepit is to introduce a fundamental part of the syntax: the accusative. The direct object is not textually mentioned; this does not mean that it is not present. The metaphorical syntax can supply the missing textual accusative giving the grammatical function to the artwork itself. The “visual accusative” embodied by the artwork is integrated with the inscription by the verb that syntactically demands a direct object that is absent in the linguistic form. Primarily, this strategy aims to emphasize the message carried by the whole composition, entailing a dynamic reading process that demands from the audience a scrutiny encompassing both its visuality and its textuality. The verb cepit, in conjunction with Hinnad, contains in itself the energetic action that gives life to the composition, this being the result of a series of events. The creation of this kind of monumentum starts from the consulship obtained by an individual through which he was able to overcome enemies of Rome, expressed by the location in which the artwork was originally situated. The energetic action On the campaign, Polyb. 2.34–5; Plut., Marc. 6–8; Frontin., Str. 4.5.4; Eutrop. 3.6; Oros. 4.13.15; Zonar. 8.20; Livy Per. 20; Naev., Fr. Praetext. In Varro, Ling. 9.78. For the conquest of Syracuse see MRR I, 273. The perception of Syracuse as a powerful and beautiful city exalts the propaganda around Marcellus’ victory. Marcellus obtained this victory as a consul; therefore, the military and victorious aspects are contained in the meaning of the consulship. See Livy 26.21; Holliday 2002, 112; Cadario 2005, 154. On the comparison between Marcellus and Nobilior’s triumphs, see Östenberg 2009a, 208–12. 34

110

Fabio Luci

expressed by the verb ends with the creation of the composition. The statue assumes a new role: now, it commemorates the deeds that produced it and emphasizes the exemplum of the individual’s virtus as a model to admire and follow.35 Thus, the verb, in its indispensable function of expressing the action of the sentence and clarifying its meaning, has a privileged role compared to the other elements of the monument in that it renews and perpetuates the political message of the action. The emphatic tone of cepit is strengthened by the ablative of separation, which in most cases is placed immediately before it.36 In Marcellus’ inscription, Hinnad is where the action of the verb takes place, recalling the Sicilian military campaign and, thus, the victory of the Roman command. The word Hinnad is a Latin word, although archaic, and is a transcription of the toponym Ἔννα.37 The name of the place from which the Greek artwork was plundered creates a picture in the viewers’ mind of the setting of the victorious military campaign conducted against a powerful enemy of Rome. The phrase Hinnad cepit makes clear the meaning also demonstrated by the act of putting a piece of Greek captured art on display in Rome. The artwork would have been Greek, and likely had a Greek inscription in its original location; in Rome it was carved a Latin inscription, which “translated” and altered its meaning. Both the artwork and the inscription were re-functionalized in favour of Marcellus’ self-presentation strategy.38 It is not surprising that the massive haul of loot taken from the Sicilian campaign was used in this way by Marcellus to decorate the Temple of Honos and Virtus.39 This behaviour was not uncommon: the columna rostrata dedicated to C. Duilius after his naval victory in 260 BC against Hannibal and the Carthaginians was adorned with the ramming “beaks” of captured enemy ships.40 Through the appropriation and re-functionalization of concrete objects, the superiority of Rome over powerful and foreign enemies was reaffirmed by specifying the location where the military actions were performed by the Roman commanders. Hölscher distinguished the function of expressing “the political entities, states and statements” through their visuality.41 Accordingly, the monuments are “power and weapons” leaving no chance for an interpretation other than political importance, Cf. Hölkeskamp 2004, 180–3. CIL I2 608: …Hinnad cepit; CIL I2 608: …Leucado cepit; CIL I2 615: …Ambracia cepit; CIL I2 616: …Aetolia cepit; AE 1993, 643: …Heraclea cepit; CIL I2, 2926: …Scarphea cepit. Cf. CIL I2 626. Another group of inscriptions, although they feature ablative absolutes, follow the same structure: CIL I2 631: …Achaia capta; CIL I2 630: …Corintho capta; CIL I2 625: …Carthagine capta. 37 Clackson and Horrocks 2007, 131–2; Wallace 2011, 18. 38 For political implications of displaying spoils of war see Holliday 2002; Hölscher 2006; Östenberg 2009a. 39 Cf. Livy 25.24.11, 26.21. The display of spoils was not limited to the temple, but was also in Marcellus’ house. For instance, the disposition of the sphaerae of Archimedes: one, the finest, kept in his house as the unique loot that was personally held; and another sphaera displayed in his temple for public consumption (Cic., Rep. 1.21). See Bravi 2012, 29–32. 40 Plin., HN 34.11.20; Sil., Pun. 6.663–6; Quint., Inst. 1.7.12. 41 Hölscher 2006, 27–48 is fundamental, especially 27–8. See also Holliday 2002. 35 36

5.  Epigraphic strategies of communication

111

Fig. 5.3: Votive dedication by M. Claudius Marcellus in Rome (National Archaeological Museum of Naples).

whether it was accepted and celebrated or opposed and destroyed. Textual elements, as part of the monument, represent these political entities and states in a more direct way. The foreign textual name of a place from which the artwork was taken, Hinnad in this case, informed the viewer about the geographical military expedition, even before he/she looked at the formal and stylistic features of the foreign artwork. Even when spectators did not know where Enna was, it had the same effect: by not knowing, the viewers acknowledged that the place was far away or even exotic. Another inscription concerning Marcellus presents a different strategy, although the construction is similar (Fig. 5.3). Martei / M(arcus) Claudius M(arci) f(ilius) / consol dedit [vov(it)].42 To Mars, Marcus Claudius (Marcellus), son of Marcus, consul, gave [(vowed)]. The composition comes from the same event, but its textual elements suggest a slightly different meaning. The verb dedit (“gave”), was carved over an already existing inscription: vovit (“vowed”). The word vovit was not clearly erased and is still 42

CIL I2 609.

112

Fabio Luci

visible on the epigraph. This might suggest the desire to show the different events simultaneously: the vow to Mars beforehand, and the accomplishment of the vow through dedit afterwards. Initially, Marcellus expressed a vow to honour Mars and once his achievement had been accomplished the wording was replaced with the verb dedit (to give); the dative of interest, Martei, underlines the recipient: Mars. Unlike in the case of CIL I2 608, in which the dative of interest must be inferred from the context, the textual dative of interest and the topographical position of the whole composition emphasize the relationship between Marcellus and Mars.43 Although the artworks cannot be investigated, it is evident from the different textual elements of the two inscriptions that the political meaning carried by the two compositions followed different strategies and must be read in different ways. CIL I2 608 emphasizes Marcellus’ legitimation as consul and his military victory; CIL I2 609 adds to the former inscription the pietas of Marcellus through the dedication to Mars. The dedication set up by L. Quinctius Flamininus at Praeneste in 192 BC, already presented above (Fig. 5.1), gives another example of how the verb(s) of an inscription work(s) with the visual object: [L. Quinctius L(uci) f(ilius) Le]ucado cepit / [eidem conso]l dedit.44 Lucius Quinctius, son of Lucius, took from Leucadus, he gave as consul. Quinctius Flamininus was a legatus under the command of his younger (and betterknown) brother Titus, who conquered Leucas in 197 BC during the war against Philip V of Macedon. He set up this dedication in the Temple of Fortuna in Praeneste, but only during his consulship in 192 BC. The inscription consists of two lines, each with a different verb: cepit (“took”) and dedit (“gave”). The syntactic strategy used by Flamininus relies on the double integration of the visual accusative in the sentence meanings. This double integration might be explained considering the different status of Flamininus between the victory and his consulship. The first integration of the visual accusative in the sentence in which the verb “take” appears works as a reminder of the exceptional military deed achieved by Flamininus during the war, although he was not consul. The second integration of the visual accusative represented by the verb “give” emphasizes his generosity when he finally obtained the consulship. This analysis has underlined the existence of a pattern followed by many honorific dedications set by commanders during the third and second centuries BC: artworks dedicated with short inscriptions including the donor’s name, filiation, and office along with a verb indicating the action commemorated and often either an ablative of separation or a dative of interest. The direct object of the verb, the artwork itself, is almost always omitted from the inscription. These inscriptions are similar in their structure as they all omit the accusative direct object, but they also differ slightly 43 44

On Marcellus’ piety, Val. Max. 1.1.8. CIL I2 613.

5.  Epigraphic strategies of communication

113

from each other. The syntactic strategies used by the patrons in their dedicatory texts show a significant amount of dynamism. This is because patrons had different political strategies according to their own life experiences and backgrounds. These strategies, visible as result of a single narrative created by the syntax between text and art, influenced the way the audience understood, interacted with, and responded to these compositions. In other words, monuments and inscriptions are intertwined and cannot be separated by carrying out two different analyses, and their syntax created potential “guidelines” on how to “read” the whole composition and translate it into a single narrative. At this point, the political significance of the events that produced the whole composition was effectively communicated to the audience, because they actively participated in restoring the meanings of the compositions.

Visual accusative and the audience So far, I have argued that Roman monuments and their inscriptions are strongly interconnected. They are complementary to each other, rather than being two separate and distinct objects merely assembled together. Their connection is established by the combination of artworks and inscriptions, the result of which is single and consistent sets. To understand the mechanism behind these combinations it is necessary to look at the way monuments and inscriptions interact with each other. Such a mechanism can be analyzed by considering artworks and their inscriptions as part of the same grammatical sentence. As in all sentences, the parts follow rules regulated by a specific syntax. This mechanism is, however, a passive agent and needs a catalyst to be triggered. The catalyst are the people who approach the monuments. The restoration of a sentence, created by the combination of a monument and its inscriptions, can be accomplished only by the viewers, who were induced to complete the sentence by logically placing each textual and visual element in the right grammatical order. Once the syntax is respected, the narrative of the inscribed monument is restored, delivering its precise and unmistakable message to the audience. Accordingly, the active participation of the audience was fundamental to accomplish the mechanics of the interaction between the artwork and its inscription. As we have seen in the previous examples, omitting the direct object in the inscriptions of plundered artworks that were rededicated in Rome was a deliberate choice made and a syntactic strategy adopted by dedicators to push the viewers to look at their artworks as the direct objects of the inscriptions. The relationship between the visual accusative and the audience is discussed here according to three main points: first, the visual accusative mechanism within the honorific practice and its strength in being adopted by the audience; second, the attitude and reaction of the audience towards the use of the visual accusative for restoring the meaning of honorary dedications; and, finally, the result and its effect on the audience.

114

Fabio Luci

The strength of the visual accusative relies on the easy way in which it can engage, and be engaged by, the audience. The honorific system in which the visual accusative is developed as a tool for its syntactic strategies has a peculiar characteristic: it is marked by very simple and concise syntaxes. In the previous example, all the inscriptions paired with plundered artworks are composed of only a very few lines with a very few and simple words. On the one hand, this particular aspect eased the process of integration of the visual artworks as the direct object of the sentence as understood by the audience. A simple text is easy to approach, and the integration of the visual accusative delivers a straightforward message: a certain object has been taken by the military actions of a certain consul of the Republic. On the other hand, the simple and concise syntax responded to the need to address the vast majority of people, even those who were illiterate, in a way that all would understand, by the repetition of the honorific practice and the use of simple inscriptions. By writing a very simple sentence structure on an ongoing basis – subject and transitive verb – the absence of the direct object was evident. In the previous example the honorific inscriptions are formed by subject, filiation, apposition of the subject (i.e., his title), and the missing direct object. It would be immediately clear to the viewers what the object was that had been plundered and rededicated in one of Rome’s public places or in a temple. Looking at Marcellus’ inscription, Marcus Claudius (Marcellus), son of Marcus, consul, took from Enna, by not including the direct object, the viewers would naturally think about what Marcellus took from Enna. To complete the sentence’s meaning, a viewer had to literally “look around” for the missing direct object. This task can be achieved only by considering the artwork itself, which is set up over the inscribed base, as the direct object of the sentence. As matter of fact, there cannot be any other elements that could fit into the syntax. The catalyst for the completion of the sentence meaning is the viewers. Using the visual object as the direct object they rebuild the syntax of the sentence and, therefore, its meaning. The process of integrating the artwork as the direct object of the inscription syntax of the visual object in the textual syntax is triggered by the reaction of the viewers in approaching an incomplete inscription. The incompleteness of a very simple and concise inscription prompts a “tension” in the audience and pushes them to find a way to complete it. The simplicity of these inscriptions is fundamental in triggering such tensions in the audience because it significantly emphasizes the missing direct object and, at the same time, avoids the requirement for special knowledge or philological skills to reintegrate it. It was only necessary to look at the artwork to find the missing element of the sentence. Finally, the tension created by the incompleteness of the inscription secured the reaction of the people by inducing them to actively participate in the syntactic restoration. The reaction of the people towards an incomplete sentence is prioritized here. It is true that nothing would have stopped the viewers from looking first at the visual object and only afterwards at its inscription. It is almost certain that for anybody walking towards a composition consisting of an artwork and inscription, the first

5.  Epigraphic strategies of communication

115

element to be noticed would be the artwork. In looking at a composition, the visual object is prioritized, but mostly because it is far more scenic than an inscription, especially if approached from a distance. This scenario is similar to what happens when we step into a museum. We immediately look at the physical objects – the archaeological finds, paintings, and statues – and usually only afterwards do we focus on the description of the various items. The main difference is that with regard to the rededicated spoils of war during the Roman Republic, their inscriptions were not captions, but were an inseparable part of the whole composition, and no different from a torso or arm fragments that were once part of a statue. A caption describes an item, but an inscription defines it. Even by looking first at the artwork and only after this at its inscription, a syntactic reconstruction is required. If a viewer first looked at an artwork, he/she would not easily understand the meaning of that dedication. An inscribed monument could be erected for many reasons and by many agents: by a benefactor from the same city; by another community; by a foreign king; or by a commander after his military campaign, as in our cases. The inscription is fundamental in giving a meaning to the whole composition and offering a full experience to the viewers who approach it. In addition, in Latin, the accusative case can also be used at the beginning of a sentence. Similarly, the visual accusative could have been the first element of the syntax that was approached by the viewers; nevertheless, the rest of the sentence must be integrated as well to complete the significance of the artwork. The result of using the visual accusative as a virtual tool for comprehending honorific dedications shows how the dialogue between artworks and their inscriptions relied on the response of the audience, whose main task consisted of combining these visual and textual elements to form a single and harmonious set. In this case, such dialogue was firmly secured by the synergy between the active participation of the audience and the visual and textual components of the monuments themselves. By using the visual accusative, the viewers transformed two apparently separate and different elements – the artwork and the inscription – into a single narrative, which was, ultimately, embodied by the whole honorific dedication. In other words, viewers completed the final stage of a rededicated plundered artwork and, therefore, the message of its dedicator. The effect that the syntactic reintegration of the visual accusative within the inscription had on the audience is at the core of the syntactic strategies adopted by patrons. The process of reconstructing the syntax of the monument by the audience also restored its message. As has been argued in the previous sections, the patron’s preferences and self-presentation strategies decided the kind of message. What is interesting to observe here is the effect that this process had on the audience. The viewers, by restoring the message created by the combination of textual and visual elements of a single monument did not only rebuild the message of the patron; the active participation of the audience in re-establishing the full meaning of the sentence helped to reproduce, acknowledge, and fix in the memory the event behind the

116

Fabio Luci

dedication. The memory of a historical fact is renewed in the collectivity. For example, the restoration of Marcellus’ inscription analyzed before imparts fundamental information: that Marcellus conquered and plundered Enna. Further, consequences are deduced by the historical fact of Marcellus’ conquest of Enna, such as his military abilities, his virtus, his success against an enemy of Rome and his generosity towards the people of Rome in decorating the city. Marcellus’ victory is constantly repeated over time by whoever engages in reading his dedication. This kind of honorific dedication worked as “hubs” of information from which the people of Rome could acknowledge the military achievements of an illustrious member of the nobilitas. The process of acknowledgement stems from the active participation of viewers in reconstructing the syntax of the visual and textual elements of these monuments. In other words, the audience renewed the events behind the dedications by (re-)building them every time they engaged with the compositions. The consequences of their active participation strengthened the shared knowledge of the society about specific events because the restoration process through the syntax could, without a doubt, be performed by anyone and easily shared. Most importantly, these monuments are fundamental to understanding the relationship between history and its transformation into memory, in this case, the creation of the Roman Republican cultural memory and the topographical “landscape” of memory in Rome.45 Such “monumental memory” was a distinctive feature of the Roman Republic, and its topography focused on the most important public space in Rome, because an interplay between locations and history, and between public and religious functions of monuments and past events in Rome was vital for making sense of this type of memory. Viewers walking through the monumental cityscape and looking around obviously reinforced the construction of the cultural memory of Rome, especially through the exposition of historical events monumentalized and fixed in the urban topography of Rome. Rituals and processions, whether triumphs or public funerals, had a significant role in constructing the cultural memory of Rome.46 It is true that these processions were fundamental in reinforcing the cultural memory and identity of Rome, especially because their routes passed through the monumental cityscape. However, in much of modern scholarship, it is not stressed enough how the audience were “reading” the spectrum of mythical or historical events embodied by honorific dedications either during the processions or on normal days.47 The visual accusative gives a more definite

See especially Hölkeskamp 2006a; 2006b. Hölkeskamp 2006b. 47 There is a wide bibliography on the spectacle, ceremonies, and representation of power. A selective list: Veyne 1967; Scullard 1981; Price 1984; Hopkins 1991; Bergmann 1999; Beacham 1999; Flaig 2003; Bell 2004; Flower 2004; Sumi 2005; Hölkeskamp 2004; 2006a; Hölscher 2006; Östenberg 2009a; Galinsky 2016. However, none of these works engages with the practical involvement of the audience with the monuments and their inscriptions. 45 46

5.  Epigraphic strategies of communication

117

answer to this because it avoids sweeping statements about viewers simply “walking through the monumental landscape of Rome and looking around”.48 Finally, the syntactic model also offers a further distinction between the role of dedications, such as spoils of war, and that of other kinds of monuments, such as honorific statues or temples. Through the syntactic reconstruction of their messages by the audience, the events behind them were evoked and fixed into the viewers’ memory, once, of course, the task of integrating the visual accusative within the inscription was completed. This series of information was complementary to other forms of monumentalization, such as statues of individuals and temples. The heterogeneity of the monuments created a historical and celebrative “map”, simple to understand, and that worked according to the needs of the patrons.

Conclusion The syntactic methodology expounded shows the potential for considering monuments and inscriptions as a single consistent and inseparable set. Words and visual elements work synergistically to produce a single narrative and fulfil the specific communication strategies of the dedicators. The examples discussed show the importance of reconstructing the connection between art and text, especially with regard to understanding how crucial the role of inscriptions was in reshaping the meaning of the monuments and its interaction with the audience. Further considerations originate from this kind of approach. The assumption that Rome’s Republican artworks are completely lost can be contested. Inscriptions and artworks were conceived as a whole by their dedicators and consumed as a whole by their recipients. It is thus possible to rebuild what the monument was, in terms of significance and impact to its audience. The inscriptions played a fundamental role in defining artworks’ meanings, rather than being merely a decoration or a legend. This process shows how inscriptions are syntactically linked with their artworks. The interplay between texts and images pushes the audience to participate actively in completing the meaning of their sentence, rebuilding the narrative that the whole composition intended to convey. The Roman Republican inscriptions of honorific monuments were concise, simple in their choice of the lexicon, and relatively standardized in their formula. Words like consul, cepit, dedit, were immediately recognized by their vast presence in Rome and its institutions, as well as having a simple, yet strong, meaning. Reading such short texts, syntactically combined with fine foreign artworks, enhanced their communicative power and resonance. Accordingly, the interplay between text and artwork can improve our understanding of how the 48 I here refer to Hölkeskamp 2006a, 483, who argues that: “not only can such a landscape be ‘read’ like a text, since it stores the full spectrum of myths, historical, etiological and other stories – it can also be experienced directly, by Roman citizens as viewers, in the concrete sense of walking through it and looking around” (emphasis mine). Hölkeskamp is right to assert the “readability” of landscape, but has not given a sufficiently strong account of how it was read by viewers.

118

Fabio Luci

audience approached and interacted with the honorific practice. Patrons shaped their messages in a way that best suited their own needs and purposes by adopting specific syntactic strategies. Subsequently, when the viewers approached a monument, a complex or any brand new or recently refurbished building, they searched for the inscriptions. Once found, the viewers could fully appreciate and understand the significance of the whole composition and memorialize the historical event behind it. Finally, when the artwork was paired with an inscription, as was normal practice in the ancient world, the relationship between art and text became so strong that the two narratives merged into a single and more consistent whole.

Bibliography

Angeli Bertinelli, Maria Gabriella. 1983. “Luni, Museo archeologico. Scheda 13: CIL I2 623 = XI 1339 = ILLRP 325.” In Marmora Lunensia erratica: mostra fotografica delle opere lunensi disperse, edited by Antonio Frova and Maria Gabriella Angeli Bertinelli, 68. Sarzana: Lions Club. Beacham, Richard C. 1999. Spectacle Entertainments of Early Imperial Rome. New Haven, London: Yale University Press. Beck, Hans. 2011. “Consular Power and the Roman Constitution: The Case of Imperium Reconsidered.” In Consuls and Res Publica: Holding High Office in the Roman Republic, edited by Hans Beck, Antonio Duplá, Martin Jehne and Francisco Pina Polo, 77–96. Cambridge: Cambridge University Press. Bell, Andrew J.E. 2004. Spectacular Power in the Greek and Roman City. Oxford, New York: Oxford University Press. Bergmann, Bettina. 1999. “Introduction: The Art of Ancient Spectacle.” In The Art of Ancient Spectacle, edited by Bettina Bergmann and Christine Kondoleon, 9–35. Washington, New Haven: Yale University Press. Bourdieu, Pierre. 1984. Distinction. A Social Critique of the Judgment of Taste. Translated by R. Nice. Cambridge, MA: Harvard University Press. Bravi, Alessandra. 2012. Ornamenta Urbis. Opere d’arte greche negli spazi romani. Bari: Edipuglia. Cadario, Matteo. 2005. “I Claudi Marcelli: strategie di propaganda in monumenti onorari e dediche votive tra III e I sec. a.C.” Ostraka 14, no. 2: 147–77. Cadario, Matteo. 2011. “Teatro e propaganda, trionfo e mirabilia: considerazionisul programma decorativo del teatro e della porticus di Pompeo.” Stratagemmi. Prospettive teatrali 19: 10–68. Cadario, Matteo. 2014. “Preparing for Triumph. Graecae Artes as Roman Booty in L. Mummius’ Campaign (146 BC).” In The Roman Republican Triumph: Beyond the Spectacle, edited by Carsten Hjort Lange and Frederik Juliaan Vervaet, 83–101. Roma: Quasar. Clackson, James and Geoff Horrocks. 2007. The Blackwell History of the Latin Language. Oxford: Blackwell. Coarelli, Filippo. 1996. Iuppiter Victor, Templum. Lexicon Topographicum Urbis Romae 3: 161. Cooley, Alison E. 2008. “Text et image: du Musée capitolina au British Museum. Tradition et interprétation.” Ktèma 33: 433–46. Corbier, Mireille. 2013. “Writing in Roman Public Space.” In Written Space in the Latin West, 200 BC to AD 300, edited by Gareth Sears, Peter Keegan and Ray Laurence, 13–48. London: Bloomsbury. Demma, Filippo. 2011. “Leucado Cepit: Praeneste Roma e la conquista dell’oriente.” Rendiconti della Pontificia Accademia Romana di Archeologia, Series III, 83: 3–57. Edwards, Catharine. 2003. “Incorporating the Alien: The Art of Conquest.” In Rome the Cosmopolis, edited by Catharine Edwards and Greg Woolf, 44–70. Cambridge: Cambridge University Press. Elsner, Jaś. 2002. “Introduction: The Genres of Ekphrasis.” Ramus 31: 1–18.

5.  Epigraphic strategies of communication

119

Elsner, Jaś. 2007. Roman Eyes: Visuality and Subjectivity in Art and Text. Princeton, Oxford: Princeton University Press. Elsner, Jaś. 2010. “Art History as Ekphrasis.” Art History 33: 10–27. Flaig, Egon. 2003. Ritualisierte Politik. Zeichen, Gesten und Herrschaft im Alten Rom. Göttingen: Vandenhoeck & Ruprecht. Flower, Hariet I. 1996. Ancestor Masks and Aristocratic Power in Roman Culture. Oxford: Clarendon Press. Flower, Harriet I. 2000. “The Tradition of the Spolia Opima. M. Claudius Marcellus and Augustus.” Classical Antiquity 19: 34–64. Flower, Harriet I. 2004. “Spectacle and Political Culture in the Roman Republic.” In The Cambridge Companion to the Roman Republic, edited by Harriet I. Flower, 377–98. Cambridge: Cambridge University Press. Galinsky, Karl. 2016. Memory in Ancient Rome and Early Christianity. Oxford: Oxford University Press. Goldhill, Simon. 2007. “What is Ekphrasis for?” Classical Philology 102: 1–19. Graverini, Luca. 2001. “L. Mummio Acaico.” Maecenas 1: 105–48. Hölkeskamp, Karl-Joachim. 2004. Senatus Populusque Romanus. Die politische Kultur der RepublikDimensionen und Deutungen. Stuttgart: Franz Steiner Verlag. Hölkeskamp, Karl-Joachim. 2006a. “History and Collective Memory in the Middle Republic.” In A Companion to the Roman Republic, edited by Nathan Rosenstein and Robert Morstein-Marx, 478–95. Oxford: Blackwell. Hölkeskamp, Karl-Joachim. 2006b. “Rituali e cerimonie ‘alla romana’. Nuove prospettive sulla cultura politica dell’età repubblicana.” Studi Storici 47: 319–63. Holliday, Peter J. 2002. The Origin of Roman Historical Commemoration in the Visual Arts. Cambridge: Cambridge University Press. Hölscher, Tonio. 1987. Römische Bildsprache als semantisches System. Heidelberg: Winter. Hölscher, Tonio. 2004. The Language of Images in Roman Art. Translated by Annemarie KunzlSnodgrass. Cambridge, New York: Cambridge University Press. Hölscher, Tonio. 2006. “The Transformation of Victory into Power: From Event to Structure.” In Representations of War in Ancient Rome, edited by Sheila Dillon and Katherine E. Welch, 27–48. Cambridge: Cambridge University Press. Hopkins, Keith. 1991. “From Violence to Blessing: Symbols and Rituals in Ancient Rome.” In City States in Classical Antiquity and Medieval Italy, edited by Anthony Mohlo, Kurt Raaflaub and Julia Emlen, 479–98. Stuttgart: Franz Steiner Verlag. Lahusen, Götz. 1985. “Das Bildnis des Konsul Cn. Cornelius Lentulus Marcellinus.” Archäologischer Anzeiger, 113–5. Lewis, R. Geoffrey. 2006. Asconius. Commentaries on Speeches of Cicero. Oxford: Oxford University Press. Ma, John. 2013. Statues and Cities. Honorific Portraits and Civic Identity in the Hellenistic World. Oxford: Oxford University Press. Oakley, Stephen P. 2005. A Commentary on Livy, Books VI–X, vol. II: Book IX. Oxford: Oxford University Press. Östenberg, Ida. 2009a. Staging the World. Spoils, Captives, and Representations in the Roman Triumphal Procession. Oxford: Oxford University Press. Östenberg, Ida. 2009b. “Titulis Oppida Capta Leget: the Role of the Written Placards in the Roman Triumphal Procession.” Mélanges de l’Ecole française de Rome 121–2: 463–72. Panciera, Silvio. 2012. “What is an Inscription? Problems of Definition and Identity of an Historical Source.” Zeitschrift für Papyrologie und Epigraphik 183: 1–10. Papini, Massimiliano. 2004. Antichi volti della Repubblica: La Ritrattistica in Italia Centrale Tra IV e II Secolo A.C. Rome: L’Erma di Bretschneider. Petrovic, Andrej, Ivana Petrovic and Edward Thomas, eds. 2018. The Materiality of Text: Placement, Perception, and Presence of Inscribed Texts in Classical Antiquity. Leiden, Boston: Brill.

120

Fabio Luci

Price, Simon R.F. 1984. Rituals and Power: The Roman Imperial Cult in Asia Minor. New York: Cambridge University Press. Revell, Louise. 2013. “The Written City: Political Inscriptions from Romana Baetica.” In Written Space in the Latin West, 200 BC to AD 300, edited by Gareth Sears, Peter Keegan and Ray Laurence, 231–46. London: Bloomsbury. Riggsby, Andrew M. 2006. Caesar in Gaul and Rome. War in Words. Austin: University of Texas Press. Robillard, Valerie. K. and Els Jongeneel, eds. 1998. Pictures into Words: Theoretical and Descriptive Approaches to Ekphrasis. Amsterdam: VU University Press. Roby, Courtney. 2016. Technical ekphrasis in Greek and Roman Science and Literature: The Machine between Alexandria and Rome. Cambridge, New York: Cambridge University Press. Ruck, Brigitte. 2004. “Das Denkmal der Cornelia in Rom.” Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 111: 477–93. Scullard, Howard H. 1981. Festivals and Ceremonies of the Roman Republic. London: Thames and Hudson. Sears, Gareth, Peter Keegan and Ray Laurence, eds. 2013. Written Space in the Latin West, 200 BC to AD 300. London: Bloomsbury. Sehlmeyer, Markus. 1999. Stadtrömische Ehrenstatuen der republikanischen Zeit. Stuttgart: Franz Steiner Verlag. Spannagel, Martin. 1999. Exemplaria principis: Untersuchungen zu Entstehung und Ausstattung des Augustusforums. Heidelberg: Verlag Archäologie und Geschichte. Squire, Michael. 2015. “Ecphrasis: Visual and Verbal Interactions in Ancient Greek and Latin Literature.” Oxford Handbooks Online, n.p. Sumi, Geoffrey S. 2002. “Spectacles and Sulla’s Public Image.” Historia 51 : 414–32. Tarpin, Michel. 2011. “Inscriptions républicaines et triomphe: ritual et obligations sociales.” In Corolla Epigraphica. Hommages au professeur Yves Burnand, edited by Carl Déroux, 683–99. Collection Latomus 331. Brussels: Latomus. Veyne, Paul. 1967. Le Pain et le Cirque: Sociologie Historique d’un Pluralisme Politique. Paris: Le Seuil. Wallace, Rex. 2011. “The Latin Alphabet and Orthography.” In A Companion to the Latin Language, edited by James Clackson, 9–28. Malden, Mass.: Blackwell. Warmington, E.H., ed. 1940. Remains of Old Latin, Volume IV: Archaic Inscriptions. Loeb Classical Library 359. Cambridge, MA: Harvard University Press. Webb, Ruth. 1999. “Ekphrasis ancient and modern: the invention of a genre.” Word & Image 15: 7–18. Zanker, Paul. 1988. The Power of Images in the Age of Augustus. Translated by Alan Shapiro. Ann Arbor: University of Michigan Press.

Chapter 6 Inscribing the artistic space: blurred boundaries on Romano-British tombstones Hanneke Salisbury

Introduction: the material and new approaches The art of commemoration is a rich source of imagery and epigraphy from the ancient world (perhaps providing up to 75% of our surviving epigraphic material),1 and tombstones from Roman Britain are no exception. Stone commemorative monuments with images and text were a new introduction to the region, and tombstones from the Empire’s most north-westerly province have long been the subject of academic interest. However, they have been viewed primarily as potential sources of information about the people of the region, while their artistic importance has been underestimated. This chapter seeks not only to exercise new approaches to the relationship between image and inscription on stone carved surfaces, but in doing so to demonstrate the potential for provincial material to further our understanding of funerary material more generally. To this end, in this section, I briefly outline the broader conversations taking place about the relationship between art and text and the nature of funerary epigraphy specifically, before turning to a series of case studies from Roman Britain. In my choice of objects, I focus primarily on those with inscriptions and images that include the human form. While we thus have a smaller dataset, it is one which, through the presence of both text and a sculpted body, foregrounds the relationship between the function of these monuments to commemorate the deceased and to communicate with viewers and mourners.2

  Beltrán Lloris 2014.   The presence of a carved body on a tombstone has a special significance in a funerary context, given the connection between it and the body of the deceased which may or may not have been present in some form, depending on the ritual used – cremation or inhumation. 1 2

122

Hanneke Salisbury

Breaking down boundaries between image and text In this chapter, I look at tombstones from Roman Britain through the lenses of two related trends in recent scholarship: the approach to text as a visual and material practice, and the more recent push towards a less dichotomous treatment of figural imagery and other decoration, including text. Key to both approaches is the insistence on the potential for text to have a combined visual and verbal impact. This can be framed in different ways. For example, Zahra Newby and Ruth Leader-Newby’s 2007 edited volume, Art and Inscriptions in the Ancient World, acknowledges the need for greater integration in the study of art and text, while contributions tended still to examine image and text separately, though in dialogue. For example, Glenys Davies’ chapter on Roman funerary art, particularly relevant to the current study, explores the frequent apparent disjuncts between art and inscription in that sphere, but does not go so far as to think of them as inextricably connected.3 For Davies, the impacts of text and image are still separate, even though they are given equal significance. Since then, others have explored the potential for bridging the gap between image and text. Michael Squire, in Image and Text in Graeco-Roman Antiquity, argues for a new understanding of the relationship between art and text, one which unmakes the divisions set up between them by Protestant theology’s influence on culture and academia. He explores not only the relationship between text and image where they exist together, but also how they work “through and against each other”, in different and related ways from one another.4 A similar approach continues in edited volumes by Verity Platt and Michael Squire, and by Squire and Nicholas Dietrich (The Frame in Classical Art (2017) and Ornament and Figure in Graeco-Roman Art (2018)). These volumes demonstrate the need to break down the divides between certain kinds of artistic material – between non-figural and figural art, and between the frame and the framed. It is that same impetus that drives this chapter, in an attempt to unite the surfaces of tombstones and to start treating them as artistic wholes that include text, figure, and non-figural decoration. One way to start doing this, and dissolve unproductive categories, is to examine text as a visual and material phenomenon, not just a verbal one. This is an approach that has been tried and tested with monumental architectural inscriptions. For example, Liz James has explored the ways in which text could function visually, as architectural decoration on Byzantine churches, both in the form of writing itself and in how it interacts with other forms of decoration.5 A similar approach was taken by Patricia Butz with regards to several architectural inscriptions from the Classical and Hellenistic Greek worlds. In this, she focuses particularly on letter forms and the 3   Davies 2007. At the end of her chapter, she does raise the possibility of connections between image and text which we simply do not appreciate (59). I would suggest that focusing on the visual dialogue between text and image is one way to get closer to those connections. 4   Squire 2009, 11. 5   James 2007.

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

123

visual effects of text such as the visual rhythm set up by spacing, the ways in which letter forms can echo some of the forms of architectural ornament.6 Another, related, approach to bringing the study of text into the realm of the visual is that taken by Kathryn Piquette and Ruth Whitehouse, and their contributors, in the edited volume Writing as Material Practice (2013). In this they argue that material – stone, paper, wood, wax – is more than a support for writing; it actually plays an active role in the construction of meaning.7 This volume brings together studies with a wide range of cultural and historical contexts. Chapters explore different writing styles, such as cursive, as well as different materials, like Minoan seals. Whitehouse’s own chapter is on funerary writing on Iron Age tombstones from Northern Italy. The writing on these stones goes around the sides of stone, and leads the viewer-reader up and down and around the stone. This reminds us that our concern should not only be with the materiality of writing, but with the physicality of reading – these require the involvement of the whole body, not just the eyes, to read.8 As this section has shown, there are many approaches out there that demonstrate the huge potential in opening up categories of image, ornament, and text. As we now turn to the material with which this chapter engages, we will see that the nature of tombstones from Roman Britain is such that they require us to ask new questions, creating the perfect environment for testing out new approaches to text and image. Approaches to Roman tombstones, their texts, and images It is not just that tombstones from Roman Britain are a useful case study through which to explore the issues of this chapter. Rather, the nature of these monuments, their particular commemorative properties and social associations, demand that we find new and better ways of dealing with the images and texts upon them. Scholarship on funerary material culture from the Roman Empire in general, as well as Roman Britain in particular, has often focused on what that material, designed to commemorate the dead, can tell us about the living. This can take the form of demographic studies,9 but also those that see the regional variation visible in this material as evidence for relative levels of “romanisation”.10 More recently, a chapter by Laura Chioffi in The Oxford Handbook to Roman Epigraphy focuses largely on how funerary epigraphy can be used to study self-representation and social status, as well as social history and demography, only devoting a small section to what it might tell us about the relationship to death and the dead.11   Butz 2009. See in particular the sections on a dedication by Alexander at Priene (32–3) which focuses on letter forms as themselves being ornamental and on the memorial of Theodoros, where letter forms are seen to echo ornamental shapes (37). 7   Piquette and Whitehouse 2013, 3. 8   Whitehouse 2013, 286. 9   E.g., Saller and Shaw 1984. 10   Henig 1995, 48; Hope 1997, 247–8. 11   Chioffi 2014. 6

124

Hanneke Salisbury

The idea that funerary epigraphy might tell us about the social status of the living has been used before with provincial material. The most extensive studies of this type have been conducted by Valerie Hope, on tombstones from North-Western provinces, including Britain. In a study of the funerary monuments of Mainz, Aquileia, and Nimes, Hope recognizes the importance of an approach that does not only focus on one aspect of these monuments, e.g. the inscription, but rather brings different elements together.12 To bring epitaph and image together is also her approach in earlier work on the tombstones of Roman Britain. In an article fittingly entitled “Words and Pictures: The Interpretation of Romano-British Tombstones”, Hope concludes that tombstones were largely a medium used by outsiders in the province as a way of asserting their identities in a new place.13 This line of inquiry, using tombstones to think about social structures and identity, is one explicitly followed by Geoff Adams and Rebecca Tobler in their 2007 monograph on the tombstones of Roman Britain.14 By looking at the whole dataset of tombstones and often taking a statistical approach, they are able to offer a broad picture of the kinds of familial and social systems that lie behind the images and epigraphy that survive on these funerary monuments. Their work is very useful for gathering information about who was being commemorated and by whom, as well as what form that commemoration took. There are, therefore, studies that have taken image and text together when examining tombstones from Roman Britain.15 A key difference in this chapter is the shift in focus from what tombstones can tell us about the society of the living to thinking about how image and text work together as effective commemorative tools, designed to captivate viewers and engage them in remembering the dead. In order to do this, it will be necessary to deal with their visual impact, something which means thinking about the forms and styles in which image and text are presented on these monuments. Style has long been a concern in studies of the epigraphy and art of the Roman provinces. How their style measures up against that of Rome has often been a primary focus. In Martin Henig’s 1995 synthesis of the artistic material from Roman Britain, the commemorative function of tombstones often takes a back seat to consideration of their style and their artistic merit. While he notes the dual function of funerary art both to commemorate and to enhance the status of the dedicator, more time is spent on tombstones’ style and artistic quality.16 The tombstone of Marcus Aurelius Nepos and his wife from Chester (Fig. 6.1), for example, is said to have “the charm and child-like quality of some seventeenth- and eighteenth-century village tombstones

  Hope 2001, 7.   Hope 1997, 255. 14   Adams and Tobler 2007. 15   In scholarship on the provinces, “romanisation” and cultural change as reflected and created by epigraphy loom large as concerns. See e.g., Häussler 2008. 16   Henig 1995, 64. 12 13

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones but it is hardly in the ‘high Roman manner’”.17 Henig then notes, “there is better work from Chester”.18 This kind of analysis leaves little room for how this tombstone worked to commemorate Nepos, or how the imagery conveyed meaning. Not only are images considered separately from their texts, but the function and effect of both is side-lined in favour of questions of quality. This has remained an issue for scholars. When investigating gender identity as manifested in images on tombstones, Maureen Carroll wrote that, “Often all we can say is that the inhabitants of Roman Britain wore tunics, coats and capes generally similar to those worn throughout Roman Gaul.”19 She focuses on the tombstone of Regina from South Shields, which has much more detailed depiction of adornment, thus suiting her purpose. Perhaps, as Carroll suggests, the less-detailed images on many tombstones from Roman Britain mean that these monuments can tell us little. This chapter, however, contends that both text and image on tombstones can tell us a great deal despite (or in fact because of) their appearance. We just need to change our questions.

Approaching the material

125

Fig. 6.1: Tombstone of Marcus Aurelius Nepos. Chester. Early third century AD. RIB 491; CSIR I.9, 33. © West Cheshire Museums, Grosvenor Museum.

So how do we actually go about examining image and text on the same surface while paying attention to the visual impact of both in such a way that we unite that artistic surface?20 That unity is crucial; seeing the whole of the tombstone at once allows all its elements to speak louder, and starts to move us away from the frequently piece-meal approaches of past scholarship. Thinking about images in terms of their form and style provides the basis for which we might then approach the text that accompanies them. This is in line with the approaches taken to monumental inscriptions before, but here it is tied more pointedly to how these texts work on the same plane as relief sculpture. In order for this approach to have real impact,   Henig 1995, 47.   Henig 1995, 47. 19   Carroll 2012, 290–1. 20   The use of the word “surface” here is in danger of flattening the three-dimensional materiality of the tombstones being examined. While the physicality of viewing and reading these monuments is, as we will see, crucial to understanding them, they are also in a sense planar in that they are covered in surfaces. This is an angle that could be a starting point for future study, to investigate further the ways in which these monuments’ three dimensionality affects the approach taken here. 17 18

126

Hanneke Salisbury

this must then be framed in the context of the object, which is in this case funerary. This brings us to the eventual, overarching question: how do the workings of text and image perform a commemorative purpose? There is significant stylistic variety among the tombstones of Roman Britain. This has been primarily explored with reference to the figural imagery upon them, as was seen above. There is a similar variety in the styles of lettering in the inscriptions accompanying this imagery. Lettering styles vary from the practically cursive to the fairly standard classically monumental, with a great deal in between. Script styles vary by region and time period throughout the Roman empire, for a variety of reasons. It has been noted, for example, that there appears to have been a tendency to move away from the consistent use of the classical “monumental square capitals” from around the first century AD.21 The stylistic variety in Britain, on tombstones that date from the first to the third centuries AD, should not then surprise us, nor should deviation from square capitals be necessarily considered as a mark of crudeness. In order to better appreciate the significance of each inscription, this chapter considers the effects that different kinds of text might have had in terms of their visual impact, rather than worrying as much about the circumstances of their production (textual errors, for example). Something that quite literally creates a boundary to uniting the surface of the tombstone is framing. When image or text is framed, it creates a separate space for that element, and appears to demand that we treat that element separately. Sometimes both are framed, apart from one another, so perhaps we can forgive the scholarly tendency to follow suit. The vast majority of tombstones from Roman Britain that have both figural imagery and an inscription have separate frames for image and text, but treating each separately ignores the multivalency of the act of framing.22 What if we think of frames not as dividing lines, but as linking, perhaps liminal, spaces between images and texts? Platt and Squire, in their edited volume on frames and framing in classical art, acknowledge the role of frames in delineating and categorizing space.23 But they also hint that frames might have the potential to call the divisions they help create into question.24 Examining inscriptions within mosaics, Sean Leatherbury has demonstrated how frames can in fact integrate texts into wider decorative schemes, not just separate them out.25 Frames certainly guide how we view, but they do so in ways that need not be clear-cut. By examining the visual impact of inscriptions on tombstones from Roman Britain, the aim here is to better explore how their whole inscribed and carved   Edmonson 2014, 124–5.   It is most common for both text and image to be framed. There are several examples in which the image is framed, but not the text, and only one in which the text is framed while the image has no frame. 23   Platt and Squire 2017, 12. 24   Platt and Squire 2017, 38. 25   Leatherbury 2017, 545–6. 21 22

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

127

surfaces functioned as commemorative monuments. This involves bringing together approaches that consider the materiality of these tombstones and their inscriptions, and what it means to carve into stone in specific ways, as well as work on the visuality of text, its styles, and forms. Exploring text and image beyond their content, and focusing on visual impact, will unite these artistic surfaces and help these understudied monuments to speak on their own terms.

Illustrating the issues: the tombstone of Flavia Augustina from York The shift I am suggesting, that we take the whole surface of a monument as one united whole, not only applies to the tombstones of Roman Britain but rather has the potential to be valuable in a much wider set of material. Sticking with Roman Britain, though, the approach can be well illustrated through the analysis of one funerary stele from York. It is a rich artistic and epigraphic surface; it is also a monument that not only demonstrates the pitfalls of an approach that foregrounds content and separates text and image, but can also guide us into asking different questions of such material. The tombstone pictured in Figure 6.2 (RIB 685) is from the city of York, and its front surface contains a relief with four figures within a double-arched niche with rosette motifs, below which is a framed inscription. I begin here by testing out a content-based examination of this tombstone’s text and image, beginning with the image, which meets the viewer roughly at eye-level. There are four figures, of various heights – we assume that the two tallest figures are adults, and that the smaller ones are children. Their clothing is barely differentiated from one another, an impression compounded by the use of dense linear patterning throughout, creating the sense of a single surface. But there are differences. The two figures on the viewer’s right are wearing a tunic under a pointed mantle, and a cloak over their shoulders, which they each hold in their right fingers at the neck. The adult on the left is wearing the same combination without the pointed mantle, and the child on the left appears to be wearing a tunic and cloak, though this figure does not hold their cloak at the neck. All wear boots. Both adults hold what appears to be a scroll; the taller child holds a round object, perhaps a ball or a piece of fruit, and the smaller one holds what may be a small bird. The faces of these figures are lost, which adds to a sense of sameness that is created by the repeated clothing assemblages and hand-positions. This in turn sets up a visual connection between these figures, who look, by virtue of this aesthetic unity, like a family. Who these figures are, including details of age (beyond relative age), gender, social status, occupation, and ethnic origin, is not communicated clearly by the relief on this tombstone. The inscription that accompanies this relief, in a separate frame below, gives the viewer a great deal more specific information about the deceased:

128

Hanneke Salisbury D M FLAVIAE AUGUSTINAE VIXIT AN XXXVIIII M VII D XI FILIVS SAENIVS AVGVSTINVS VIXIT AN I D III […]A VIXIT AN I M VIIII D V G AERESIUS SAENVS VET LEG VI VIC CONIVGI CARI [..]SIMAE ET SIBI F C

To the spirits of the departed (and) of Flavia Augustina; she lived 39 years, 7 months, 11 days; her son, Saenius Augustinus, lived 1 year, 3 days, and […]a, (her daughter), lived 1 year 9 months, 5 days; Gaius Aeresius Saenus, veteran of the Sixth Legion Victrix, had this set up for his beloved wife and himself.26

In this inscription, we are given the names of the deceased (though one child’s name is lost, the gender is strongly implied by the name ending in ‘a’), as well as their ages at death. This tombstone, we now know, commemorates a mother and her two young children, and was set up by her husband, who had served in the Roman army. The ages tell a story of children born one after the other, who did not survive infancy, and a mother also taken fairly young. Their names, too, have information to convey – this tombstone is one of only three from Britain to give evidence of children’s nomina being formed from their father’s cognomina.27 This is an alteration of the traditional Roman naming practice of the nomen being handed down through the generations, and also appears to be a factor in naming practices in the German provinces.28 This suggests the family were Roman citizens, but may not have their origins in Italy. Image and inscription both give us information, but of different kinds. The way in which the figures are depicted in the relief is such that shows them to be a family unit through visual cues of sameness and unity. This is compounded by the shapes of their bodies – the so-called “bell shape” – which is repeated, and in turn echoes the shape of the double-arched niche that houses this family. But there is no specificity or individuality in this image – and that appears to be the point. The inscription gives information about the specific relationships between the people commemorated by this monument (including the dedicator, whose memory is just as preserved as those whom he had lost). But there is a disconnect here. It is not only the case that different information is given by text and image, but that this information is not compatible.29 From the inscription we have learned that both children died in infancy, yet neither is represented as an infant. This in itself is not unusual in art of the Greco-Roman world, in which it was not uncommon, even in the Roman period, for children to   Text and translation taken from RIB.   Birley 1966, 229. The other examples are RIB 690 and RIB 67. 28   Birley 1966, 229. 29   Rinaldi Tufi 1983, 24. Rinaldi Tufi suggests that this disparity is down to the monument having been pre-fabricated, and the inscription added at the point of commission. This is unlikely given the extreme rarity of tombstone reliefs from Britain with as many as four people depicted – this is one of two (see also RIB 688). 26 27

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

129

Fig. 6.2: Tombstone of Flavia Augustina and Family. York. Mid-third century AD. RIB 685; CSIR I.3, 39. Image courtesy of York Museums Trust, https://yorkmuseumstrust.org.uk, CC BY-SA 4.0.

130

Hanneke Salisbury

appear as little adults.30 The issue with marrying text and image becomes clearer, though, when we ask the question of gender for this image. It has been suggested that pointed cloaks are a male clothing item. RIB identify the adult figure on the right as male, and this would suggest that the child dressed in the same clothing should also be male. Looking again at the inscription, there is a disconnect – not only are the children not depicted as infants, but (if we have the gendering right) the child who was died youngest is depicted as significantly taller than his sister. Perhaps, given the artistic tradition of depicting children as little adults, their age is entirely irrelevant and we should not worry that the relative ages of the children do not seem to match in image and text. But issues continue to be raised if we think a little harder about the gender of these figures. Identification of male figures on the basis of a pointed cloak is widespread, but by no means watertight as a method. A tombstone from Ilkley (RIB 639), showing a woman seated and wearing a pointed cloak, casts some doubt over this as a reliable marker of gender. The objects held by the adult figures are also worth a closer look. The editors of RIB identify the figure on the right as male, and holding a scroll. They then suggest that the figure they see as a woman is holding a bird. In fact, there is no sign of a bird in her hands; she (if she is a she), appears to be holding a scroll as well. The figure on the left is also the taller of the two, which might lead some to imagine this figure as the man of the family. If this image tells us anything, it should caution us about using certain criteria to “gender” figures represented in art. But this does all set up a further disjunct between text and image – the inscription is very sure of who is being commemorated, and of the precise details of their ages, down to the day they died. The image insists that we do not, and cannot, know.31 One way in which the relief and the inscription do appear to marry up, if we put our worries about gendering the figures to one side, is in putting a focus on one figure in particular. Flavia Augustina’s name is in larger lettering than the rest of the inscription, and occupies the first line.32 If we accept that the taller figure on the left represents Flavia Augustina, the wife and mother, then her presence is emphatic in both image and inscription. Otherwise, we are left with a text and an image that project very different pictures from one another. So how are we to reconcile them, given that the aim here is to unite the surfaces of tombstones, to break down the boundaries between text and image? Each is separately framed here, and so are 30   Uzzi 2005, 12–14 traces the development of images of children from classical Greece to the Roman Empire. Mander 2013, 19–20 notes the frequent ambiguity in the depiction of age in representations of children in Roman funerary art. Vout 2010 examines the case of the children in the Laocoon group, exploring reasons why it might be more effective in this case to represent them not as children, but as youthful adults. 31   A further caveat exists in the fact that the faces of these figure are very worn – perhaps there would have been more certainty if they had survived – a beard on one of the adult’s faces would clearly help us a great deal here. Hairstyle can potentially be of use here, while worn, the hair of the figure on the left looks long, and might suggest this figure is female. 32   These letters are around 5.8 cm high, the rest are around 3.8 cm.

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

131

potentially asking to be looked at one after the other. And yet the inscription and the relief together make up the sum total of this monument to a lost family. It is time now to turn away from content, and instead look to form and visual impact in search of an answer to the question of how we might bring these apparently disparate pieces of commemorative surface together. What happens if we put aside our desire to read and learn from text and images and focus instead on their visual effects? The overriding sense conveyed by the relief here is one of unity – this is the effect of depicting figures with bodies of the same shape, with very similar gestures, all standing frontally, looking out at the viewer. The inscription below is dense, and is quite uniform, apart from the size difference in some of the letters. On a funerary monument, an epitaph as lengthy and as dense as this is effective as a demonstration of the magnitude of this loss – the weight of the writing en masse has a great impact by itself. The preponderance of numbers among the dense text immediately speaks of the kinds of details included. There are interpuncts in the form of dots in between each word, and the repetition of this indicates, even without reading or understanding a great deal of the content, that there is a lot to know about the unified family in the image above. Of course, it is tempting to mine this inscription for information – the inclusion of the dedicator, with details of his occupation, signals a desire to convey his status, as well as to commemorate those himself and those he has lost. The young ages of those who have died tug at the heartstrings. But we miss so much if we ignore the potential for the visual impact of the writing to have an emotional force as well. The density of the writing does suggest that there is a lot of information being imparted, and this aspect of the writing’s visual impact contrasts with the unity projected by the relief. But they do have something in common: both image and inscription confront the viewer with the volume and the weight of the loss of this family. The tombstone of Flavia Augustina and her family demonstrates how, even where text and image appear in some ways at odds, there are varied and multifaceted ways in which text and image in the same space can actually work together to present a unified message. The rest of this chapter continues to prioritize visual impact over legible content to explore the productiveness of looking at text and image on tombstone surfaces in explicitly visual terms. The next section looks at one particular element – framing – to explore how frames and framing devices influence the ways in which we view, and demand that we see text and image in a more integrated way. After that I turn to the style and form of text, both on the level of letter forms and inscriptions as a whole, to see how they fit into the overall carved surface of tombstones.

The frame as divider and unifier Frames of many kinds abound on the surfaces of tombstones from Roman Britain, as they do in the visual and material culture of the classical world more broadly. As the recent volume edited by Verity Platt and Michael Squire outlines and addresses, the

132

Hanneke Salisbury

significance of these practically ubiquitous devices has been under-appreciated for the roles they play in how we view images and inscriptions.33 This section explores how frames guide how we view tombstones, their reliefs, and inscriptions. The types of frames used can be significant on their own, but function is more important than type here.34 Frames, in this chapter, mean not just lines that surround image or text or both. Here I mean also the kinds of decorative elements often considered secondary to figural and textual adornment, which can also fulfil a framing role. I begin with an example that demonstrates the potential for such decorative elements to unite the artistic space of the tombstone, bringing harmony to compositions that include text and image. The tombstone of Vellibia Ertola (RIB 1181; Fig. 6.3) was found in Corbridge in 1937, and is dated to the late third or early fourth century AD by the style of the letter forms.35 In a gabled niche stands a figure wearing a belted tunic and holding a round object. Below this is an inscription, beginning with the formula “D M” or Dis Manibus, which is inscribed into a protruding ledge. The rest of the inscription is below this, three lines of it within a rectangular panel and the final two below, outside the panel.36 The framing of this stone’s surface is therefore quite complex. The gabled niche is a common frame type in tombstones from Roman Britain, and suggests a kind of architectural space, perhaps even a sacred space, which the dead now inhabit.37 This motif has a long history, going back to Attic grave stelai.38 While the figure might thus be imagined as inhabiting a space within the stone, the inscription remains less clear-cut, given the way in which it twice breaks any frame – once with “D M”, again by expanding out of the frame below. This suggests that any interpretation of this tombstone that deals with image and text separately is deviating from the unity that the spatial aspect of the text demands – there is not, in fact, a defined space for text as separate from image. It breaks the bounds of the frame set out for it, leaking into the rest of the stone. The presence of “D M” on the ledge between text and image planes is also significant. Not only does this provide a physical bridge between relief and inscription (just as the address to the spirits of the dead asks to create a bridge between this world and the next), but the use of non-figural decoration compounds this effect. Phillips suggests that the holes above and below the niche were intended

  Platt and Squire 2017a.   Examples exist from British tombstones of tabulae ansatae used as frames for epitaphs (see RIB 961 and RIB 2003 from Carlisle and Castlesteads respectively). This form of frame has been extensively studied for its implications in different contexts by Leatherbury 2017; 2018. 35   Phillips 1977, 28. 36   It appears there was an earlier inscription beneath this one, which fitted into the rectangular panel, and was erased. The relationship between the content of text and image here cannot, therefore, be taken for granted. This should not prevent us from exploring the visual unity of this artistic surface, and its visual impact as it stands. 37   Thomas 2000, 17–18. 38   Squire 2017, 160–2. 33 34

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

133

Fig. 6.3: Tombstone of Vellibia Ertola. Corbridge. Late third or early fourth century AD. RIB 1181; CSIR I.1, 71. Now in Corbridge Museum (CO23338); image reproduced with the permission of the curator.

134

Hanneke Salisbury

to contain hooks on which garlands could be supported.39 This would create a living version of the rosettes so often seen in funerary art, from Britain and beyond. But they might also have an effect beyond that. The repetition of rounded forms in the depiction of the figure – their small circular head, rounded shoulders, and round object held within the hands form a harmonious geometrical group – are echoed by the round holes in the gable of the niche and on the ledge below. These holes are a visual bridge between the forms of commemoration here, between the image and the text. They ensure that the surface is a fluid one, in which text and image demand to be seen as part of the same space, more because of than in spite of their frames. In the case of Vellibia Ertola’s tombstone, not only did the text break its bounds, but non-figural decoration created a sense of a united surface. In our next example, RIB 538 from Chester, text and image are framed in ways that make such a fluid and united space seem impossible (Fig. 6.4). The use of frames in this monument helps to guide our viewing in ways that can only be appreciated when we look at the surface as a whole. In this stone, a bust is framed in a niche at the top, flanked by two open-mouthed lions. The form of the bust, and its framing, immediately set up a relationship to sculptural display within the monument, a kind of mise en abyme, an image within an image.40 Below this, in a separate frame, is a scene in which a horse and rider appear to move to the viewer’s right, with a captive or attendant walking in front. The “sedate pace” at which horse and rider advance makes this a different kind of cavalry tombstone from those which show a rider with spear raised, and horse rearing.41 What is most interesting about the way in which the inscription is framed is the emphatic inclusion of horizontal lines between the lines of the text – the guidelines put in to help the inscriber keep it straight have been made permanent and significant parts of the inscription here.42 The appearance of lettering and image echo one another to an extent; diagonal lines criss-crossing in the horse’s legs and the “X”s and “V”s in “SEXTUS SEXTI FILIVS”. And yet the insistence of framing not just image and text separately, but each line of the inscription, has an effect other than uniting the field. Instead, the guidelines and the use of mise en abyme suggest that the focus here is on the artifice, the process of making, and the craftsmanship, of this stone. It wears its materiality on its sleeve, placing a heavy weight on the surface of the stone as a space of artistic and epigraphic display. The text has enough solid materiality here to support the weight of the horse and rider that walk over a base, which is, in appearance, just another epigraphical guideline. All this is watched over by Sextus in   Phillips 1977, 28.   A particularly emphatic example of this kind of reference to sculptural display on a funerary monument is the Tomb of the Haterii from Ostia; Trimble 2018 explores the role of framing, ornament, and mise en abyme there. 41   Henig 2004, 19. Examples of the latter type from Britain include RIB 3185 from Lancaster, RIB 121 from Gloucester, and RIB 201 from Colchester. For a detailed study of this iconography see Kramer 2014. 42   Such guidelines, as they survive in the epigraphic record, testify to the act of ordinatio, of laying out the space for, and the text of, an inscription before the task of carving it could begin (see Edmonson 2014, 117–21). 39 40

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones bust form from his architectural façadelike tomb top. We have seen how frames and other framing devices guide our viewing, and create monuments with very different effects. But what about the relatively few cases in which either text or image, or both, have no frame? One such example is the tombstone of Lucius Valerius Aurelius from Gloucester (RIB 3074; Fig. 6.5). Lucius, with heavy drapery held up in his hands, and lines on his face that match up with his veteran status described in the inscription, stands facing to the viewer’s left. Like so many other figures, he stands within a gabled niche. Unlike so many, the inscription that devotes this stone to his memory is above, not below, the frame in which his body is depicted. In a similar way to the tombstone of Vellibia Ertola above, the formula “D M” is positioned on a boundary between the planes occupied by body and text, this time the “roof ” of the gabled niche. This means that, rather than being read at the beginning of the epitaph,43 the formula interrupts the name of the deceased:

135

Fig. 6.4: Tombstone of Sextus Simil[…]. Chester. Second or third century AD. RIB 538; CSIR I.7, 54. Photo: Corentyn Smith © West Cheshire Museums, Grosvenor Museum.

L VAL D M AURELIUS VET LEG XX

The inscription, thus broken up, frames the head of the deceased with his name and status, creating an inextricable bond between the figure in the image and the inscription. In this case, then, the lack of a frame, as well as the unusual positioning of the text, makes for a very effective monument to the memory of this man. The depiction of lines on his face, unusual in funerary images from Britain, suggests a greater level of individuality than in many other tombstones, which reinforces the focus on this man and his identity created by having his epitaph framing his head. The epitaph at the top of the stone is not 43   Henig 1993, 49 writes about this inscription as though “D M” is above the rest (and it is similarly presented in RIB) when in fact it is on the same line as the name of the deceased.

136

Hanneke Salisbury

the only inscription on this monument. There is another below the figure of Lucius, carved in a tabula ansata frame. Henig presumes this to be the name of the dedicator (only the letters “IVL” survive), which seems a reasonable prospect.44 This part of the dedicatory inscription is not only separate from that which gives the deceased’s name and status, but is also in a separate frame, one which is heavy with potential meanings.45 The use of space, and of framing (both framing text and using text as a frame) ensures that each part of the inscription is allowed to convey its different force. The upper text, surrounding Lucius’ head, ensures the identification of the deceased, and frames his body as the focus of the monument. The lower inscription fulfils the role of Fig. 6.5: Tombstone of L. Valerius Aurelius. ensuring the memory of the generous Gloucester. Second century AD. RIB 3074; CSIR dedicator and possible heir, which was I.7, 142. Figure from Hassall and Tomlin 1984, no. another function of a tombstone, besides 1, reproduced with permission. commemorating the dead.46 Frames delineate and define artistic and inscribed space, but they also have visual properties themselves, and have the ability to make connections as well as divisions. Here, they have worked alongside text and image to convey meaning, as well as bringing text and image together. Frames therefore can be and do so much, and even though the next section is not explicitly focused on them, they nevertheless remain in the background, helping to guide the viewing and commemorating happening on the surfaces of Roman Britain’s tombstones.

Style and surface The style of text and image, particularly image, can be a troublesome line of inquiry in scholarship on Roman provincial art. So often, it dissolves into questions of skill   Henig 1993, 49.   The tabula ansata began as a form of votive tablet in wood, and made its way into metal and stone in official monuments, military inscriptions and, eventually, tombstones (Leatherbury 2017, 556–9). Leatherbury 2017, 559 suggests that the form itself could signal the presence of a text. 46   Henig 1995, 64; Adams and Tobler 2007, 12–13. RIB 2003, commemorating Gemellus, makes use of both features explored here. The names of both deceased and dedicator are framed in a tabula ansata, which is placed behind Gemellus’ head. 44 45

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

137

and artistic quality, and of how these pieces measure up against the art of Rome.47 Here, then, the individual stylistic elements of a given piece are not at issue, nor is their relationship to the Greco-Roman artistic tradition. Instead, the focus is on how the styles of text and image create an artistic and inscribed surface that reinforces certain commemorative strategies. The first example given leans heavily on styles and motifs from Classical art, but that does not make its insights any more valuable, or its effect any more pointed, than those of the second which looks very distant indeed from the art of Greece and Rome. The tombstone of Philus (RIB 110; Fig. 6.6), from Cirencester and dated to the first century AD,48 is a mixed bag in many ways both in terms of its imagery and its inscription. The text is in Latin, as the vast majority of epitaphs from Roman Britain are,49 but it provides us with the information that Philus is a “CIVIS SEQV(anus)”, a citizen of a region in Eastern France, near the border with Switzerland. The image of Philus is housed within a heavily classicizing architectural frame, and executed with a relatively high level of naturalism, particularly in the treatment of the drapery, and yet he is wearing a cloak commonly seen in the art of the North-Western provinces. So how are we to understand the visual impact of this tombstone? The Corinthian pilasters and pediment complete with rosette decoration and pinecone finial that frame Philus here are very much in line with the style of the text in which his epitaph has been carved. It may be that the herringbone patterns on their side of the pediment are supposed to represent a roof in perspective, which would add another dimension, quite literally, to the architectural façade depicted.50 The letters are not quite the monumental square capitals of Augustan epigraphic history, but they are in line with the general trend towards the use of “librarial script” through the first century AD in monumental inscriptions, with narrower letters and a less rigid and geometric form of lettering.51 They are somewhere in between these categories (which are fluid in any case), and what matters is that they appear monumental, and official. The kind of lettering used on this tombstone fits in with the elaborate architectural façade that has been carved into this stone. In fact, the monumental lettering style ensures that the rest of the architectural façade appears as such, and also appears as a separate artistic space from that within the frame, where Philus stands “at the entrance of his tomb”.52 Non-figural framing elements, including text, here create the effect that Philus exists within a defined architectural space, one that is apart from 47   A key, and recent, example is Ben Croxford’s creation of a “class system” (Croxford 2016, 604). Other work, which acknowledges that the work of unskilled craftsmen is still useful, has been done by Catherine Johns (Johns 2003). Johns still suggests that the ideals of Greco-Roman art are “the only criteria available” for judging the art of the provinces (Johns 2003, 18). 48   Henig 1993, 49. 49   Two examples in Greek from London are likely later imports (RIB 2318 and 2320). There is also the famous Palmyrene inscription on the tombstone of Regina from South Shields (RIB 1065). 50   Henig 1993, 49. 51   Edmondson 2014, 124–5. 52   Henig 1993, 49.

138

Hanneke Salisbury

the one in which the viewer is standing. This is a version of an artistic commemorative strategy that is widespread in Roman (and Greek) funerary art, which is to place the image of the deceased in a frame that suggests the tombstone is a separate space: sometimes a pedimented temple, as here, other times a house, with truncated figures represented as though looking out of a window.53 A different effect is created by the style of relief and inscription in our next, and final, example. The carving on the tombstone of Mamma (RIB 3398; Fig. 6.7) from Carvoran, a fort on Hadrian’s Wall, has none of the classicizing features or naturalism of that of Philus.54 The head of the carved figure is missing, and the level of detail on the rest is low. The figure is wearing a short tunic, and could be male or female, though Coulston and Phillips lean towards female.55 They have their right arm raised, and the left next to the body, holding what may be a bag. The feet are shown in profile, facing the viewer’s Fig. 6.6: Tombstone of Philus. left. Below, in a roughly rectangular recessed frame, Cirencester. First century AD. RIB is the inscription, dedicated to Mamma, the coniunx 110; CSIR I.7, 141. © Gloucester sanctissima of Balbius Duianus. The letter forms Museums Service. here are narrower than those of Philus’ epitaph, and have a great deal more in common with cursive script, particularly in the style of “M”s and “A”s, which are repeated in the name of the deceased, following “D M”, in the first line of the inscription.56 The crooked style of the lettering echoes that of the relief above, particularly the line of the figure’s left arm which zig-zags down the side of the body – the frame also follows this line. The cohesion between the styles of text and image here is significant, but not necessarily meaningful in itself. Instead, the way in which the panel of text recedes into the stone, and the flat carving of the figure, practically invisible from the side, sets up the surface of this stone just that – a surface, in which what remains of the figure in the relief resides. The cursive style of the writing draws attention to its nature as writing, and to the act of writing, as opposed to a monumental

  Erasmo 2008, 160.   Coulston and Phillips 1988, 90 call the carving “shallow and crude”. 55   Coulston and Phillips 1988, 90. 56   Cursive here means not “joined up” writing, but the forms of individual letters that have the appearance of a scrawl, and might look “scruffy” (Johnston 2013, 193). 53 54

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

139

inscription which becomes an object in itself.57 The effect of writing and relief here creates a very different vessel for the memory of Mamma than that set up for Philus, but both use the visual impact of their texts and images to fulfil their commemorative function. The use of certain styles in text and image can usually tell us little on its own, but when elements of a tombstone are put into dialogue, as here, and considered for their visual impact, style can speak alongside other elements in order to convey meaning and to create monuments that various artistic strategies in order to commemorate the deceased most effectively. Crucially, the ways in which a tombstone’s surface is composed and styled sets up what kind of space that surface entails, and thus what space the deceased’s representation now inhabits.

Conclusions and new directions

Fig. 6.7: Tombstone of Mamma. Carvoran. Third century AD. RIB 3398; CSIR I.6, 90. Image Credit: R.S.O. Tomlin, reproduced with permission.

We have seen how focusing on the content of images and inscriptions, and neglecting their capacity to make visual statements, can seriously restrict the meanings that we can extract from these monuments. And yet, as the chapter has gone on, content has proved difficult to leave aside. Ultimately, a greater balance is required in how we approach this material, so that we can better marry content, form, and style together. In attempting to provide a corrective to approaches that neglect the visual, this chapter has shifted the weight to visual elements in the hope that their value can now be appreciated, and more balanced readings can be offered in the future. By focusing in on framing, as well as the visual impact of form and style with regards to text and image, this chapter has united the surfaces of the tombstones looked at here. Crucial to this has been treating frames as bridges, rather than as dividing lines, and showing that what we think of as a frame (a linear, often rectangular border) is a restrictive category. Decorative features, and text itself, can have a framing function, and by recognizing this we are better equipped to appreciate tombstones

57   Leatherbury 2017, 547 makes a comparable argument about tabulae ansatae frames and those that resemble architectural facades, calling them “object frames”.

140

Hanneke Salisbury

from Roman Britain as artistic wholes. Once we do this, and allow the visual effects of these monuments to be felt, we can see how the appearance of lettering and images can drastically alter the viewer’s relationship to a given monument and thus their relationship to the dead commemorated by it. This has by no means been a comprehensive look at the ways in which text and image work on the tombstones of Roman Britain. Rather, this chapter aims to provide a template for how we might better approach inscribed spaces as united wholes as opposed to fragmented into images, ornamentation, and text. Tombstones from Roman Britain have proved to be a fruitful area of study in this regard, but they are by no means unique in the ways in which they use image and text to commemorate the dead. This chapter is novel not only in the ways it approaches its material, but in taking the provincial as a starting point and using data from Roman Britain to prompt and to pave the way for more studies. Much, if not all, of the ideas put forward here for how we read and view funerary monuments could equally be applied to such material from different times and places in the ancient world, potentially beyond it. But the new approaches outlined in this chapter are necessary if we are going to take provincial material culture on its own terms. It has been shown here that, when we are able to do this, the material has a great deal to say.

Bibliography

Adams, Geoff W. and Rebecca Tobler. 2007. Romano-British Tombstones Between the 1st and 3rd Centuries AD: Epigraphy, Gender, and Familial Relations. British Archaeological Reports 437. Oxford: Archaeopress. Beltrán Lloris, Francisco. 2014. “Latin Epigraphy: The Main Types of Inscriptions.” In The Oxford Handbook of Roman Epigraphy, edited by Christer Bruun and Jonathan Edmonson, 89–110. Oxford: Oxford University Press. Birley, Eric. 1966. “Review of The Roman Inscriptions of Britain I: Inscriptions on Stone, by Robin George Collingwood and Richard Pearson Wright.” Journal of Roman Studies 56: 226–31. Butz, Patricia. 2009. “Inscription as Ornament in Greek Architecture.” In Structure, Image, Ornament: Architectural Sculpture in the Greek World: Proceedings of an international conference held at the American School of Classical Studies, 27–28 November 2004, edited by Peter Schultz and Ralf von den Hoff, 30–9. Oxford: Oxbow Books. Carroll, Maureen. 2012. “‘The Insignia of Women’: Dress, Gender, and Identity on the Roman Funerary Monument of Regina from Arbeia.” Archaeological Journal 169 (1): 281–311. Chioffi, Laura. 2014. “Death and Burial.” In The Oxford Handbook of Roman Epigraphy, edited by Christer Bruun and Jonathan Edmonson, 627–48. Oxford: Oxford University Press. Collingwood, Robin George and Richard Pearson Wright. 1965. The Roman Inscriptions of Britain. Oxford: Clarendon Press. Coulston, Jon C. and Edward John Phillips. 1988. Corpus Signorum Imperii Romani, Great Britain Vol. 1 fasc. 6. Hadrian’s Wall West of the North Tyne and Carlisle. Oxford: Oxford University Press. Croxford, Ben. 2016. “Art in Roman Britain.” In The Oxford Handbook of Roman Britain, edited by Martin Millett, Louise Revell and Alison Moore, 599–618. Oxford: Oxford University Press. Davies, Glenys. 2007. “Idem ego sum discumbens, ut me videtis: Inscription and Image on Roman Ash Chests.” In Art and Inscriptions in the Ancient World, edited by Zahra Newby and Ruth E. Leader-Newby, 38–59. Cambridge: Cambridge University Press.

6.  Inscribing the artistic space: blurred boundaries on Romano-British tombstones

141

Edmonson, Jonathan. 2014. “Inscribing Roman Texts: Officinae, Layout, and Carving Techniques.” In The Oxford Handbook of Roman Epigraphy, edited by Christer Bruun and Jonathan Edmonson, 111–30. Oxford: Oxford University Press. Erasmo, Mario. 2008. Reading Death in Ancient Rome. Columbus: Ohio State University Press. Hassall, Mark W.C. and Roger S.O. Tomlin. 1984. “Roman Britain in 1983. II. Inscriptions.” Britannia 15: 333–56. Häussler, Ralph, ed. 2008. Romanisation et épigraphie: études interdisciplinaires sur l’acculturation et l’identité dans l’Empire romain. Montagnac: Monique Mergoil. Henig, Martin. 1995. The Art of Roman Britain. London: Routledge. Henig, Martin. 2004. Corpus Signorum Imperii Romani, Great Britain Vol. 1 fasc. 9. Roman Sculpture from the North West Midlands. Oxford: Oxford University Press. Henig, Martin, Graham Webster, and Tom F.C. Blagg. 1993. Corpus Signorum Imperii Romani, Great Britain Vol. 1 fasc. 7. Roman Sculpture from the Cotswolds Region. Oxford: Oxford University Press. Hope, Valerie. 1997. “Words and Pictures: The Interpretation of Romano-British Tombstones.” Britannia 28: 245–58. Hope, Valerie. 2001. Constructing Identity: The Roman Funerary Monuments of Aquileia, Mainz and Nimes. Oxford: J. and E. Hedges. James, Liz. 2007 “‘And Shall These Mute Stones Speak?’: Text as Art.” In Art and Text in Byzantine Culture, edited by Liz James, 188–206. Cambridge: Cambridge University Press. Johns, Catherine. 2003. “Art, Romanisation, and Competence.” In Roman Imperialism and Provincial Art, edited by Jane Webster and Sarah Scott, 9–23. Cambridge: Cambridge University Press. Johnston, Alan. 2013. “Straight, Crooked, and Joined-up Writing: An Early Mediterranean View.” In Writing as Material Practice: Substance, Surface, and Medium, edited by Kathryn E. Piquette and Ruth Whitehouse, 193–212. London: Ubiquity Press Limited. Kramer, Jessica Colleen. 2014. The Roman Riders: Ethnicity and Iconography on Roman Cavalrymen Tombstones. MA diss., Brigham Young University. Leatherbury, Sean V. 2017. “Writing, Reading, and Seeing Between the Lines: Framing Late-Antique Inscriptions as Texts and Images.” In The Frame in Classical Art: A Cultural History, edited by Verity Platt and Michael Squire, 544–81. New York: Cambridge University Press. Leatherbury, Sean V. 2018. “Framing Late Antique Texts as Monuments: The Tabula Ansata between Sculpture and Mosaic.” In The Materiality of Text: Placement, Perception, and Presence of Inscribed Texts in Classical Antiquity, edited by Andrej Petrovic, Ivana Petrovic, and Edmund Thomas, 380–404. Leiden: Brill. Mander, Jason. 2013. Portraits of Children on Roman Funerary Monuments. Cambridge: Cambridge University Press. Newby, Zahra and Ruth E. Leader-Newby, eds. 2007. Art and Inscriptions in the Ancient World. Cambridge: Cambridge University Press. Phillips, Edward John. 1977. Corpus Signorum Imperii Romani, Great Britain Vol. 1 fasc. 1. Corbridge, Hadrian’s Wall East of the North Tyne. Oxford: Oxford University Press. Piquette, Kathryn and Ruth Whitehouse, eds. 2013. Writing as Material Practice: Substance, Surface, and Medium. London: Ubiquity Press Limited. Platt, Verity and Michael Squire, eds. 2017a. The Frame in Classical Art: A Cultural History. New York: Cambridge University Press. Platt, Verity and Michael Squire. 2017b. “Framing the Visual in Greek and Roman Antiquity: an introduction.” In The Frame in Classical Art: A Cultural History, edited by Verity Platt and Michael Squire, 3–101. New York: Cambridge University Press. Rinaldi Tufi, Sergio. 1983. Corpus Signorum Imperii Romani, Great Britain Vol. 1 fasc. 3. Yorkshire. Oxford: Oxford University Press.

142

Hanneke Salisbury

Saller, Richard P. and Brent D. Shaw. 1984. “Tombstones and Roman Family Relations in the Principate: Civilians, Soldiers, and Slaves.” Journal of Roman Studies 74, 124–56. Squire, Michael. 2009. Image and Text in Graeco-Roman Antiquity. New York: Cambridge University Press. Squire, Michael. 2017 “Framing Bodies: Introduction.” In The Frame in Classical Art: A Cultural History, edited by Verity Platt and Michael Squire, 256–69. New York: Cambridge University Press. Squire, Michael and Nikolaus Dietrich, eds. 2018. Ornament and Figure in Graeco-Roman Art: Rethinking Ontologies in Classical Antiquity. Berlin: De Gruyter. Thomas, Thelma K. 2000. Late Antique Egyptian Funerary Sculpture: Images for this World and the Next. Princeton: Princeton University Press. Trimble, Jennifer. 2018. “Figure and Ornament, Death and Transformation in the Tomb of the Haterii.” In Ornament and Figure in Graeco-Roman Art: Rethinking Ontologies in Classical Antiquity, edited by Michael Squire and Nikolaus Dietrich, 327–52. Berlin: De Gruyter. Uzzi, Jeannine Diddle. 2005. Children in the Visual Arts of Imperial Rome. Cambridge: Cambridge University Press. Vout, Caroline. 2010. “Laocoon’s Children and the Limits of Representation.” Art History 33 (3): 396–419. Whitehouse, Ruth. 2013. “‘Tombstones’ in the North Italian Iron Age: Careless Writers or Athletic Readers?” In Writing as Material Practice: Substance, Surface, and Medium, edited by Kathryn Piquette and Ruth Whitehouse, 271–88. London: Ubiquity Press Limited.

This is an Open Access article, distributed under the terms of the Creative Commons 3.0 Unported Licence http://creativecommons.org/licenses/by/3.0/. A copy may be downloaded from http://books.casematepublishing.com/Dynamic_Epigraphy.pdf

Chapter 7 When poetry comes to its senses:* inscribed Roman verse and the human sensorium Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

Introduction Some mistakes are very easy to make, very easy to overlook, and at the same time very obvious, as soon as one starts thinking about them. One of these mistakes – not of great consequence in the grander scheme of things, but sufficiently problematic within our academic field – is the tacit assumption that the way in which we encounter inscribed monuments and objects, and the way in which we perceive ancient Roman texts and art, is somehow more or less identical to the way in which they were approached, perceived, handled, and consumed in the ancient world. The mistake, resulting from a lack of reflection, becomes apparent as soon as one starts considering the – very obvious! – fact that no Roman ever read Vergil or Ovid in the beautifully produced, thoroughly researched editions of the Bibliotheca Teubneriana or the Oxford Classical Texts series, much less in bilingual editions such as those of the Loeb, Budé, or Tusculum libraries, or even in translation, as those of us who consume the provisions of the Penguin or Reclam libraries for quick and easy access. In fact, even if these and other esteemed publishing houses and series had already existed in the ancient world (for which there is no credible evidence), the experience would have been altogether different due to the different media formats and types that were in common use in the ancient Mediterranean. It may be tempting to think that, while this is evidently the case for the literary and technical texts of the Roman world that were transmitted through a manuscript tradition, the same may not be true for texts transmitted on durable objects such * This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 832874 – MAPPOLA). Early stages of this paper were developed during a British Academy Mid-Career Fellowship held by P.K. in 2014–5 and, in part, presented on occasion of the 2018 Celtic Conference in Classics, at the University of St. Andrews: comments and feedback by the audience were gratefully received.

144

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

as inscriptions or coins (or, to a lesser extent perhaps, on papyrus). This, too, is a mistake, however. First and foremost, students of ancient inscriptions today typically encounter the objects of their studies through editions, either digitally or in print. The number of individuals who encountered inscriptions through the ancient counterparts of our editions was, though not zero (there are several ancient sources that transmit inscriptions as part of their text),1 insignificant. The substantially more common experience, however, was one of an encounter in situ, just as we navigate the lettered world that we inhabit ourselves today. At the same time, it is important to exercise caution: even cultural practices that we might be inclined to consider universal and lasting – such as reading, writing, or singing – may not necessarily have been (and, in fact, in many ways were not) carried out, perceived, and thought of in the same way that we imagine and implement them in our own societies today. An especially obvious and well-researched, yet still often neglected and underappreciated case is that of the cultural practice known by the deceptively familiar term “reading”. In most contexts, from our own experience, we are used to individuals reading in silence, and, as anyone who has ever encountered anyone mumbling and reading aloud in a library or in public transport will know, violations of that unwritten rule of cultural behaviour are met with significant dismay. As a number of important studies have established with certainty, this was not the way in which those who were able to read in the Greco-Roman world(s) read: the act of reading typically, and for the longest time, was one of reading out loud,2 adding an acoustic element to the visual experience of the practice of reading that we ourselves (presumably) do not commonly associate with it, and certainly do not associate with it to the same degree that the people(s) of the Greco-Roman world(s) did. This issue becomes even more complex and important in such circumstances in which a number of cultural practices are intertwined, most notably in the case of, though not at all restricted to, the practice of inscribing verse: here the production and presentation of verbal art manifests itself in the epigraphic habit, and it is vital to get closer to an improved, more thorough understanding of how such art was perceived, encountered, and met by those exposed to it – those, who did not read it in digital or printed editions, potentially in translation, and those who encountered these texts in situ, in their original contexts, rather than in a book or in display in a potentially beautiful, yet still de- and re-contextualizing museum display. How were these texts experienced and “felt” by those who were trying to make sense of

1 There is substantial literature on the use(s) of inscriptions, actual and fictive, in ancient literature. A useful and still up-to-date collection of contributions on this matter, from a wide range of perspectives, can be found in the edited volume of Liddel and Low 2013. 2 See the important study by Gavrilov 1997 (with Burnyeat 1997). For a more recent, complex treatment of the matter see Busch 2002. On reading in non-typographical societies cf. also Berti et al. 2015.

7.  When poetry comes to its senses

145

the inscribed environment that they inhabited?3 These are the questions that merit attention and further study, and it seems sensible to approach this matter in a way that follows the five senses of the human sensorium, re-establishing the sensuous experience, or at least its conceptualizations, of encounters with inscribed poetry (and in many cases: with the inscribed world in general).

Sight The most obvious, primary way in which anyone, at any given time, would encounter inscribed poetry is, of course, visual. Sight is, however, a significantly more complex experience than commonly considered, and productively taken into account, in relevant scholarship on the matter.4 Recent Carmina Epigraphica research has investigated a wide range of aspects related to the visual and graphic presentation of inscribed poetry. In particular, scholarship has focused on the paradigms and strategies behind the ways in which poetic texts were commonly presented, both on their own and in relation to the inscribed monuments as well as, where applicable, prose elements of more complex textual compositions.5 The (narrow) focus on the actual object, the micro-context of an actual inscription and its wording in relation to the inscribed monument, is only one aspect that matters when it comes to the visual experience of inscribed poetry, however – and in that it also is the one that is most closely related to our own visual experience, as we still encounter texts from up close, in situ just as much as in museums, collections, and galleries.6 This would not, however, be the way in which people in Roman times initially encountered both inscribed objects and, more specifically, instances of verbal art. Rather, the entire encounter is a complex process, escalating (potentially) through a significant number of steps from distance to intimate proximity and familiarity. Approach and distance/proximity are only two aspects, however, in a much more complicated framework still. Monuments, portable objects, mosaics, once they are placed (or at least considered) in their original context, interact with their experienceable surroundings in a complex and intricate manner, and they

Further on the notion of a “lettered world” in Roman antiquity see Kruschwitz 2016a. Important observations on enclosed spaces, such as columbaria tombs, and the specifics of locally found Carmina Latina Epigraphica can be found in Massaro 2006. 4 Landmark monograph-length studies on this aspect were produced by Squire 2009 and Eastmond 2015, respectively. More generally on the role of sight in the context of ancient discourse(s) about the human sensorium see Squire 2016. A number of aspects considered in this section have also been treated by Schwitter 2019. 5 Further on the relevance of text layout in the study of the Latin verse inscriptions see del Hoyo 2002; Kruschwitz 2008; and, most notably, the extensive study by Limón Belén 2014. 6 On the meaning of the context for the act of reading and the comprehension of the text see Berti et al. 2015, esp. 641–2, 644–6. 3

146

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

are subject – looking at visual matters alone for now – to a (potentially) carefully designed and shaped presentation and (potentially) changing light conditions.7 From the ever-changing and ever-shifting sunlight to artificial light sources such as lamps, candelabras, and torches, these variable sources of light provision contribute to the experience of visual encounters with inscribed poetry just as much as they impact on one’s mood and perception: of course it matters a great deal whether one reads a collection of poems in romantic candlelight whilst seated in a comfortable armchair, potentially accompanied by a glass of wine, or under cold, fluorescent striplight in the noisy, draughty waiting area of an airport. It is not unreasonable to assume that, from the way in which letters were cut and text was presented on a monument down to the actual placement of texts, e.g. in tombs or in inhabited quarters, lighting matters were quite carefully considered by those who designed inscribed objects and monuments. This, however, is radically and emphatically different from modern-day experiences of inscriptions, especially in sterile museum environments or (sometimes less sterile) archives and storage facilities, in which the light commonly provided is steady, neutral, and unlike any ancient type of light. The situation does not fundamentally change in research settings, in which tools such as torchlights are employed to bring out details of letters and the monuments overall. One could push this catalogue of once important, now commonly neglected, visual factors even further, and thus increase the complexity and accuracy of our appreciation of ancient monuments, by including into the mix the way in which texts interact with sculpted or paved ornaments and even the properties of the materials that were employed in crafting them. The relevance of such aspects is obvious, of course, and it is even possible to adduce evidence for an ancient awareness of such relevance. Perhaps one of the most striking examples is a text from third-century AD Bowness-on-Solway in the province of Britannia:8

[- - - ?] [Ant?]onianus dedico [s]ed date, ut fetura quaestus suppleat uotis fidem: aureis sacrabo carmen

Some considerations on the environmental factors that could influence the reading of an inscription are discussed in Susini 1968, 74–8; cf. also Susini 1982, 56, where he refers to CIL V 74 as an example of an “iscrizione incisa su superficie aggettante e inclinata, in modo da meglio incontrare l’occhio del lettore” (“an inscription engraved on a protruding, slanted surface in order to support the reader’s eye”). More specifically, cf. Di Stefano Manzella 1987, 158–9, on the red colouring (rubricatura) of incised letters. Here, he mentions the possibility of colour and/or plastering employed to correct mistakes in the text. This, in turn, is explored further by Orlandi and Panciera 1999, 583 (with n. 29). Finally, Bolle 2019, 193–249 analyzes the difference between “materiality” and “presence” and studies the latter in several contexts, such as public urban spaces (focusing on the case studies of Aquileia and Ostia), churches and that of invisible inscriptions. 8 See Kruschwitz 2020, 194–5 for a recent discussion of this piece. 7

7.  When poetry comes to its senses

147

5 mox uiritim litteris. I, [Ant?]onianus, dedicate [- - -]. But give that an increase in earning fills my vows with credibility, (and) soon I will dedicate this song with golden letters, (executed) one by one. (CIL VII 952 = CLE 229 = RIB I 2059)

Here, the dedicant, one Antonianus (?), apparently a tradesman or merchant with a desire to increase his fetura quaestus through prayer, marked his initial request for higher profits with a verse inscription, thus making the stone and its poetic decoration a valuable and valid gift to the (in the piece’s present state of preservation) unspecified god(s). At the same time, he promised a second votive of this kind, offering to boost the object’s value through a precious embellishment of the text by means of filling each individual letter with gold.9 Colour, material, and ultimately value of every letter counts, and a piece of art, such as a poem,10 increases in value when it is executed in a more precious material.11 It is difficult, though by no means impossible, to accumulate additional useful data, and to extract further meaningful information, about visual aspects from the monuments that survive in situ and even from those in collections. What is substantially more difficult, however, is to show conclusively that the motivations behind certain decisions in design and presentation were, in fact, the same ones that are stated in the inscription from Bowness – and that they were the same ones, consistently, in all relevant cases, even if no such thing has explicitly been stated. Methods of experimental archaeology (or epigraphy, more specifically) might be usefully employed in future research on this aspect. A fine example of how experimental epigraphy is able to yield important clues, and one that is especially relevant to the study of inscribed verse, lies in an experimental approach to the oft-repeated assumption that certain texts placed visual markers at their beginning to reach out, and to reel in, their future audiences – using phrases such as asta ac pellege, which are attested in numerous variations hundreds of times across the empire. If one puts this view to the test, simply by walking down a Gräberstraße at a reasonable (i.e., not-too-high and not-too-low) pace, walking past a number of 9 One can be certain that this refers to another (promised) inscription, not the present one. No further inscription by the same dedicant has been found so far, which could either mean that the deities did not find in Antonianus’ favour or, equally likely, that someone eventually found what appeared to be a better use for the golden letters than permanently to remain on display. 10 The idea that texts themselves become precious manifestations of art is a concept that is developed much further in subsequent periods and contexts; see e.g., Rhoby 2017. 11 This notion also finds its way into funerary carmina, where it is used as a form of laudatio. See, for example, CIL XI 6551 = CLE 1088: (...) si meritis possem dare | munera tantum | quanta tibi debent|ur praemia laudis | aureus hic titulus et | littera nominis auro | condecorata legi deb|et (...) (“If only I were able to present you with such gifts as you deserve as prizes for the praise that you receive, this inscription would have to be made of gold, and the writing of your name should be read in golden decoration”). Further on this topic see Cugusi 1980–1981 and Hernández Pérez 2001, 185–6 (§ 199).

148

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

inscribed monuments, e.g., at the cemeteries of a town like Pompeii, one realizes that not only it is impossible to make out any such phrase with sufficient certainty, but that it would be even more impossible to perceive this as any form of imagined competition of tombs reaching out for attention over one another.12 So what is the point of this phrase (and its like) then? Is it possible to imagine that their actual point was not, in fact, a visual one at all? Much recently published research has focused on visual clues in the text layout to communicate the poetic nature of inscribed texts to their respective audiences, separating prose from verse, and highlighting metrical and other compositional structures to an audience.13 Significantly less attention has been paid to the various forms of reading that may occur, for the first encounter with an inscribed poem, one may safely assume, was hardly ever one of a thorough reading of the text in its entirety and with sufficient attention to detail. It is a long, not always linear, not always successful road from (i) a first encounter and realization that there is, in fact, a text to be read to (ii) the realization that parts of the text, if not all of it, were actually written in verse to (iii) a potential appreciation of the underlying rhythm (which, incidentally, may be a completely overrated aspect in modern-day research as compared to the relevance that this aspect held for those who read the text in the ancient world) to (iv) an initial skim-reading to (v) a potential more detailed (though not necessarily complete, but still selective) engagement with the actual wording of the text.14 A second aspect that deserves significantly more attention than it has thus far received is the way in which inscribed poetry shapes, and interacts with, its material and immaterial surroundings. This does include, but is by no means restricted to, the under-researched area of text-monument and text-image interactions in the Carmina Epigraphica:15 this important aspect affects the way in which poetic texts enable(d) their readers to read the entire surrounding landscape just as much as to elucidate structure and notable elements of the texts themselves. A particularly obvious case are acrostics, mesostics, and telestics: allusions to these elements in the texts themselves are frequent,16 similar, in a way, to the not altogether uncommon For further considerations on this topic in the context of Pompeii, see Kruschwitz 2016b. See above, n. 5 and cf. esp. Limón Belén 2014, 43–53 and 96–100 on the devices used to distinguish verse from prose in the verse inscriptions from Rome and Hispania. 14 For a more detailed discussion of the levels of proximity involved in ancient modes of reading inscribed texts, as well as their reflection in terminology that is used in this context, see now Kruschwitz 2019, esp. 346–55. 15 Further on this see Feraudi-Gruénais 2017. 16 Besides the list of acrostics included by Bücheler in the indexes to his two volumes (CLE p. 920), see Krummrey 1963, 285 n. 20 and Barbieri 1975, 364–71, who updated that corpus. Further on the topic, with new updates and extensive bibliography, Sanders 1979 = 1991, 183–205. On acrostic Carmina Latina Epigraphica see the superficial study by Amante 1913 and Zarker 1966; on Greek and Latin acrostics, see Courtney 1990. There are a few cases of verse inscriptions that exhibit both acrostics and telestics: Colafrancesco 2012 studies CLE 1977 as one such example (Mauretania Caesariensis). See also del Hoyo 2014a, who discusses two relevant medieval carmina from Hispania. 12 13

7.  When poetry comes to its senses

149

poetic desire, including in verse inscriptions, to draw attention to puns, wordplays, and etymologies.17 There is more, however. Note, for example, direct references to “aboves” and “belows” vel sim. in texts themselves. An especially remarkable example of that is the following one, from Brixia (Brescia):18 5

L(ucius) Naeuidius C(ai) f(ilius) Fab(ia tribu) (sex)uir Aug(ustalis), sibi et Vitali, Faustae, Cnomini, Fidae libertis t(estamento) f(ieri) i(ussit). ḍẹịṇde hoc elogium breue: [Diu uiuen]ḍo multa uîdi încom(m)od(a). [apto requieui f]essus âeuo tempore.

Lucius Naevidius, son of Gaius, of the tribus Fabia, sevir Augustalis, ordered to have this made according to his testament for himself and the (or: his?) freedwomen (?) Vitalis, Fausta, Gnome, and Fida. Then this little poem: During my long life, I encountered many adversities. At an appropriate time, tired of this age, I retired. (CIL V 4445 = CLE 142 = Inscr. It. X 5.234 = EDR090234)

Line 7 preserves an element that in the context of the inscription itself is completely meaningless: deinde hoc elogium breue, “then this little poem”. Any reader can see that (i) it is a poem, (ii) it follows what was written before, and (iii) it is short. The only reasonable explanation is, of course, that this was meant to be a directive given to the stonecutter to indicate the tentative layout intended by those who created the overall text and contemplated its final design for the purpose of ideal consumption by subsequent audiences. At the same time, it gives us a clear indication that linear nature and consumption of inscribed texts and poetic elements were carefully planned and structured.19 Touching on visual matters and interactions between monuments and readers, it may well be possible to advance even further, however, and to explore if there was any concept of gazes that were, in fact, mutual, directed not just from the reader at the monument, but also in the other direction. A fascinating case in that regard can be seen in the following monument, originating from Beja in the province of Lusitania. Fashioned in the (epigraphically) somewhat rare(r) rhythm of Phalaecian

Further on etymologising names in the Latin verse inscriptions see Michalopoulos 1997. Further on this inscription see Masaro 2017, 378–81 no. 96. 19 This notion is perhaps explored to the fullest within the realm of figurative poetry. See, for example, Schmidt 2019 on CIL VIII 28110. 17 18

150

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

hendecasyllables, this piece renders its readers almost observed in their responses and actions, as the deceased comments on likely emotions with a set of instructions:20 Quisq(uis) praet[eris hic] sitam, uiato[r, postquam] termine legeri[s mori] me aetatis uicesim[o] 5 dolebis. etsi sensus er[it] meae quietis qu(a)e lasso tibi dulcius precabor. uiuas pluribus et diu [se] nescas qua m[ihi non] 10 [l]icu[it] fruare uita. [si] [t]e flere iuuat quitn(i) inge[m] [i]scis? ann[- - -]. Inachus han[c] m[e] [ri]to fac(it). i, potius propera: nam [tu] legis, ipse legeris: i! Nice a(nnos) XX u(ixit). Whoever passes me by, as I lie here, traveller, you will feel pain [after] you have read that I [died] at my life’s twentieth boundary-marker. Even though you will experience a sensation of my own rest, I will pray for a sweeter one for yourself, as you are weary. May you live for many years and grow older still for a long time: enjoy a life which I was denied. If you feel like crying, why don’t you start? […] years. Inachus had (this monument) made deservedly. Go! Nay: hurry! You read, you will be read yourself! Go! Nice lived 20 years. (CIL II 59 = CLE 1553 = IRCP 270 = HEp 2006.440 = AE 2006.564 = HEpOl 21142)

The poetic I in this text represents young Nice who had died at the age of 20, expressed both factually and through the remarkable phrase termine … peremptam … aetatis uicesimo, at life’s 20th boundary marker. Nice instigates communication with her audience in a common first-person narrative. The passer-by thus becomes a reader, imagined to interrupt their hurry to another place. The speaker thus purports to know that the traveller must be tired and weary, which makes her wishing them “a sweeter rest” than her own. Obviously, the wordplay with quies, denoting “rest” just as much as “death”, cannot be lost on any reader, and it will remain an ominous presence throughout the poem. The highpoint of the exchange, however, lies in a double suggestion of the speaker’s awareness of the reader’s disposition and action(s), resulting in both a description of what is imagined to be perceived and in instructions given to the reader. First, Nice predicts the reader’s emotional disposition, resulting in her inviting them, while maintaining her overall compassionate and comforting tone, to express their sadness and mourning if they so feel. The observation that the reader may well 20 Further on this piece see Pena and Carbonell 2006, 157–73. We follow their edition of the text (rather than Bücheler’s outdated and more problematic version).

7.  When poetry comes to its senses

151

feel like crying at this point (yet potentially feel ashamed to do so), [si] | [t]e flere iuuat, is followed by an encouraging permission: quitn(i) inge[m]|[i]scis (ll. 10–12). Secondly, there is the poem’s closing part, which prima facie would see to resort to a topical message: potius propera nam | [tu] legis ipse legeris, “Hurry! You read, you will be read yourself ”, releasing the audience from the exchange and restoring them to their busy everyday lives. This commonplace phrase, however, has been framed in a double imperative i, “go”. Thus the speaker has been fashioned by the author of this text to exhibit an awareness of how a reader of the text, only just invited to stall their journey, might, in fact, now be very reluctant to leave. In its syntactic and pragmatic design, the next carefully varies and regulates the pace of the instructions given, in triplicate, to the reader through the set of directive expressions: “go, nay: hurry … now go!”

Sound More than in virtually all other contexts of Roman epigraphy, with the possible exception of curse tablets, the acoustic aspect21 is of the utmost importance when it comes to inscribed verse. Since the process of close and careful reading in the Greco-Roman world was – as already stated above – one of reading the text out loud, there is no need to discuss the general relevance of the acoustic dimension of inscribed Roman verse. One must bear this aspect in mind, however, and ponder its implications, when interpreting texts such as the aforementioned Nice epitaph: if an inscribed text, in verse no less, is written with a first-person speaker role, this will, by default, and inevitably so, become a role, almost theatrically, to be taken on by, and associated with, the reader. In funerary environments, such as the one imagined in the Nice epitaph, the living reader would thus, in actual, become the deceased incarnate – a living surrogate of, and proxy for, the deceased in the environment of the burial. But first things first. The most obvious manifestations of sound in Roman verse inscriptions must, of course, be seen in their extensive use of stylistic devices such as assonances, alliterations, homoioteleuta, and so forth – features that link inscribed manifestations of Roman verse with general practice in Greco-Roman poetic style. These elements become all the more apparent in instances such as the following one, from Pompeii (Fig. 7.1). In this piece visual clues of a figurative poem (here: a snake, to invoke the serpentis lusus of a performer called Sepumius) are combined with sound elements (here: repeated use of hissing S-sounds, wherein art emulates serpents’ nature): 21 For a recent and comprehensive collection of studies on this topic see Butler and Nooter 2019. Cf. also Webster 2019, 119–24, specifically on how magic words on healing amulets or curse tablets in the Greco-Roman world had no specific meaning, but reproduced sounds whose repetition would grant the desired effect.

152

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

Fig. 7.1: CIL IV 1595. Line drawing from CIL. [Ser]pentis lusus si qui sibi forte notauit Sepumius iuuenis quos fac(i)t ingenio spectator scaenae siue es studiosus equorum sic habeas [lan]ces se[mp]er ubiq[ue pares]. Should someone happen to notice the serpent’s playfulness that young Sepumius makes in his witty manner, be you a spectator of stage performances or keen on horses, may you thus have your scales in balance, always, everywhere. (CIL IV 1595 (cf. p. 209, 463) = CLE 927 = EDR170352)

The soundscape of Roman verse inscriptions is neither one-dimensional nor exclusively linked to the decoration of the text by means of striking stylistic features, however. To appreciate the significance – and power – of texts that are being read out loud, it is instructive to ponder the mythical story of the apple of Cydippe: Cydippe’s creepy admirer Acontius threw an inscribed apple in front of her, and when Cydippe picked up this apple and read the inscription (aloud!), this immediately set in motion a spell.22 It is with that in mind that one ought to consider the relevance of sound, not only for the readers themselves, but for those surrounding them. The (ancient) reality of experiencing epigraphic verse through an acoustic channel is explicitly mentioned in an also otherwise extraordinary inscription from Sulmo (Sulmona), Ovid’s birthplace. In this piece, a man called Murranus presents himself 22

For a comprehensive treatment of this story cf. Kruschwitz 2010a, 46–51.

7.  When poetry comes to its senses

153

as a barbarian from Pannonia, who tragically lost all of his six children, and even contemplated ending his own life, had it not been for the survival of a grandson. In this (in every respect) unique text, with an underlying iambic rhythm, Murranus eventually addresses his audience in second- or third-century AD Sulmo: (...) rogo at nu[nc] inprecamus deos ut si quis hoc sephulcr[um] 40 aut hunc titulum laeserit in[tulerit illi?] fortuna mala et quod mer[itu]m sit [hunc] titulumque quicumque legerit aut lege[ntem] ausculta(ue)rit alleuet illos for[tuna] superior et ualeant semper [in aeterno?] 45 quicumque in hoc titulo scrip[ta legerit] quietis sit uobis terra leuis et [- - -] desperatum qui superant [- - -] tempore obito sit [- - -]. I ask: let us now implore the gods that, if anyone damages this tomb or this inscription, they may thrust ill fate on such a person and whatever else is deserved; but whoever reads this inscription or listens to someone reading it out, may a more desirable fate comfort them and may they flourish forever and ever (?): whoever reads what is written in this inscription, may you find peace and may earth be light on you. (… ) (CIL IX 7164 = Suppl. It. 4 (1988), Sulmo 58 = AE 1989.247 = EDR114466, ll. 38–48)

Not only does this passage contain a vital clue as to how those whose literacy levels were below the required standard to engage with inscriptions and inscribed verse as readers were still able to engage and to participate in a meaningful way (namely acoustically):23 it also (i) demonstrates that reading out texts aloud was a common, anticipated practice, and (ii) provides a useful idea of how a more substantial audience could be attracted by, and get involved in, the re-enactment of a text that initially served a very specific purpose in the act of (in this case) a burial. What is more, once this acoustic dimension is taken into account, assigning an audible performance to a monument and its text, it also becomes abundantly clear that objects do not by themselves reach out to their audiences: phrases such as asta ac pellege, as mentioned above in the section on sight (see also n. 12), become specific and meaningful as soon as one imagines a reader who produces this text in a form that allows others to hear what they are reading (and therefore saying).24 The ritual dimension of such acoustic productions and re-enactments of monumental verse (and texts more generally) must not be underestimated. This affects a wide range of issues, not least the widely observed habit to produce iscrizioni This was duly noted by Bodel 1993. To this one should add the more than likely presence of ambient noises such as crying, lamenting, or even songs and music from a nearby pompa funebris. Further on these, as elements of the funerary ceremony, see for example Wille 1967, 65–73 and, more recently, Carroll 2011. 23 24

154

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

parlanti, i.e., texts that would appear to speak in the first-person:25 this habit becomes a lot less puzzling once one bears in mind that the text would be read out loud by someone who thus would make a statement in the first person about themselves (or the individual they represent in their role as readers). This leads to a second, even more significant matter, namely the observation that there are several instances in which Carmina Epigraphica in the funerary record do not only imagine, but clearly articulate, the notion of the living speaker effectively taking on the (almost theatrical,26 certainly ritualized) role of the deceased, restoring their voice in the act:27 carpis si qui [uia]s, paulum huc depone la[borem]. cur tantum proper(as)? non est mora dum leg(is), audi lingua tua uiuum mitique tua uoce loquentem. oro libens libe[n]s releg(as), ne taedio duc(as), amice If you there seize these ways, let go of the stress for a short while: why such a rush? There is no time wasted while you read: listen to a living person who talks in your tongue and with your gentle voice. I ask you to read this favourably, favourably, so that you will not derive dislike, my friend. (CIL XI 627 (cf. p. 1236) = CLE 513, ll. 1–4)

This passage does not only provide evidence for an imagination and conceptualization of the dead as communicating through the voices of the living. It goes beyond that through its qualification of the reading voice as mitis … uox, a “gentle voice” to be employed by the reader, adding tone (warmth and calm) to the disembodied and reembodied transmission from beyond the grave. The types of voice modulation requested by the dedicants of readers of funerary monuments vary, and in several instances they are seen to reflect the circumstances and personality of the deceased. An excellent example of that is the (unfortunately lost) inscription for Calethyce (sic!), from Hispalis (Sevilla), in the province of Baetica:28 5

[C]alethyc(h)e ann(orum) XXII. Siste gradum quicumque precor paulumque morare [di]sces summisso carmina maesta sono infelix primo flore sepulta soror mirum [- - -] uiuus et bis ut undenos inuida morte premor [M]anibus ara m[eis - - -]ta in[- - -] [- - -] funera nostra [- - -]

On “iscrizioni parlanti” see Agostiniani 1982 and, for a historical perspective, Gregori 2008. For a comprehensive study of interlocutor arrangements in the Latin verse inscriptions, see Socas 2002. 26 In that regard, the use of death masks in certain circles may have added to the theatrical, performative nature of the act; cf. also Flower 1996, 114–15. 27 Further on this notion see Carroll 2007–2008. 28 Further on this inscription see Martín Camacho 2010, 249–62 (SE7). 25

7.  When poetry comes to its senses

155

10 [- - -]la quicumque ui[debit] [- - -] cineri terra sit usq(ue) [leuis] [- - -] miseris etiam solacia pra[estat - - -] [et quos turb]a terit laeta frequenter [- - -] infelix quod sum [- - -] uide 15 [- - - - - -] S(it) t(ibi) t(erra) l(euis). Calethyce, aged 22. Stall your step, please, whoever you are, and rest a little: you shall hear sad verses, subdued in tone. My unfortunate sister has been buried here, in the flower of her youth … wondrous… alive … So, as at the age of twice eleven years, I am in envious death’s firm possession, … an altar to the Spirits of the Departed … our death … whoever beholds …: ‘May the earth forever rest lightly on your ashes.’ … (offer) the wretched these words of consolation also . … crushing … happy frequently, as I am ill-fated … behold … May the earth rest lightly on you. (CIL II 1094 = CLE 1195 = HEpOl 499)

Passers-by who chose to engage with Calethyce’s monument encountered a deeply moving poem, in sad verses, and they did so equipped with the advice to read those lines softly with a subdued voice (summisso … sono, l. 4). The reader, reading the text by and for themselves, though within earshot of further passers-by, thus adopts the voice of young Calethyce, keeping not only her memory, but her very voice alive from beyond the grave. The dead person thus remained an audible, though obviously not tangible, presence even after their burial, giving wayfarers a credible reason to express the ubiquitous wish “may earth rest lightly on you” directed at the deceased. Very little is known about performative practicalities behind the apparent connection of verse (especially when produced by the human voice) and music. Verse inscriptions that were read out loud thus fall somewhere into a not altogether well understood grey area between rhythmical text and song. Songs and music, however, are complementary aspects of an integrated Roman soundscape that is now almost completely lost to us. Work songs, theatre, holy music, children’s rhymes and funerary music were, as in every society, part of everyday life.29 It is perfectly possible that individuals produced verse inscriptions with some kind of melody (in addition to a verse rhythm) – an aspect that is significantly more tangible in the Greek hemisphere, whence annotations of verse inscriptions are known. This aspect was not irrelevant to the authors of Latin verse inscriptions, however, as a funerary monument from the Roman cemetery of Aquincum (Budapest), in the province of Pannonia goes to prove. This text takes its readers to a world of musical professionals in the context 29 These aspects are covered, passim, in the already mentioned unsurpassed evidence collection produced by Wille 1967. For a more recent approach to this topic see Perrot 2020.

156

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

of the Roman army. Its text, consisting of hexameters and a closing elegiac couplet, reads as follows:30 5

Clausa iacet lapidi (!) coniunx pia cara Sab[[d]]ina. artibus edocta superabat sola maritum. uox ei grata fuit, pulsabat pollice cordas (!). set (!) cito rapta silet. ter denos duxerat annos, heu male quinque minus, set (!) plus tres meses (!) habebat, bis septemque dies uixit. Hec (!) ipsa superstes spectata in populo hydraula grata regebat. sis felix quicumque leges (!), te numina seruent, et pia uoce cane: Aelia Sabina uale. T(itus) Ael(ius) Iustus hydraularius salariarius leg(ionis) II Ad(iutricis) coniugi faciendum curauit.

Locked under this tombstone my dear and devoted wife Sabina does lie. She was educated in the arts, she was the only one who managed to overcome her husband. She had a charming voice, she plucked the strings with her fingers, but now, snatched quickly, she is silent. Twenty five years she had lived, alas!, and three months and fourteen days. While still alive, she was popular, admired as a magnificent organist she ruled everyone. You who read this, whoever you are, be happy: may the gods protect you, and may you sing with a respectful voice: Farewell, Aelia Sabina. Titus Aelius Iustus, water-organ player and salariarius of the legio II Adiutrix, had this made for his wife. (CIL III 10501 = CLE 489 = HD068458)

The funerary monument, datable to the early third century AD, was dedicated by one Titus Aelius Iustus, ex-soldier of legio II Adiutrix, who had been recalled to his unit, probably with the non-military task of playing the hydraula. Iustus praises his wife Aelia Sabina, who, according to his commemoration of her, surpassed him in the art of playing the hydraula as well as in her singing. Aelia Sabina’s death is described by him in a brief and evocative way, explicitly referring to matters of acoustics and music a number of times. Sound and music come to an abrupt end, however, through Sabina’s sudden death: sed cito rapta silet. But this silence is not a lasting one. In the last two verses of the inscription, forming an elegiac couplet (and thus a change in tune also from a rhythmical standpoint), Iustus addresses his readership and asks them to greet his dead wife, and to do so through singing with a respectful intonation of the voice: pia uoce cane. Manifestations of the human voice are not the only factors, however, that constitute the soundscape that surrounds the Latin verse inscriptions. Ambient noises contribute a great deal, whether they are those of nature or those of settlements, and, of course, these matters are taken into consideration and conceptualized on occasion. One such circumstantial trigger, responsible for an entire set of verse inscriptions, in Greek and

30 Further on this piece see TitAq II 519; Courtney 1995, 114–15, 324 no. 115; Cugusi and Sblendorio Cugusi 2007, 82–5 (CLEPann 36); cf. also http://lupa.at/3025 (last accessed: October 2020).

7.  When poetry comes to its senses

157

Latin, is the “singing” of the Memnon statue at the Theban necropolis at Luxor.31 To mention but one example from this context:32 Memnonis [Aethiopis uocem] clarumque sonor[em] exanimi inanimem mi[ssum] de tegmine bruto 5 auribus ipse meis cepi sumsique canorum praefectus Gallorum al[ae] praefectus item Ber(enices) Caesellius Quinti f[il(ius)] 10 A Barmo[- - -]. (or: A Bararo?) I myself absorbed and took in with my own ears the voice of Aethiopian Memnon and the distinct harmonious sound, animate though emitted from an insensible, inanimate surface – I, Caesellius … , son of Quintus, prefect of the troop of Gauls and prefect of Bere(?)nice. (CIL III 55 = CLE 272; transl. E. Courtney, modified)

This piece would be remarkable enough in and of itself for its extensive reference to sounds and imaginations of animated objects, of course. In the present context, however, what is especially noteworthy is the clear reference to the sound as something that must be seen in connection with (i) the presence of the author and (ii) the production of the very poem, which responds to what was heard from an object that is imagined to keep making such noises in the future.33 Acoustic elements are mentioned as essential for the setting and experience of verse inscriptions in conjunction with additional sensory factors in a number of occasions and in a number of contexts. An example of that can be seen in verse inscriptions that originate from, and are related to, the Roman bathhouses. Here, both tactile and auditory elements appear simultaneously (and are presented as complementary),34 e.g. when texts refer to the cold water (with a power to extinguish the warming fire)35 in For a recent edition and discussion of these texts (after that of Bernand and Bernand 1960) see Rosenmeyer 2018, who provides a comprehensive overview of the extant texts on pp. 211–40. 32 Further on this piece see Bernand and Bernand 1960, 54–6 no. 14; Courtney 1995, 90–1, 298–9 no. 76; Cugusi and Sblendorio Cugusi 2011, 178, 216–17 (CLEOr 39). 33 An especially remarkable case is that of CIL III 77 = III 12076 = CLE 271 from Talmis (Kalabsha, Egypt), which, like many other inscriptions, refers to the sounds of the Memnon colossi that, as is implied, went silent, and resumed activity, in response to political changes. Further on this inscription cf. Courtney 1995, 50–3, 245–6 no. 26. 34 On verse inscriptions from, and more generally ancient poetry related to, bathhouses see Busch 1999. 35 Cf. e.g., CIL VIII 25362 (= CLE 1754 = 2039 = ILS 8960 = ILCV 787 = ILTun 1154 = ILPBardo 432 = AE 1908.29 = 78 (add.) = 1999.1758 (add.) = 2016.1832 (add.), from Tunis), IRT 918 (= 919 = Zarker 1958, no. 21 = Cugusi 2012, 90 = Cugusi 2012, 116 = Cugusi 2014, 4 = AE 1929.7 = 1987.993 (add.) = 1995.1641 (add.) = 1999.1760 = 2014.1476 (add.) = 2016.1834 (add.), from Bu Njem), and Riese, AL 212 (= Monceaux 1906, 264 no. 160, from Carthage). 31

158

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

contrast with descriptions of natural springs from where water flows freely, falling onto the rocks, thus producing harmonious sounds.36 In the context of bathhouses one must, of course, also imagine the soundscape that is related to human activities there: the ancient sources relate vivid depictions of substantial noise and big crowds in these settings, especially in the big thermal complexes of the city of Rome.37

Touch The topic of sensations of hot and cold temperatures leads on to the much wider realm of tactile experiences – an area that has remained altogether under-researched.38 What is more, considering the wide range of contexts in which inscribed verse can be found, and considering the breadth of inscribed materials in the context of Rome’s inscribed world in general, any attempt at describing tactile interactions with verse in its material form is bound to remain incomplete or overly generalizing: an inscribed cup or spoon brings about forms of sensory experiences that are necessarily different from those one would imagine in the context of a mosaic, a tombstone, a graffito, and so forth. An aspect that might meaningfully be explored through future archaeological research is the way in which specific wear and tear of inscribed objects might give us important clues as to how individuals in the ancient world physically interacted with their monuments and inscribed objects. Without such datasets, one must necessarily rely on speculation, common sense, and a small number of verbal clues that might help to reconstruct physical interactions involving the sense of touch. One important aspect that unites a wide range of text types, as well as their monumental manifestations, is the important matter of their being sacred objects, i.e., objects that are marked as belonging to, as being the rightful property, of the divine sphere and its fearsome representatives.39 This does not only include inscriptions of an obvious religious nature (such as votives and dedications), but also tombstones and 36 See for example an inscription from the mosaic floor of the frigidarium in a small bathhouse in Sullectum (Africa Proconsularis), published for the first time in AE 1968.610: En! perfecta cito baiaru(m) grata voluptas | undantesque fluunt aq(uae) saxi de rupe sub ima | nisibus hic nostris prostratus libor (!) anhelat | quisquis amat fratrum veniat mecumq(ue) laetetur. (“Look! The pleasant delight of the bathhouse has recently been completed and the waters flow freely from the highest rock. Struck down by our efforts, Envy struggles for breath. Whoever of the brothers loves me, should come and bathe with me”). In the small thermal complex, the building was combined with a natural spring discharging its water from a towering rock, whence fresh water fell, forming waves upon hitting the surface (undantes aquae); further on this inscription see Busch 1999, 235–9. 37 See e.g., CIL VI 9797 = VI 33815a = VI 41107a = CLE 29 = ILS 5173 for a lively image of everyday life at the thermae in Rome; cf. Schmidt 1999 (= AE 1999.207) for an interpretation of this piece. Topically, Seneca (ad Lucil. 56.2), complained about the manifold noises that came with living above the baths, which made it impossible for him to concentrate. 38 For a recent collection of relevant essays, though not specifically on epigraphic material, see Purves 2018. 39 Further on this cf. e.g., Scheid 2019, 68 on private votive altars and objects, and 78–9 on necropoleis and tombstones.

7.  When poetry comes to its senses

159

graves in general (thus being relevant the vast majority of all instances of inscribed verse of the Roman world). References to the sacred nature of an object and its environs are a common occurrence, generally in Roman epigraphy as well as in the verse inscriptions in particular. This extends to specific appeals to individuals and groups to refrain from physical interference with such structures, especially with a view to forms of desecration through additional burials, to violation of the existing burial, to defecation, and to vandalism.40 A straightforward example – one of many – can be seen in a late antique text from the city of Rome addressing its audience thus: Sic tibi perpetuo sint, lector, uota secunda: parce pios Manes sollicitare manu. Thus, dear reader, your prayers shall be answered in perpetuity: refrain from bothering the pious Spirits of the Departed with your hand. (CIL VI 36655 = ICVR VIII 23359 = CLE 1468 = EDB35220)

A more complex approach to the same directive – “do not desecrate!” – is found in the following piece from the city of Rome: Dis Manibus Maeuiae Sophes. C(aius) Maenius Cimber, coniugi sanctissimae et conseruatrici, desiderio spiritus mei, quae uixit mecum an(nos) XIIX, menses III, dies XIII. 5 Quod uixi cum ea sine querella; nam nunc queror aput (!) Manes eius et flagito Ditem aut et me reddite coniugi meae, quae mecum uixit tan (!) concorde (!) ad fatalem diem. Meuia Sophe, impetra, si quae sunt, Manes, ni 10 tam scelestum discidium experiscar diutius. hospes, ita post obitum sit tibe (!) terra leuis, ut tu hic nihil laeseris, aut, si quis laeserit, nec superis comprobetur nec inferi recipiant, et sit ei terra grauis. To the Spirits of the Departed of Maevia Sophe. Gaius Maenius Cimber to the most blessed and caring wife, beloved by my soul, who lived with me 18 years, 3 months and 13 days.  As I lived with her without any fight, now I beg her spirits of the departed and implore Ditis: send me also to join my bride, who lived with me in such harmony until the last day. And you, Mevia Sophe, implore your spirits, if they exist, that I do not have to suffer for so long a cruel separation. Stranger, may the earth be light on you after Further on this see Kruschwitz 2010b and del Hoyo 2014b. See also Hernández Pérez 2013, who studies the inscriptions that try to dissuade electoral propagandists from vandalising the monuments, and Caruso 2004. This protection was also sought after through the iconography. Often, the sacrality and inviolability of the funerary space was represented by an ascia, symbol of protection of the grave (see Cenati 2017 for this aspect in the city of Rome). For a general analysis of this symbol and its origins, see Couchoud and Audin 1952. 40

160

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz your death, if you do not violate this grave: if someone violates it, may neither the Gods of Heaven accept him, nor the Infernal Spirits welcome him, and may the earth weigh heavy upon him. (CIL VI 7579 = CLE 2170 = EDR107816)

Examples can easily be multiplied,41 and they are not restricted to the funerary sphere.42 What is more, they are not always restricted to requests to avoid interference altogether. An especially interesting piece of evidence in that regard is the following inscription from Eboracum (York) in Britain, in which reference to touch is made in an iambic senarius that was added at the end (mostly in draft form, not fully executed in monumental script beyond a mere tracing of the letters):43 5

D[eo sancto] Silua[no s(acrum)]. L(ucius) Celerini(?)us Vitalis corni(cularius) leg(ionis) VIIII His(panae) u(otum) s(oluit) l(aetus) l(ibens) m(erito). et donum hoc, donum adpertiniat: cautum attiggam.

Sacred to the saint god Silvanus.  Lucius Celerin(i?)us Vitalis, cornicularius of the legio IX Hispana, has happily, gladly, and deservedly paid his votum. And this gift, let the gift pertain: I shall touch it with care and respect. (RIB I 659 = HD069924)

No explicit instructions, in the negative as well as in the positive, would have been worth inscribing at some expense, of course, if they had not been of relevance in practice (in practical terms of preventing destruction as well as on a level of showing respect for sacred items). Cf. e.g., CIL VI 10237 (cf. p. 3502, 3908 = CIL XIV *417 = CLE 371 = ILS 07870 = EDR126012, from Rome): (...) impensae causam titulum qui perlegis audi | et iustam quaeso pietatis percipe curam | qui(bu)s vera ut cupiant concorde uiuere mens est | hos animos spectent atq(ue) haec exempla sequantur | haec loca dum uiuent libeat bene cuncta tueri | post obitumq(ue) suum tradant tum deinde futuris | ne deserta uacent ignotis deuia busta | sed tuta aeterno maneant si dicere fas est (“listen to the cause of these expenses, you who read this epitaph, and understand, please, the inevitable concern that arises from devotion. Those for whom the true intention is to live in harmony, as they wish, shall observe these souls and follow their examples. As long as they live, they shall protect all these places, and after their death, they shall then leave them to their successors, so that these isolated graves will not be abandoned in the hands of unknown people, but remain protected forever, if it is possible to say so”). 42 Bücheler suggested that the fragment that is CIL VI 30117 (cf. p. 3736) = CLE 1469 might have pertained to a fountain of some description. It is much more likely, however, that it originally was part of a funerary inscription: cf. Orlandi ad EDR142152. 43 Further on this inscription see Kruschwitz 2017; 2020, 184–5. 41

7.  When poetry comes to its senses

161

This latter observation has a number of significant implications that must be explored further in a different context: what does it mean, for example, if a human, through touch, creates a connection to a sacred object and thus, ultimately, to the divine sphere, from a ritual point of view? How does a tactile experience relate to the immediate environs, and human motion within this space,44 adding a multidimensional (spatial) facet to the initially predominantly superficial and twodimensional tactile experience, in which ritual(s), motion, and inscribed memory become a whole. It is certainly possible to push the aspect of touch much further still, e.g., through a careful examination of references to common tactile experience such as “heavy” and “light”, “hard” and “soft”, “hot” and “cold”.45 This must be left aside for now, however, not only because of space restrictions, but because these matters, though mentioned with some frequency, are rarely filled with meaning beyond the topical, preventing, to an extent, a clear understanding of their impact on the experiences, actions, and behaviours of those who encountered inscribed verse in the Roman world in terms of their practical significance.

Smell Spaces are not only filled with tangible, experienceable objects and sounds, of course, but also – in many cases – marked by specific constant or recurring olfactory experiences.46 These, too, at least to an extent contribute, and relate, to ritual and memory. It is a well-known fact that scent can be a trigger that may bring back powerful emotions and memories. Precisely this phenomenon appears to be the background of an emphatically sensuous memory of the smell of the deceased’s homeland as it is mentioned in a fragmentary inscription from Cirta (Constantine, Numidia, Algeria). The inscription commemorates one Publius Sittius Optatus, and it does so by invoking a combination of smell, colour, and perception of nature:47

Qui properas, quaeso, tarda, uiator, iter, ut paucis discas cum genus (!) exitium.

Note, for example, instances in which the reader is requested to “rise” (surgere vel sim.) before departure such as CIL VI 9938 (cf. p. 3471 = XI *104.2 = CLE 989) or CIL VI 21521 (cf. p. 3526 = VI 34137 = CLE 1109). 45 See, for example, CLE 1147.3-4: nunc si qu(i)d Manes sapiunt in mollibus umbris conprecor ut matris sit tibi gratus honos (“now, if the spirits of the departed are able to discern in the soft shadows, I pray that your mother’s offering is pleasing to you”). Here, umbrae (the underworld), something untouchable, are referred to as molles (soft), which is an adjective that is linked, in its most denotative meaning, to the tactile world. The choice of this word could have been influenced, in turn, by the semantic proximity with leuis, used in this funerary context within expressions as sit tibi terra leuis and such. 46 For a recent collection of articles on smell and the human sensorium in the ancient world see Bradley 2015. This collection contains an article by Butler 2015 that aims to make “scents” of poetry, but does not cover its inscribed dimension(s). 47 Further on this piece see Hamdoune et al. 2011, 206–9 no. 123. 44

162

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz non externa satus Scythi5 ca de gente Syrorum, [s]um satus Aethna, uiros ub[i] cingunt Anspagae moles. cognitus est locus amoenis simus Alba, in qua frondicoma 10 odoratur ad mare pinus, Daphne pudica ui[ret, sa]lit et loco uitrea Na[is]. (...) Wayfarer, as you are rushing along, please stall your journey, so that you may learn in a few words about my end in conjunction with my origins. Not a scion of the exotic Scythian branch of Syrians, I am the offspring of Etna, hailing from where the waves of Ampsaga enclose men. This most beautiful place is known as Alba, where the leafy pine tree exudes its scent towards the sea, where bashful Daphne [i.e., laurel] is green, and where the crystalline Naiad leaps. (...) (CIL VIII 7759 (cf. p. 966) = VIII 19478 = ILAlg II.1.831 = CLE 1327 = AE 2006.145, ll. 1–11)

The link between scent and the emotive response(s) they trigger is not only important, of course, in bringing back powerful memories. To the present day, smellscapes, at least in certain contexts, are deliberately designed to affect our emotions and to shape our behaviours and responses. Their relevance in, and their impact on, our reading habits specifically, however, certainly as far as the ancient world is concerned, have not yet been thoroughly researched. Yet, as anyone who has ever picked up, e.g., a freshly printed newspaper or an old book, with the typical vanilla-scent of its slowly disintegrating paper, or who has entered an antiquarian bookshop with its unique, overwhelming smell, knows and can confirm, reading experiences and our sense of smell are by no means separate entities. As books in the Roman world were made of organic material and then treated with fragrant substances to prevent their decay, and as they were in many instances stored in libraries with purpose-built furniture and even purpose-built and designed rooms (to stall decay even further), experiences comparable to modern-day ones existed beyond any reasonable doubt. But what about monumental environments, monumental settings for encounters with poetry and verse?48 It is fair to say that (similar to, say, lighting conditions) smellscapes are as varied as there are numbers of settings, and they may, of course, also vary within a very short period of time. In that regard, it will remain impossible to construct an accurate picture of such settings. If one were to imagine, for example, the smellscape around a dining or living room mosaic, with its imagery, decoration, and a poetic addition, both light and smell would vary depending on the time of day, and, for example, the provision of foodstuffs and drinks, resulting in a potentially very complex scenario in 48

Some aspects of this section have previously been presented to a wider audience by Kruschwitz 2015.

7.  When poetry comes to its senses

163

which perceptions of read texts (and their aesthetics and wording) might potentially change rather significantly. There are, however, scenarios for which at least a certain stability may be surmised, and there is evidence for such stability even being imagined and expressed in the surviving evidence. Unsurprisingly, this is especially true for the funerary record.49 In the funerary sphere, there are two concurrent, yet contradictory aspects to be addressed: unpleasant smells, related to the inevitable presence of mortal remains, and pleasant smells, either related to the natural environment or designed and planned by those who were responsible for the burials. Mentions of both aspects can be found in the surviving record, usually in conjunction, as the latter often was a response arisen as a necessity out of the former. A first clue in that regard is a line in an inscription from Hadrumetum (Sousse, Tunisia). The remains of one Lucius Ummidius (who had died at a relatively young age of 32 and a half years, and was buried by his brother Peregrinus) were described as having been buried as follows: Sola quies retinet tumulo tellure manentem. ||50 condidimus cineres latebris et odoribus ossa. This is the only rest that holds you now, remaining in this earth as your tomb. We covered the ashes and bones with shelter and fragrances. (CIL VIII 22971 = CLE 1829, ll. 9–10)

Here, latebrae – adding a visual component of obscurity combined with a notion of a hiding place – is employed alongside odores, “fragrances” or scents, to envelop the cineres of the deceased, bedding the remains in a safe, olfactorily pleasant environment, for him to remain in this spot forever (retinet … tellure manentem).51 Similarly complex experiences are mentioned in the following piece from the city of Rome, in a highly moving epitaph for one Marcus Lucceius Nepos that, despite its length, is worth quoting in full:52 49 Similarly, for a non-funerary example of smell and scent as an important aspect for the general setting in the Latin verse inscriptions see CIL XIV 3565 = CLE 1504, ll. 12–22, a phallic herm dedicated to the Genius of Priapus in a Hercules sanctuary at Tibur. Here the fragrant garlands donated to Priapus (olentes coronae) are opposed to criminals that are imagined to be dirty with mud and blood. 50 The first nine lines are aligned to the left, to make an acrostic stand out; lines 10 ff. are moved further to the right. For an image see http://db.edcs.eu/epigr/bilder.php?bild=$CLEAfrique_00002.jpg (last accessed: December 2020). 51 A similar idea is expressed in CIL IX 7447 = CLE 1321 (Aternum, Samnium): (...) Ninnius et cinerem spargit odore gemens (‘Ninnius scattered among laments the ashes with perfume’). Furthermore, see CIL VI 30102 = CLE 1508 (Rome) where the wish of the dedicant is mentioned to adorn the grave with garlands and to keep an oil lamp with nard: parcas oro uiro puella parcas | ut possit tibi plurimos per annos | cum sertis dare iusta quae dicauit | et semper uigilet lucerna nardo. 52 See Bianchini et al. 2020 for a recent study of this piece.

164

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz Memoriae M(arci) Lucceì M(arci) f(ili) Nepotis Sex(tus) Onussanius Se[x(ti)] f(ilius) Com[- - -] || Quum praematura raptum mihi morte Nepotem flerem Parcarum putria fila querens et gemerem tristi damnatam sorte iuuentam 5 uersaretque nouus uiscera tota dolor, me desolatum, me desertum ac spoliatum clamarem largis saxa mouens lacrimis, exacta prope nocte suos quum Lucifer ignes spargeret et uolucri roscidus iret equo, 10 uidi sidereo radiantem lumine formam aethere delabi. Non fuit illa quies, sed uerus iuueni color et sonus, at status ipse maior erat nota corporis effigie. ardentis oculorum orbes umerosq(ue) nitentis 15 ostende(n)s roseo reddidit ore sonos: “adfinis memorande, quid o me ad sidera caeli ablatum quereris? desine flere deum. || ne pietas ignara superna sede receptum lugeat et laedat numina tristitia. 20 non ego Tartareas penetrabo tristis ad undas, non Acheronteis transuehar umbra uadis, non ego caeruleam remo pulsabo carinam nec te terribilem fronte timebo, Charon, nec Minos mihi iura dabit grandaeuus et atris 25 non errabo locis nec cohibebor aquis. surge, refer matri ne me noctesque diesque defleat ut maerens Attica mater Ityn. nam me sancta Venus sedes non nosse silentum iussit et in caeli lucida templa tulit.” 30 erigor et gelidos horror perfuderat artus; spirabat suauui tinctus odore locus. dìe Nepos, seu tu turba stipatus Amorum laetus Adoneìs lusibus insereris. || seu grege Pieridum gaudes seu Palladis [arte], 35 omnis caelicolum te chor[u]s exc[ipiet]. si libeat thyrsum grauidis aptara co[rymbis] et uelare comam palmite, Liber [eris]; pascere si crinem et lauro redimire [- - -] arcum cum pharetra sumere, Ph[oebus eris]. 40 indueris teretis manicas Phrygium [decus, Attis] non unus Cybeles pectore uiuet a[mor]. si spumantis equi libeat quatere ora [lupatis], Cyllare, formosi membra uehes e[quitis]. sed quicumque deus, quicumque uocaber[is heros], 45 sit soror et mater, sit puer incolu[mis]. haec dona unguentis et sunt potiora c[orollis], quae non tempus edax, non rapi[t ipse rogus]. Sextus Onussanius Com…, son of Sextus, to the memory of Marcus Lucceius Nepos, son of Marcus.

7.  When poetry comes to its senses

165

 When I was lamenting my loss of Nepos through premature death, complaining of the easily-snapped threads of the Fates, and was bemoaning his manhood condemned by a cruel destiny, and pain not previously experienced was torturing my whole heart; when I was bewailing my bereft, abandoned, deprived state, moving the rocks with my floods of tears; almost at the end of night, when the dewy Dawn-Star was spreading his rays and riding his swift horse, I saw a shape, glowing with stellar light, glide down from the sky. That was no dream, but the man had his actual complexion and voice, though his stature was greater than the familiar shape of his body. Showing the blazing orbs of his eyes and shining shoulders he spoke from his rosy lips. ‘My noble kinsman, why do you complain that I have been snatched away to the stars of the sky? Cease to bewail a god, lest your affection, unaware that I have been welcomed in the celestial abode, may mourn and by its sorrow distress a supernatural being. I shall not gloomily make my way to the underworld streams and shall not as a ghost be ferried across the waters of Acheron; I shall not with my oar drive forward the dark boat nor shall I fear Charon with his terrifying countenance, nor will ancient Minos pass judgment on me; I shall not wander in those dark places nor be pinned in by the rivers. Rise, tell my mother not to lament me night and day, as the mourning Attic mother does Itys. For holy Venus has forbidden me to know the abodes of the silent and has carried me to the bright halls of heaven’. I jumped up, and trembling had pervaded my cold limbs; the place was fragrant, redolent with a sweet smell. Sanctified Nepos, the whole heavenly chorus will welcome you, whether, escorted by a crowd of amorini, you happily mingle with the amusements of Adonis, or you rejoice in the crowd of the Muses or in the artistic skill of Athena. If you should want to fasten heavy clusters of ivy-berries to the thyrsus and veil your hair with vine-shoots, you will be Bacchus; if you should want to grow your hair and garland it with bay and take up bow and quiver, you will be Apollo. Put on fine sleeves and a Phrygian (cap), more than one love will quicken in Cybele’s breast. Should you desire to shake the mouth of a foaming horse with the bridle, then Cyllarus will carry the body of a handsome rider. But whatever god, whatever demigod you shall be called, may your sister, mother and young son be safe and sound. These gifts, which gnawing time and [the pyre?] do not take away, are better than perfume and garlands. (CIL VI 21521 (cf. p. 3526) = VI 34137 = CLE 1109 = EDR176675; transl. E. Courtney)53

The highly impressive visual world that is created and invoked towards the beginning, with sensuous elements of light and darkness, of shiny elements and colour, yet combined with utter gloom, with sound elements of lament and also silence, is eventually enriched by two further aspects, references to temperature (cold limbs are mentioned here) and, important in the present context, references to smell: “the place was fragrant, redolent with a sweet smell”. In a captivating, rapid sequence, further references are made to sound and to singing, as well as to taste, only to conclude with another reference to the smellscape (now in a dismissive form): “these gifts, which gnawing time and [the pyre?] do not take away, are better than perfume and garlands”.

A picture of the only preserved fragment of this inscription is available at https://catalogo. museivaticani.va/index.php/Detail/objects/MV.8720.0.0 (inv. 8720).

53

166

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

The presence of perfume and garlands (and flowers more generally) in funerary contexts is well attested in a wide range of sources.54 Yet, in actuality, such experiences were far more complex still. The epigraphical record preserves conceptualizations of the ancient smellscape beyond the actual burial. In doing so, they take stock of both the immediate context of the built structure and its decoration and the landscape in a wider sense. An excellent example of that is the following piece, a funerary inscription from Theveste / Tébessa (Numidia, now Algeria) demonstrates: Inter odoratos nemorum ubi laeta recessus mater pingit humus et lectis dedala Tellus floribus exultat gratisque et frondibus almum ix patitur cum sole diem, hic prouide felix 5 Florentine decus cum coniuge sancta pudica Hostiliana tua et Splendonillae natoque || (...) Where in the fragrant seclusion of the groves mother Earth cheerfully brings colours to herself and ingeniously rejoices in exquisite flowers and with her treetops barely yields the day’s nourishing sunlight, here, provident, felicitous Florentinus, your splendour, alongside your saintly, bashful wife Hostiliana and … Splendonilla’s … boy … reunited (?) … (CIL VIII 2035 (cf. p. 1590) = ILAlg I 3550 = CLE 469, ll. 1–6)

References to the monument’s emotionally charged, aromatic environs are combined with an appreciation of colour and light effects, caused by the flowers and the treetops (which, of course, are also responsible for the emission of the fragrances), stimulating the reader’s senses and suggesting happiness (laeta), indulgence (gratis … frondibus), and good cheer (exultat). In this instance, the conceptualized smellscape is merely circumstantial. There are more carefully thought-through examples, however, as the final example for this section demonstrates.55 The monument of the Flavii at Cillium (Kasserine), in the province of Africa Proconsularis, imagines living inhabitants of the structure in addition to the dead ones – and with them comes a very striking and specific smell:56 quid non docta facit pietas: lapis ecce foratus luminibus multis hortatur currere blandas intus apes et cerineos componere nidos ut semper domus haec thymbraeo nectare dulcis sudet florisapos dum dant noua mella liquores. 54 See Lattimore 1962, 134–6 for Greek and Latin epitaphs and Cumont 1949, 44–8 for a more general view. Šterbenc Erker 2011, 45–6 mentions a relief in which a man appears standing next to the dead body of a matron whom he is about to adorn with a garland. 55 Much more could be mentioned here, of course, such as, e.g., CIL II2/7.116 = CLE 1851 add. and CIL II2/7.575. 56 Further on this piece see Les Flavii du Cillium (1993) and Courtney (1995) 186–193, 399–406 no. 199A.

7.  When poetry comes to its senses

167

What does a sense of filial duty not achieve: behold, the gaping stonework, with many a light crack, invites enchanting bees to go inside and to build their waxy nests, so that this home forever will exude a sweet scent from the nectar of thyme, when new honey produces flower-dripping juices. (CIL VIII 212 = 11300b = CLE 1552a + CIL VIII 213 = VIII 11300c = CLE 1552, ll. A, 86–90 = HD064997)

Here, the confident and profoundly reflective individual behind the creation of the monuments reveals his imagination that the structure will eventually be populated by bees, to build their nests in the cracks between the stones, whose activity will bring a specific, pleasant smell to its surroundings and to be perceived by those stopping by. The entire description of this imagination is loaded with sensuous experiences, from the scent to the taste to the texture of the substance in question. The material gathered in this section, due to its more or less consistent generic background, covers only a segment of the actual olfactory experience involved in the encounters with, and consumption of, inscribed poetry: many further aspects become obvious once one considers the environs of the inscriptions when in situ, alongside their specifics. Helpfully, verse inscriptions in the funerary record occasionally reflect on this aspect (and even their planful design). Absence of such reflections in other contexts, or relative lack thereof, does not mean, however, that similar considerations did not occur or were of little or no relevance. What is particularly striking in those instances that were mentioned here is the observations that references to the olfactory experience are quite commonly combined with reflections of simultaneous appeals to other senses.

Taste It may seem counterintuitive to include, in a paper on inscriptions, a section on taste:57 ingestion of monumental texts, and an appreciation of their taste, would appear to be quite beyond regular human capabilities. No fundamental argument to the contrary shall be attempted. At the same time, it would not be appropriate to omit this aspect altogether. Rather, two aspects must be mentioned at least in passing, in addition to noting that, obviously, sensations of smell and taste are closely linked (as anyone knows who has ever had a blocked nose even during a common cold). The first aspect that must be addressed is derived from experiences that most readers of this paper will have had themselves: it is, of course, perfectly possible to ingest texts, though not necessarily for a sensuous, taste-related experience as such. To swallow a secret note, however, is not beyond the scope of what a human being could experience (and would do, too, if under a certain amount of pressure). There is no particular reason to believe that the same could not be done with, say, a papyrus note or even non-sharp-edged inscribed amulets. What is lacking primarily is the 57

Taste and the ancient senses are the topic of a recent edited volume by Rudolph 2018.

168

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

textual evidence to prove that this has, in fact, happened. Yet, absence of evidence is not the same as evidence of absence. And the notion of “devouring texts” was, at least in a metaphorical way, certainly already known in the Roman world.58 The second aspect is more pertinent, and it is also more easily backed up by the evidence: it is a known fact that edible matter was on occasion inscribed. One such example has already been mentioned, above, namely the mythical apple of Cydippe.59 The best-known actual, rather than mythical, examples might be seen in the stamped, carbonized bread loaves of Pompeii and Herculaneum,60 which, when eaten, would have made someone devour a genuinely edible text. What is more, it seems entirely plausible to assume that entire foodstuffs were even formed to resemble the shape of letters and potentially even texts.61 No instance specifically of poetry on ancient baked goods survives. What does survive, however, is credible evidence for their production, most notably through the discovery of a mould that appears to have served as a patrix, revealing a poetic inscription, designed for the production of cake moulds that in turn would serve as matrices for actual poetic cakes.62 And with that, even though some may find it hard to swallow, it becomes altogether inevitable to contemplate a taste-related dimension of Roman poetry! While inscribing solid matter is perfectly fathomable, poetic drinks may be less easily imagined. These, too, existed, however, most notably through the use of inscribed cups and jars, exhibiting erotic and sympotic themes.63 And on an even Cf. Cic., Att. 126 [= 7.3.2] S-B and Att. 86 [= 4.11.2] S-B. See above, “Sound” and n. 23. 60 See CIL X 8058.18 for Pompeii. Loreti 1994, 652–3 studies the actual bronze stamp used for that inscription, found centuries after the loaf. Manganaro 2001, 194 defends the same use for many of the bronze stamps with names that have been found, such as those studied by Sotgiu 2000, 1018–19. For Herculaneum, see Allroggen-Bedel 1975, 99–101. Pliny the Elder (NH 33.26) laments that at his time everything was stamped, even food and drinks, blaming it on the proliferation of robberies: quae fuit illa uita priscorum, qualis innocentia, in cui nihil signabatur! Nunc cibi quoque ac potus anulo uindicantur a rapina (“what lives people led in the olden days, what innocence there was, when nothing was stamped! Now even foodstuffs and drinks are claimed with a seal, to protect them from being stolen”). 61 See a clay cake mould from Apulum (CIL III 6287 = IDR III/6, 422): Accipio | annum | nouum | felicem (“I receive a happy new year”). An especially noteworthy case is Hor., Sat. 1.1.25–6, where Horace would appear to mention an alleged habit of teachers to provide their pupils with cake – thought to be in the shape of letters – as a way to instill a love for the alphabet; further on this cf. Kruschwitz 2016a, 31. 62 See Alföldi 1945, 71 and, more recently, Thüry 2008, 297; cf. also Cugusi and Sblendorio Cugusi 2007, 72–3 (CLEPann 28). 63 A short text preserved on a set of “speaking” objects, a ceramic wine service found in a large Roman villa in Ulcisia (Pannonia Inferior) has some poetic echo as well. The set consists of five cups and a jar, all inscribed (IIL 180), which, according to the stratigraphy, are to be dated to the end of the third century AD. On the cups very fragmentary texts are to be read: [a]m[a] me; am[a me? -as? -at?] (“Enjoy it”); v[i]t[a] or [i]u[a]t (“Darling” or “That makes you happy”). The text on the jug is more complex and appears to show a dactylic rhythm (non amat me cu[pid]us). The interpretation of the text on the jug is challenging. Thüry 1998 proposes to translate this as “a greedy person (who does not mix wine with water) does not like me”. Further “speaking” glasses and jars with erotic or sympotic messages can be found across the Empire: cf. e.g., the following instances from the Gallic and Germanic provinces, to name but a few: accipe m[e si]tie(n)s et trade sodali (“receive me when you are thirsty and then hand me 58 59

7.  When poetry comes to its senses

169

larger scale one could mention the dactylic graffito placed at the stave of a wine barrel that was discovered at Castra Regina (Regensburg) in the province of Raetia:64 Caui alios conui(u)as orca sto `[- - -]ia D´ I have been careful with other guests, so I am still a full barrel … `[- - -]ia D´ (ILGIL 998)

The wooden barrel was found in the eastern vicus of the military castellum, and it has been dated to the late first century AD. Additional inscriptions on the barrel indicate the names of the wine merchants involved. The text on the stave is secondary, presumably an addition from the time when the barrel was already in Castra Regina – in a tavern perhaps? – and soldiers deployed to the Raetian limes were able to enjoy its contents. The barrel is introduced as a first-person speaker,65 and it addresses its audience (so presumably the patrons of a tavern) with the welcome news that there was plenty of wine left (a matter that, regrettably, has since changed for the worse). With that, however, one reaches the very limits of linking inscribed verse and experiences of taste.

Synaesthesia Much more could be said about, and added to, the evidence gathered for each of the human senses in relation to the experience of inscribed verse. Much, if not all of it, also bears significance and relevance to non-poetic encounters with inscriptions, with substantial implications for the ways in which individuals and groups who navigated the lettered world of ancient Rome interacted with the writing that surrounded them. Instead of concluding this section with a paragraph or two that merely repeat and reassert some of the more salient points of what has already been said, it seems worth exploring yet another sensory dimension that, to an extent, is in danger of falling through the cracks of an all-too-rigid categorical approach. As a logical extension to our focus so far, placing emphasis on individual facets of the human sensorium, it seems reasonable to devote a few concluding lines to scenarios in which the divisions between the senses are blurred, and in which sensory experiences are merged into more complex, compound clusters – approaching, at least to an extent, over to your companion,” CIL XIII 10016.4); de(s) et do (“give me (wine) and I will give it to you”, CIL XIII 10018.64); fero uinum tibi, dulcis (“I bring you wine, sweet thing”, CIL XIII 10018.86); a me, dulcis amica, bibe (“drink from me, sweetheart”, CIL XIII 10025.199). For bibliography on “speaking” objects see above, n. 26. 64 The end of the second line, given in inverted commas, was added by another hand. 65 Cf. Thüry 1995, 301: “Von anderen Gästen habe ich mich gehütet. (So) bin ich (noch) ein volles Faß” (or “So stehe ich Dir noch zur Verfügung”). See also Frei-Stolba 2017, 192; Thüry 1995, 301–2.

170

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

the phenomenon that is known as synaesthesia (though it is impossible for us in some cases to have a clear distinction between that and “merely” multisensory experiences). Already in what was discussed before, there were numerous occasions in which texts provided their readers with complex, intersecting sensescapes. One particularly fascinating example of that was the epitaph of Nepos:66 this composition combines a range of expressions referring to a range of senses, and while it is remarkable with a view to the extent to which this happens, it is not at all uncommon to find such clusters that in turn are employed to create vivid and highly effective relatable poetic images. Is it possible, though, to discern instances of (at least imagined) synaesthesia in the Carmina Epigraphica – instances in which the experience of one sensory stimulus is inextricably linked to another sense in the poetic expression? As it has been argued that ancient colour perception was often the result of several cognitive stimuli,67 the visual component seems to be an especially promising starting point. And, indeed, there appears to be a certain amount of evidence for such conceptualizations and experiences. One such example is the following from a fragmentary piece discovered at Thamallula in the province of Mauretania Caesariensis (Algeria). The passage in question reads as follows: (...) [qualia p]allente[s d]eclinant l(i)lia culm[os] [pubent]esq(ue) rosae primos moriuntur ad [austros] [aut ubi] uer(na) nouis expirat purpura pra[tis] 5 [- - - m]ortis erat pallentis imago O[- - -] [- - - l]ibante pietatis [- - -] (...) (...) Like lilies droop their paling stalks, and juicy roses die with the first winds of the south or as spring’s purple fades away in fresh meadows, [such?] was paling death’s appearance … pouring … of piety ... (CIL VIII 20588 = CLE 1787 = AE 1894.95 = HD028810)

Traditionally, the value of this inscription has been somewhat underappreciated by its editors, presumably because at least ll. 2–4 were regarded merely as a direct quotation of Stat., Silv. 3.3.128–30, a consolatio poem to Claudius Etruscus upon the passing-away of his father. The passage in question, however, addresses the early death of Etruscus’ mother in the same poem, with its imagery based on a Vergilian model (with a long, subsequent poetic tradition).68

See above, “Smell” (with nn. 53–4). See Bradley 2013, 130–2 and, more generally, the edited volume of Butler and Purves 2013 on representations of, and notions of, synaesthesia in the ancient world. 68 A useful discussion of this passage and its models (in Vergil and in Statius’ earlier work) can be found in Coffee 2019, 187–9 (who does not, however, comment on the inscription at hand). 66

67

7.  When poetry comes to its senses

171

The purpose of Statius’ lines in the context of Statius’ Silvae, is to provide poetic similes for an untimely death, of a “death in full bloom”, so to speak. That is not entirely (and solely) their purpose in the present inscription, however, as far as its fragmentary state allows its modern readers to arrive at any bold conclusions. In Statius, colour is a proxy for vitality – the lilies’ fading stalks are put in contrast with the juicy, youthful roses: freshness and vitality are facing languishing exhaustion at too early a stage. In the Thamallula epitaph, by contrast, the very same, complex image that, in a sensuous concerto, links experiences of colour and touch (and, in actual fact, scent!) is linked, expressis uerbis and with a clear verbal reference, to the subsequently introduced notion of the [m]ortis … pallentis imago. In that, fading, vanishing colour experiences and the disappearance of life’s juices accumulate to grisly effect that is very much absent from Statius’ poem. The way in which all of this would sit in the natural, experienceable environment of a burial, as explored in several contexts, above, adds an even more striking experience to the poem’s imagery. Is it possible to move even further, and to go beyond holistic experiences of a multitude of cognitive stimuli in the Latin verse inscriptions that derive their power from the visual component? The answer to that question may well lie in a final piece that must be introduced here, viz. a poeticizing, rhythmical, though not altogether metrical,69 piece from Arelate (Arles) in the province of Gallia Narbonensis. This bilingual inscription commemorates a young girl called Secundilla – nicknamed Aroma or Aromation (!) – who died an untimely death at the age of three and a half years. Its text reads as follows:70 5

Iacet sub hoc signno (!) dulcissim(a) Secundilla qu(a)e rapta parentibus reliquit dolorem ut tam dulcis erat tanquam (!) aromata desiderando semper mellea(m) uita(m). qu(a)e uixit annis III men(sibus) VI die(bus) XVI. Ἀρωμάτι ∙ ταῦτα.

Sweetest Secundilla lies underneath this memorial: when she was snatched, what she left behind for her parents was pain: she was as sweet as fragrant herbs in her desire for a honey-sweet life. She lived 3 years, 6 months, 16 days. For Aroma(tion?): This (is it). (CIL XII 874 = IG XIV 2475 = CLE 1851 = IGF 57 = SEG 13.258)

In an etymologizing, code-switching wordplay on the girl’s Grecian nickname,71 Aroma, the author of this piece first calls Secundilla dulcissima (l. 1), introducing a sense of taste (through a common place term of endearment for close family members), then reasserts this impression through the use of the dulcis (l. 3) compared to aromata (also Cf. Fernández Martínez 2018, 303. Further on this piece see Mullen 2013, 296. 71 On this phenomenon in the Carmina Latina Epigraphica more broadly, cf. Michalopoulos 1997. 69 70

172

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

in l. 3), viz. herbs of a fragrant, spicy scent (rather than flavour). Finally, a longing for mellea(m) uita(m) is introduced, through which sweet flavour and aromatic (herbal) scents are merged, just as sweetness and beautiful scent are apparently imagined to have formed a beautiful unity in Secundilla a.k.a. Aroma(tion). The bespoke image, related to the recipient of this poeticizing text and her nickname, is already sufficiently striking in terms of the multisensory experience that is expressed, verging on synaesthesia-like blurrings and crossings of sensory experiences. Once one relates it back to the type of imagined smellscapes, however, that are described in the aforementioned passage from the Flavii monument at Cillium,72 the full potential of the sensuous experience becomes all the more apparent. It would be a wonderful idea, as well as a great, yet worthwhile challenge, to recreate such complexities of reading experiences, especially, but not exclusively, for monumental poetry for its modern readers and students, who now typically encounter these items in (more or less) sterile museum environments.

Bibliography

Agostiniani, Luciano. 1982. Le “iscrizioni parlanti” dell’Italia antica. Florence: L.S. Olschki. Alföldi, Andreas. 1945. “Tonmedaillons und runde Kuchenformen aus Pannonien und Dacien.” Folia archaeologica 5: 66–73. Allroggen-Bedel, Agnes. 1975. “Der Hausherr der ‘Casa dei Cervi’ in Herculaneum.” Cronache Ercolanesi 5: 99–101. Amante, Antonio. 1913. “Gli acrostici nella poesia sepolcrale latina.” Athenaeum 1 (3): 288–94. Barbieri, Guido. 1975. “Una nuova epigrafe d’Ostia e ricerche sugli acrostici.” Miscellanea Greca e Romana 4: 301–403. Bernand, André and Étienne Bernand. 1960. Les inscriptions grecques et latines du Colosse de Memnon. Paris: Adrien-Maisonneuve. Berti, Irene, Christian D. Haß, Kristina Krüger, and Michael R. Ott. 2015. “Lesen und Entziffern.” In Materiale Textkulturen. Konzepte – Materialien – Praktiken, edited by Tomas Maier, Michael R. Ott, and Rebecca Sauer, 639–50. Berlin: De Gruyter. Berti, Irene, Katharina Bolle, Fanny Opdenhoff, and Fabian Stroth, eds. 2017. Writing Matters. Presenting and Perceiving Monumental Inscriptions in Antiquity and the Middle Ages. Berlin: De Gruyter. Bianchini, Gianmarco, Dylan Bovet, and Gian Luca Gregori. 2020. “CIL VI, 21521 = 34137 (CLE 1109): un sogno in forma poetica.” Espacio, Tiempo y Forma, serie II Historia Antigua 33: 213–32. Bodel, John. 1993. “Review of Literacy in the Roman World, edited by John H. Humphrey. Journal of Roman Archaeology Supplementary Series 3. Ann Arbor: Journal of Roman Archaeology.” Journal of Roman Studies 83: 183–4. Bolle, Katharina. 2019. Materialität und Präsenz spätantiker Inschriften. Berlin, Boston: De Gruyter. Bradley, Mark. 2013. “Colour as Synaesthetic Experience in Antiquity.” In Synaesthesia and the Ancient Senses, edited by Shane Butler and Alex Purves, 127–40. Durham: Acumen. Bradley, Mark, ed. 2015. Smell and the Ancient Senses. London, New York: Routledge. Burnyeat, Myles. 1997. “Postscript on Silent Reading.” Classical Quarterly n. s. 49: 74–6. Busch, Stephan. 1999. Versus Balnearum. Die antike Dichtung über Bäder und Baden im römischen Reich. Stuttgart, Leipzig: Teubner. 72

Cf. “Smell” (with n. 57).

7.  When poetry comes to its senses

173

Busch, Stephan. 2002. “Lautes und leises Lesen in der Antike.” Rheinisches Museum für Philologie 145: 1–45. Butler, Shane. 2015. “Making Scents of Poetry.” In Smell and the Ancient Senses, edited by Mark Bradley, 74–89. London, New York: Routledge. Butler, Shane and Sarah Nooter, eds. 2019. Sound and the Ancient Senses. London, New York: Routledge. Butler, Shane and Alex Purves, eds. 2013. Synaesthesia and the Ancient Senses. Durham: Acumen. Carroll, Maureen. 2007–2008. “‘Vox tua nempe mea est.’ Dialogues with the Dead in Roman Funerary Commemoration.” Accordia Research Papers 11: 37–80. Carroll, Maureen. 2011. “‘The Mourning Was Very Good’. Liberation and Liberality in Roman Funerary Commemoration.” In Memory and Mourning. Studies on Roman Death, edited by Valerie M. Hope and Janet Huskinson, 126–49. Oxford, Oakville: Oxbow Books. Caruso, Carlotta. 2004. “Dolus malus abesto et ius civile. Prevenzione e repressione di azioni fraudolente.” In Libitina e dintorni. Libitina e i luci sepolcrali. Le leges libitinariae campanae. Iura sepulcrorum: vecchie e nuove iscrizioni. Atti dell’ XI Rencontre franco-italienne sur l’épigraphie., edited by Silvio Panciera, 384–91. Rome: Quasar. Cenati, Chiara. 2017. “Note sulla collezione epigrafica di Giovanni Battista de Rossi al Pontificio Istituto di Archeologia Cristiana.” Rivista di Archeologia Cristiana 93: 191–201. Coffee, Neil. 2019. “Intertextuality as Viral Phrases: Roses and Lilies.” In Digital Classical Philology. Ancient Greek and Latin in the Digital Revolution, edited by Monica Berti, 177–200. Berlin, Boston: De Gruyter. Colafrancesco, Pasqua. 2012. “Una memoria familiare: in margine all’iscrizione di «Aelia Secundula».” Invigilata Lucernis 34: 69–76. Couchoud, P.L. and Aimable Audin. 1952. “Requiem aeternam… L’ascia, instrument et symbole de l’inhumation.” Revue de l’Histoire des Religions 142.1: 36–66. Courtney, Edward. 1990. “Greek and Latin Acrostichs.” Philologus 134: 3–13. Courtney, Edward. 1995. Musa Lapidaria. A Selection of Latin Verse Inscriptions. Atlanta: Scholars Press. Cugusi, Paolo. 1980–1981. “Un possibile tema dei Carmina Latina Epigraphica: l’Aureus titulus.” Annali della Facoltà di Lettere e Filosofia dell’ Università di Cagliari, n. s. 3.40: 5–9. Cugusi, Paolo. 2014. Carmina Latina Epigraphica Africarum provinciarum post Buechelerianam collectionem editam reperta cognita. Faenza: Lega. Cugusi, Paolo and Maria Teresa Sblendorio Cugusi. 2007. Studi sui carmi epigrafici. Carmina Latina Epigraphica Pannonica (CLEPann). Bologna: Pàtron. Cugusi, Paolo and Maria Teresa Sblendorio Cugusi. 2011. “Carmina Latina Epigraphica delle province greco-orientali (CLEOr).” Epigraphica 73: 161–245. Cugusi, Paolo and Maria Teresa Sblendorio Cugusi. 2012. I Carmina Latina Epigraphica non-Bücheleriani delle province africane. Bologna: Pàtron. Cumont, Franz. 1949. Lux Perpetua. Paris: Geuthner. Di Stefano Manzella, Ivan. 1987. Mestiere di epigrafista. Guida alla schedatura del materiale epigrafico lapideo. Rome: Quasar. Eastmond, Antony, ed. 2015. Viewing Inscriptions in the Late Antique and Medieval World. Cambridge: Cambridge University Press. Feraudi-Gruénais, Francisca. 2017. “Das synaktive Potential von Beischriften.” In Writing Matters. Presenting and Perceiving Monumental Inscriptions in Antiquity and the Middle Ages, edited by Irene Berti, Katharina Bolle, Fanny Opdenhoff, and Fabian Stroth, 43–76. Berlin: De Gruyter. https:// doi.org/10.1515/9783110534597-004. Fernández Martínez, Concepción. 2018. “CIL XIII 2180 y AE 2006, 820: magnam enim partem ex iambis nostra constat oratio.” Anuari de Filologia Antiqua et Mediaevalia 8: 300–10. Les Flavii de Cillium. Étude architecturale, épigraphique, historique et littéraire du mausolée de Kasserine (CIL VIII, 211–216) (Publications de l’École française de Rome 169), Rome 1993.

174

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

Flower, Harriet I. 1996. Ancestor Masks and Aristocratic Power in Roman Culture. Oxford: Clarendon Press. Frei-Stolba, Regula. 2017. Holzfässer. Studien zu den Holzfässern und ihrer Inschriften im römischen Reich mit Neufunden und Neulesungen der Fassinschriften aus Oberwinterthur/Vitudurum. Zürcher Archäologie 34. Zurich, Egg: Baudirektion Kanton Zürich. Gavrilov, Aleksandr K. 1997. “Techniques of Reading in Classical Antiquity.” Classical Quarterly n. s., 47: 56–73. Gregori, Gian Luca. 2008. “Sulle origini della comunicazione epigrafica defunto-viandante.” In La comunicazione nella storia antica. Fantasie e realtà. Atti del III Incontro Internazionale di Storia Antica (Genova 23–24 novembre 2006), edited by Maria Gabriella Angeli Bertinelli and Angela Donati, 83–115. Rome: G. Bretschneider. Hamdoune, Christine, Laure Échalier, Jean Meyers, and Jean-Noël Michaud, eds. 2011. Vie, mort et poésie dans l’Afrique romaine d’après un choix de Carmina Latina Epigraphica. Collection Latomus 330. Brussels: Latomus. Hernández Pérez, Ricardo. 2001. Poesía latina sepulcral de la Hispania romana: estudio de los tópicos y sus formulaciones. Valencia: Departament de Filologia Clàssica, Facultat de Filologia, Universitat de València. Hernández Pérez, Ricardo. 2013. “Versos epigráficos contra los abusos de la propaganda electoral en el mundo romano.” In Ex officina. Literatura epigráfica en verso, edited by Concepción Fernández Martínez, María Limón Belén, Joan Gómez Pallarès, and Javier del Hoyo Calleja, 157–69. Sevilla: Universidad de Sevilla. Hope, Valerie M. and Janet Huskinson, eds. 2011. Memory and Mourning. Studies on Roman Death. Oxford, Oakville: Oxbow Books. del Hoyo, Javier. 2002. “La ordinatio en los CLE Hispaniae.” In Asta ac pellege. 50 años de la publicación de Inscripciones Hispanas en Verso de S. Mariner, edited by Javier del Hoyo and Joan Gómez Pallarès, 143–62. Madrid: Signifer. del Hoyo, Javier. 2014a. “Acerca de dos carmina medievales de Hispania con acróstico y teléstico.” In Manipulus studiorum en recuerdo de la profesora Ana María Aldama Roy, edited by María Teresa Callejas Berdonés, Patricia Cañizares Ferriz, María Dolores Castro Jiménez, María Felisa del Barrio Vega, Antonio Espigares Pinilla, and María José Muñoz Jiménez, 311–26. Madrid: Escolar y Mayo. del Hoyo, Javier. 2014b. “Ne velis violare. Imprecaciones contra los violadores de tumbas.” In El mundo de los difuntos: culto, cofradía y tradiciones, edited by Francisco Javier Campos y Fernández de Sevilla, vol. 2, 809–24. San Lorenzo del Escorial: Ediciones Escurialenses. del Hoyo, Javier and Joan Gómez Pallarès, eds. 2002. Asta ac pellege. 50 años de la publicación de Inscripciones Hispanas en Verso de S. Mariner. Madrid: Signifer. Krummrey, Hans. 1963. “Zu dem akrostichischen Grabgedicht für Pilarus aus Sicca.” Helikon 3: 278–300. Kruschwitz, Peter. 2008. “Patterns of Text Layout in Pompeian Verse Inscriptions.” Studia Philologica Valentina 11: 225–64. Kruschwitz, Peter. 2010a. “Tree Inscriptions: Restoring a Lost Facet of the Graeco-Roman Epigraphic Habit.” Zeitschrift für Papyrologie und Epigraphik 173: 45–62. Kruschwitz, Peter. 2010b. “Attitudes towards Wall Inscriptions in the Roman Empire.” Zeitschrift für Papyrologie und Epigraphik 174: 207–18. Kruschwitz, Peter. 2015. “More than Meets the Eye: Fragrance, Sensuousness, and Inscribed Latin Poetry.” The Petrified Muse. thepetrifiedmuse.blog/2015/06/24/more-than-meets-the-eyefragrance-sensuousness-and-inscribed-latin-poetry/. Kruschwitz, Peter. 2016a. “Inhabiting a Lettered World. Exploring the Fringes of Roman Writing Habits.” Bulletin of the Institute of Classical Studies of the University of London 59.1: 26–41. Kruschwitz, Peter. 2016b. “Hello Stranger, or: Pompeian Greetings from Beyond the Grave!” The Petrified Muse. https://thepetrifiedmuse.blog/2016/11/04/hello-stranger-or-pompeiangreetings-from-beyond-the-grave/.

7.  When poetry comes to its senses

175

Kruschwitz, Peter. 2017. “‘I Shall Touch It with Care and Respect’: À Propos a Hitherto Neglected Senarius from Roman Britain (EE VII 928 = RIB 659).” Zeitschrift für Papyrologie und Epigraphik 204: 24–6. Kruschwitz, Peter. 2019. “How the Romans Read Funerary Inscriptions: Neglected Evidence from the Querolus.” Habis 50: 341–62. Kruschwitz, Peter. 2020. “Poetry on the Advance: The Emergence and Formation of a Poetic Culture in Roman Britain.” Greece & Rome 67.2: 177–202. Lattimore, Richmond. 1962. Themes in Greek and Latin Epitaphs. Urbana: University of Illinois Press. Liddel, Peter and Polly Low, eds. 2013. Inscriptions and Their Uses in Greek and Latin Literature. Oxford: Oxford University Press. Limón Belén, María. 2014. La compaginación de las inscripciones latinas en verso: Roma e Hispania. Rome: L’Erma di Bretschneider. Loreti, Ersilia Maria. 1994. “Signacula bronzei dell’Antiquarium di Roma.” In Epigrafia della produzione e della distribuzione: actes de la VIIe Rencontre Franco-Italienne sur l’Épigraphie du Monde Romain; Rome, 5–6 juin 1992, 645–53. Rome: L’Erma di Bretschneider. Maier, Thomas, Michael R. Ott, and Rebecca Sauer, eds. 2015. Materiale Textkulturen. Konzepte – Materialien – Praktiken. Berlin: De Gruyter. Manganaro, Giacomo. 2001. “Tra archeologia ed epigrafia: due note.” Zeitschrift für Papyrologie und Epigraphik 137: 189–95. Martín Camacho, Jesús. 2010. Carmina Latina Epigraphica Baeticae ex schedis: edición y comentario. Seville: Universidad de Sevilla. Masaro, Gabriele. 2017. Iscrizioni metriche e affettive della X regio augustea. Rome: Aracne. Massaro, Matteo. 2006. “Epigrafia metrica in alcuni colombari romani della prima età imperiale.” In Temptanda viast. Nuevos estudios sobre la poesía epigráfica latina, edited by Concepción Fernández Martínez and Joan Gómez Pallarès, 1–32. Bellaterra: Spuab. Michalopoulos, Andreas N. 1997. “Etymologising on Proper Names in Latin Epigraphic Verse.” Sandalion 20: 125–37. Monceaux, Paul. 1906. “Enquête sur l’épigraphie chrétienne d’Afrique.” Revue Archéologique 7: 177–92. Mullen, Alex. 2013. Southern Gaul and the Mediterranean. Multilingualism and Multiple Identities in the Iron Age and Roman Periods. Cambridge: Cambridge University Press. Orlandi, Silvia and Silvio Panciera. 1999. “Due note di epigrafia tardoantica.” Atti della Accademia Nazionale dei Lincei, Classe di Scienze morali, storiche e filologiche. Rendiconti, ser. IX, 10: 575–94. Pena, María José and Joan Carbonell. 2006. “Un interesante carmen epigraphicum de Pax Iulia (Portugal).” Humanitas 58: 157–73. Perrot, Sylvain. 2020. “Ancient Musical Performance in Context: Places, Settings and Occasions.” In A Companion to Ancient Greek and Roman Music, edited by Tosca A.C. Lynch and Eleonora Rocconi, 89–102. Hoboken, NJ: Wiley. Purves, Alex, ed. 2018. Touch and the Ancient Senses. London, New York: Routledge. Rhoby, Andreas. 2017. “Text as Art? Byzantine Inscriptions and Their Display.” In Writing Matters. Presenting and Perceiving Monumental Inscriptions in Antiquity and the Middle Ages, edited by Irene Berti, Katharina Bolle, Fanny Opdenhoff, and Fabian Stroth, 265–84. Berlin: De Gruyter. https:// doi.org/10.1515/9783110534597-011. Rosenmeyer, Patricia A. 2018. The Language of Ruins: Greek and Latin Inscriptions on the Memnon Colossus. Oxford, New York: Oxford University Press. Rudolph, Kelli C., ed. 2018. Taste and the Ancient Senses. London, New York: Routledge. Sanders, Gabriel. 1979 “L’au-delà et les acrostiches des Carmina Latina Epigraphica.” Roczniki Humanistyczne 27.3: 57–75. Sanders, Gabriel. 1991. Lapides memores: Païens et chrétiens face à la mort: Le témoignage de l’épigraphie funéraire latine. Faenza: Lega.

176

Chiara Cenati, Victoria González Berdús, and Peter Kruschwitz

Scheid, John. 2019. La religion des Romains, 4th ed. Malakoff: A. Colin. Schmidt, Manfred G. 1999. “Ursus togatus (CIL VI 9797).” Zeitschrift für Papyrologie und Epigraphik 126: 240–2. Schmidt, Manfred G. 2019. “The Magic Square from Madauros (Numidia, 2nd/3rd century AD).” Custos Corporis, https://custos-corporis.com/the-magic-square-from-madauros-numidia-2nd3rd-century-ad/. Schwitter, Raphael. 2019. “Funkelnde Buchstaben, leuchtende Verse – Die Materialität der Inschrift und ihre Reflexion in den Carmina Latina Epigraphica.” In Antike Texte und ihre Materialität. Alltägliche Präsenz, mediale Semantik, literarische Reflexion, edited by Cornelia Ritter-Schmalz and Raphael Schwitter, 119–38. Berlin: De Gruyter. Socas, Francisco. 2002. “Materiales para una tipología de los epitafios latinos trazada a partir de sus voces e interlocutores.” In Asta ac pellege. 50 años de la publicación de Inscripciones Hispanas en Verso de S. Mariner, edited by Javier del Hoyo and Joan Gómez Pallarès, 183–204. Madrid: Signifer. Sotgiu, Giovanna. 2000. “Due ritrovamenti epigrafici dalla Sardegna.” In Ἐπιγραφαί: miscellanea epigrafica in onore di Lidio Gasperini, edited by Gianfranco Paci, 1011–19. Tivoli: Tipigraf. Squire, Michael. 2009. Image and Text in Graeco-Roman Antiquity. Cambridge: Cambridge University Press. Squire, Michael, ed. 2016. Sight and the Ancient Senses. London, New York: Routledge. Šterbenc Erker, Darja. 2011. “Gender and Roman Funeral Ritual.” In Memory and Mourning. Studies on Roman Death, edited by Valerie M. Hope and Janet Huskinson, 40–60. Oxford, Oakville: Oxbow Books. Susini, Giancarlo. 1968. Il lapicida romano. Introduzione all’epigrafia latina. Rome: L’Erma di Bretschneider. Susini, Giancarlo. 1982. Epigrafia romana. Roma: Soc. Ed. Jouvence. Thüry, Günther E. 1995. “Zur Deutung einer römischen Fassinschrift aus Regensburg.” Bayerische Vorgeschichtsblätter 60: 301–2. Thüry, Günther E. 1998. “Wasser im Wein. Zur Deutung einer Spruchbecherinschrift aus Szentendre (Ungarn).” In Mille fiori. Festschrift für Ludwig Berger zu seinem 65. Geburtstag, Forschungen in Augst 25, edited by Eckhard Deschler-Erb, Renate Ebersbach, and Gertrud Grossmann, 207–210. Augst: Römermuseum. Thüry, Günther E. 2008. “Die erotischen Inschriften des instrumentum domesticum: ein Überblick.” In Instrumenta Inscripta Latina. 2, Akten des 2. Internationalen Kolloquiums, Klagenfurt, 5.-8. Mai 2005, edited by Manfred Hainzmann and Reinhold Wedenig, 295–304. Klagenfurt: Verlag des Geschichtsverein für Kärnten. Webster, Colin. 2019. “The Soundscape of Ancient Greek Healing.” In Sound and the Ancient Senses, edited by Shane Butler and Sarah Nooter, 109–29. London, New York: Routledge. Wille, Günther. 1967. Musica Romana: Die Bedeutung der Musik im Leben der Römer. Amsterdam: Schippers. Zarker, John W. 1958. Studies in the Carmina Latina Epigraphica. Princeton: Princeton University Press. Zarker, John W. 1966. “Acrostic Carmina Latina Epigraphica.” Orpheus 13: 125–51.

Chapter 8 Lassi viatores: poetic consumption between Martial’s Epigrams and the Carmina Latina Epigraphica Alessandra Tafaro

Martial’s Epigrams create witty dialogues with a multifaceted reading public and create numerous scenes of readerly consumption. Patrons, friends, the emperor and the anonymous readers are all imagined as consumers of Martial’s poetic work, creating what Fitzgerald has suggestively termed a heterogeneous “society of the book”.1 With the expansion of the book trade and a proliferating commodification of literature, Martial reflects on the material circulation of his poetic work and imagines a democratization of literary circulation and consumption.2 The most disparate layers of society, from élite to lower classes, women to slaves, as Martial proudly boasts in a post-Ovidian gesture, engage with his epigrams: ille dabit populo patribusque equitique legendum, / nec nimium siccis perleget ipse genis, “He will give you to people and Fathers and knights for them to read, and will peruse you himself with eyes not altogether dry” (12.2.15–6).3 As Fowler suggests, no other ancient poet imagines the material circulation and consumption of poetry in the vivid way Martial does.4 The epigram had had a long-standing tradition in the celebratory exchanges between poet-clients and patrons in Rome.5 Carried around baths, theatres, and banquets, this nugatory poetry, which Martial ironically represents as ioci and nugae sitting at the bottom of the literary hierarchy, perfectly suited the convivial and festive contexts of Roman social life, from the arena spectacles of the Liber spectaculorum, to the Saturnalian

Fitzgerald 2007, 139. For the expansion of the book trade and book market see discussion in Citroni 1995 and White 2009. 3 Martial’s texts are quoted from the Loeb edition (Shackleton Bailey 1993). Translations throughout are taken from the Loeb Classical Library, unless otherwise specified. 4 Fowler 1995. 5 Fitzgerald 2007, 27; Rimell 2008, 6. 1 2

178

Alessandra Tafaro

gift-tags of the Xenia and Apophoreta.6 When the poet issued his liber primus, however, he challenged the expectations of a literary genre hitherto identified with ephemeral verses, embedded in specific social contexts and directed to immediate uses.7 While the Xenia and Apophoreta, tied as they were to the occasion of the Saturnalia, extensively play with their “non-literary status” and practical usefulness, Martial’s publication of epigrammatic collections disrupts the traditional consumption of the epigrams as individual compositions, dashed off in the hic et nunc for social entertainment.8 Martial, following in the footsteps of Ovidian exilic poetry, responds to and reflects on the new marketization of literature by offering his epigrams to multiform audiences. Anonymous readers are eagerly committed to Martial’s witty poems and offer him love, devotion and glory. Yet, they are also often imagined to be fickle in their consumption of epigrams, whose number may result in boredom. Although recognized as one of the most distinctive features of his poetics, Martial’s construction and imagination of his anonymous public needs reconsideration in the light of epigraphic poetry. While scholars have suggested that Martial’s conceptualization of his anonymous readership is much inspired by Ovid’s exile poetry, I shall argue that the unprecedented centrality of the anonymous lector entertains a key intertextual dialogue with the anonymous reader par excellence, the viator encountered within inscriptions. In response to the new circulation of his epigrams in the book form, Martial often depicts his reader as meta-poetically satiated by the length and number of epigrams. When at the beginning of Book Two Severus is compared to a lassus viator, who is unwilling to pursue his reading journey through the book, after having scanned just two pages, Martial explicitly introduces the image of the reader/passer-by who tirelessly engages with the epigrams. Similarly, three Latin verse inscriptions address lassi viatores, who, supposedly ill-disposed to stop and read, are wittily dismissed by the speaking tombstones. Höschele has demonstrated how addressing the epitaphic passer-by, which is ingrained into epigraphic communication, survives metamorphosed within epigrammatic collections, where the poems similarly strive to attract the reader’s attention.9 As scholars have noted, throughout his production Martial morphs the passer-by reading epitaphs on stone into a meta-literary “wanderer” within the literary space of the epigrams.10 The poet re-invents the epigraphic dialogues with moving readers, but also reworks the inscriptions’ engagement with readers and opens up the possibilities offered by epitaphs to the mordant wit typical of epigrams, putting into play the paradoxes already encapsulated in inscriptions.

Gowers 1993, 245. Roman 2001, 113. 8 Citroni 1988, 13. 9 Höschele 2007. 10 Höschele 2007, 333–69. 6 7

8.  Lassi viatores

179

In this chapter, I shall explore parallel concerns in the readerly reception of and the engagement with epigrams within their stone and book realizations, in order to reconsider the role played by different media in both the poetic creation and consumption. In what follows, I shall address how Martial’s satirical representations of readers difficult to please wittily interlock with anxieties already voiced by roadside epitaphs, whose imaginings of poetic reception are as complex as Martial’s. In the context of Martial’s poetry, the reader, who meta-poetically moves through the epigrams and physically within the city, carrying around the epigrammatic libellus, evokes the reader of many inscribed epigrams. While the epigraphic viator is called upon by inscriptions to stop and read, the epigrammatic reader is attracted by numerous poetic snippets and encouraged to pursue his reading activity. Yet, he is potentially distracted by the otiose and convivial contexts which are so congenial for epigram. Martial thanks his studiosus lector for bestowing him literary success in life. Nevertheless, Martial’s satirical representation of otiose readers unwilling to engage with epigrams in contexts of otium reworks epigraphic dramatization of dialogues between tombstones and neglectful viatores. At the beginning of Book Two and in the concluding sequence of Book Eleven, Martial’s construction of his reader as a lassus viator (2.6.14), a weary traveller in his epigrammatic journey, wittily manipulates the epitaphic memento mori, which threatens tired passers-by. By exploiting the wit already embedded in the epigraphic interaction with the reading public, Martial harnesses the potential of epigraphic language to represent the need to fend off the reader’s boredom. In contrast to previous scholarship which sets out to trace verbal parallels between epigrams and inscribed poems, my comparison of scenes of readerly reception in Martial’s epigrams and verse inscriptions disrupts discourses of directionality that have so far hindered our appreciation of the relationship between literature and epigraphy. Furthermore, it shall illuminate the literary wit of inscriptions, suggesting a dynamic interaction in the creation and appreciation of epigrammatic poetry across different material contexts. Reading Martial’s work through an epigraphic lens, this chapter shall prompt a new appreciation of the potential embedded in the epigraphic language. Simultaneously, it shall encourage us to re-frame Martial’s exploration of the paradoxes concealed in the act of reading, his addresses to lazy readers and the double-edged attitudes towards his readerships as influenced by epigraphic production, whose paradoxical potentials and witty brevitas are made distinctively epigrammatic.

Praising and attacking: Martial’s lector As early as Book One Martial reworks key tropes of epigraphic poetry. By giving new prominence to an anonymous reader, who is imagined in many respects as a passerby, Martial advertises the wide circulation of his own poetry, whose portability has achieved a newfound dissemination, one which increases the possibilities already

180

Alessandra Tafaro

tested by Horatian and Ovidian poetics of the book. When the reader embarks in Book One, Martial introduces him with a multifaceted audience: the anonymous reader (1 Praef.; 1.1; 1.2), the emperor (1.4–5) and named patrons (1.7) all make their entrance on the epigrammatic stage: epigrammata illis / scribuntur qui solent spectare Florales, “Epigrams are written for those who are used to watch Flora’s games” (1 Praef. 16–17). The anonymous reader plays here a crucial role. He is identified as the main dedicatee of the book in the preface and embodies the reason for Martial’s worldwide renown within the first epigram: cui, lector studiose, quod dedisti / viventi decus atque sentienti, / rari post cineres habent poetae, “Devoted reader, the glory you have given him while he lives and feels comes to few poets in their graves” (1.1.4–6). Furthermore, he is represented as a voracious consumer, who, physically moving through the city, wants Martial’s books to become his journey companions: qui tecum cupis esse meos ubicumque libellos / et comites longae quaeris habere viae, “you who want my little books to keep you company wherever you may be and desire their companionship on a long journey” (1.2.1–2). The addressee of epigram 1.2 is a reader in motion, whom Martial redirects to the bookshop of Secundus to buy his portable codex edition: ne tamen ignores ubi sim venalis et erres / urbe vagus tota, me duce certus eris, “but in case you don’t know where I am on sale and stray wandering all over town, you will be sure of your way under my guidance” (1.2.5–6). Not only does Martial productively align his reader to a wanderer in the city, but also, by comparing his books to journey companions, he advertises the portability of his own epigrams, which, unlike still epitaphs, can move together with the reader in the physical space of Rome.11 Furthermore, by virtue of intertextual links and juxtapositions, single epigrams can travel separately both within and outside the book, acquiring themselves a movability that was previously unknown to epigraphic poetry.12 Across his books Martial frequently constructs the reader/passer-by as a benevolent, affectionate reader, fond of his epigrams, who allows the poet to live a glorious poetic career before rather than after death.13 Like the epitaphic reader, who rescues the deceased from forgetfulness through the act of reading, the epigrammatic lector is endowed with the power to perpetuate Martial’s name, letting him live on across the centuries: lector, opes nostrae: quem cum mihi Roma dedisset, ‘nil tibi quod demus maius habemus’ ait. 11 Martial’s poetics of the portable books rework both Horace’s (Ep. 1.20) and Ovid’s models (Tr. 1.1; 3.1). See Citroni 1986; Hinds 2007. See also Mart. 9.99.5–6: tu qui longa potes dispendia ferre viarum, / i, liber, absentis pignus amicitiae, “Book, who can bear long stretches of travel, go, pledge of absent friendship.” 12 Bing 2009, 141: “For epigrams on scrolls – unlike their stationary archetypes – could move about, readily disseminated, becoming themselves the ‘wayfarers’ (ὁδοιπόροι, παροδῖται, etc.) that their readers had by necessity been before. Because of this newfound mobility, epigram begins to display for the first time a pervasive – indeed, exuberant – intertextuality: poets avidly respond to and vary each other’s book epigrams in a manner rarely seen between inscribed ones.” 13 Citroni 1975, 16.

8.  Lassi viatores

181

‘pigra per hunc fugies ingratae flumina Lethes et meliore tui parte superstes eris. Mart. 10.2.5-8 Reader, you who are my riches. When Rome gave you to me, she said: “I have nothing greater to give you. Through him you will escape ungrateful Lethe’s idle waters and survive in the better part of yourself.”14

When Martial reflects upon the transcendence and long-lasting legacy of his epigrams in comparison to the crumbling physical monuments of Licinius and Cripsus, he identifies the reader with the ultimate source of his own immortality, the one through whom he will escape the fatal Lethe and survive in his better part.15 Similarly to inscriptions, for which the act of reading is essential to prevent the memory of the deceased from fading away, Martial acknowledges the crucial role of his reader, through and thanks to whom his work and his own authorial persona survive. Indeed, poetry without readers can and does die. As Martial provocatively suggests, non scribit, cuius carmina nemo legit, “he is not a writer, he whose poems nobody reads” (3.9.2).16 Nevertheless, Martial’s relationship with his public is fundamentally double-edged. Martial praises his devoted reader, who allows him to live a stellar poetic career. Yet, he often satirically represents him as bored or difficult to please.17 On these occasions, rather than representing the small poetic Callimachean domain and the aesthetic principle of the μέγα βιβλίον μέγα κακόν, brevitas becomes instrumental in encouraging the reader to pursue his reading journey and is associated with the need to fend off the readerly boredom. In these contexts, Martial deliberately plays with the epigraphic counterpart of the reader, the viator or hospes, who is normally required to take a moment to just read a few lines: paulisper requiesce (CLE 1591), paulisper consiste (CLE 2024), pauca legas (CLE 1125) are common expressions in the epitaphic rhetoric which strive to stimulate the reader in his reading activity. In all these requests a crucial aspect of the reader’s engagement with inscriptions emerges: the dimension of time involved in the act of reading. By imagining his lazy reader as a hasty traveller passing by an inscription, Martial voices paradoxical concerns about his readerly reception that hark back to the epigraphic fear of being ignored and remaining un-read. By aiming at a reluctant reader, as inscriptions aimed at a supposedly indifferent audience, Martial’s attitude towards the bored audience seems to respond to a self-deprecatory strategy as well as a mocking attitude towards his readership. The epilogues of Books One (epigrams 1.117–8) and Eleven (11.106–8) and the proemial section of Book Two (2 Praef. and 2.6) host several consumers of epigrammatic The translation here provided is my own adaptation of the Loeb Edition 1993. Similarly, Mart. 8.3; see Williams 2002, 422. On the Ovidian intertext see Williams 2002 and Rimell 2008, 53–98. 16 The translation here provided is my own adaptation of the Loeb Edition 1993. 17 Williams 2004, ad 2.1.11 and 2.6. 14 15

182

Alessandra Tafaro

poetry: the avid reader, the busy patron, and the bored viator all mingle together in different imagined scenes of poetic reception. Martial strongly alludes to the epigraphic model through his dialogue with these distracted and potentially weary readers in an attitude of affected modesty. The construct of the reader/passer-by allows Martial to defend but also to advertise his poetry, which is obsessed with its own reception. As Canobbio suggests, the epilogues of Book One and Eleven host readers who show indifference or are ill-disposed towards Martial’s work.18 By creating precise verbal and thematic correspondences between the readers represented in epigrams 1.117 and 1.118 and those who inhabit the final sequence of Book Eleven, 11.106–8, Martial, as critics argue, gestures towards a fresh poetic start under the auspices of Nerva’s reign. This traditional view, however, has underplayed the more complex ways in which epigraphic poetry is construed: the apostrophe to the wayfarer is a constitutive element of verse inscriptions through which we can grasp the profoundly innovative structure and literary qualities of poetry on stone. It is not surprising that these collections of Martial are more deeply concerned with readerly responses to the epigrammatic book. While in Books One and Two satirical scenes of reception respond to the desire to defend the new consumption of epigrams in book form, the preoccupation in Book Eleven with poetic reception emerges anew in connection with a mutated political climate.19 Domitian’s downfall in AD 96 trembled with new insecurities for Martial’s role as a court poet and, ultimately, posed new challenges to his own position in the system of imperial literary patronage. When Martial publishes Book Eleven, he negotiates his poetic position with the new political power by addressing his affectionate anonymous readership, whilst anticipating his political and poetic dislocation from the city of Rome.20 A close comparative reading of epigrams across Books One, Two, and Eleven, alongside verse inscriptions attacking neglectful readers will illuminate how Martial manipulates and appropriates the witty potentials of epigraphic poetry to ironically attack his otiose readerships.

Sapis / sapisti! – satis / satur. Stopping and passing by At the end of Book One we encounter the satirical representation of a failed commercial transaction between author and reader where both contracting parties end up defeated (1.117). A character simply identified as Lupercus, who represents the hypostasis of the common reader, asks Martial to lend him the latest epigrammaton libellum, a periphrasis which provocatively evokes and advertises the title of the poet’s entire work (argutis epigrammaton libellis, 1.1.3).21 In response to Lupercus’ stinginess, who would send a slave-boy to pick up the latest copy of the epigrams, Martial depicts an impervious itinerary to his own apartment in the Quirinal (ad Pirum, 6) and redirects Canobbio 2007, 220. Citroni 1988; Merli 1993; Canobbio 2007. 20 Fearnley 2003; König 2018; Rimell 2018. 21 Citroni 1975, ad loc. 18 19

8.  Lassi viatores

183

both Lupercus and his anonymous public to the book shop in the nearer area of the Argiletum. There, as Martial tells us, Atrectus sells a refined copy of the book for only five denarii. Precisely as in epigram 13.3, Martial ascribes a precise price to his own work, which can be purchased at Trypho’s shop for four denarii: Omnis in hoc gracili Xeniorum turba libello / constabit nummis quattuor empta tibi. / quattuor est nimium? Poterit constare duobus, / et faciet lucrum bybliopola Tryphon, “The entire assembly of Mottos in this slender little book will cost you four sesterces to buy. Is four too much? It could cost two, and bookseller Trypho still make a profit” (13.3.1–4). In a materialistic way, Martial aligns his poetry to a material object, exchangeable, acquirable, and priced. In a ring composition with epigram 1.2 and in juxtaposition with epigram 1.113, the reader becomes acquainted with the booksellers Secundus, who owns Martial’s new codex edition and Valerius Pollio, thanks to whom Martial’s juvenilia can live on. Nevertheless, the affair between Lupercus and Martial wittingly ceases, because Lupercus, even Martial admits in a tone of devaluation of his own poetry, is not so stupid as to acquire the book: Occurris quotiens, Luperce, nobis, ‘vis mittam puerum’ subinde dicis, ‘cui tradas epigrammaton libellum, lectum quem tibi protinus remittam?’ non est quod puerum, Luperce, vexes. longum est, si velit ad Pirum venire, et scalis habito tribus, sed altis. quod quaeris propius petas licebit. Argi nempe soles subire Letum: contra Caesaris est forum taberna scriptis postibus hinc et inde totis, omnis ut cito perlegas poetas. illinc me pete. †nec† roges Atrectum – hoc nomen dominus gerit tabernae – de primo dabit alterove nido rasum pumice purpuraque cultum denarîs tibi quinque Martialem. ‘tanti non es’ ais? sapis, Luperce. Mart. 1.117 Whenever you come my way, Lupercus, you say straight off: “May I send a boy for you to give him your little book of epigrams? I’ll return it to you as soon as I’ve read it.” Lupercus, you need not trouble the boy. It’s a long way for him to come to Ad Pirum and I live up three flights of stairs, long ones too. You can look for what you want closer to hand. No doubt you often go down to Argiletum. Opposite Caesar’s Forum there’s a shop with its doorposts completely covered by advertisements, so that you can read the entire list of poets at a glance. Look for me there. Ask for Atrectus (that being the name of the shop’s proprietor), and he will hand you from the first or second pigeonhole a Martial, shaved with pumice and smart with purple, for five denarii. “You’re not worth it,” you say. You’re a man of sense, Lupercus.

184

Alessandra Tafaro

The concluding ironic statement of this epigram, in which Martial himself paradoxically agrees with Lupercus’ “you are not worth it,” wittily recalls the poet’s response to another selfish reader at the end of Book Four, who refuses to pay up for Martial’s nugae: ‘aes dabo pro nugis et emam tua carmina sanus? / non’ inquis ‘faciam tam fatue.’ nec ego, “‘Am I to give cash for rubbish’, you say, ‘and buy your verses in my right mind? I’ll be no such fool.’ No more will I” (4.72.3–4). In these satirical scenes Martial spotlights the downgraded status of his poetry, which, identified with cheap entertainment, is not deemed worth paying for. Lupercus’ haughty “tanti non es” (1.117.18), which reverberates in the later “non faciam tam fatue” (4.72.4), is vividly evoked by epigram 11.106. The pointe encapsulated in the verb sapere, which verbally connects these epigrams, links the Lupercus willing to read Martial but not ready to pay for it and the busy patron Vibius Maximus, who is given permission to skip even the four lines dedicated to him. In recreating the same fulmen in clausula with the verbs sapis, sapisti, Martial self-deprecatingly shares the reluctant attitude of his readers. Vibi Maxime, si vacas havere, hoc tantum lege: namque et occupatus et non es nimium laboriosus. transis hos quoque quattuor? Sapisti. Mart. 11.106 Vibius Maximus, if you have no time to say hello, read only this; for you are a busy man and not over-industrious. Do you pass-by these four verses too? You show your sense.

The lexical texture of this poem and the satirical representation of a reader in motion unwilling to engage with reading explicitly interact with epigraphic poetry. Martial is here mocking his own work, which is not worth paying for in 1.117 and not even worth reading in the present poem, but he is also teasing Vibius, who will miss the joke if he does not read it. In arguing against White’s libellus theory,22 according to which Martial presented private patrons with small collections of epigrams before their publication in book format, Nauta convincingly shows that the current closural position of the epigram heightens the pointedness and wit underlying the well-known literary trope of the busy patron (namque et occupatus, 2), whose negotia prevent him from engaging with Martial’s nugae. Since Vibius is too busy, Martial paradoxically White 1974, 4 argues that this epigram is ill-suited in the present position, for Vibius would have had to unroll the entire volume before finding the piece dedicated to him and before learning that he was allowed to limit his reading to these four lines. Kay 1985, 284 supports White’s theory that the epigram should have been placed at the head of a private smaller collection of epigrams privately presented to the patron. The current position of the epigram, at the end of the book, creates a joke for further intersecting readerships, not for Vibius. Citroni 1988, 58–9 and Merli 1993, 253 believe that the epigram gains its significance if considered as part of a previous unpublished private collection. Contra and more convincingly, Fowler 1995, 47 and Nauta 2002, 119. 22

8.  Lassi viatores

185

encourages a desultory reading of his book, and allows him to read only this one epigram. The punchline expressed by sapisti enhances the epigrammatic humorous tone: indeed, Vibius, who is imagined passing-over everything else in the book (hos quoque), is complimented by Martial for having neglected the poem he has just read.23 With this epigrammatic satirical scene not only does Martial underscore his role as poet-client who seeks the patronage of the influential man, but also paradoxically harnesses the epigraphic language to mock the patron’s readerly attitude. As Fitzgerald emphasizes, Vibius is “here hailed not only as though by a petitioner with a libellus (think of Augustus and the graeculus), but also as a passer-by is hailed by an epitaph”.24 Martial’s epigram seems to function as an epitaph trying to escape readerly indifference. Like funerary inscriptions, the epigram calls out for the passerby’s attention, putting into play the dimension of time involved in the engagement with monuments and the notion of reading small-scale poetry at times of leisure. Both the verb transire at line four and the request hoc tantum lege at line two echo linguistic features likely to be found in verse inscriptions and call into play the paradoxes at the core of the epigraphic requests to the reader to stand still and read.25 The syntagm hoc tantum lege points to the epitaphic mode of attracting the viator by reassuring him that reading will take just a few moments. Epitaphic compression and shortness become – as in Martial – instrumental to preventing the readers’ boredom or unwillingness to approach the epigrams. A few examples from the Carmina Latina Epigraphica will show the ways in which the poet appropriates epigraphic expressions to spur his readers in their reading activity. Tu qui secura procedis mente, parumper siste gradum quaeso uerbaque pauca lege. illa ego quae claris fueram praelata puellis, hoc Homonoea breui condita sum tumulo, cui formam Paphie; Charites tribuere decorem, quam Pallas cunctis artibus erudiit. nondum bis denos aetas mea uiderat annos, iniecere manus inuida fata mihi. nec pro me queror hoc, morte est mihi tristior ipsa maeror Atimeti coniugis ille mei. ‘sit tibi terra leuis, mulier dignissima uita quaeque tuis olim perfruere bonis’. CLE 995, ll. 1–12 You who make your way with nothing on your mind, halt briefly, please, and read a few words. I, that Homonoea who had been given the palm over famous women, am laid to rest in this small tomb. To me Venus gave beauty, the Graces comeliness; Athena trained me in every accomplishment. My youth had not yet seen twenty years when grudging destiny laid Nauta 2002, 119. Fitzgerald 2007, 144. 25 Höschele 2007, 353. 23 24

186

Alessandra Tafaro hold of me. I do not make this complaint for myself; that grief of my husband Atimetus is more bitter to me than death itself. “May earth lie light on you, woman who deserved life and enjoyment of your blessings long ago.”26

This elegant Roman epitaph, dating to the first century AD, set up by Atimetus Anterotianus for himself and his wife Claudia Homonoea, hosts a complex series of speech acts: the opening dialogue between the deceased woman and the wayfarer (lines 1–2 and 11–12) is followed by an exchange between husband and wife (lines 13–18 and 1926). The voice of the deceased Claudia, speaking through the stone, requires the traveller to stop for a moment (parumper siste gradum) and read only a few lines (uerbaque pauca). The following verses encapsulate the biographical information of Claudia Homonoea, snatched away from life at the age of not yet 20 by the envious hand of fate, nondum bis denos aetas mea uiderat annos, / iniecere manus inuida fata mihi, “My youth had not seen twenty years when grudging destiny laid hold of me” (7–8). To the initial request for sympathy, the viator is imagined uttering a good wish for the deceased: “sit tibi terra leuis, mulier dignissima uita / quaeque tuis olim perfruerere”, “May the earth lie light on you, woman who deserved life and enjoyment of your blessing long ago” (11–12). The concluding section of the poem displays the husband’s sufferings, who, having outlived Claudia, invokes a speedy death to reach his wife, at nunc, quod possum, fugiam lucemque deosque / ut te matura per Styga morte sequar, “but as matter stands in reality, I shall shun the light of day and the gods, which is all I can do, so that I can follow you over the Styx in speedy death” (17–18). The dialogue between Claudia Homonoea and the viator, whose act of reading performs a commemorative function, is a crucial feature of epitaphs, which, lining the roadsides, seek to be rescued from oblivion.27 The aforementioned epitaph shows the direct address of the deceased to the reader. There are, however, instances in which it is the gravestone itself speaking in the first person, displaying a more ironical attitude towards the passer-by. The earliest attestation from the Roman world dates to the Republican times: hospes, quod deico paullum est, asta ac pellege heic est sepulcrum hau pulcrum pulcrai feminae nomen parentes nominarunt Claudiam. suom mareitum corde deilexit souo. gnatos duo creauit: horunc alterum in terra linquit, alium sub terra locat. sermone lepido, tum autem incessu commodo, domum seruauit, lanam fecit. dixi: abei. CLE 52 26 Texts of verse inscriptions are quoted from F. Buecheler and E. Lommatzsch. 1895–1926. Carmina Latina Epigraphica, vols. I-III. Leipzig. The translations provided are my own, unless otherwise specified. For the translation of CLE 995, see Courtney 1995, 378–9. 27 Lattimore 1942, 230–7; on the figure of the viator as addressee of epitaphs, see the landmark work by Tueller 2008; see also Gregori 2008, 3–115.

8.  Lassi viatores

187

Stranger, what I have to say is short, stand still and read. This is the unlovely tomb of a lovely woman. Her parents gave her the name Claudia. She loved her husband with all her heart. She gave birth to two sons; one of them she leaves on earth, the other she places beneath it. She was charming in conversation and modest in gait. She kept to the house and made wool. This is all I have to say; be on your way.28

The famous epitaph for Claudia, a devoted wife and mother of two sons, has attracted much scholarly attention for its unconventional engagement with the readership.29 Unlike the previous carmen, the tombstone is here directly addressing the wayfarer (hospes, 1), who is invited to stand still (asta, 1), read through (pellege, 1) and then leave (abei, 8). The conciseness of the poem (quod deico paullum est, 1) and the related promise of the few instants involved in the act of reading urge the traveller to engage with the text. After having accomplished his duties, the passer-by is hastily dismissed by the concluding future imperative abei, which is likely to anticipate the more assertive and “aggressive” attitudes towards the potentially careless readers which mark later imperial epitaphs.30 As Lattimore suggests, the address to the viator is so integral to epigraphic communication that it was preserved when the epigram migrated from stones to papyrus, revealing how epigraphy had exerted its influence on literature.31 Literary epitaphs, which extensively deploy fictitious dialogues between the deceased and the wayfarer as a distinctive epigraphic feature, increase the focus on the poetic recipient, whose role in “the creation of meaning” and whose power over the text is heightened over the Hellenistic period.32 By editing the book down to size, controlling the pace of reading and consuming anew or ignoring poems, the reader of the epigrammatic book dominates the reception of the poetic text. The (often double-edged) epitaphic dialogue between the voice of the stone and the wayfarer is cleverly reinvented by the epigrammatic authors and becomes quintessential to the numerous sepulchral epigrams within Book Seven of the Palatine Anthology and the overly famous literary epitaphs of Pacuvius and Ennius. Rather than confining the address to the reader/viator to funerary contexts,33 however, Martial opens up the construct of a travelling and (supposedly reluctant) audience, the question of time involved in the act of reading and the paradoxes ingrained in the epitaphic dialogues to the satirical representations of a lazy and neglectful readership. As we have appreciated, in many epitaphs brevitas is used as a Leitmotif to compel an (otherwise disinterested) passer-by to read: cur tantum Translation by Courtney 1995, 234. The speaking tomb: CLE 848, Adulescens, tam etsi properas, hic te saxsolus / rogat ut se aspicias, deinde ut quod scriptust legas; CLE 53, Rogat ut resistas hospes te hic tacitus lapis. For further Republican examples see Lattimore 1942, 232 n. 128 and Massaro 1992, 88–90. 29 On CLE 52 see Massaro 1992, 78–114; Courtney 1995, 234–6; Kruschwitz 2013, 193–5. 30 Kruschwitz 2013, 194. 31 Lattimore 1942, 230. 32 Höschele 2007, 339. 33 Henriksén 2006, esp. 358–61. 28

188

Alessandra Tafaro

properas? “why do you hasten?” (CLE 513.2). The expression parumper siste addressed to the traveller in the epitaph CLE 995 highlights how, paradoxically, only the passerby who has already stopped in front of the inscription and read will be receiving the request of roadside monuments.34 In epigram 11.106, therefore, Martial appropriates the epigraphic conventional modes of addressing an indifferent reader moving along a road. The verb transire in epigram 11.106.4 acquires, in fact, a meta-literary function. In the epigraphic contexts the periphrasis tu qui transis (akin to tu qui praeteris or praeteriens) refers to the viator, the (potential) reader who, in passing by, confronts himself with a roadside epitaph. Such indications are normally combined with the requests to read (lege); look upon the tomb (aspice); utter a good wish that earth may lay light upon the dead (dice sit tibi terra levis); or shed tears in an act of sympathy towards the deceased (flete meos cineres).35 Martial’s epigram 11.106 deploys the verb transire to signify the meta-poetic movement of the influential patron Vibius Maximus, who, moving in the literary space created by the epigrams, is represented as passing by the poem addressed to himself. However, unlike inscriptions in which only the deceased and dedicatees are named, the epigraphic anonymous addressee is here hailed by name. The question with which the reader is left is: who is here playing the role of the traditional viator? The anonymous reader, who reads the poem pretentiously ignored by the busy patron, embodies here Martial’s meta-literary viator. Frequens uiator saepe qui transis lege:

CLE 123.1

Assiduous traveller who always pass by, read.

The fragmentary epitaph CLE 123, a funerary stele for the auriga Aelius Hermerotus from Hispania Citerior, offers a clear example of the use of the verb transire as referred to a traveller in a road called out by an epitaph.36 Both in epigram 1.117 and 11.106 Lupercus and Vibius, thus, sapiunt: they are not just wise for not having bought or read Martial, but they also have good taste. Martial is here possibly playing with the double meaning of sapere as both “to have knowledge of, to have discernment” and “to taste, to have flavour”, hinting at the metaphor of poetry as ephemeral and throw-away food, whose existence is coextensive to the duration of a banquet or dinner-party (OLD s.v. II.2). In Martial’s epigram 11.106 the address to Vibius juxtaposes different readerships of the epigrammatic book, both the intended addressee of the epigram and the generic anonymous reader. While roadside inscriptions invite acts of sympathy from the passer-by and voice resentment for those who disregard such prayers, Martial ironically concludes that Vibius’ heedless engagement with his epigram-epitaph is wise. The epigraphic model is debased to Höschele 2007, 344. Lattimore 1942, 230–7; Morelli 2017, 123–5. 36 Similarly, CLE 1452: dic rogo qui transis sit tibi / terra leuis; CLE 1152: qui uia Flaminia transis / resta ac relege. 34 35

8.  Lassi viatores

189

increase the final humorous joke, in which Martial self-deprecatingly agrees that his poetry is pointless. Vibius Maximus’ indifferent attitude towards Martial’s poetry is closely echoed by the Septicianus in the following poem, 11.107, who pretends to have read Martial throughout and gives him back his book quasi perlectum, 2: Explicitum nobis usque ad sua cornua librum et quasi perlectum, Septiciane, refers. Omnia legisti. Credo, scio, gaudeo, verum est. Perlegi libros sic ego quinque tuos. Mart. 11.107 Septicianus, you give me back my book unwound right to his horns as if you read it all through. And you have read it all through, of course. I believe you, I know it, I’m pleased, it’s true! I read your five books through in precisely the same way.

The epigram creates a ring composition with the opening section of the book. Like Parthenius, who has left the book unread and unrolled, vadas et redeas inevolutus, “go and return unrolled” (11.1.4), Septicianus has returned Martial’s manuscript unwound (11.107.2). Underlying both these satirical representations of busy patrons and lazy readers are the different conditions of poetic production and imperial patronage after Domitian’s assassination, which prompt Martial to refashion his poetic position against the backdrop of the imperial power.37 Parthenius, who had been previously asked to introduce Martial’s books into Domitian’s library, had been involved in the palace conspiracy against the last of the Flavians.38 Martial’s decision to address him at the outset of a book dedicated to the new emperor Nerva crucially illuminates the shift in the poet’s political engagement with the imperial court, (pretentiously) forgetful of the Domitianic past and welcoming a new era of unrestrained epigrammatic and Saturnalian frankness. Previously imagined reaching magnas Caesaris manus (6.1.5), the book, now in the Porticus Quirini, will content itself with the manus minores (11.1.8) of a few potential readers, who will eventually turn to the epigrams, once the chariotracing gossip will have ceased: sunt illic duo tresve, qui revolvant / nostrarum tineas ineptiarum, / sed cum sponsio fabulaeque lassae / de Scorpo fuerint et Incitato, “there are two or three people there who may unwind the moths of my triflings, but only when the gossip and betting on Scorpus and Incitatus has petered out” (11.13–16). As we have appreciated, Martial’s construction of his readers as potential distracted viatores both at the end of Books One and Eleven responds to strategies of poetic representation and to a mutated literary and political climate. 37 Book Eleven, published in December 96, is addressed to Nerva, acceded to the throne immediately after Domitian’s assassination in September 96. The lascivious and obscene tone of Book Eleven is interpreted as a (political and poetic) unrestrained freedom achieved after 15 years of Domitian’s censoria potestas. 38 Mart. 8.28. On Martial’s relationship to his friend Parthenius see also Mart. 4.45; 4.78; 8.28; 9.49. See Kay 1985, 52–4.

190

Alessandra Tafaro

Avid consumers and meta-poetic satiety In opposition to this (ironic) depreciation of his literary opus, Martial’s satirical stance towards his readership is counterbalanced by epigram 11.108, where a voracious reader invites Martial to offer up more epigrams. Quamvis tam longo possis satur esse libello, lector, adhuc a me disticha pauca petis. sed Lupus usuram puerique diaria poscunt. lector, solve. Taces dissimulasque? Vale. Mart. 11.108 Reader, although you might well be satiated with so long a little book, you ask me for a few couplets more. But Lupus demands his interest and the boys their rations. Pay up, reader. You say nothing and pretend not to hear? Goodbye.

Ironically disregardful of Martial’s post-Callimachean principle of the oligostichia, an avid reader, an anonymous lector wishes for more couplets. As the Lupercus of Book One, however, he pretends not to hear, as soon as Martial asks for money to pay his debts back to Lupus: lector, solve. Taces dissimulasque? Vale. With this ingenious epilogue, Martial draws our attention to the materiality of the epigram, which is debased to the level of a purchasable object, whose excessiveness in number runs the risk to turn the book into a μέγα κακόν. As Martial points out, what is the advantage of composing and consuming epigrams, whose main quality is brevitas, if they are collected into a book (Mart. 8.29)? The opening line quamvis tam longo possis esse satur libello, 11.108.1, illustrates Martial’s concern for an overly long book, which may result in a bored reader.39 The final section of Book Eleven, therefore, combines a tone of mock modesty with a clear teasing attitude towards the readership, in which Martial ironically assumes that the reader will have had enough of his poetic work. Martial’s ironic attack towards the readerly indifference calls into play a further epigraphic model, in which the traditionally consolatory motif of the memento mori morphs into a malignant curse.40 Eus tu uiator, ueni hoc et queiesce pusilu Innuis et negitas? Tamen hoc redeudus tibi. CLE 120 You, traveller, come here and rest shortly. You nod and then say no? However, you have to come back here. Martial repeatedly reflects on the concept of the book of epigrams. If the main advantage of epigram is its shortness and compression, a book of epigrams may be counterintuitive. Martial problematises the epigrammatic brevitas in both 8.29 (Disticha qui scribit, puto, vult brevitate placere. / quid prodest brevitas, dic mihi, si liber est?, “He who writes distichs wishes, I suppose, to please by brevity. What use is brevity, tell me, if it’s a book?”) and 1.118. 40 Lattimore 1942, 256–8. 39

8.  Lassi viatores

191

The engraved poem asks the traveller to move closer to the tomb (ueni hoc, 1) and physically rest for a short time (queiesce pusilu, 1). The act of reading, therefore, seems to offer a physical pause to the reader in motion, precisely as Martial’s epigrams offer the reader entertainment at times of otium. At a closer analysis, what seems a conventional address to the wayfarer turns out to be veined with a sarcastic tone.41 The voice of the dead emanating from the tombstone interrogates the viator, anticipating and attacking the expectation of readerly hesitance to approach the monument: “you nod and then say no? However, you have to come back here.” The concluding sententia resounds with black humour, for the dead, by evoking the ineluctability of human fate, threatens the viator that he will inevitably (and figuratively) return to that tombstone. The traditional consolatory theme that death is more tolerable because it is a condition common to all mankind, is subverted into an attack against a wayfarer unwilling to perform the expected acts of sympathy. Epigraphic poetry hosts manifold curses directed against the disrespectful or neglectful passer-by. Occasional intruders are warned to stay away from the private space of the tomb with highly formulaic expressions. Formulae such as heredes non sequentur or h(uic) m(onumento) m(alus) a(besto) not only demarcated the religious character of the grave, but also prevented the passer-by from dishonouring it.42 Numerous epitaphs request the occasional reader not to physically outrage the tomb (ne velis violare) and curse the potential trespassers with a set of colourful invectives: anger divinities are invoked to punish the interlopers, physical uneasiness and painful deaths may be threatened to those who do not respect the grave.43 The concerns related to the protection of the tomb from both disrespectful behaviours and neglect cleverly appear as early as the Republican period. The verse inscription CLE 120 wittily expresses the fear of remaining ignored by undermining the rhetoric of the memento mori, which is transformed into a powerful admonishment that death is indeed inevitable. As the passive periphrastic construction highlights, returning to that sepulchral monument represents indeed an ineluctable destiny. The ending punchline seems to be anticipated by the request queiesce, which encloses a double-edged meaning, if read in pair with the concluding threatening remark tamen redeudus tibi. The verb quiesco, which occurs repeatedly in funerary contexts to indicate the peaceful rest of the deceased, seems here to conceal not only the idea that reading the lines engraved in the monument will offer a physical pause to the traveller’s journey, but also a disquieting allusion to an eternal (mortal) sleep, as if the deceased speaking through the tomb were attracting the viator in its deathly atmosphere. This provocative epigraphic attitude towards a hesitant and potentially indifferent reader appears in epigrams 5.16 and 11.108. On both occasions, Martial puts forward the persona of the indigent and impoverished client, whose career as a practitioner Morelli 2017. Lattimore 1942, 106–24, 250–6. 43 Del Hoyo 2014, 809–24. 41 42

192

Alessandra Tafaro

of a “lesser genre” does not afford him a remunerative life. Martial’s expectations for a financial reward for his poetry are constantly disappointed by selfish readers: sed non et veteres contenti laude fuerunt, / cum minimum vati munus Alexis erat. / ‘belle’ inquis ‘dixti: iuvat et laudabimus usque’. / dissimulas? facies me, puto, causidicum, “But the men of old were not content with praise, in the days when an Alexis was a trifling present for a poet. ‘Prettily said,’ you say. ‘We like it and will continue to praise.’ You pretend not to understand? Methinks you’ll make a barrister of me yet” (5.16.11–14). His lector amicus, as Martial tells us, has spurred him to pursue the poorly remunerative poetic career in opposition to the more fruitful legal path. Sarcastically belittling the value of his work, Martial admits that his page is appreciated only when it comes free of charge, et tantum gratis nostra pagina placet (5.16.10). Martial’s rhetorical interrogation dissimulas? of poem 5.16 is variated in epigram 11.108.4, lector solve. taces dissimulasque? Vale (11.108.4). The indifferent and elusive approach of the epitaphic reader provides Martial with a set of possibilities which he turns into witty attacks against his anonymous readerships in contexts of defence of his own poetry. Within the epitaphic rhetorical and mordant question innuitas et negitas? of CLE 120 lies the key to understand Martial’s paradoxical representation of his disengaged readership. If the inscription openly threatens a lazy reader-traveller that he will sooner or later deal with a gravestone, Martial reacts to the readerly indifference in a typical epigraphic gesture: vale, “farewell” (11.108.4). One last final connection can be detected between the end of Book One and Eleven, which are both reflexive on their own reception. The voracious reader of 11.108 winks at the Caedicianus of 1.118 who, dissatisfied with just a hundred epigrams, is imagined asking for more: cui legisse satis non est epigrammata centum, nil illi satis est, Caediciane, mali. Mart. 1.118 He for whom reading a hundred epigrams is not enough, will never have enough of a bad thing, Caedicianus. Quamvis tam longo possis satur esse libello, lector, adhuc a me disticha pauca petis. Mart. 11.108.1–2 Reader, although you might well be satiated with so long a little book, you ask me for a few couplets more.

The etymological connection between satis of 1.118 and satur of 11.108 makes the relationship between these two consumers even more evident.44 Besides, satis and 44

Canobbio 2007, 227.

8.  Lassi viatores

193

satur meta-poetically signal the end of the books. These readers are both satiated with Martial’s poetry. The ambiguity concealed in sapis in epigrams 1.117 and sapisti of 11.106, is reinvigorated in satis, satur, both suggesting the metaphor of poetry-asfood, of fleshy and material epigram, well-suited to convivial entertainment.

Lassus tam cito deficis viator? Martial’s otiose readers To some extent, the theme of readerly satiety with which the reader was greeted at the end of Book One is brought to light again at the beginning of Book Two. The prefatory epistle creates a fictional communicative situation where Martial imagines himself and his patron, the philosopher Decianus, conversing about the necessity of prefaces for a book of epigrams. The proemial section of this book is concerned with the theme of readerly boredom and how to prevent it by making brevitas an instrument of self-defence. After agreeing with Decianus that a long prefatory epistle would be unnecessary for a book of epigrams, which are able to speak for themselves, Martial follows his friend’s advice, so that “anyone who chances upon this book will have you to thank that they don’t come through to the first page worn out” (2 Praef. 14–16): Debebunt tibi si qui / in hunc librum inciderint quod ad primam paginam / non lassi pervenient. Mart. 2 Praef. 14–16

The language deployed here brings forth the idea of an occasional and accidental reader-viator, who happens (inciderint) to read Martial as an occasional traveller who accidentally confronts an inscription. Through the metaphoric use of the epigraphic language, Martial represents his poetry in an attitude of affected modesty, as if his worldwide success were still to be ascertained. On the grounds of recusatio, he is here suggesting in a prose preface that there will not be a prose preface. The depiction of the potentially bored reader as viator to explain the risk of being full of and bored by a long book of epigrams is even clearer in epigram 2.1, where Martial enumerates the advantages of writing and publishing a brevis libellus, suggesting the idea of a reader who occasionally happens to read his epigrams and who will be at least consoled by their brevitas: tertia res haec est, quod si cui forte legeris / sis licet usque malus, non odiosus erit, “the third thing is that if you happen to be read to somebody, you may be thoroughly bad, but you won’t be a bore” (2.1.7–8). The hemistich si cui forte legeris mirrors the epigraphic expression found both in CLE 1085.1 si quis forte legit titulum and CLE 1086.1 si quis forte leget titulum (“if anyone happens to read this epitaph”). The fil rouge of readerly boredom or satiety, present in 1.118, is pursued both in the preface of Book Two and in epigram 2.1, and is strategically resurrected in epigram 2.6. It responds to Martial’s initial concerns pertaining the new kind of circulation of his epigrams in the form of the published book. While in the preface Martial shows a tone of affected modesty (si cui forte legeris), and in 2.1 he claims that many readers will find the book overwhelmingly long (ei mihi, quam multis sic quoque longus eris!, “Ah

194

Alessandra Tafaro

me, how many will think you long even so!”, 12), in epigram 2.6 he harshly attacks Severus’ weariness: i nunc, edere me iube libellos frames the epigram, becoming a refrain directed against Severus, who, despite urging Martial to publish his epigrams, fails now to appreciate them.45 The previously hinted metaphor which associated a bored reader with a viator who has no inherent interest in reading inscriptions, patently emerges in epigram 2.6.14–16, where the relationship between the reader of epigrams and the epitaphic passer-by bears Martial’s sphragis: I nunc, edere me iube libellos. lectis vix tibi paginis duabus spectas eschatocollion, Severe, et longas trahis oscitationes. haec sunt, quae relegente me solebas rapta exscribere, sed Vitellianis, haec sunt, singula quae sinus ferebas per convivia cuncta, per theatra; haec sunt aut meliora si qua nescis. quid prodest mihi tam macer libellus, nullo crassior ut sit umbilico, si totus tibi triduo legatur? numquam deliciae supiniores. lassus tam cito deficis viator, et cum currere debeas Bovillas, interiungere quaeris ad Camenas? I nunc, edere me iube libellos. Mart. 2.6 Go ahead, tell me to publish my little books! You have hardly read a couple of pages, Severus, and you are looking at the final sheet and fetching lengthy yawns. These are the pieces you used to grab as I read them and to copy, on Vitellian tablets at that. These are what you used to carry in your pocket one at a time to every dinner party, every theatre. These are they, or perhaps some better that you don’t know of. What good is it to me, a little book so slender that it is no thicker than any roller stick, if it would take you three days to get through all of it? Never was an aesthete so languid. Does the traveller flag so quickly? When you ought to drive to Bovillae, do you want to change horses at the Camenae? Go ahead, tell me to publish my little books!

The enthusiastic Severus, who used to enjoy Martial’s epigrams, carry them around at the theatre and to banquets in his pockets (singula quae sinu ferebas) and hastily copy them (rapta), is now an otiose reader, yawning after having read just two pages of the newly published book (edere). He embodies the “languid” lover, (deliciae supiniores) – deliciae is frequently used in Roman love poetry, as in Catullus’ deliciae meae puellae (Catull. 2) – and a tired viator in his reading journey, who immediately 45

Williams 2004, ad loc.

8.  Lassi viatores

195

looks for the eschatocollion (3). Just as the traveller stops at Camena in his extremely brief journey to Bovillae, so Severus loses interest in Martial’s brief collection after merely two pages, the shortness of the journey being cleverly compared to the brevity of the book. Although Severus used to admire Martial’s individual epigrams when they were recited in theatres (per convivia cuncta, per theatra, 8), he now fails to enjoy reading the same poems collected in the book format. Although perfectly suited to being consumed and listened to in sympotic contexts and social gatherings by a diner as the ultimate in light entertainment, Martial’s epigrams are rejected as a work of literature by Severus, who is depicted as an otiose lover unwilling to read and having no time to pay attention to the bookish poems, when more entertaining divertissements are provided by the symposia. Eager Severus, who used to “seize” Martial’s epigrams (rapta, 6), who carried the epigrams in his lap (sinus, 7), has now ceased his love affair with the poems. By deploying the erotic language evident in rapta, referring to Severus’ eagerness for Martial’s epigrams at line 6 and opposed to deliciae supiniores at line 15, Martial, in a clever intertext with Ovid’s erotic poetry, alludes to the physical and erotic consumption of his epigrams, where the lazy loverreader, looking for entertaining poetry during periods of otium, can embody his worst audience if taken to the extreme.46 In comparing his Severus to a lassus viator, Martial appropriates and manipulates the anxieties voiced by funerary inscriptions, the concern about facing a traveller not inclined to read. As three sepulchral epigrams testify, the nexus lassus viator was certainly part of the epigraphic repertoire. Quamuis la[ss]e uiator, rogo ne graueris et tumulum contempla meum, lege et moraris, iam aliquid resciueris. CLE 77, ll. 1-2 Although tired, traveller, I beg you, do not regard it as a burden and contemplate my tomb, read and linger, you will already have learnt something. Heus tu, uiator lasse, qu[i] me praetereis, cum diu ambulareis, tamen hoc ueniundum est tibi. CLE 119 You, weary traveller, who pass me by, although you will walk for a long time, however, you will have to come back here. tu qui praeteriens, [legis]ti, lasse uiator, sit tibi lux dulcis et mihi terra leuis. CLE 1125, ll. 10-11 46

Rimell 2008, 134–6.

196

Alessandra Tafaro Tired traveller, you who passing by read (this), may life be delightful for you and earth light on me.

These three verse inscriptions address a lassus viator, who is asked to stop his journey and read. The act of reading creates an occasion for the physical rest of the passer-by, who has an excuse to interrupt his own journey.47 Both the epitaphs CLE 77 and CLE 1125 combine the pathetic appeal to read and rest (moraris) with the good wish for the wayfarer (uale et bene facito uitae, dum fatum uenit, CLE 77.6; sit tibi lux dulcis, CLE 1125.2). Like CLE 120, the Republican CLE 119 paradoxically engages with its readership and threatens the neglectful audience with a malignant and far from consolatory memento mori, tamen hoc veniundum est tibi (2).48 Martial heightens the paradoxes already concealed by the epitaphic dialogue with the anonymous wanderer at the outset of Book Two, deploying the figure of a tired traveller and otiose lover to depict his potentially hostile readership, whilst bringing forth the concerns for the publication of his epigrammatic libelli. The Severus of Martial’s 2.6, precisely as a viator, has come to a standstill in his reading “journey”, dismounting the horses before reaching Camenae, just outside the Porta Capena. Nonetheless, by exploiting the ambivalent meaning of lassus, Martial complicates and enriches the scenario offered by inscriptions, where the act of reading offers the traveller an excuse to pause and physically rest. The epigram becomes now the cause of the reader’s tiredness. The otiose sympotic context for which Martial imagines his reception in both epigrams 2.1 and 2.6 may result, conversely, in the risk of overly lazy readers. The idea of stopping before inscriptions (lege et moraris, CLE 77) is paradoxically adopted by Martial, who wants his travelling reader to pause before his book and read. At the same time, he encourages him in moving through (and with) the epigrams in a metaphorical poetic journey. But, like the anonymous lector at the end of Book Four, Severus is tired, satiated after a few epigrams: Iam lector queriturque deficitque, iam librarius hoc et ispe dicit ‘ohe, iam satis est, ohe, libelle’. Mart. 4.89.7-9 Already the reader grows querulous and weary, already the very copyist says “whoa, there’s enough, whoa, now, little book!”

Lattimore 1942, 233. Lattimore 1942, esp. 257; Morelli 2017, 120 on moriundust. For hoc veniundum est tibi see CLE 83; CLE 242; CLE 1097. 47 48

8.  Lassi viatores

197

Conclusion As demonstrated in the present cases, Martial imports the construct of his reader as a lassus viator from inscriptions and appropriates an epigraphic concern about potentially bored readers, tired by the excessive length of books and meta-poetically satiated. As in inscriptions, brevitas becomes the ultimate defence against lazy, reluctant readers. Moreover, by re-reading Martial’s scenes of poetic reception in an epigraphic frame of reference, we can truly appreciate the already complex and witty nature of the inscriptions’ engagement with the reading public: the reader-traveller is not only requested to read in acts of sympathy, but also sarcastically cursed and attacked in ways that anticipate the epigrammatic mordacity. However, the contexts in which Martial deploys the epigraphic construct of the lassus viator present a stratification of meanings which enriches the epitaphic rhetoric, complicates the scenario offered by inscriptions, and manipulates the epigraphic repertoire by making it distinctively epigrammatic. As demonstrated, both sapis and sapisti at the end of Book One and Eleven play on the double meaning of the verb sapere as to know and to have taste, good sense, knowledge; satur and satis of epigrams 1.118 and 11.108 both etymologically hint at the idea of a reader who has had too much. More explicitly, epigrams 2.1 and 2.6 suggest that his poetry can be consumed at convivia, as ephemeral, impromptu poetry-food. With a tone of mock modesty, Martial engages with the epigraphic tradition and brings to light a bored reader who overlaps with the figure of the hasty viator, the privileged interlocutor of inscriptions. In the end, Martial’s Epigrams continuously play with the tension between the ephemerality of poetry as material, fleshy and the transcendent ambition for poetic monumentality, where his poems, precisely as outlasting marble inscriptions, would want to escape the menace of eternal oblivion and the encounter with the lazy passer-by. Are we therefore to believe Martial’s request not to pass-by his epigrammatic monuments, as in poem 11.13 noli nobile praeterire marmor, “do not pass this noble marble by”, or the ephemeral, material nature of his poetry as claimed in Xenia 13.3.8: praetereas, si quid non facit ad stomachum?, “if anything is not to your taste, just pass it by”? By appropriating epitaphic addresses to the lassus viator, Martial ironically attacks his bored languid readers in a self-deprecatory attitude, especially in epigram 2.6, where Severus embodies the passive reader par excellence. While inscriptions offer a physical rest for the exhausted viator, Martial’s epigrams subvert and distort this possibility, becoming themselves the cause of the reader’s tiredness. Martial’s reader/ passer-by is therefore the new wanderer in both his epigrammatic books and the new Flavian monumental Rome, the figure through which the poet tests his poetic (after)life and re-frames the “epigraphic” in his epigrammatic world.

Bibliography

Bing, Peter. 2009. The Scroll and the Marble. Studies in Reading and Reception in Hellenistic Poetry. Ann Arbor: University of Michigan Press.

198

Alessandra Tafaro

Boyle, Antony J. and William J. Dominik, eds. 2003. Flavian Rome. Culture, Image, Text. Leiden, Boston: Brill. Canobbio, Alberto. 2007. “Dialogando col lettore. Modalità comunicative nei finali dei libri di Marziale.” In Dialogando con il passato. Permanenze e innovazioni nella cultura latina di età flavia, edited by Alessia Bonadeo and Elisa Romano, 207–31. Grassina: Le Monnier Università. Citroni, Mario. 1975. M. Valerii Martialis Epigrammaton Liber Primus. Introduzione, testo, apparato critico e commento. Firenze: La Nuova Italia. Citroni, Mario. 1986. “Le Raccomandazioni del poeta: apostrofe al libro e contatto col destinatario.” Maia 38: 111–46. Citroni, Mario. 1988. “Pubblicazione e dediche dei libri in Marziale.” Maia 40: 3–39. Citroni, Mario. 1995. Poesia e lettori in Roma antica. Forme della comunicazione letteraria. Roma: Laterza. Colafrancesco, Paolo and Francesco Massaro. 1986. Concordanze dei Carmina Latina Epigraphica. Bari: Edipuglia. Courtney, Edward. 1995. Musa Lapidaria: a Selection of Latin Verse Inscription. Atlanta, GA: Scholar Press. Del Hoyo Calleja, Javier. 2014. “Ne velis violare. Imprecaciones contra los profanadores de tumbas.” In El mundo de los difuntos: culto, cofradías y tradiciones, edited by Francisco J. Campos and Fernández de Sevilla, 809–24. San Lorenzo de El Escorial: Instituto Escurialense de Investigaciones Históricas y Artísticas. Fearnley, Hannah. 2003. “Reading the Imperial Revolution. Martial, Epigrams 10.” In Flavian Rome. Culture, Image, Text, edited by Antony J. Boyle and William J. Dominik, 613–35. Leiden, Boston: Brill. Fitzgerald, William. 2007. Martial and the World of Epigram. Chicago, London: Chicago University Press. Fowler, Don P. 1995. “Martial and the Book.” Ramus 24: 31–58. Gowers, Emily. 1993. The Loaded Table. Representations of Food in Roman Literature. Oxford, New York: Clarendon Press. Gregori, Gian Luca. 2008. “Sulle origini della comunicazione epigrafica defunto-viandante: qualche riflessione sulla documentazione urbana d’età repubblicana.” In La comunicazione nella storia antica: atti del III incontro internazionale di storia antica (Genova 23-24 novembre 2006), edited by Maria G. Bertinelli Angeli and Angela Donati, 83–115. Rome: G. Bretschneider. Henriksén, Crister. 2006. “Martial’s modes of mourning. Sepulchral epitaphs in the Epigrams”. In Flavian Poetry, edited by Ruurd R. Nauta, Harm-Jam Van Dam and Johannes Jacobus Louis Smolenaars, 349–367. Leiden, Boston: Brill. Hinds, Stephen. 2007. “Martial’s Ovid / Ovid’s Martial.” Journal of Roman Studies 97: 113–54. Höschele, Regina. 2007. “The Traveling Reader: Journeys Through the Ancient Epigram Books.” Transactions of The American Philological Association 137: 333–69. Kay, Nigel M. 1985. Martial Book XI: A Commentary. London: Duckworth. König, Alice. 2018. “Reading Frontinus in Martial’s Epigrams.” In Roman Literature under Nerva, Trajan, Hadrian: Literary Interactions, AD 96–138, edited by Alice König and Christopher Whitton, 233–59. Cambridge: Cambridge University Press. König, Alice and Christopher Whitton, eds. 2018. Roman Literature under Nerva, Trajan, Hadrian: Literary Interactions, AD 96–138. Cambridge: Cambridge University Press. Kruschwitz, Peter. 2013. “Memento mori: the use(s) of the future imperative in the carmina latina epigraphica.” In Ex Officina. Literatura epigráfica en verso, edited by Concepción Fernàndez Martinez, María Limón Belén, Joan Gómez Pallarés, and Javier del Hoyo Calleja, 193–216. Sevilla: Secretariado de Publicaciones Universidad de Sevilla. Lattimore, Richmond. 1942. Themes in Greek and Latin Epitaphs. Urbana: Illinois University Press. Massaro, Matteo. 1992. Epigrafia metrica di età repubblicana. Bari: Istituto di Latino dell’Università di Bari. Merli, Elena. 1993. “Ordinamento degli epigrammi e strategie cortigiane negli esordi dei libri I-XII di Marziale.” Maia 45: 229–56.

8.  Lassi viatores

199

Morelli, Alfredo M. 2017. “Di alcuni carmi epigrafici in senari giambici del Latium adiectum.” In Le Epigrafi della Valle di Comino. Atti del quattordicesimo convegno epigrafico cominese, Atina, Palazzo Ducale, 27–28 Maggio 2019, edited by Heikki Solin, 111–47. Arezzo: F&C Edizioni. Nauta, R. Ruurd. 2002. Poetry for Patrons. Literary Communication in the Age of Domitian. Leiden: Brill. Rimell, Victoria. 2008. Martial’s Rome. Empire and the Ideology of Epigram. Cambridge: Cambridge University Press. Rimell, Victoria. 2018. “I Will Survive (You): Martial and Tacitus on Regime Change.”. In Roman Literature under Nerva, Trajan, Hadrian: Literary Interactions, AD 96–138, edited by Alice König and Christopher Whitton, 63–85. Cambridge: Cambridge University Press. Roman, Luke. 2001. “The representation of literary materiality in Martial’s Epigrams.” Journal of Roman Studies 91: 113–45. Shackleton Bailey, David R. 1993. Martial. Epigrams (III vols). Cambridge (MA), London: Harvard University Press. Tueller, Michael A. 2008. Look Who’s Talking: Innovations in Voice and Identity in Hellenistic Epigram. Leuven: Peeters. White, Peter. 1974. “The presentation and dedication of the Silvae and the Epigrams.” Journal of Roman Studies 64: 40–61. White, Peter. 2009. “Bookshops in the literary culture of Rome.” In Ancient Literacies: the Culture of Reading in Greece and Rome, edited by William A. Johnson and Holt N. Parker, 268–87. New York: Oxford University Press. Williams, Craig. A. 2002. “Ovid, Martial and poetic immortality: traces of Amores 1.15 in the Epigrams.” Arethusa 35.3: 417–33. Williams, Craig. A. 2004. Martial Epigrams Book Two; edited with Introduction, Translation and Commentary. Oxford, New York: Oxford University Press.

Chapter 9 Epigraphy and critical fabulation: imagining narratives of Greco-Roman sexual slavery Deborah Kamen and Sarah Levin-Richardson

Inscriptions, by their very nature, “set in stone” a particular moment in time. Simultaneously, they are only one stage in a longer process, from the situation that caused the act of inscription in the first place, to the scenarios that might be called into being afterwards. Employing Saidiya Hartman’s concept of “critical fabulation” – used by her to tell stories of individuals on the trans-Atlantic slave route (see more below) and Black women in early twentieth-century New York and Philadelphia – we propose a model that imagines, sometimes in multiple and conflicting ways, the other moments on this timeline.1 This approach is especially important for accessing the lived experiences of individuals whose presence is fleeting in the epigraphic record, such as enslaved individuals and other marginalized groups.2 In her work on slavery, Hartman responds to the problems she identifies in the archive in general, and archives pertaining to the trans-Atlantic slave trade in particular: first, that the archive is full of gaps and fissures;3 second, that it is constituted by fabrications and fantasies as much as by facts;4 and third, that the individuals she is most interested in – those forced into slavery – are present in the archive only as the commodities of white slave traders or in their moment of death.5 She therefore proposes a new methodology she calls “critical fabulation”, a mode of storytelling rooted in archival material and approached through a critical lens. Hartman describes this as “writing at the limit of the unspeakable and the unknown”.6 Hartman 2007; 2008; 2019. For other scholarship that tries to access marginalized individuals through epigraphy, see e.g., for the Greek world: Zelnick-Abramovitz 2005; Kamen 2012; Eidinow 2012; for the Roman world: Joshel 1992; Levin-Richardson 2011; 2013; Milnor 2014, 191–223; Perry 2014; Williams 2014, 510–12. 3 Hartman 2007, 16; 2008, 10–11; 2016, 209. 4 Hartman 2008, 9. 5 Hartman 2007, 17; 2008, 10; 2016, 208. 6 Hartman 2008, 1. 1 2

202

Deborah Kamen and Sarah Levin-Richardson

It involves “listening for the unsaid, translating misconstrued words, and refashioning disfigured lives” while “respect[ing] the limits of what cannot be known”.7 As regards the experiences of enslaved individuals, she writes: “[H]ow might it be possible to generate a different set of descriptions from this archive? To imagine what could have been?”8 Hartman explores answers to these questions in a chapter of her 2007 book, Lose your Mother. The raw materials for her critical fabulation are the transcripts of a 1792 trial of a slave-ship captain accused of murdering two of the enslaved women that he was transporting (supplemented by a speech delivered before the House of Commons that used the case to argue for abolishing the slave trade). “By advancing a series of speculative arguments and exploiting the capabilities of the subjunctive (a grammatical mood that expresses doubts, wishes, and possibilities),”9 Hartman imagines the scenarios leading to the death of one of these young women from multiple perspectives:10 that of the slave-ship captain accused of murder; the thirdmate of the ship, who provided testimony in the case; the ship’s surgeon (who also provided testimony); the abolitionist who addressed the House of Commons; the English insurer of the ship and its enslaved cargo; and finally, the deceased young woman herself.11 Each re-telling reframes the same set of archival details to explore the physical and emotional points of view of those involved in the case. Always keeping in mind the flaws of the archive and acknowledging the inevitable failure of any attempt to fully represent the experiences of a historical individual,12 Hartman is nevertheless able to weave vivid stories, often brutally vivid ones, about the life and death of enslaved individuals on the Atlantic slave route. Historians of slavery have found Hartman’s approach useful in examining their own archives.13 For instance, the historians Marisa Fuentes and Brian Connolly coedited a special issue of History of the Present that takes Hartman as its starting point in exploring the value of speculative or poetic approaches to slave archives.14 The payoff for the contributors to the special issue is not only a richer understanding of their area of study (though it does provide that), but, as Stephanie Smallwood explains it, a fundamental reframing of “the subaltern not as the archive’s object but rather as its historical subject”.15 Connolly and Fuentes also note that this approach calls into question traditional modes of history-writing, which purport to be “transparent” Hartman 2008, 2, 4. Hartman 2008, 7. 9 Hartman 2008, 11. 10 She returns to the story of the second young woman, called Venus, in Hartman 2008. 11 Hartman 2007, 138–52. 12 Hartman 2007, 137; 2008, 11–12, 14. 13 In addition to using critical fabulation in this chapter, we also use it in Kamen and Levin-Richardson forthcoming. 14 Connolly and Fuentes 2016b. 15 Smallwood 2016, 126. 7 8

9.  Epigraphy and critical fabulation

203

or “empirical”, but which in fact replicate the perspective of enslavers and other dominant groups.16 Ancient historians and epigraphers confront even more pervasive lacunae in the archive, from fragmentary inscriptions to lost works of literature to the (nearly) missing voices of those on the margins of society. Hartman’s approach seems fruitful for probing these gaps, allowing us to flesh out the past in new, critical ways.17 If this seems a radical step, we might consider that scholars routinely fill in gaps in the historical and epigraphic record, especially when reconstructing the lives of famous individuals or the chronology of famous events from Greco-Roman history. We extend the same courtesy to less famous individuals and events, taking seriously the lives of those on the margins rather than relying on lacunae as an excuse not to explore the past. If the narratives that this methodology conjures from the past are multiple, conflicting, and unverifiable, they also reveal that all narratives from antiquity are, to some degree, fictions and fantasies created by agendas past and present. As scholars of slavery and sexuality, we tested out this methodology on material with which we are most familiar in order to assess its interpretive payoffs.18 Not only has this exercise enabled us to envision the emotions and embodied experiences latent in our epigraphic corpora, it has also raised new questions and made us rethink some of the material we thought we knew so well. For the purposes of this essay, we limit ourselves to two case studies: the manumission inscriptions of Hellenistic Delphi and the graffiti of the purpose-built brothel of Roman Pompeii.

Case study I: The manumission inscriptions of Delphi Our first case study draws from what is probably the largest epigraphic archive pertaining to Greek slavery: namely, the manumission inscriptions from Delphi, which date from around 200 BC to AD 100, and which detail the manumission of around 1400 enslaved persons.19 For the most part, these inscriptions record the sale of enslaved persons by their owners to Apollo “on the condition of freedom”.20 Because these sales were designed to secure the enslaved person’s freedom from his or her owner, many scholars have called them “fictive sales” – that is, manumissions that were simply

Connolly and Fuentes 2016a, 111. For other critiques of history-writing, see especially Trouillot 2015 [1995], who stresses the various moments at which silences enter the production of history. 17 For other creative approaches to the lacunae of antiquity, see e.g., Tringham 1991 (brought to our attention by Sarah Morris) and Ebbott 2017 (brought to our attention by Olga Levaniouk). 18 Kamen 2011; 2014a; 2014b; Levin-Richardson 2011; 2013; 2019; 2021; Marshall and Kamen 2021. See also Downs 2018 on the importance of historians using imaginative approaches in their study of slavery and sexuality. 19 Mulliez 2016, §3. 20 On this form of manumission, see Hopkins 1978, 133–71; Zelnick-Abramovitz 2005, 81–91, 96–9, and passim; Kamen 2014a, 285–9; Sosin 2015; Mulliez 2016; Zanovello 2017. It was most likely the slaves themselves who supplied the money, although this is never stated outright; cf. Zanovello 2017, 41–82. 16

204

Deborah Kamen and Sarah Levin-Richardson

framed as sales.21 Others have preferred to view them as real sales, where the god did in fact become legal owner of the enslaved person, but the latter had usufruct of his or her freedom.22 But whichever way we classify these sales – as fictive or as real – the outcome is the same: for all intents and purposes, the enslaved person became free. These inscriptions – which were publicly displayed at Delphi, primarily on the polygonal blocks making up the terrace wall of the temple of Apollo, but also elsewhere at the site – provide quite a lot of detail: they list the name of manumittor, the name and sometimes the origin of the enslaved person, the sale price, provisions for the safeguarding of the freed person’s new status, and, where applicable, any conditions attached to this freedom (referred to in Greek with the verb paramenō, “to remain beside”, or the related noun paramonē, “remaining beside”).23 Despite this level of detail, however, the Delphic inscriptions never spell out the situation underlying the manumission – that is, what kind of relationship existed between the enslaver and the person he had enslaved, and what prompted the former to set the latter free. We would like to suggest, however, that it is sometimes possible to read between the lines of the inscription’s official narrative to tease out these details. Indeed, that is precisely what we would like to try to explore through critical fabulation. As we will see shortly, one of the details that can be clarified by this approach is the ramifications of the so-called paramonē clause. While most scholars hold that those under paramonē were free, albeit with further obligations to their former enslaver,24 it has recently been argued that they remained enslaved until they finished their paramonē service, at which point (and only at which point) they became free.25 But the premise of this argument – namely that individuals could be only enslaved or free, not somewhere in between – is to our minds faulty. After all, the concept of “conditional freedom” exists in many slave societies (including ancient Rome and colonial Brazil), with individuals occupying a limbo status of semi-freedom until they have completed their remaining obligations.26 There is no reason to think the same concept did not exist at Delphi, and one of the things we explore is what it would have meant for someone to occupy this liminal status. E.g., Kamen 2014a. Mulliez 2006, §1; 2016, §3. Cf. Sosin 2015, who also argues that these sales are real, but who contends that the terms of the sale are that the god must set his new possession free at the point denoted in the agreement. 23 In the second century BC at Delphi, close to three quarters (72%) of manumitted slaves were granted full freedom, that is, freedom without paramonē, but by the end of the first century BC, that percentage decreased to 48% (Hopkins 1978, 141–2). 24 E.g., Zelnick-Abramovitz 2005, who calls them freed slaves who are “not wholly free”; for a survey of views on this topic, see Zelnick-Abramovitz 2018. 25 Sosin 2015; but see the important critique of Zelnick-Abramovitz 2018. 26 For ancient Rome, see e.g., Digest 40.7 on statuliberi (enslaved people manumitted conditionally); see also Huemoeller 2020 on the obligations Roman freedwomen-wives owed to their patron-husbands, rendering the former, in effect, somewhere between slave and free. On conditional versus unconditional freedom in colonial Brazil, see Higgins 1999, 154–6. 21 22

9.  Epigraphy and critical fabulation

205

For the purposes of this case study, we focus on the sexual relationship between an enslaver and the woman he had enslaved,27 as illuminated by a pair of Delphic manumission inscriptions (FD III.3.329 and III.3.333).28 The first inscription (FD III.3.329) is undated in Fouilles de Delphes, but Dominique Mulliez has recently suggested a date in the penultimate decade BC.29 It should be noted that this inscription is somewhat unusual for Delphi, in that it records not a straightforward sale to the god but the dedication of the owner’s right of ownership to the god.30 Nonetheless, although the process is slightly different, the substance seems to be essentially the same as what we see in normal manumission-sales. The inscription records the manumission of a woman named Eisias: When Theoxenos son of Philaitolos, by adoption son of Babylos son of Aiakidas, was archon, in the month of Elaios, when Aiakidas son of Eukleidas and Neikon son of Neikaios were councilors, Kleomantis son of Dinon dedicates the right of ownership (ōnan)31 of Eisias [sc. to Apollo], on the condition of freedom (ep’ eleutheriai), at a price of two mnas of silver. Guarantor in accordance with the laws of the city: Aiakidas son of Eukleidas. And let Eisias remain by (parameinatō) Kleomantis the whole time of his life doing everything ordered, like a slave (hōs doula). If Eisias does not remain, or does not do what is ordered, let Kleomantis have authority in punishing [her] in whatever way he wishes, by whipping her, and shackling her, and selling her. If someone seizes Eisias for re-enslavement, let the guarantor present [the sale] to the god as secure. And let another also be authorized to rescue her as free, being unpunished and not subject to any punishment or penalty, since he rescued her. If anything mortal happens concerning Kleomantis [i.e., when he dies], let Eisias be [completely] free, belonging to no one in any way […]. Eisias […] doing all the things pertaining to his funeral […] Witnesses are the priests Philon son of Stratagos, Polemarchos son of Damon and the private citizens Eukleidas son of Aiakidas, Babylos, […], Melission, Lykos, Da[…].32

Then, in another, presumably later inscription from Delphi, we find the following: When Diokles son of Philistion was archon, in the month of Eilaios, Kleomantis son of Dinon, being of sound mind, released his own female fosterling [slave] (threpta) Eisias from paramonē, and I [Kleomantis] hold the money33 that was written down in the paramonē agreement, On sex with slaves in Greece, see e.g., Davidson 1997, 98–108; Klees 1998, 161–75; Fischer 2010; Walin 2012; Todd 2013; and the chapters on Greek culture in Kamen and Marshall 2021. On slaves’ “erotic experiences”, see Cohen 2014. On sexual violence against female slaves, including but not limited to domestic slaves, see Paradiso 1999. 28 Discussion of these inscriptions can be found in Tucker 1982, 230–31; Mulliez 2016, §32–3; Hunt 2018, 106–7; Zelnick-Abramovitz 2018, 383–8. 29 Mulliez 2016, §32. 30 Cf. Mulliez 2016, §32, who refers to this as a manumission by consecration. 31 We owe this translation of ōnē to Dominique Mulliez (personal communication). 32 FD III.3.329. 33 A number of scholars have rendered the phrase apechō to…chrēma as something like “I remit the money”: that is, they assert that Kleomantis actually remitted the two mnas that Eisias had originally handed over for her sale to Apollo, and moreover that she owed nothing in addition for her release (Wiedemann 1981, 49; Tucker 1982, 231; Zelnick-Abramovitz 2005, 236 n. 115; Hunt 2018, 107). But the idea that Kleomantis gave back the purchase money requires, to our mind, a strained interpretation of the verb apechō as meaning “to remit”. The verb in this context is more likely to have its attested meaning 27

206

Deborah Kamen and Sarah Levin-Richardson and [I released from paramonē] the son Nikostratos born from her during paramonē, whom I renamed Kleomantis by adoption, in order that they be free in every honorable way and belonging to no one at all in any way. And whenever Kleomantis should suffer something mortal [i.e., dies], let everything left behind by him be for Sosula for her use. And if Sosula suffers anything, let everything be Eisias’ and Kleomantis’, and let them belong to no one else in any way. Let Eisias do all the things pertaining to his funeral, just as descendants also do. Witnesses are the priests of Apollo, Diodoros son of Philonikos, Dionysios son of Astoxenos, Damos son of Polemarchos, and the archons Laiadas son of Melision, Nikon son of Nikaios, and the private citizens Stratagos son of Philon, Xenagoras son of Habromachos, Lamenes son of Eukrates, Euangelos son of Megartas, Agon son of Poplios.34

Drawing from these two inscriptions, we will use critical fabulation to explore the moments in time that led to Eisias’ manumission (and therefore to these inscriptions), including her initial purchase, her relationships with her owner and her owner’s wife, and her status after being freed.35 Following Hartman’s methodology of fashioning a narrative from various perspectives, we begin from the point of view of the owner (Kleomantis), before turning to that of the owner’s wife (most likely Sosula) and finally that of the enslaved woman (Eisias) herself. After imagining these three perspectives, we return to a discussion of what was rooted in the archive and what we had to speculate about. We start with Kleomantis. Kleomantis One day, Kleomantis, in need of a new slave, purchased a young slave-girl, only two or three years old, whom he decided to call Eisias. He and his wife Sosula had no children of their own, and so, despite their best efforts, they found themselves growing attached to the girl. When Eisias got a little older, six or seven, Sosula began to train her to be a domestic slave, and soon Eisias set to work running errands for her masters, cooking them meals, and cleaning the house. The years went on, and Kleomantis in particular continued to feel very affectionately toward her. When she was a teenager, he wanted to show her how much he appreciated her and came up with an idea: he would offer her her freedom, and all he’d ask for in exchange was the relatively small amount he’d spent to purchase her (two mnas). It was actually a great bargain for her, he thought, especially given how much he’d spent to feed and clothe her over the years. Eisias happily accepted the offer of freedom, and they travelled to the neighbouring town of Delphi to sell her to the god. And because Kleomantis

“to receive (in full)” (LSJ s.v. apechō IV) (and indeed this is how it is translated in Zelnick-Abramovitz 2018, 385). In this case, it would mean that Kleomantis received the payment (chrēma) that had been stipulated for Eisias’ release from paramonē. 34 FD III.3.333. 35 Hunt 2018 also recognizes the potential of these two inscriptions to “illuminate possibilities and allow us to imagine several possible life stories” (107), briefly offering a few suggestions: Eisias may have been raped, she may have seduced Kleomantis, or their relationship may have been reciprocal; Eisias and Sosula were most likely rivals, but it is also possible they were on friendly terms.

9.  Epigraphy and critical fabulation

207

considered Eisias part of the family, he decided to include a paramonē clause in the agreement, stipulating that she continue living with them. Soon after they returned home from Delphi, Kleomantis began viewing Eisias in a new light: less as a daughter and more as a beautiful young woman, one who was considerably more attractive to him than his now-aging wife Sosula. He would visit Eisias during the night when Sosula was asleep, certain that his feelings toward the slave-girl were mutual. Before long, Eisias became pregnant and bore a son she named Nikostratos, who, like his mother, would also remain in the household. As time passed, Kleomantis began thinking about his waning years, and he realized that he needed to take measures to secure an heir for himself. An obvious solution suggested itself to him: he would formally adopt Nikostratos as his son, renaming him after himself, and give both the boy and his mother their complete freedom. His only condition was that Eisias hand over a little more money for their release. When Eisias agreed, Kleomantis had their arrangement inscribed on stone at Delphi, specifying that his wife would inherit his property, but that when the wife died, Eisias and his son were next in line – an arrangement Kleomantis felt was beneficial to all parties involved. Sosula Sosula was unable to bear children, a fact that caused her a lot of grief, and so it was bittersweet when her husband brought home a small child they would raise and keep as their slave. It was her duty as woman of the house to care for the little one, and as joyful as it was to have a young child’s laughter in the house, it was also deeply painful for her. For this reason, Sosula tried to distance herself from the girl, reminding herself that this was not her child but her slave, and as soon as the girl was able to perform basic tasks around the house, Sosula put her to work. Eisias was a diligent slave, hardworking and competent, and Sosula was pleased that she’d trained her well. But when Eisias started developing into a young woman, Sosula became suspicious of the way her husband looked at her; she knew that hungry look in his eyes. And so when Kleomantis announced to Sosula one day that he was going to set Eisias free, as a reward for the “diligent service” she’d provided them, Sosula couldn’t help but think Kleomantis had ulterior motives, that he was trying to make Eisias grateful so that she’d willingly do things for him in return. But Sosula couldn’t stop her husband, and when they returned from Delphi, Sosula began to see that her suspicions were right; as subtle as her husband thought he was being, she could hear the door creak when he sneaked around the house at night, and sometimes she would catch him pawing at Eisias after dinner. Once or twice Sosula confronted him, often framing it as a joke – “you’re just trying to get me out of the room so you can make a move on Eisias!”36 – but she also knew there was no way she’d be able to stop him. The day Sosula realized Eisias was pregnant was an incredibly difficult one for her: not only because it revealed what she already knew, that her husband had been 36

This is inspired by an episode related in Lys. 1.12.

208

Deborah Kamen and Sarah Levin-Richardson

sleeping with the girl, but also because Eisias had been able to provide Kleomantis with something, a child, that she never could. In her rage, Sosula refused to acknowledge that the baby was her husband’s and simply treated the child as a servant. As time went on, Sosula could not bear to have the daily reminder of her husband’s infidelity in their house, and she insisted that he release both Eisias and her son from paramonē. He agreed to do so, but only if he could adopt the boy and make both him and the boy’s mother heirs to his property, next in line after Sosula. This official recognition of her husband’s whore and bastard son infuriated Sosula, but since she knew it was the only way of getting rid of Eisias and the boy, she gave her consent. Eisias When Eisias was just a few years old, she was torn from her mother’s arms and carried to the house of strangers. She was of course too young to know what was going on, but in time she found herself growing attached to her foster parents. Not knowing any differently, she wasn’t surprised when she was put to work around the house at the age of six. As she got older, she began to notice that Kleomantis was looking at her differently than he had when she was a little girl. He would also touch her more than was necessary, always making excuses to come near her when she was bent over the stove or sweeping the floors. It didn’t feel like the behaviour of a father, and it made her uncomfortable. And so when he offered to give her her freedom, she couldn’t believe her good luck, and she quickly agreed. She then realized there was a catch, or rather multiple catches: first, she’d have to pay for her freedom (since she had no money, this would mean having to collect loans37), and second, although she’d technically be free, she’d have to continue to live with Kleomantis and work for him. Nonetheless, she thought, being legally free would give her some privileges she currently lacked – she’d be able to work outside Kleomantis’ house for money, she could get married and start a family – and she figured that incomplete freedom was better than no freedom at all. But when they got to Delphi, and a circle of witnesses gathered round, Kleomantis’ tone changed dramatically. He declared sternly that paramonē meant that Eisias had to do whatever he asked – “just like a slave”, he added.38 He then reminded her that if she didn’t do what he asked, he could do whatever he wanted to her, threatening the use of whips, shackles, even sale back into slavery.39 At that moment, Eisias realized that her position in Kleomantis’ household was now, ironically, more precarious than The collection of loans for manumission is well attested in the Greek world: see e.g., the enslaved prostitute Neaira, who collected money from her past and present clients ([Dem.] 59.30–1). 38 Cf. Sosin 2015, 337, 340 who translates this phrase as “as a slave”; but cf. Zanovello 2017, 81 who argues, convincingly to my mind, that this language should be read metaphorically. 39 The language of whipping, shackling, and sale present in our first manumission inscription is rare, showing up only a handful of times in the Delphic corpus (forms of psopheō appear three times, mastigeō six times, dideō eight times), and sale shows up only four times (see Mulliez 2016, §32 n. 65 on the infrequency of sale as a punishment). 37

9.  Epigraphy and critical fabulation

209

it had been before. It had of course never been a secret that she was enslaved, but her status had never been put in such stark terms before. She now felt that, despite the fact that she was technically free, she couldn’t say no to anything Kleomantis asked. And so when Kleomantis started touching her – first stealing a kiss here and a grope there, and then forcing himself on her at night while he thought (stupidly) that his wife wouldn’t know – Eisias felt she couldn’t resist. After all, if she did, he could make the experience even more painful, or, god forbid, he might sell her into slavery to an even worse master. So she endured his nightly “visits” and eventually bore him a son she named Nikostratos. At first, Kleomantis refused to acknowledge the boy was his, even accusing her of sleeping around, especially when they were in the presence of Sosula. Things in the house were becoming increasingly intolerable, and so Eisias began asking, when she was alone with Kleomantis, if she and her son might be granted release from paramonē. Knowing that she needed to use whatever she had at her disposal, she wasn’t averse to using her body in her attempts to persuade him. Eventually he agreed, provided that she hand over more money for their early release. He did promise them an inheritance, but they’d inherit only after Sosula died, and so Eisias felt pretty sure they’d never actually see the money. Even so, the deal gave her great relief: she was now, finally, fully free, and she and her son and could go wherever and do whatever they pleased.40 There are some details from our inscriptions we can accept more or less at face value: namely, that Kleomantis sold an enslaved woman named Eisias to Apollo and required her to remain with him under paramonē until he died; he later released both her and the son born to her during paramonē; and he included Eisias and her son in his will, to inherit only after a woman named Sosula died. We have had to speculate about many details in the narrative, with varying degrees of fabulation. For instance, we think it is a safe assumption that Sosula was Kleomantis’ wife,41 and that Kleomantis Jr. was Kleomantis Sr.’s biological child,42 even if those details are not explicitly spelled out. And even though we are not told how old Eisias was when Kleomantis bought her as

40 We can actually track what happened to her afterwards. Mulliez 2016, §32 discusses, alongside these two inscriptions, two more dealing with Eisias and Kleomantis Jr., which he dates to the 20s or 30s AD. In FD III.6.39, Eisias “of Kleomantis”, along with (her partner?) Aristion son of Eukleidas, with Kleomantis Jr. also giving consent, sells to Apollo a girl (korasion) named Sostrata. The document specifies that any children born to Sostrata during paramonē will be free, but will stay with Eisias and Aristion, who reserve the right to sell these children in case of need; Sostrata also has to give a two-year-old baby to Kleomantis if she wants her complete freedom. In FD III.6.40, Eisias and Aristion release Sostrata from her paramonē obligations. When Aristion dies, Kleomantis Jr. and Sostrata will take care of the funerary rites and will share what he leaves behind. See also Zelnick-Abramovitz 2018 on these four inscriptions. 41 See also Tucker 1982, 231; Zelnick-Abramovitz 2005, 169; Mulliez 2006, §25; 2016, §32; Hunt 2018, 107. Another possibility, however, is that she is Kleomantis’ sister (Lene Rubinstein, personal communication). 42 See also Tucker 1982, 230–1; Zelnick-Abramovitz 2005, 169; 2018, 385; Mulliez 2016, §30; Hunt 2018, 107.

210

Deborah Kamen and Sarah Levin-Richardson

enslaved chattel, she must have been young, since she is called a fosterling (threpta);43 nor are we told how old she was when she was freed, but she must have been of childbearing age. Other elements of our story required a bit more imagination, including teasing out, from various perspectives, the nature of the relationships between Eisias, Kleomantis, and Sosula. Adopting, if only briefly, the perspective of Kleomantis has forced us to think through how an average slaveowner (if not necessarily the real historical Kleomantis) might have viewed his rights vis-à-vis his chattel – and to reconstruct how he, like many slaveowners of his day, might have justified his (mis)treatment of them. Perhaps even more fruitful was envisioning the perspective of other family members – in this case, Kleomantis’ wife – in a situation like the one we have posited here, where her husband has sex with and impregnates one of the household’s enslaved women. We can grasp intellectually that men in the ancient world (for the most part) did not have qualms about using sexually those they enslaved, but we do not often give thought to the ramifications, emotional or otherwise, this might have had for other (free and unfree) members of the household – even though both Greek and Roman sources attest that this behavior made at least some women jealous or angry.44 Imagining what this might have looked like gives us a clearer sense for the complex emotional landscape that wives like Sosula – and enslaved individuals like Eisias – would have had to negotiate in the household.45 Indeed, most illuminating for us has been to try to reconstruct the experiences of Eisias herself. Imagining a young girl’s inevitable disorientation at entering a new household has compelled us to consider what an enslaved person’s “natal alienation” might have felt like.46 In addition, having to imagine how someone might feel to be offered her freedom, but only her partial freedom, with brutal punishments stipulated for any misbehaviour, clarified for us the reality of what it meant to be freed with On the various meanings of threptos, see Cameron 1939; on threptoi in the context of slavery and manumission, see Zelnick-Abramovitz 2005, 173–80 and passim. That “threptos indicated a special relationship created by foster-care of children”, see Zelnick-Abramovitz 2005, 174. 44 The anger of wives at husbands who impregnate enslaved women of the household is a not infrequent trope in Greek literature, especially Greek tragedy (e.g., Euripides’ Andromache); see Gaca 2021 on this topic, who argues that wives in such cases were not simply “jealous” but justifiably angry. See also Ripat 2014 on a similar phenomenon in the Roman world; she focuses on curse tablets, likely written by Roman wives, that target (enslaved or freed) rivals for their husbands’ attention. For instance, in two secondcentury BC curse tablets from Pompeii (CIL IV 9251), the female author curses a rival woman named Philematium, slave of Hostilius (who may be the author’s husband: Ripat 2014, 342). 45 Comparative evidence can be useful here. See e.g., Dusinberre’s description (2009, 31) of what happened after Harriet Jacobs (author of Incidents in the Life of Slave Girl) was bequeathed to a new mistress (Mrs. James Norcom): “Jacobs’s presence soon became a source of marital discord within the Norcom household. Dr. Norcom eyed the twelve-year-old girl with interest, and by the time she was fourteen, he was whispering suggestive things into her ear and devising various schemes to be alone with her. Never beating her himself (until later), and never letting anyone else punish her, he tried to persuade Jacobs that her life could be transformed for the better if she were kind to him; and he drove his wife into fits of unfounded suspicion that Jacobs had accepted his advances.” 46 On natal alienation, see Patterson 1982. 43

9.  Epigraphy and critical fabulation

211

paramonē – more specifically, the ways in which the state of being freed was sometimes closer to slavery than to freedom. Finally, being forced to envision how an enslaved woman like Eisias might have come to bear her enslaver’s child compelled us to confront in a new way the everyday horrors an enslaved person might have faced.

Case study II: The purpose-built brothel at Pompeii The second archive is constituted by the almost 150 texts and images scratched into the walls of Pompeii’s purpose-built brothel (VII.12.18–19; CIL IV 2173–296, 3101a, on which see especially Levin-Richardson 2019, 40–63, 153–62). Dating to the seven years before the eruption of Mount Vesuvius in AD 79,47 these graffiti vary in content. Most common are names, but there are also death notices, greetings, a snippet of the Aeneid, drawings of phalluses, ships, people, and birds, and, unsurprisingly, sexual boasts. It is a subset of the sexual boasts on which we focus here. Graffiti with the formula “[male name], you fuck well” constitute nine of the structure’s 134 textual graffiti.48 Above the masonry platform in room f, Sollemnes was praised twice – S[ol]lemnes | b[e]ne futues, “S[ol]lemnes, you fock w[e]ll” (CIL IV 2185; see also 2186) – and this formula was used in the same room with the names Felix, Vitalio, and December (CIL IV 2176, 2187, 2219). Syneros was praised with the same formula, but in Greek, above the platform of room e (Cυνέρωc | καλὸc · βινεῖc, CIL IV 2253).49 Victor was praised three times total – twice in room d, and a third time in room f – with Victor | valea qui bene | futues, “Victor, may you fare well who focks well!” (CIL IV 2260 Add. p. 216; see also 2274 Add. p. 216 and 2218). In employing secondperson grammatical forms, these graffiti present themselves as written by someone other than the named clients. Readers of these graffiti, both then and now, are made to think that prostitutes wrote these statements, since prostitutes would have had the knowledge to comment upon and praise the sexual capabilities of clients. While the voices of these statements are certainly those of prostitutes, who actually wrote them is impossible to know (as is true of all graffiti).50 Based on evidence from Pompeii and the wider Roman world, it has been argued that we can and should consider prostitutes as the possible authors of some of the brothel’s graffiti.51 In including texts possibly written by prostitutes – that is, members of the marginalized The impression of a coin pressed into the wet plaster of one of the brothel’s walls dates the last preeruption renovation of the brothel to AD 72 (Fiorelli 1862, 52); the graffiti scratched into these walls must date from AD 72 to the eruption, then. 48 This paragraph and the next draw heavily on Levin-Richardson 2019, 115–16. 49 For the grammar of this statement and the possibility that Syneros was a male prostitute, see LevinRichardson 2019, 137–8. 50 Graffiti can create personas and spin fictions like any literary genre, and we can thus never prove authorship beyond a doubt, even in self-conscious statements like …scribit Narciss[us], “Narcissus writes [this]” (CIL IV 1841). For all we know, someone else wrote that graffito in the voice of Narcissus. On authorship of graffiti, see also Williams 2014, 494, 498, 502–4. 51 Levin-Richardson 2013. 47

212

Deborah Kamen and Sarah Levin-Richardson

groups we seek to focus upon – the brothel’s archive differs from that of the previous case study, as well as from the texts that constitute Saidiya Hartman’s archive. In what follows, we imagine the scenarios leading to the writing of one of these statements – “Victor, may you fare well who focks well!” – from the perspective of the client, from that of the brothel’s madam, and from that of a prostitute. Doing so allows us to explore the physical and emotional experiences of the various individuals involved in brothel prostitution, and to fill in the other moments on the timeline of this graffito. In addition, we use this mode of storytelling as a first attempt at tracing how different levels of literacy and degrees of authorship among the brothel’s prostitutes impacted their own experiences and those of others in the structure. We begin with the client’s imagined perspective. Victor Victor had lost count of how many times he visited the brothel. He tried to go at least once a month, but sometimes it was several months between visits. He always looked forward to it, and knew most of the women and boys who worked there, though there were occasionally new faces added to the mix. He had his favourite girls, though – Felicla and Fortunata, especially. He went in after work finished in the late afternoon and chit-chatted with the madam in the front room like always. He didn’t see Fortunata – she might be the one moaning with a client behind the curtain in the next room – but Felicla came out from another room and took the madam’s place on the bed while the madam brought him a glass of wine. As he drank his wine, Felicla asked how business was going. Victor was pleased to share that he was able to double his monthly business after buying two more slaves. Felicla told him how important he must be in town, and once he drained the dregs of the wine, took his hand and led him to one of the smaller rooms for “the usual”. They found the madam back in the front room when they were done. She looked Victor over and asked, “how was it?”, to which Victor replied “Felicla was amazing…as always”. He looked expectantly at Felicla, hoping she would return the compliment. After a moment’s hesitation and a piercing look from the madam, Felicla regained her composure, turned to look Victor in the eyes, and purred, “And you were amazing, too, lover-boy”. “Can we write it on the wall, so everyone else knows?” he asked with excitement. “I noticed earlier that someone wrote that Vitalio fucks well, and everyone knows he only has one testicle!” All three laughed at that, and the madam took out the metal hairpin that helped secure her abundant curls and handed it to Victor so he could use it to write on the wall. “You two have fun – I’m going to go check on Fortunata.” Twenty minutes later, after joking about the wording and figuring out the perfect spot, Victor helped Felicla put the finishing touches on the statement: “Victor, may you fare well who fucks well!” He threw a wink to Felicla as he left, promising to bring her something special as a gift the next time.

9.  Epigraphy and critical fabulation

213

A few weeks later, Victor’s friend Iarinus came by the fullery to drop off some tunics. Right before leaving, Iarinus gave him a rascally smile and said, “I saw that you’ve been living up to your name at the brothel – apparently you really know how to use your sword!” Victor couldn’t stop grinning all day. The madam The madam noticed Victor come in and motioned for him to join her. He was a regular customer, and as such, helped keep the brothel afloat financially. As the manager of one of the local fulleries, Victor also regularly gave her a deal on her laundry. It’s good to cultivate clients like these, she thought. One of her best girls, Felicla, popped her head in and started chatting with Victor, allowing the madam to pour some wine. Reaching past Felicla, she handed the glass to Victor; she wouldn’t waste her wine on any of the girls or boys anyway, but she was growing certain that some of them were stealing sips of it when she wasn’t around. The fact that she had done the same in their position, about 15 years ago now, convinced her that her explanation for the slowly dwindling wine supply was correct. Returning her thoughts to the present, she cast a sharp eye at Felicla, hoping to communicate that Victor was to be given special attention. While she waited for them to finish, the madam stood outside the brothel, catching up with the neighbours on their way to running errands and teasing strangers and regulars alike with the pleasures that awaited them inside. She nipped back inside just before Felicla and Victor emerged from their room, both sweaty. “How was it?” she asked Victor, wiggling her eyebrows at Felicla to remind her of what to do next; she hoped the extra time and effort the girls spent with the clients would lead to more repeat customers. Before long, she could hear Victor laughing as Felicla wrote about Victor’s supposed manliness on the wall. He promised to bring something extra next time for both herself and Felicla, but it’d be too late to help her with the semi-annual rent due next week. Felicla Felicla had already been with five clients that day and her body ached. When Victor showed up, she sighed, smoothed out her dress, put on her practiced smile, and went in to greet him. He wasn’t so bad, as clients went. He sometimes held her too tightly, leaving bruises on her arms, but he also knew how to slip her some extra money without the madam finding out. She remembered that one of his kids had been sick, so she steered clear of asking after his family, and turned instead to his work, which she knew he took great pride in. It was the right question – he had soon emptied his glass of wine and was talking more and more excitedly about the changes he had made at the fullery. She nodded at the appropriate moments and encouraged him with the requisite “oohs” and “ahhs” and “tell me mores”, then reached for his hand and led him back to one of the other rooms. While he thrusted away, she told him how great he was, and thought about what she and the other girls would do for dinner that night. After he finished, he slipped

214

Deborah Kamen and Sarah Levin-Richardson

her an extra coin. She took it with her to the bathroom, and as she was cleaning up, hid it in a secret compartment under the toilet seat. She hadn’t saved much, but it was enough to treat herself to a new bottle of perfume, something she was able to do a couple times a year, ever since she was sold to the madam seven years ago, when she was eight years old (or so – she didn’t really know). She then joined back up with him and they went together to the madam. Victor asked Felicla if she’d write about how good he was in bed, and when the madam gave her a pointed glance, she assented. One of the reasons she hesitated, though, was that she couldn’t read or write. Once the madam left, she found an empty spot on the wall and asked Victor what he thought the statement should say. He read some of the other praise on the wall, and wanting to be better than all the others, said, “Have it say, ‘Victor, may you fare well who fucks well!’” They both got a good laugh out of that, and then touching his hand and playing the shy girl, Felicla flattered him: “I bet your handwriting is much better than mine, since you run a successful business. I want to make sure it looks as good as possible – what if you write it while I watch?” As he happily started to scratch the words into the wall with pride, she let out a small sigh of relief behind his back. There are as many facts as possible in these critical fabulations. The graffito about Victor is real, and Felicla was indeed a prostitute in the brothel, with two graffiti about her (CIL IV 2199: Felicla ego f, “I f-ed [=fucked] Felicla [= Felicula]”; 2201 Add. p. 215: Feliclam ego hic futue, “I focked Felicla [= Felicula] here”). Fortunata is another prostitute attested in at least three of the brothel’s graffiti (CIL IV 2224: Felix cum | Fortunata, “Felix with Fortunata”; 2259: Fortunata XI XV XI, “Fortunata 11 15 11”, 2275: Fortunata fellat, “Fortunata sucks”).52 Two small glass cups and a glass bottle were among the archaeological finds, and these were probably for alcohol.53 It has been argued recently that wine was served to clients,54 and that prostitutes may have stolen some for themselves as a form of self-care.55 The other males mentioned in the critical fabulation from Victor’s perspective – Vitalio and Iarinus – are attested male names from the brothel (CIL IV 2187: Vitalio | bene · futues ·, “Vitalio, you fock well”; 2181 Add. p. 215: Iarinus; 2220: Iarin[us]; 2251: Iarnius; and possibly 2205: Iar[inus?]). There is a toilet at the back of the structure, but its wooden seat is not preserved, so the hiding spot Felicla uses is made up. We have had to create backstories for both Victor (i.e., his occupation and family) and Felicla (her age, status, and history of sale). The graffiti in the brothel provide no other details about either individual.

52 If Fortuna is taken to be a variation of the name Fortunata (as Guzzo and Scarano Ussani 2009, 122 n. 606 have it), we can add CIL IV 2266: Ver[---] | felas | Fortuna sic, “Ver… you suk. Fortuna likewise”. 53 Levin-Richardson 2019, 36. 54 McGinn 2013, 624–9, Levin-Richardson 2019, 36, 100–1. 55 Levin-Richardson 2019, 128.

9.  Epigraphy and critical fabulation

215

Whether there was a madam or pimp is not known – there is no direct proof of one – but it seems very likely.56 Since room f is the largest, and may have even been a type of waiting room,57 that is where we have placed her. The brothel may not have operated at full capacity or been an economic success,58 and so we brought out those economic concerns in the critical fabulation from the madam’s perspective. The detail about rent being due comes from the likelihood that the owner of the structure was not the same as the individual who leased and ran the brothel,59 and the standard practice of rent being due at sixth-month intervals.60 It seems possible, if not probable, that a female manager of a brothel would have worked in a brothel herself when she was younger.61 What object was used to scratch graffiti into the brothel’s walls remains an open question. None of the objects found in the structure could be put to use effectively to write wall graffiti. Polly Lohmann, using replicas of ancient tools and a plastered wall, argues that metal styluses were the most likely implements for producing Pompeian graffiti, with iron nails and metal hairpins also being possibilities.62 Lohmann notes that writing even with the best of the implements was a slow and careful process. Looking at this one inscribed text through the lens of critical fabulation has allowed us to explore what may have led to the writing of the graffito, and what the repercussions were afterwards for various individuals. The graffito most likely resulted from a sexual encounter between client and prostitute, with their interaction continuing afterwards and involving banter and ultimately with both parties participating in the creation of the text. Already, we can see sexual and emotional labour lurking under the text of this one graffito. We see, too, different possible instigators and motivations, whether the client came up with the idea for writing on the wall (to compete with other clients), or whether the madam did (as part of a business strategy). We can imagine some of the later moments on the timeline, too, from the client being congratulated outside the brothel for his sexual prowess, to future revenue for the madam if graffiti like these succeeded in encouraging repeat customers (or possible economic failure if it did not), to continuing interactions between specific pairs of clients and prostitutes (and from there, more sexual and For the management of Roman brothels, see McGinn 2004, 37–40. Levin-Richardson 2019, 51. 58 Levin-Richardson 2019, 50, 146. 59 Levin-Richardson 2019, 94; for ownership and management of Roman brothels in general, see McGinn 2004, 30–40. 60 Frier 1977, 29. 61 See e.g., Strong 2012, 124; Marshall 2013, 177 (on this situation in New Comedy); Richlin 2017, 105 (on Plaut., Cist. 38–9); Petrova 2019, 168 (on Plaut., Most. 197–202); on this phenomenon in the modern period, see e.g., Patterson 2012, 333. 62 In Lohmann’s experiment (2018, 247–55), only metal implements could penetrate the wall plaster. Of the metal objects found in the brothel, the five bronze coins and bronze shell-shaped basin did not have the right shape to be used to write on the walls, while the structure’s iron blade would be too flexible to be effective (cf. Lohmann 2018, 250). 56 57

216

Deborah Kamen and Sarah Levin-Richardson

emotional labour on the part of prostitutes, and, at the same time, the possibility of the client helping the prostitute with gifts or money). Critical fabulation has also encouraged us to think about the bodily and emotional experiences of those present in the brothel. Through this methodology, we can recognize the various ways in which clients’ bodies and desires were privileged over the bodies and desires of others: not only did clients get to decide what they did with prostitutes, clients could also drink wine at leisure, and it seems as though their desire to have their masculinity affirmed was met. We envision how managing a brothel must have brought the madam into regular contact with her neighbours and the community, and we glimpse her economic vulnerability through concern for paying the rent (an anxiety that may underlie the common Roman stereotype of the “greedy” pimp or madam).63 Perhaps most importantly, we cannot ignore the aching body, bruises, young age, and limited hopes of Felicla and her fellow prostitutes. Finally, we get a sense of how various levels of literacy may have played out for Pompeii’s prostitutes (as well as for others with limited literacy). In the three scenarios above, Felicla is the sole writer in one (the fabulation involving the madam), is helped by Victor in another (the client’s fabulation), and covers up her illiteracy in the third (the prostitute’s fabulation). These scenarios may help us think in a more nuanced manner about literacy and authorship in antiquity, especially in terms of epigraphic texts like graffiti. These critical fabulations of course do not and cannot exhaust all possible narratives.64 For example, none of the narratives in the first case study explores the possibility that there were other enslaved persons in Kleomantis’ house, which could have changed the dynamics of the household. Likewise, none of the three variations in the second case study presents Felicla as the instigator of wall writing (though we see her possibly benefiting from it through the promise of future gifts in the narrative from the client’s perspective). However, even if we do not imagine all possibilities, this exercise nonetheless yields three distinct perspectives, including some that are difficult to access through traditional historical methods. The divergences between these perspectives, moreover, complicate the idea of history as a series of singular, univocal, uncontested facts.

Conclusions Applying this more imaginative methodology yields results in four areas. First, it allows us to reconstruct the events that might have led to – as well as followed – a particular inscription or set of inscriptions. A version of this kind of fleshing out of a timeline is of course something we routinely do, if not consciously so, with public For the greedy pimp or madam, see e.g., the representation of the pimp Ballio in Plautus’ Pseudolus; Myers 1996, 8, 12–16. 64 Cf. Hartman 2016, 209–13. 63

9.  Epigraphy and critical fabulation

217

inscriptions like laws or decrees. But it is less often attempted with documents like manumission records or graffiti, which may appear to have lower stakes (at least from a political or legal perspective) but which have just as important stories to tell. Second, critical fabulation demands that we ask questions that may not have knowable answers (and so are usually left unasked in traditional scholarship), providing important glimpses into experiences that otherwise remain below the threshold for scholarly inquiry.65 For instance, while scholars generally do not explore why any given manumission took place (since this is not made explicit in the records themselves), critical fabulation forces us to think through what might have changed in the relationship between enslaved person and owner that prompted the latter to free his chattel. Similarly, the lack of material evidence for a manager of the purposebuilt brothel means that she or he is usually left out of discussions of the structure. Critical fabulation from her perspective encourages us to consider what the madam might be doing throughout the day, what her history might have been, where she might be physically in space, and how she might interact with the prostitutes and clients. In these ways, critical fabulation (and our essay) takes part in a broader set of epistemological interventions – including by practitioners of Black feminist theory,66 of critical Indigenous studies,67 and of counterhistory68 – that challenge how knowledge is created, how it is valued, and the form in which it is presented. Relatedly, critical fabulation helps us to grasp, in a more visceral way than traditional historical analysis does, the experiences and emotions of individuals removed 2,000 years in time from us. It thus provides a new path into accessing ancient emotions, an avenue of inquiry that is growing in popularity in Classical studies.69 Finally, the payoff of this approach is not merely intellectual. As Saidiya Hartman has argued, we have an ethical obligation to tell the stories of those who are marginalized and oppressed, even if those stories are inevitably some combination of fiction and history.70 Having followed her exhortation to “paint as full a picture of the lives of [enslaved individuals] as possible”,71 we were forced to confront the dehumanizing behaviour of the slave-holding societies that we study. At the same time, through a combination of our imagination and our archives, we were able to add the experiences and hopes of enslaved women like Eisias and Felicla to the historical record.

Cf. Ferrer 2019 and Johnson 2019 on the new types of questions that speculative approaches allow us to ask about the experiences of slavery and freedom in Cuba and Haiti. 66 E.g., Collins 2009 [1990]. 67 E.g., Barker and Dumont 2006. 68 E.g., Gallagher and Greenblatt 2000. 69 See e.g., Chaniotis 2012. 70 Hartman 2008, 2–4, 11–12. 71 Hartman 2008, 11. 65

218

Bibliography

Deborah Kamen and Sarah Levin-Richardson

Barker, Joanne and Clayton Dumont. 2006. “Contested Conversations: Presentations, Expectations, and Responsibility at the National Museum of the American Indian.” American Indian Culture and Research Journal 30: 111–40. Cameron, A. 1939. “ΘΡΕΠΤΟΣ and Related Terms in the Inscriptions of Asia Minor.” In Anatolian Studies Presented to William Hepburn Buckler, edited by W.M. Calder and Josef Keil, 27–62. Manchester: Manchester University Press. Chaniotis, Angelos. 2012. “Moving Stones: The Study of Emotions in Greek Inscriptions.” In Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, edited by Angelos Chaniotis, 91–129. Stuttgart: Franz Steiner Verlag. Cohen, Edward E. 2014. “Sexual Abuse and Sexual Rights: Slaves’ Erotic Experience at Athens and Rome.” In A Companion to Greek and Roman Sexualities, edited by Thomas K. Hubbard, 184–98. Chichester, Malden, MA, Oxford: Wiley-Blackwell. Collins, Patricia Hill. 2009 [1990]. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York: Routledge. Connolly, Brian and Marisa Fuentes. 2016a. “Introduction: From Archives of Slavery to Liberated Futures?” In Slavery and the Archive, edited by Brian Connolly and Marisa Fuentes, 105–16. Special Issue of History of the Present 6. Connolly, Brian and Marisa Fuentes, eds. 2016b. Slavery and the Archive. Special Issue of History of the Present 6. Davidson, James N. 1997. Courtesans and Fishcakes: The Consuming Passions of Classical Athens. New York: St. Martin’s Press. Downs, Jim. 2018. “When the Present is Past: Writing the History of Sexuality and Slavery.” In Sexuality and Slavery: Reclaiming Intimate Histories in the Americas, edited by Daina Ramey Berry and Leslie M. Harris, 189–204. Athens, GA: University of Georgia Press. Dusinberre, William. 2009. Strategies for Survival: Recollections of Bondage in Antebellum Virginia. Charlottesville: University of Virginia Press. Ebbott, Mary. 2017. “Seeking Odysseus’s Sister.” Michigan Quarterly Review 56: 268–84. Eidinow, Esther. 2012. “‘What Will Happen to Me if I Leave?’ Ancient Greek Oracles, Slaves, and Slave Owners.” In Slaves and Religions in Graeco-Roman Antiquity and Modern Brazil, edited by Dick Geary and Stephen Hodkinson, 244–78. Newcastle upon Tyne: Cambridge Scholars Publishing. Ferrer, Ada. 2019. “Slavery, Freedom, and the Work of Speculation.” Small Axe 23: 220–8. Fiorelli, Giuseppe. 1862. Giornale degli scavi di Pompei. Naples: Stamperia della R. Università. Fischer, Josef. 2010. “Unfreiheit und Sexualität im Klassichen Athen.” In Unfreiheit und Sexualität von der Antike bis zur Gengenwart, edited by Josef Fischer and Melanie Ulz, 58–82. Hildesheim, Zurich and New York: Georg Olms. Frier, Bruce. 1977. “The Rental Market in Early Imperial Rome.” Journal of Roman Studies 67: 27–37. Gaca, Kathy L. 2021. “Controlling Female Slave Sexuality and Men’s War-Driven Sexual Desires.” In Slavery and Sexuality in Classical Antiquity, edited by Deborah Kamen and C.W. Marshall, 40–65. Madison: University of Wisconsin Press. Gallagher, Catherine and Stephen Greenblatt. 2000. Practicing New Historicism. Chicago, London: University of Chicago Press. Guzzo, Pier Giovanni and Vincenzo Scarano Ussani. 2009. Ex corpore lucrum facere: La prostituzione nell’antica Pompei. Rome: ‘L’Erma’ di Bretshneider. Hartman, Saidiya. 2007. Lose Your Mother: A Journey along the Atlantic Slave Route. New York: Farrar, Straus, and Giroux. Hartman, Saidiya. 2008. “Venus in Two Acts.” Small Axe 26: 1–14. Hartman, Saidiya. 2016. “The Dead Book Revisited.” In Slavery and the Archive, edited by Brian Connolly and Marisa Fuentes. Special Issue of History of the Present 6: 208–21.

9.  Epigraphy and critical fabulation

219

Hartman, Saidiya. 2019. Wayward Lives, Beautiful Experiments: Intimate Histories of Social Upheaval. New York: W.W. Norton & Company. Higgins, Kathleen J. 1999. ‘Licentious Liberty’ in a Brazilian Gold-Mining Region: Slavery, Gender, and Social Control in Eighteenth-Century Sabará, Minas Gerais. University Park, PA: Pennsylvania State University Press. Hopkins, Keith. 1978. Conquerors and Slaves. Cambridge: Cambridge University Press. Huemoeller, Katharine P. D. 2020. “Freedom in Marriage? Manumission for Marriage in the Roman World.” Journal of Roman Studies 110: 123–39. Hunt, Peter. 2018. Ancient Greek and Roman Slavery. Hoboken, NJ, Chichester: Wiley-Blackwell. Johnson, Sara E. 2019. “‘He Was a Lion, and He Would Destroy Much’: A Speculative School of Revolutionary Politics.” Small Axe 23: 195–207. Joshel, Sandra. 1992. Work, Identity, and Legal Status at Rome: A Study of the Occupational Inscriptions. Norman: University of Oklahoma Press. Kamen, Deborah. 2011. “Slave Agency and Resistance in Martial.” In Reading Ancient Slavery, edited by Richard Alston, Edith Hall, and Laura Proffitt, 192–203. London: Bristol Classical Press. Kamen, Deborah. 2012. “Manumission, Social Rebirth, and Healing Gods in Ancient Greece.” In Slaves and Religions in Graeco-Roman Antiquity and Modern Brazil, edited by Dick Geary and Stephen Hodkinson, 174–94. Newcastle upon Tyne: Cambridge Scholars Publishing. Kamen, Deborah. 2014a. “Sale for the Purpose of Freedom: Slave-Prostitutes and Manumission in Ancient Greece.” Classical Journal 109: 281–307. Kamen, Deborah. 2014b. “Slave-Prostitutes and ἐργασία in the Delphic Manumission Inscriptions.” Zeitschrift für Papyrologie und Epigraphik 188: 149–53. Kamen, Deborah and C.W. Marshall, eds. 2021. Slavery and Sexuality in Classical Antiquity. Madison: University of Wisconsin Press. Kamen, Deborah and Sarah Levin-Richardson. Forthcoming. “Approaching Emotions and Agency in Greek and Roman Slavery.” In Les lectures contemporaines de l’esclavage: problématiques, methodologies et analyses depuis les années 1990, edited by Adam Pałuchowski. Franche-Comté: Presses universitaires de Franche-Comté. Klees, Hans. 1998. Sklavenleben im klassischen Griechenland. Stuttgart: Franz Steiner Verlag. Levin-Richardson, Sarah. 2011. “Facilis hic futuit: Graffiti and Masculinity in Pompeii’s ‘Purposebuilt’ Brothel.” Helios 38: 59–78. Levin-Richardson, Sarah. 2013. “fututa sum hic: Female Subjectivity and Agency in Pompeian Sexual Graffiti.” Classical Journal 108: 319–45. Levin-Richardson, Sarah. 2019. The Brothel of Pompeii: Sex, Class, and Gender at the Margins of Roman Society. Cambridge: Cambridge University Press. Levin-Richardson, Sarah. 2021. “Sex and Slavery in the Pompeian Household: A Survey.” In Slavery and Sexuality in Classical Antiquity, edited by Deborah Kamen and C.W. Marshall, 188–210. Madison: University of Wisconsin Press. Lohmann, Polly. 2018. Graffiti als Interaktionsform: Geritzte Inschriften in den Wohnhäusern Pompejis. Berlin: De Gruyter. Marshall, C.W. 2013. “Sex Slaves in New Comedy.” In Slaves and Slavery in Ancient Greek Comic Drama, edited by Ben Akrigg and Rob Tordoff, 173–96. Cambridge: Cambridge University Press. Marshall, C.W. and Deborah Kamen. 2021. “Mere Sex Objects?” In Slavery and Sexuality in Classical Antiquity, edited by Deborah Kamen and C.W. Marshall, 3–14. Madison: University of Wisconsin Press. McGinn, Thomas. 2004. The Economy of Prostitution in the Roman World: A Study of Social History and the Brothel. Ann Arbor: University of Michigan Press. McGinn, Thomas. 2013. “Sorting out Prostitution in Pompeii: The Material Remains, Terminology and the Legal Sources.” Journal of Roman Archaeology 26: 610–33.

220

Deborah Kamen and Sarah Levin-Richardson

Milnor, Kristina. 2014. Graffiti and the Literary Landscape in Roman Pompeii. New York: Oxford University Press. Mulliez, Dominique. 2006. “Maîtres et esclaves: Les relations familiales au sein de l’oikos. L’exemple de Delphes.” In Parenté et société dans le monde grec, de l’Antiquité à l’âge moderne. Actes du colloque de Volos (18–21 juin 2003), edited by Alain Bresson, Marie-Paul Masson-Vincourt, Stavros Perentidis, and Jérôme Wilgaux. Pessac: Ausonius Éditions. https://books.openedition.org/ausonius/9422. Mulliez, Dominique. 2016. “La loi, la norme et l’usage dans les relations entre maîtres et esclaves à travers la documentation delphique (200 av. J.-C.–100 ap. J.-C.).” In Esclaves et maîtres dans le monde romain. Expressions épigraphiques de leurs relations, edited by Monique Dondin-Payre and Nicolas Tran. Rome: Publications de l’École française de Rome. https://books.openedition.org/efr/3192. Myers, K. Sara. 1996. “The Poet and the Procuress: The Lena in Latin Love Elegy.” Journal of Roman Studies 86: 1–21. Paradiso, Annalisa. 1999. “Schiavitù femminile e violenza carnale: stupro e coscienza dello stupro sulle schiave in Grecia.” In Femmes–esclaves. Modèles d’interprétation anthropologique, économique, juridique. Atti del XXI Colloquio internazionale GIREA Lacco Ameno–Ischia, 27–29 ottobre 1994, edited by F. Reduzzi Merola and A. Storchi Marino, 145–62. Naples: Jovene. Patterson, Orlando. 1982. Slavery and Social Death. A Comparative Study. Cambridge, MA, London: Harvard University Press. Patterson, Orlando. 2012. “Trafficking, Gender and Slavery: Past and Present.” In The Legal Understanding of Slavery: From the Historical to the Contemporary, edited by Jean Allain, 322–59. Oxford: Oxford University Press. Perry, Matthew. 2014. Gender, Manumission, and the Roman Freedwoman. New York: Cambridge University Press. Petrova, Mina. 2019. “Single as a Lena: The Depiction of Procuresses in Augustan Literature.” In The Single Life in the Roman and Later Roman World, edited by Sabine R. Huebner and Christian Laes, 165–78. Cambridge: Cambridge University Press. Richlin, Amy. 2017. Slave Theater in the Roman Republic. Cambridge: Cambridge University Press. Ripat, Pauline. 2014. “Cheating Women: Curse Tablets and Roman Wives.” In Daughters of Hecate: Women and Magic in the Ancient World, edited by Kimberly B. Stratton and Dayna S. Kalleres, 340–64. Oxford: Oxford University Press. Smallwood, Stephanie E. 2016. “The Politics of the Archive and History’s Accountability to the Enslaved.” In Slavery and the Archive, edited by Brian Connolly and Marisa Fuentes, 117–32. Special Issue of History of the Present 6. Sosin, Joshua D. 2015. “Manumission with Paramone: Conditional Freedom?” TAPA 145: 325–81. Strong, Anise K. 2012. “Working Girls: Mother-daughter Bonds among Ancient Prostitutes.” In Mothering and Motherhood in Ancient Greece and Rome, edited by Lauren Hackworth Petersen and Patricia Salzman-Mitchell, 121–39. Austin: University of Texas. Todd, S.C. 2013. “Male Slave Sexuality and the Absence of Moral Panic in Classical Athens.” Bulletin of the Institute of Classical Studies 56: 37–53. Tringham, Ruth. 1991. “Households with Faces: The Challenge of Gender in Prehistoric Achitectural Remains.” In Engendering Archaeology: Women and Prehistory, edited by Joan M. Gero and Margaret W. Conkey, 93–131. Oxford, Cambridge, MA: Blackwell. Trouillot, Michel-Rolph. 2015 [1995]. Silencing the Past: Power and the Production of History. Boston: Beacon Press. Tucker, C. Wayne. 1982. “Women in Manumission Inscriptions at Delphi.” Transactions of the American Philological Association 112: 225–36. Walin, Daniel C. 2012. Slaves, Sex, and Transgression in Greek Old Comedy. PhD dissertation, University of California, Berkeley. Wiedemann, Thomas. 1981. Greek and Roman Slavery. Baltimore, London: Johns Hopkins University.

9.  Epigraphy and critical fabulation

221

Williams, Craig. 2014. “Sexual Themes in Greek and Latin Graffiti.” In A Companion to Greek and Roman Sexualities, edited by Thomas K. Hubbard, 493–507. Malden, MA: Wiley-Blackwell. Zanovello, Sara L. 2017. From Slave to Free: A Legal Perspective on Greek Manumission. PhD dissertation, Università degli Studi di Padova and University of Edinburgh. Zelnick-Abramovitz, Rachel. 2005. Not Wholly Free: The Concept of Manumission and the Status of Manumitted Slaves in the Ancient Greek World. Leiden: Brill. Zelnick-Abramovitz, Rachel. 2018. “The Status of Slaves Manumitted under Paramonē: A Reappraisal.” In Symposion 2017: Vorträge zur griechischen und hellenistischen Rechtsgeschichte (Tel Aviv, 20.–23. August 2017), edited by Gerhard Thür, Uri Yiftach, and Rachel Zelnick-Abramovitz, 377–401. Vienna: Österreichische Akademie der Wissenschaften.