Dress and Identity 9781407309422, 9781407339238

Table of contents :
Front Cover
Title Page
Copyright
IAA Interdisciplinary Series Studies in Archaeology, History, Literature and Art
TABLE OF CONTENTS
CONTRIBUTORS
CHAPTER 1. DRESS AND IDENTITY: AN INTRODUCTION
CHAPTER 2. COSTUME AS TEXT
CHAPTER 3. VEILING THE SPARTAN WOMAN
CHAPTER 4. DRESSING TO PLEASE THEMSELVES: CLOTHING CHOICES FOR ROMAN WOMEN
CHAPTER 5. THE ARCHAEOLOGY OF ADORNMENT AND THE TOILET IN ROMAN BRITAIN AND GAUL
CHAPTER 6. DRESS AND CULTURAL IDENTITY IN THE ROMAN EMPIRE
CHAPTER 7. INVESTIGATING THE EMPEROR’S TOGA: PRIVILEGING IMAGES ON ROMAN COINS
CHAPTER 8. ANGLO-SAXON WOMAN: FAME, ANONYMITY, IDENTITY AND CLOTHING
CHAPTER 9. REPRESENTING HIERARCHY AND HOMOSOCIALITY: VESTMENTS AND GENDER IN MEDIEVAL SCOTLAND
CHAPTER 10. COSMETICS AND PERFUMES IN THE ROMAN WORLD: A GLOSSARY
CHAPTER 11. THE SOCIAL LIFE OF MUSEUM TEXTILES: SOME COMMENTS ON THE LATE ANTIQUE AND EARLY MEDIEVAL COLLECTION IN THE URE MUSEUM AT THE UNIVERSITY OF READING

Citation preview

IAA Interdisciplinary Series Studies in Archaeology, History, Literature and Art Volume II Series Editor Gillian Shepherd

Dress and Identity Edited by

Mary Harlow

BAR International Series 2356 2012

Published in 2016 by BAR Publishing, Oxford BAR International Series 2356 University of Birmingham IAA Interdisciplinary Series: Studies in Archaeology, History, Literature and Art 2 Dress and Identity © The editors and contributors severally and the Publisher 2012 Fragment of tapestry band (clavus) with vine scroll. From Akhmîm, Egypt, probably 4th‐ 6th century AD. Ure Museum, University of Reading

COVER IMAGE

The authors' moral rights under the 1988 UK Copyright, Designs and Patents Act are hereby expressly asserted. All rights reserved. No part of this work may be copied, reproduced, stored, sold, distributed, scanned, saved in any form of digital format or transmitted in any form digitally, without the written permission of the Publisher.

ISBN 9781407309422 paperback ISBN 9781407339238 e-format DOI https://doi.org/10.30861/9781407309422 A catalogue record for this book is available from the British Library BAR Publishing is the trading name of British Archaeological Reports (Oxford) Ltd. British Archaeological Reports was first incorporated in 1974 to publish the BAR Series, International and British. In 1992 Hadrian Books Ltd became part of the BAR group. This volume was originally published by Archaeopress in conjunction with British Archaeological Reports (Oxford) Ltd / Hadrian Books Ltd, the Series principal publisher, in 2012. This present volume is published by BAR Publishing, 2016.

BAR PUBLISHING BAR titles are available from:

E MAIL P HONE F AX

BAR Publishing 122 Banbury Rd, Oxford, OX2 7BP, UK [email protected] +44 (0)1865 310431 +44 (0)1865 316916 www.barpublishing.com

IAA Interdisciplinary Series Studies in Archaeology, History, Literature and Art Institute of Archaeology and Antiquity University of Birmingham

The IAA Series is an interdisciplinary volume reflecting the wide geographical, chronological and disciplinary range of the Institute of Archaeology and Antiquity at the University of Birmingham. The volumes are based on a thematic research seminar series held at the IAA. Contributions from scholars outside the IAA are welcome. For further information on current and future themes, the research seminars and future volumes in the IAA Interdisciplinary Series, please contact the IAA Series Editor, Gillian Shepherd ([email protected]) or visit www.iaa.bham.ac.uk.

Table of Contents Contributors iii Chapter 1

Dress and Identity: an Introduction 1 Mary Harlow Chapter 2

Costume as Text 7 Zvezdana Dode Chapter 3

Veiling the Spartan Woman 17 Lloyd Llewellyn-Jones Chapter 4

Dressing to Please Themselves: Clothing Choices for Roman Women 37 Mary Harlow Chapter 5

The Archaeology of Adornment and the Toilet in Roman Britain and Gaul 47 Ellen Swift Chapter 6

Dress and Cultural Identity in the Roman Empire 59 Ursula Rothe Chapter 7

Investigating the Emperor’s Toga: Privileging Images on Roman Coins 69 Ray Laurence Chapter 8

Anglo-Saxon Woman: Fame, Anonymity, Identity and Clothing 83 Gale R. Owen-Crocker Chapter 9

Representing Hierarchy and Homosociality: Vestments and Gender in Medieval Scotland 95 Penelope Dransart Chapter 10

Cosmetics and Perfumes in the Roman World: A Glossary 109 Susan Stewart CHAPTER 11

The Social Life of Museum Textiles: Some Comments on the Late Antique and Early Medieval Collection in the Ure Museum at the University of Reading 117 Anthea Harris

i

Contributors Zvezdana Dode is Professor of Cultural Anthropology at Stavropol University and Senior Researcher at the Southern Scientific Centre of Russian Academy of Sciences. The main focus of her research is the archaeology of textiles and costume. Penelope Dransart is Reader in Anthropology and Archaeology at the University of Wales, Lampeter. Much of her research has focussed on Andean culture with an interest is dress and textiles. Mary Harlow is a Senior Lecturer in the Institute of Archaeology and Antiquity at the University of Birmingham. Her research interests are in the social and cultural meanings of Roman dress. Her other interests are in the life course – examining age and gender in Roman society. Anthea Harris is a Lecturer in Archaeology in the Institute of Archaeology and Antiquity, University of Birmingham. Her main research interests are in the material culture of western medieval kingdoms post AD 400, and cultural interaction with the east. Ray Laurence is Professor in the Department of Classical and Archaeological Studies, University of Kent. His research interests range from the city in antiquity, to age and ageing in the Roman Empire to cultural heritage and the classical tradition. Lloyd Llewellyn-Jones is a Senior Lecturer in the Classics Department, School of History, Classics and Archaeology, University of Edinburgh. He has published extensively on dress in the Greek and Persian worlds and on the reception of antiquity in popular culture. Gale Owen-Crocker is Professor of Anglo-Saxon Culture at the University of Manchester. Her specialisms are Old English, particularly Beowulf, interdisciplinary Anglo-Saxon research and dress/textiles. She directs a 5 year AHRC funded project: The lexis of cloth and clothing in Britain c. 700-1450: origins, identifications, contexts and change which will produce a data-base and subject related multi-lingual dictionary. She is also Co-Investigator on a Leverhulme funded project: Medieval dress and textile vocabulary in unpublished sources. Gale also co-founded and co-edits the international journal, Medieval Clothing and Textiles. She is currently Director of MANCASS (Manchester Centre for Anglo-Saxon Studies) and President of the Manchester Medieval Society. Ursula Rothe is a Leverhulme Early Career Research Fellow in the Classics Department, School of History, Classics and Archaeology, University of Edinburgh. Her research focuses on cultural identity in the Roman provinces, with a particular interest in clothing and dress behaviour. Susan Stewart is an independent scholar and librarian. Following on from her PhD, Susan published her book Cosmetics and perfumes in the Roman world in 2007. Her research focuses on beauty products in antiquity but she is also interested in ancient medicine and in classical reception studies. Ellen Swift is a Senior Lecturer in the Department of Classical and Archaeological Studies, and a director of the Centre for Late Antique Archaeology, University of Kent. She has wide interests in artefact studies in the late Roman period and has published extensively on dress and dress accessories.

iii

Chapter 1

Dress and Identity: an Introduction Mary Harlow Most of us get up in the morning and immediately make choices about what we will wear depending on the activities we intend to undertake during the day. The clothes and other accoutrements we choose to put on can be used to signify any number of identifiers: age, gender, status, occupation, ethnicity, relative wealth, social class, in an almost unending series of variation. As a form of non-verbal communication dress conveys a multiplicity of messages about the wearer to the viewer. If both parties share a similar cultural background and understand the social codes surrounding the clothed body in that context, such messages might be perceived as relatively transparent. However, the subtlety and nuances of meanings encoded in dress are rarely simple and require a range of analytical tools to unpick even in a shared cultural context. This problem is enhanced if the observer is distanced by culture and time.

with a representation that is static or chosen for a particular purpose. For example, the iconography on a funerary monument may be formulaic but it may still reflect social aspirations, ideals or indeed, social reality. Grave goods likewise can fulfil a number of roles. They may reflect the status of the individual when alive or the aspirations of the family or social group, and they may also have religious connotations. In literature the role of genre is fundamental in structuring the image which is presented. This means that identities are fixed at a specific moment in time and it is hard to assess how far they were controlled at any point by an individual. The terms used to discuss the subject: dress, costume, fashion etc. are also not without problems.2 In recent scholarship much of the history of the clothed body has assigned early research (i.e. prior to 1960s) as essentially the history of ‘costume’ as it was very much the preserve of museum curators with an interest in conservation and reconstruction or art historians with a desire to correctly label items of clothing for identification and dating purposes. Much of this scholarship was concerned with accurate description and sewing techniques with the aim of possible reconstruction or restoration.3 Unfortunately this body of work is sometimes – quite erroneously – regarded as passé. On the contrary however this work not only allows those with an interest in dress in past societies to develop a grounding in the subject but it is thriving among experimental archaeologists and re-enactors who are often very conscientious about achieving as near replication of their ‘costumes’ as possible.4 Costume as a term needs to be contextually defined and re-instated into the broader spectrum. ‘Fashion’ is also a problematic term with

The concept of social identity is also dependent on context. No individual can be said to have a single identity but rather a number of identities they can assume depending on circumstance. An individual may have an identity as part of a social grouping which can be very broad, based on gender or ethnicity, for instance. Within that grouping individuals will have nuanced self-identities. At any one time an individual has a range of identities available to them. The identity of a group is often defined differently by those who belong and those who are excluded. The visual and immediate messages of dress play a key part in presenting identities.1 Moreover, dress can be manipulated by the wearer to establish a number of associations which will send messages to the viewer about how an individual wishes to be regarded. These messages are frequently both complex and ambiguous. The evidence for the dress and overall appearance of an individual in the past is often lost to us as the literary sources and/or material remains are invariably fragmentary and often reflect the voice of the dominant group. Even when the evidence appears on the surface to present a relatively unambiguous picture, distance in space, time and culture makes interpretive analysis all the more tentative. One of the problems we face is that we are often dealing

Eicher and Roach-Higgins (1992: 8-29). For example, Arnold (1972); and the extensive work of C.W. Cunnington & P. Cunnington: C.W. Cunnington (1937): English Women’s Clothing in the Nineteenth Century.London; (1952): English Women’s Clothing in the Present Century. London; C.W. & P. Cunnington (1951): History of Underclothes. London; (1952): Handbook of Medieval Costume; (1954): Handbook of English Costume in the Sixteenth Century. London; (1955): Handbook of English Costume in the Seventeenth Century. London; (1957): Handbook of Costume in the Eighteenth Century. London; (1959): Handbook of Costume in the Nineteenth Century. London; (1959): A Dictionary of English Costume. London. For discussion of the tensions in the discipline see Cumming (2004: 15-32); Taylor (2004: 4-65); Styles (1998). 4 See for example images in Croom (2002); Hendzel et al. (2008); Nørgaard (2008). 2

3

There is an extensive literature on the relationship between dress and identity. See, for example: Barnes & Eicher (eds) (1992); Davis (1992); Kuper (1973). 1

1

Dress and Identity

implications of the short term, ephemerality and western consumerism. Here again it cannot be translated into the past without some qualification but it can be done. Readers of classical Latin texts might arguably recognise notions of fashion underlying the (male) Roman anxiety expressed in moralising literature. Changes in style, new types of fabric, especially silk, and particular colours all raised the hackles of certain Roman authors.5 In the conservative world of Roman dress, any alterations to traditional styles could be regarded as suspect, but the impact of empire brought choices for those who could afford them. As with costume, fashion needs careful contextualising but we limit our analysis if we equate fashion only with modernity. Dress, on the other hand, is much more of a catch-all word. It can be used to cover not only clothing but also footwear, headgear, jewellery and other bodily adornments. It can encompass both costume and fashion and is arguably less value laden. It is used in that sense throughout this volume.

sociology, cultural studies, gender studies, museum studies, economics, social history and social policy. For each of these, and more, the study of dress and bodily adornment has become a powerful analytical tool. Most recently (2010) Berg has produced the Encyclopedia of World Dress and Fashion in ten volumes (Joanne Eicher, editor in chief). While this is aimed primarily at the fashion industry and related university/college courses it also has extensive sections on dress in the past in a world perspective. In 2002 and 2004 Lou Taylor, and also in 2004 Valerie Cumming, produced handbooks for the study of dress and fashion history respectively which defined the directions taken since the end of the twentieth century.8 These developments have not gone unnoticed by historians of the ancient and medieval worlds and here too the sophistication of approach has evolved in the last thirty years. In 2002 Lloyd Llewellyn-Jones (a contributor to this volume) was justified in claiming: ‘Outside of the realm of fashion history proper, the study of dress as an indicator of a society’s mores has been almost entirely neglected by ancient historians and classicists, who despite several generations’ worth of potted histories on the construction and draping of Greco-Roman clothing… have only infrequently attempted to investigate the role of clothing in its wider cultural context’.9 In the decade since that was written the situation has altered drastically – for the better. For the Roman world the situation changed with the publication of the World of Roman Costume in 1994, co-edited by Judith Sebesta and Larissa Bonfante. This volume seems to have inspired a plethora of publications and new research projects among ancient historians and classicists in English.10 Lloyd Llewellyn-Jones has been instrumental in pushing the study of Greek dress forward with his own research.11 He is by no means a lone scholar and a wider range of specialist studies have been published. In 2005 Linda Roccos published Ancient Greek Costume: an annotated bibliography 1784-2005 which contained 603 entries. Dress has always held a peripheral interest for scholars of antiquity but it has tended to be studied in separate disciplines with little interaction: classical art historians, archaeologists, classicists, ancient historians, textile conservators have all approached dress from the

Understanding the language of dress is something we do unconsciously in our own immediate social surroundings, however if one moves outside that immediate comfort zone – sometimes not very far, into a new work environment for instance, or sometimes further to another country perhaps – dress codes can become harder to read, absorb and embody. The theoretical framework for reading the language of dress is perforce multi and interdisciplinary. It uses the methodologies of disciplines for whom the body in society is key: sociology, anthropology and psychology. The notion of dress as a means of communication had been noted and as early as the nineteenth century by Thomas Carlyle (1833-4) and since then, now seminal works have developed a discourse around the semiotics and semantics of clothing and the body.6 Methodological approaches have become increasingly sophisticated. Over time disparate disciplines have made overt use of each other’s methodologies and adapted them to suit different subject interests. This cross disciplinary approach has created a burgeoning of interest in the study of dress, in the past and the present. This is manifest in the number publications over the past thirty years and in teaching in academic institutions. For instance, the extensive list from the catalogue of Berg Publishing demonstrates the range of disciplines that have an interest in fashion and dress; Berg have also produced the journals Fashion Theory: Journal of Dress, Body and Culture (since 1997) and Textile: Journal of Cloth and Culture (since 2003).7 Berg’s list is evidence of the interdisciplinary nature of new dress history. It is impossible to pin down many of the volumes to a single discipline: there is an intermingling of anthropology,

Cumming writes from the perspective of museum collections. In 2004 she was not optimistic about the co-operative approaches (2004: 9-10). It may be that work between those in museums and various academic settings needs more encouragement but personal experience would suggest this is more a question of time constraints rather than lack of will. She is also highly critical of collection of essays such as this which she views as ‘fragmentary’ (39) or written for research assessment purposes (8), or at worst a form of ‘intellectual tourism’ (7). 9 Llewellyn-Jones (2002: vii). 10 This is not the place (or space) for a full bibliography but the separate bibliographies of each chapter in this volume are primarily full of post 1994 publications. 11 Llewellyn-Jones, (2002): ‘A woman’s view? Dress, eroticism and the ideal female body in Athenian art’, in Llewellyn-Jones (ed). Women’s Dress in the Ancient Greek World. 171-202; (2003): Aphrodite’s Tortoise: The Veiled Women of Ancient Greece. Swansea; (2005): ‘The fashioning of Delilah. Costume design, historicism and fantasy in Cecil B. DeMille’s Samson and Delilah (1949) in L. Cleland, M. Harlow & L. J. LlewellynJones (eds) The Clothed Body in the Ancient World. 14-29. Oxford; (2007): (with Liza Cleland and Sue Blundell) Greek and Roman Dress from A-Z. London. 8

E.g.: Tertullian De cultu feminarum; Clement of Alexandria, Paidagogos 2.10. 6 Thomas Carlyle Sator Resartus (1833-4); Roland Barthes, The Fashion System (1967); Pierre Bourdieu, Outline of a Theory of Practice (1977). See also Carter (2003) for discussion of early theorists. 7 For Berg see: http://www.bergpublishers.com/; online fashion library: http://www.bergfashionlibrary.com/ and Fashion Theory: http://www. bergpublishers.com/BergJournals/FashionTheory/tabid/524/Default.aspx; Textile: http://www.bergpublishers.com/BergJournals/Textile/tabid/518/ Default.aspx 5

2

Mary Harlow: Dress and Identity: an Introduction

point of view of their own disciplines.12 In the last few years the willingness of scholars to interact and to absorb the methodologies of tangential disciplines has produced a range of interesting publications (many by contributors to this volume) and an increasing self-awareness of ways dress can be used to shed light on many aspects of life in antiquity.

In academic departments the study of dress and textiles remains peripheral to most, central to a few. However, even in traditional departments of Classics, Ancient History and Archaeology (however they are configured) more and more courses are introducing aspects of the study of dress and textiles into more mainstream modules. Such drip feeding has been enhanced by interdisciplinary approaches in much the same way as gender studies has become implicitly embedded in many courses.18

Interaction between archaeologists, textile conservators and textile craftsmen and women has created advances in the preservation of material and new understandings of the techniques of production (e.g. papers in Gillis and Nosch 2007).13 This research has in turn highlighted the relationship of textiles and their production to the groups who consumed them. More recently the interaction of these disciplines with what I call (for want of a better collective noun) social and cultural historians has brought a new dynamic to the study of dress in the past. An example of this interaction can be seen in European funded international and highly cross and interdisciplinary Dress ID project.14 This project, subtitled ‘Clothing and identities: new perspectives on textiles in the Roman world’ encompasses a number of themes and working groups: Elements of Dress; Materials and Techniques; Quality; Colour and Dating; Experimental Archaeology; Production and Trade; Gender and Age; Self and Society; Dress and Religion; Dress in the Roman Provinces. Dress ID has brought together interested parties from across Europe including academics from a wide range of disciplines and craftspeople meeting in a series of workshops, seminars and conferences. It will culminate in a series of major exhibitions and members have already published extensively. Across Europe there is a positive outpouring of scholarship on dress, textiles and related subjects and the international networks are demonstrated by contributors to conferences and various volumes.15 In Denmark the Centre for Textile Research (CTR) at the University of Copenhagen brings together experts from across the world to encourage interdisciplinary research.16 In a recent paper Lise Bender Jørgensen has wondered whether the study of textiles will become a subject in its own right in the Humanities and Social Sciences (2007: 11).17

In Canada and the United States the Sebesta and Bonfante effect can be seen in the works of Kelly Olsen and others,19 most recently in the publication of another set of seminar papers Roman Dress and the Fabrics of Roman Culture edited by Jonathan Edmondson and Allison Keith from the University of Toronto (2008). The volume demonstrates the impact the study of dress is having on more traditional areas of study in classics, ancient history and archaeology. Of the fourteen contributors, twelve are identified as working in departments of classics and classical studies (the remaining two are defined by Humanities and Greek and Roman Art). In their extensive introduction Edmondson and Keith quite rightly state ‘Dress studies is clearly now well established as a recognised field of intellectual enquiry. It can no longer be accused of being a frivolous or light weight topic’ (2008: 3).20 Academics are using the skills of their own discipline and absorbing those of others to examine the meanings of dress in the ancient world. With the ever growing bibliography on the subject it is impossible to resist the need to take an interdisciplinary approach. The study of dress has perforce to be holistic. This has been a very brief introduction to the history of dress and the debates that surround it as a discipline. The short survey of the state of widening interest in studying dress among historians of antiquity and the Middle Ages addresses only a tiny part of a very thriving academic community. It is of necessity succinct and partial, but interested parties need only look at the review sections of relevant journals to see the quantity and quality of current research. This volume

For example: Wilson, L, (1924) The Roman Toga. Baltimore & (1938): The Clothing of the Ancient Romans. (Baltimore); Beiber, M. (1977): Ancient Copies. New York; Wild, J. P. (1968): ‘Clothing in the north-west provinces of the Roman Empire’, BJ 168: 166-239; (1970): Textile Manufacture in the Northern Roman Provinces. Cambridge; (1985): ‘The clothing of Britannia, Gallia Belgica and Germania Inferior’ ANRW II.12.3: 362-423. 13 CTR and the Danish National Research Fund have sponsored the revamped Archaeological Textiles Newsletter since 2008. 14 http://www.dressid.eu/. 15 Again, this is not the place for a full bibliography but to offer a sample: von Rummel, P. (2007): Habitus barbarous: Kleidung und Repräsentation spätantiker Eliten im 4 und 5 Jahrhundert. Berlin; Goette, H. R. (1990): Studien zu römischen Togadarstellungen. Mainz; contributors to D. Cardon & M. Feugère (eds) (2000): Archéologie des Textiles des origins au Ve siècle. Montagna; contributors to Chausson, F. & Inglebert, H. (eds) (2003): Costume et Societé dans L’Antiquité et le haut Moyen Age. Paris; Larsson Loven, L. (2002): The Imagery of Textile Making: Gender and Status in the Funerary Iconography of Textile Manufacture in Roman Italy and Gaul. Göteborg; contributors to Antiquité Tardive 12 (2004); contributors to Gillis and Nosch (2007). 16 http://ctr.hum.ku.dk/ 17 Presumably outside Colleges of Fashion and Art. 12

This collection of papers arose from a seminar series held

For a recent survey of gender teaching in UK universities see S. Blundell (2009): ‘Gender and the Classics Curriculum: A Survey’ Arts and Humanities in Higher Education 8.2: 135-158; for gender in archaeology, R. Whitehouse (2009): ‘Where have all the men gone? Sex, gender and women’s studies’ in E. Herring & K. Lomas (eds) Gender Identities in Italy in the First Millenium BC. 7-12. Oxford. 19 Olsen, K. (2002): ‘Matrona and whore: the clothing of women in Roman antiquity’. Fashion Theory 6.4: 387-420; (2008): Dress and the Roman Woman. London; (2008): ‘Dress and the appearance of the young Roman girl’ in Edmondon and Keith (2008: 139-57). 20 Despite this optimistic view, however, this editor could still have an otherwise positive reviewer of a book proposal comment that ancient historians might need persuading as to why they should read a book on dress (anonymous reviewer 2010). 18

3

Dress and Identity

by the Institute of Archaeology and Antiquity, University of Birmingham in 2005.21

period to examine the possibility of female agency. Roman women obviously had the opportunity to make choices in their dress, despite the moral codes which underlie much of the male-authored writing that survives. Harlow attempts to see how far this evidence can be used to look at the social reality of such choices, and how far women might have internalised or subverted male ideas of how they should look.

The present volume covers a wide chronological and geographical span: from archaic Greece to medieval Scotland by way of the Roman Empire and Anglo-Saxon England. The contributors come from a number of different academic disciplines: history, archaeology and classics. The authors have differing approaches dependent on their dominant discipline and/or period but the use of dress as an analytical tool and the assumption of methodologies from companion disciplines are common to all. There are a number of shared themes which characterise the volume: the examination of the ideology and symbolic nature of dress; the relationship between the literary, visual and archaeological evidence and social reality – as far as it can be accessed; and the construction and conservation of material. Explicit in some papers and implicit in others are the economics of dress: clothing as stored wealth in the form of luxury goods and symbolic capital; the notion of exchange both in terms of trading of cloth but also of ideas; and, of course, ideas about gender and national, regional and individual identities.

The theme of female identity through appearance continues with Ellen Swift’s paper on the archaeology of adornment or the toilet in Roman Britain. Here the representation of women on funerary monuments from Italy and Gaul is used as a framing for the finds of female toilet items in Roman Britain. Swift takes an integrated approach which considers how far the construction of an identity for an individual in the funerary ritual is symbolic, and how the uptake of Roman funerary practice reflects a change in the social realities of the Romano-British. The results demonstrate the evolving integration with Roman culture across time. First the assumption of the process of the toilet by indigenous elites, succeeded over time and with increasing occupation and contact with the empire by a more identifiably Roman cultural identity.

Zvezdana Dode, in the translation of a paper first presented in Russian, covers the methodological issue of reading dress and the various approaches that are used by researchers. Her paper considers the tensions in the relationship between meanings given to dress by its original creators and wearers and the use of dress by researchers to model an historical and cultural reality for clothing ensembles and individual pieces. The paper discusses the conceptual tools required to reconstruct past costumes: the knowledge of available materials and their provenance; understanding of the symbolic meaning of costume items in their own time, in short the evolution of a syntax of dress. Dode’s interest is in the Caucasus in the 7th to 10th centuries but her discussion of approaches speaks to researchers in all disciplines.

Cultural interaction in the Roman provinces is also tackled by Ursula Rothe. She makes the point that despite research into the ‘hidden transcripts’ of archaeological evidence, little attention has been given to the part dress may play in the construction of identity in the provinces. Rothe’s research here and in the publication of her thesis (2009) adds to the existing work on textiles and dress description by examining the meaning of dress choices displayed on funerary monuments in the north western provinces. This chapter, like that of Dode, is primarily concerned with methodologies and similarly stresses the plural here, making the point that those who study dress are required to step outside the comfort zones of discrete disciplines.

Lloyd Llewellyn-Jones is well known for his work on Greek dress and here the focus is on Spartan women. He analyses a range of texts and images to conclude that despite accepted social differences between Spartan society and the rest of the Greek world (accepted by modern scholarship, that is), Spartan women did in fact veil themselves when the occasion demanded. He argues that this aspect of Spartan gender identity has been overlooked so far because scholars tend to privilege literary evidence, while the material culture produced by Sparta itself tells a different story. The paper raises the issue of how far Sparta, for all its apparent differences, shared the gender ideologies of its time.

Ray Laurence made a deliberate choice to study images on coinage and explicitly aligns himself with the methodologies of ‘new dress history’. The emphasis on numismatics is partly because of the role dress might play in the obvious propaganda programme of coins but also to act as a foil to the privileging of sculpture over almost any other form of evidence in most studies of dress in antiquity. Laurence makes the case that despite its lack of detail, coinage can offer an interesting dynamic to communication between emperor and subject. Though a series of case studies on particular themes (annona to alimenta; congiarium to liberalitas; adlocutio to exercitus; imperial children) he discusses the implications of the imagery of the toga-clad emperor and the various groups with whom he interacts and the new symbolic meaning this may have given the toga in the 1st and 2nd centuries AD.

Female dress is also the focus of Mary Harlow’s chapter. This attempts to unpick the tensions between the literary and visual evidence for female dress in the early imperial

Gale Owen-Crocker commences her chapter by pointing out that for the Anglo-Saxon period, as with antiquity, the impetus of feminist scholarship and interest in the

It has taken time to come to fruition and that is entirely the editor’s responsibility. I would like to thank all the contributors and the publishers for their patience and forbearance. I would also like to thank Gillian Shepherd for all her additional assistance and editorial support. 21

4

Mary Harlow: Dress and Identity: an Introduction

history of women has refocused academic attention on previously rather overlooked material. Burial evidence for early period (mid 5th – early 7th centuries) often indicates gender distinctions in adornment (harder to locate in the later period) and grave assemblages allow for tentative reconstructions of female dress. Examination of surviving grave goods offers some insight into ethnicity, sometimes the age group and status of the individual and their belief system. The information gathered is never complete but the range and often quantity of grave goods suggest women, and not just of the elite, were held in some regard. For the later period, although there is some evidence from the wills of those wealthy enough to have something to bequeath, and from illuminated manuscripts, more tangible evidence of dress is harder to locate. As in Dode and Swift the interpretation and re-interpretation of grave goods with a focus on the clothed individual raises interesting questions about the symbolic nature of dress.

demonstrates the wealth of information that can be extracted even from fragments. Bibliography Arnold, J. (1972) 2nd ed.: Patterns of Fashion, Englishwomen’s Dresses and their Construction c. 16601860. London. Barnes, R. & Eicher, J.B. (eds) (1992): Dress and Gender, Making and Meaning. Oxford & New York. Bender Jørgensen, L. (2007): ‘The world according to textiles’, in C. Gillis & M-L Nosch (eds) Ancient Textiles: Production, Craft and Society, 7-12. Oxford. Carter, M. (2003): Fashion Classics from Carlyle to Barthes. Oxford & New York. Croom, A. (2000): Roman Clothing and Fashion. Stroud. Cumming, V. (2004): Understanding Fashion Theory. London. Davis, F. (1992): Fashion, Culture and Identity. Chicago. Edmondson, J. & Keith, A. (eds) (2008): Roman Dress and the Fabrics of Roman Culture. Toronto. Eicher, J. B. (ed) (2010): Berg Enclyclopedia of World Dress and Fashion. Oxford & New York. Eicher, J.B. & Roach-Higgins, M. (1992): ‘Definition and classification of dress’, in Barnes, R. & Eicher, J.B. (eds) Dress and Gender, Making and Meaning, 8-28. Oxford & New York. Gillis, C & Nosch, M-L (eds.) (2007): Ancient Textiles: Production, Craft and Society. Oxford. Hendszel, I., Istvánovits, E., Kulcsár, V., Ligeti, D., Óvári, A. & Pásztókai-Szeőke, J. (2008): ‘On the borders of east and west”: A reconstruction of Roman provincial and Barbarian dress in the Hungarian National Museum’, in M. Gleba, C. Munkholt & M.L. Nosch (eds) Dressing the Past, 29-42. Oxford. Kuper, H. (1973): ‘Costume and identity’, Comparative Studies in Society and History 15.3: 348-67. Llewellyn-Jones, L. (ed) (2003): Women’s Dress in the Ancient Greek World. Swansea. Nørgaard, A. (2008): ‘A weaver’s voice: making reconstructions of Danish Iron Age textiles’, in M. Gleba, C. Munkholt & M.L. Nosch (eds) Dressing the Past, 43-58. Oxford. Roccos, L. (2005): Ancient Greek Costume: an Annotated Bibliography 1784-2005. Jefferson, N. Carolina. Sebesta, J. & Bonfante, L. (eds) (1994): The World of Roman Costume. Madison, WI. Stewart, S. (2007): Roman Cosmetics and Perfumes. Stroud. Styles, J. (1998): ‘Dress in history: a contested terrain’, Fashion Theory 2.4: 383-90. Taylor, L. (2002): The Study of Dress History. Manchester. Taylor, L. (2004): Establishing Dress History. Manchester.

The symbolic nature of dress and its role in empowering the wearer is the central theme of Penelope Dransart’s paper. She focuses primarily on late medieval Scotland and the role and self-representation of clergy during the celebration of the Eucharist: their relationship both to the physical space of the liturgy and the lay congregation is analysed through the vestments worn by the celebrant. Dransart presents a short history of vestments and their construction which gives a context for a discussion on the social meaning which endowed spiritual power and religious authority on the wearer. Vestments were the outward sign of priestly authority, and this authority was unambiguously male. The final chapters in the volume are both catalogues. Susan Stewart presents an extensive list of Roman cosmetics and toilet items drawn almost exclusively from literary sources. This very useful glossary will provide a reference point for further study of cosmetics (see Stewart 2007) but in itself makes the point that beauty products as we might term them today were very much part of Roman life (for men and women). Personal appearance, as demonstrated also by Harlow and Swift in this volume, was very much part of the self-presentation of Romans who could afford it. Anthea Harris has catalogued the late antique textiles from the Ure Museum at the University of Reading. The chapter also examines how such material came to be in the University collection and asks questions about the nature of such assemblages and the problems lack of provenance (in some cases) present for the researcher. The collection is essentially a set of textile fragments and Harris discusses the nature of and attitudes towards collecting in late 19th century compared to the preoccupations of 21st century museum curators and conservators. The catalogue

5

Chapter 2

Costume as Text Zvezdana Dode1

When analysing a document, the historian must consider its context: ‘Text is always created by someone with the purpose of encoding events. The historian’s task is primarily that of a decoder. For him, a fact is not the starting point, but the result of an arduous effort. He produces facts of his own, striving to derive textual reality from within the text, and events from the narrative’ (Lotman 2000: 336). According to traditional historical perception, texts can be seen as the history of events, with cultural phenomena remaining in the background. Alternatively, if the historian picks a particular cultural occurrence as the focus for research, then the historical process frequently appears as a mere sequence of successive cultural forms. Meanwhile, it is quite obvious that certain cultural forms have an internal structure which may be considered as historical phenomena in any history of events. It is argued here that costume is such a phenomenon, possessing a poly-semantic information structure that goes far beyond the limits of the cultural historical context.

Before beginning to examine information derived from costume, its complex system of presentation has to be defined.2 Costume is a system including clothes, headgear, footgear, and accessories. Obviously, not all components are present in any given system. The absence of some components of the attire will make any costume conditional, leaving information voids in its text. On the other hand, the absence of information about the basic costume element – the clothes – renders reconstruction of the costume assembly totally impossible. So, in examining costume components in isolation, outside the whole complex, the text is effectively fragmentary. This is a typical archaeological situation where the basis of costume – the clothes – is not preserved and therefore its total shape is impossible to restore, even by analogy with other sources. In such cases when only components of the whole attire can be discussed, it is incorrect to apply the term costume to individual artefacts, as some archaeologists do (see Krylasova 2001; Mastykova 2002; Hairedinova 2000, 2002). Such authors’ attempts to represent assumed patterns of clothing, based only on the location of separate items of attire in burials, resemble written evidence where the reader fills in missing text fragments with his own thoughts. Consequently, costume as a source requires a corresponding critique to establish its authenticity and integrity, identify its missing or foreign components, and weed out inadequate interpolations and over-theorising.

The methodology of reading costume as text is logically derived from the etymology of the word text, traced back to the Latin textum – weaving, clothes, link, joining; textus – interlacement, structure, coherent statement; texo – to weave, plait, combine (Rudnev 1997: 305). Consequently, the semantic components that define notions of costume and text coincide, both being man-made, not natural, wherein the elements are bound together, and are built skilfully, expertly and with artistry (Rudnev 1997: 305).

Essentially, reading costume as text requires an attempt to establish functions and meanings which are not identifiable through mere artefact analysis. It is possible to determine the semantic space formed by a costume assemblage if other

While considering costume as text, it should be borne in mind that it is the result of purposeful actions by an agent who consciously seeks to endow an object with a certain quota of informative content, such as using costume to express socio-economic and/or sex and age-specific differentiation. It should also be remembered that costume as text reflects occurrences that the agent never intended, but which the researcher finds significant in modelling a historical and cultural reality.

Costume comprises not only clothes, head and footgear, but a whole set of objects that transform attire into costume proper. These include trimmings and appliqué, embroidery etc., belts, decorations, amulets, accessories (make-up, handbags, purses, etc.), weapons, cosmetics, hair styles. Accessories are usually understood as articles composing a set. Here, accessories are understood as items that supplement the costume, while their own semantic significance within the complex context is minimal, and their presence or absence does not violate the structure of the costume assembly. To describe the entire set of items that transform clothes into a descriptive semantic system with the generic term attire disregards its somewhat conditional character 2

All reconstructions are by Zvezdana Dode and drawings are by O. Lagodina. 1

7

Dress and Identity

sources – written, pictorial and folkloric – complement and enhance the meanings concealed in artefacts. In costume’s text space, four levels of information can be detected, whose nature is common to different costume assemblages. These levels are not discreet and one information level is frequently the key to understanding another within the text. Sensual-rational analysis Costume analysis at the empirical level gives a notion of the object: its utilitarian function, the form of the costume, its component parts, aesthetic or utilitarian preferences, quality of materials used etc. At this level, certain knowledge is derived about the activities necessary to produce costume forms, such as occupational data, landscape and climatic conditions determining use of certain materials, cut and colour preferences, and so on. This level connects with information about the natural environment and economic organisation within society. For instance, examination of headgear found by Leonid N. Glushkov in an Alan burial (7th – 9th centuries) in Lower Arkhyz helps to draw certain conclusions about the mastery not only of ancient seamstresses, but also of tanners and metal craftsmen.3 An appliqué covering 15 cm down from the crown is of leather cured by local furriers. The leather surface along the perforation is smooth and undamaged, proof that it was soft, sturdy and flexible at the moment of sewing. These typical features, as well as histological details of its structure, affirm that the appliqué was made of sheepskin or goatskin. Particles of yellow metal were found on its surface. Their spectrum analysis demonstrated that the coating was an alloy of gold, silver and copper. The headgear (Fig. 2.1)4 was trimmed with weasel fur (Siberian weasel or marten – mustelidae). Hence there is an assumption that the local population practiced both cattle-farming and game hunting.

Fig. 2.1: The Alan female headgear (7th - 10th c.).

important historical information. For example, the origin of textile fibres used by the population of the Northern Caucasus to manufacture fabrics is related to important issues of economic and cultural development in the North Caucasus area during the early Middle Ages.

The principal fabric in the headgear, Chinese taffeta, can be examined in the context of material exchange contacts. Its emergence in the Alan burial was the result of the Great Silk Road route shifting to the passes of the Northern Caucasus in the 6th century. Silk threads used to attach the item’s sections and decorations suggest that not just fabrics, but also threads and, possibly floss, were among imported goods. Moreover, leather pieces were sewn together with threads of cotton fibre, no doubt also imported. The shape of the headgear and its decoration will be further considered at the semiotic representation level (Dode 2001: 20-22).

Examination of 213 textile samples from Alan cemeteries of the North Caucasus of the 6th – 9th centuries, determined that 67.1 % were of flax bast fibres, while 13.1% were of hemp, and 14.5% were a mixture of hemp and flax, with a small percentage of cotton fabrics, cotton-flax, and flaxwool mixtures (Dode 2001: 36-51). The conclusion is that in the early Middle Ages (c.7th to 10th centuries) the burials containing linen fabrics were within the Kolchis flax distribution area. The zone was humid enough to cultivate and process the crop. Related settlements were located in river valleys – a good spot to ret and lay flax stalks, a process requiring almost no other special structures.

The technological examination of woven material provides The gendered assumption about the medieval seamstresses, is based on the fact that sewing-related artifacts (needles, scissors, loom fragments, and unfinished needlework in caskets) are only found in female graves of the medieval Alan burials; an important factor in research on the gender roles in the medieval Alan society. 4 The Alan female headgear (7th - 10th c.) of silk cloth with the leather golden appliqué and fur trimming from Podorvannaia Balka ground burial in Nizhny (Lower) Arkhiz. Authentic reconstruction (L. N. Glushkov’s find, Stavropol Museum. Stavropol. Inventory # 26640/211; published by Z. Dode, 1991; 2001, p.20-22, fig. 18, p. 24, illustration 14). 3

The empirical data that the inhabitants of the Caucasus plains cultivated flax, not just for their own consumption but also for trade, is confirmed by medieval written evidence. According to the Arab historian Masudi (10th century), Circassians grew flax and used its stalks to

8

Zvezdana Dode: Costume as Text

Semiotic analysis

produce high quality cloth: ‘In their country, various fabrics are manufactured of flax of the kind known as tola-gold, which is finer and sturdier than the dabiki kind. One length of it costs ten dinars, and is exported to the neighbouring Moslem countries. Similar fabrics are exported by adjacent peoples too, but the famous kind is that exported by those [Kashaks]’ (Minorski 1963: 207).

The semiotic information contained in the costume assemblage is meaningful and richly represented. This semiotic level reveals ideas regarding the significance and meaning of an object’s features. It is considered alongside the ethno-social community structure, its religious and mystical concepts, political organisation, artistic and aesthetic attitudes, etc. At this level costume can be read as a text holding information on numerous cultural realities. In current literary theory semiotics are subdivided into three sections, so costume as a cultural form can be viewed within the context of syntax, semantics, and pragmatics.

There are probably two reasons which explain why flax thread was preferred to hemp fibre for weaving. Firstly, the flax yield from one hectare would make 2.3 times more caftans than hemp harvested from a similar area. Given the shortage of sown areas in the Caucasus plains, the potential yield of fibrous plants was of no small account. Secondly, hemp harvesting was more labour-intensive, as its bast fibres partially lose flexibility and do not have the softness or elasticity of flax. So production of high quality hemp products was less cost-effective than for linens of comparable quality.

In syntax a set of features can be identified which express the owner’s status in the cultural context and the interrelation of status and dress. So, the early medieval costume of the peoples of Northern Caucasus was uniformlike (Figs 2.2 and 2.3).5 For example, a warrior’s status in the social hierarchy was marked by gilded tops on helmet-like headgear, the fitting of his belt set, the quality of material used, and the amount of silk in the costume; the shape of female hats was a sign of changing social and age status: girl to maid to woman. Evidently, during the historical evolution of costume from antiquity through to the present, the combination of its components and the spheres they marked were subject to change. For a later period ethnographic study suggests that in North Caucasus costume the corset also marked social and age status for females: the girl started using it at the onset of puberty, and wore it until her wedding.

Theoretically, it is even possible to establish the areas of fibre crop cultivation. To do so a comparison is needed between textile and plant samples originating from certain Caucasus plains, from plants stored in herbariums. The plant’s vegetation region, and the time taken to come to ripeness are indicated by its fibrillar structure, and differences exist between the fibrillar patterns of flax and hemp (Berestenev, Fleser, Lukyanova 1982: 163). This information also helps to establish the proportion of dogbane fibres of both local and imported origin in the later Middle Ages. Unfortunately, plant collections required for such analysis are currently inaccessible, thus such research is difficult.

Semantics disclose the significance and meaning of the costume’s features (syntactic elements), and define these as feature signs and symbol signs. Feature signs are indications of the things they denote. For example, the amount of silk used in the Alan costume indicated social divisions, and consequently the wearer’s social status: chieftains donned silk caftans, commoners used silk to decorate their canvas clothes, paupers wore patched tatters of sackcloth. The

Historically, the manufacture of yarn and cloth was always done by the farmers themselves. Artefacts connected with looms and spinning are frequently found in women’s burials. This suggests spinning and weaving were household activities. Women were evidently also in charge of the fibre-producing crops. This offers an interesting angle on the fertility cult, which probably went beyond ideas about female reproductive ability, and was also associated with certain female activities. This was expressed in specific amulets of hare and fox bone attached to the hems of the skirts of Alan women, and in hazelnuts and walnuts found in needlework kits and the dress pockets of buried women.

Fig 2.2: The authentic reconstruction of the costume of a representative of Alan nobility (7th- 10th c.). A caftan of silk Iranian cloth with senmurvas (E. A. Milovanov’s find; published by Ierusalimskaja, 1992: cat.1, p. 4, fig. 1, p.35), a helmet-like headgear with a pomme of Sogd silk with “jugs” (N. I. Vorobiov’s excavations, published by Ierusalimskaia, 1992, cat 29, p. 19, fig. 5, p.36) from the ground burial Moshchevaia Balka; leather high boots from Khasaut ground burial (A.P. Runich’s find, published by Runich, 1971: 174); the analogy is on the image of a hunter with a falcon performed on silver Hungarian dish of the 10th c. (see Artamonov 1962: 261). Fig. 2.3: The apodictic reconstruction of an Alan’s female and girl’s costumes (7th-10th century). Woman’s linen dress with a breast pocket. The reconstruction is made by analogy with the forms of female dress from burial Moshchevaya Balka (see Ierusalimskaia, Borkopp 1996: Cat. 5, p.25). The reconstruction for the decoration of the breast pocket used Byzantine silk. From a burial at Hasaut (8th century, published by Ierusalimskay 1992: 21, cat. 41). The woman-mother’s headgear consists of a lower diadem, a cap with a braid-piece, a scarf (or a shawl) fastened by an upper diadem (a ribbon) at the top (based on Jerusalimskay 1992: 44, fig. 8). Reconstruction of the child’s dress is made in accordance with older models. The decoration of dress used embroidery, M. Kovalevsky‘s find in the Hasaut cemetery (published by Dode 2001: 32, fig. 26; Ierusalimskaia, Borkopp 1996: Tafel LXXXV, abb. 221). Virgin headdress consists of ribbons adorned with symbols (published by Ierusalimskaia, Borkopp 1996: 47, Cat. 35-38). 5

It was probably the increasing aridity in the Northern Caucasus from the 14th century that led to a gradual reduction of flax consumption. This trend may perhaps be traced up to 19th century by examining textile samples from later crypt burials in the Caucasus. The reconstruction of the dynamics of fibre crop cultivation is significant for the study of changes in the economy of the Northern Caucasus and the formation of new economic structures in this territory. In this context, it is particularly important to establish the role of climatic change.

9

Dress and Identity

Golden tops of helmet-like hats doubtlessly represented the sun, as the Nart epic clearly states: Sosruko is our light, His shirt is chain mail, His hat top is the sun. He set his hat straight, and he is mounting [his horse] (Narty 1974: 215). Identical costume components can act both as feature signs and as symbol signs in different contexts. The belt is one of the most semantically significant costume components. Among its numerous meanings, first and foremost is the sign that the man is armed, i.e. vested with power. The Nart epic tradition specifically emphasises that the belt was only worn with weapons. Lamenting his poverty, Chuara Nelchibievich tells his mother: ‘I have no horse to ride, nor fancy dress to wear, nor weapons to belt...’. But when his mother gave him his father’s complete outfit, he saddled his father’s horse, effortlessly picked up and donned his armour, jumped his horse over three hurdles, and set out to roam the wide world (Dalgat 1972: 326). Possessed of the might of his weapon, a belted man was clearly a danger to those around him. Belt fittings have been found in virtually every early medieval Alan burial. In certain cases a buckle is missing from the belt, although the rest of the belt set is complete. Researchers link this to the ideology of medieval people, who removed this item to render the deceased harmless: weapons attached to the unbuckled belt could not be used against the living (Pletneva 1967: 161).

Fig. 2.2: The authentic reconstruction of the costume of a representative of Alan nobility (7th - 10th c.).

Yet not weapons alone, but also power and righteousness give strength. When the notions belt and girdle are used combined with the concept of power or righteousness, the belt becomes a symbol sign of might or piety. So, the Byzantine princess Anna, daughter of Alexius I Comnenus, relates Monomach’s appointment to the post of Ducas in the following terms, literally: ‘Having received written instructions about the Ducas belt, ... the very next day he left the royal city for Epidamnos, and the province of Illyricum’ (Alexiad 1.xvi). The commentary on this text explains that ‘a special belt was the insignia of title-bearers of novelissimus, curopalatus, and magister (Anna Comnena 1996: 462: translation and commentary Ya. N. Lubarsky). According to Anna’s evidence, the handover of the belt symbolised and accompanied the act of appointment to the post of Ducas. The apostle Paul says, ‘Wherefore take unto you the whole armour of God, that ye may be able to withstand in the evil day, and having done all, to stand. Stand, therefore, having your loins girt about with truth, and having on the breastplate of righteousness; and your feet shod with the preparation of the gospel of peace’ (Ephesians 6: 13-15). In this context, the idea ‘girt’ is used as a symbol of truth and piety.

quality of cured leather makes a similar point: wealthy citizens wore morocco shoes, while commoners used goncharyk of rough pigskin. Feature signs of poverty in the Nart epics (Caucasian oral tradition) are expressed in an offensive and abusive manner. Thus Satana, displeased with Sosruko, labels him ‘crooked-nosed slipper born of a goatherd’. The phrase ‘crooked-nosed slipper’ clearly alludes to an episode in the epic where two different shoes engage in combat, and a morocco slipper challenges a raw leather shoe: ‘You loathsome piece of shit, indeed I was made by a young maid, but you are all bristle and trash!’(Narty 1957: 152). Symbol signs contain a pictorial image of their inherent meaning. When a highland maid tells her swain, ‘You short-eared wolf, you are wolf’s kin’, she implies the hero’s qualities that symbolise wolfish behaviour: The lion and eagle rely on their strength: they challenge less than their match; the wolf challenges more than his match; he complements his lack of strength with his courage, boldness, and cunning; in the dead of night does he circle the herd and the village from where death threatens him any instance... Trapped and desperate, the wolf dies silently, showing neither fear nor pain (Kavkazskie gorci 1992: 152).

Colour, especially red, is a symbol sign. Red is for blood, symbolising life, energy, strength and power. Therefore red is preferred as the symbol of high social status. In burials of the Sarmatian period (1st-4th centuries) researchers find

10

Zvezdana Dode: Costume as Text

The notional meaning of a sign not only defines identities such as sex and age, property and social differentiation, religious, ethnic, regional affiliation, etc., it also presumes the viewer’s response to the signal. A classic dialogue in the relational context of subjects who use and perceive costume as a signal system underlies the story of the Prince and the Pauper by Mark Twain. Their trading of dresses was the starting point for the adventures of Tom Kenty and Edward, Prince of Wales, who, in beggar’s disguise, was not allowed to re-enter the palace by a guard. It is known that Mary Stuart shocked the public when she appeared in white for her wedding with Henry Stuart, Lord Darnley. The guests’ amazement can only be appreciated when it becomes known that up to that moment, every bride in Europe had been married in red, while white was the colour of widows’ mourning. ‘... as she appeared in her mourning attire, the same in which she attended the funeral of her deceased spouse, the King of France, it was as if she emphasised that her second walk to the altar was not a light-headed act by a woman who had forgotten her first husband, but simply a means by which to comply with the will of her nation’ (Zweig 1994: 100). It remains an open question whether Mary’s contemporaries found such grief sincere, for the conduct of the characters upon this stage was far too histrionic. Yet for those who know the tragic outcome of the Queen of Scots’ marriage with Darnley, the colour of her wedding dress takes on a truly symbolic significance.

Fig 2.3: The apodictic reconstruction of Alan female and girl's costumes (7th-10th century).

Perception of costume as a signal system produced the saying ‘clothes maketh the man’. While responses to the context can vary from negative, as above, to respect and awe, there will always be some feedback to the signal, through corresponding behaviour or action from those on the receiving end. Otherwise the sign function of the costume is nullified, particularly in stratified societies. Essentially, costume’s communicative function lies in semiotics. In his Reminiscences of Muta’allim, Abdulla Omar-Ogli, translator and enlightener of the Daghestan Highlanders, and author of ethnographic descriptions of mountain tribes in the second half of 19th century, writes: ‘After the noon prayer, a rider came from a neighbouring village, bearing news of the Khan’s departure... Lo, the white sign has come in sight, and the villagers are all filed up. And when the Khan approached, everyone took off their hats for the traditional welcome: Karabuz han! (Let Allah bless your visit)’ (Kavkazskie gorci 1992: 30). The white flag was the signal to organize the crowd. Uncovering the head before a representative of the superior class was a common custom at different times among different nations, but invariably triggered by the costume of a relevant person.

clothes, headgear and belts of red leather, along with other items that emphasise the high social status of the people interred (Prohorova, Guguev 1992: 142-160; Simonenko 1992: 149). A helmet-shaped item of headgear of red silk with gilt leather appliqué is doubtless a symbol of the respectable social position of its owner – a woman who resided near the upper reaches of the Greater Zelenchuk river in the 7th to 9th century (Dode 2001: 20-22). Red leather shoes decorated with silk, found in the Moschevaya Balka burial (Ierusalimskaia 1996: 204), and red sepulchral coverlets uncovered in wealthy burials of the Sarmatian and Alan periods probably had the same meaning (Prohorova, Guguev 1992; Markovin 1983). This archaeological series, which could certainly be expanded, is confirmed in written sources. The Armenian historiographer Movses Khorenatsi notes: ‘As Alans set great store by red leather, so Artashes is sending down a lot of red leather and gold as dowry’ (Armianskie istochniki 1985: 32). Social insignia of red colour are distinctly traceable in ethnographic materials, particularly the Kabarda costume. Red morocco slippers, maidens’ red caps and corsets were used as costume articles particularly by feudal nobility. In male costume colour as a social sign is mainly expressed in shoes. Thus K. Kokh writes, ‘Red shoes are those of princes, nobles wear yellow, and common Circassians wear plain leather’ (Studenetskaia 1989: 33).

Anna Comnena describes an historical event that reflects the use and perception of red as a symbol of royal power. The detail in her narrative requires no additional comment: When Michael the Ducas was dethroned, his son by Empress Maria, Constantine Porphyrogenitos,

11

Dress and Identity

took off his purple sandals of his own accord, and donned common black ones. However, Nicephorus Botaniates, who had succeeded to the Ducas sceptre from Constantine’s father, ordered Constantine to replace his black sandals with shoes of various silks; it was as though Nicephorus was embarrassed by the youth, and recognised his beauty and noble birth. He did not wish Constantine’s sandals to display pure purple, but accepted them with purple flashes. When Alexius Comnenus was enthroned, Constantine’s mother, Empress Maria, at Caesar’s persuasion asked the autocrat to issue a document, signed in red and bearing the gold seal, promising that she and her son would be safe, and that, moreover, Constantine would co-reign with Alexius, be shod in red, bear the crown, and be proclaimed Emperor together with Alexius. Maria’s request was granted, and she received the chrysobullon (the Pope’s Golden Bull) to satisfy all her claims. And then and there they took off the silk sandals he had been wearing and gave him red ones (Alexiad 3.iv). In the same pragmatic context, cases can be examined where clothes were used as a symbol of humiliation, and thus punishment. When the Persians met their defeat at the hands of Byzantium in the Battle of Caucasus (590) on the Albanian plains, the Persian King Harmizd sent women’s clothes to his war leader Varam as a sign of contempt and dishonour, thus insulting him outrageously (Feofilact Simokatta 1957: 80 book 3. VIII. 1).

Fig. 2.4: The apodictic reconstruction of Alan men’s and children's clothing (7th – 10th century).

An episode in the Nart epics narrates how Nasran-aldar castigated his childless wives:

thinking, and, therefore, of human culture.6 Examination of costume for an extensive chronological period allows the nature of these dynamics to be traced and correlated with turning points in the historical and cultural process.

It could not be helped, the people forced the Nasranaldar to accept it. Home he goes. And he cut short his wives’ dress tails and hems, and he snipped off their hair plaits, and he took off their underskirts, and he brought them among the crowds... The people laughed at their prince’s wives, stoned them, muddied them, spat at them, for they had given no issue to Nasran-aldar, and left him without progeny ( Narty 1989: 319)

Cultural analysis The level of cultural relations allows an object to be viewed in the context of cultural dynamics from the moment of its formation up to its absorbance into popular culture through diachronic and synchronic interaction. The costume assemblage evolves in tandem with the formation and development of culture, a process in time and space. So, for instance, the change of old forms and the emergence of new forms of costume assemblage in the North Caucasus occurred in the course of interaction of the local Caucasus people with numerous Iranian and Turkic speaking tribes. On the one hand medieval costume of the Caucasus reflects the essence of cultural evolution as a process of continuous self-rejuvenation. On the other hand, the costume assemblage has absorbed as its components the traditions and innovations of the cultures of different north Caucasian peoples interacting at synchronic and diachronic levels.

Here, one of most severe disgraces described for a woman of the mountain tribes was never erased from memory, and was passed down through generations. In the same vein is the Russian tradition of the 15th - 16th centuries of punishing boyars by a ride around town dressed in a caftan worn back-to-front, with the then popular high collar before their faces. Here lies the origin of the Russian idiom «шиворот-навыворот» (‘his collar the wrong way round’). It is known that the ability to operate symbol signs rather than objects is an indicator of the development of abstract

Costume demonstrates most clearly that cultural evolution That is why the realism of the Paleolithic art was replaced with symbolic images in the Mesolithic era. 6

12

Zvezdana Dode: Costume as Text

is a complicated and lengthy process. At the diachronic level mutually interacting ethnic traditions are passed down. It is often hard to single out and date the time periods accurately, as traditions are preserved, developed and handed down through generations from time immemorial. For instance, the motif of Soslan’s overcoat sewn of human beards, moustaches and scalps, as represented in the Nart epics, is rooted in ancient Iranian culture, experienced by Scythians, Sarmatians and Alans in the Northern Caucasus.

change. What is now available suggests immediate contact with Polovets steppe tribes, and a resultant interference in various socio-cultural systems that was a powerful innovative drive to change the cut of traditional dress for Alan women. This emphasised further the appearance of women’s outer clothing, as the cut had previously been all but identical with men’s clothes. In the historical context cultural synthesis is key to interaction between tradition and innovation and expressed as a form of costume evolution, while at the same time such costume illustrates the shared origins and destinies of the North Caucasus peoples. Cultural dynamics and inter-cultural interaction in the North Caucasus resulted in essentially the same costume assemblage, which included various ethnic layers, and which is claimed for their ‘own’ by representatives of different ethnic groups inhabiting a large territory. It follows from this that for the culture of the North Caucasus tribes, costume became an ethno-regional symbol and an indicator of the cultural commonality of the North Caucasus region. Meanwhile, there is every reason to believe that in previous eras, such as the Iron Age, costume was a sign of variation within a culture that marked the western and eastern communities of the Koban. Hence there is a need to establish indigenous layers of the costume assemblage, and trace the dynamics of their interaction with later alien cultural components for a longer period from ancient to modern times.

At the synchronic level of cultural development transformation of poly-ethnic innovation occurs. The complex of poly-ethnic interaction in the North Caucasus in the Middle Ages comprises a number of key components: local (indigenous cultures, the Alans), Muslim (Near and Middle East, Central Asia), central Asian steppes (the Avars, Sogdians, Polovets, Mongolians), and Byzantine, with occasional Chinese and Italian influences. The male costume assemblage of the 7th century is the result of a synthesis of Central Asian (ancient Turkic), Iranian and local components (see Fig. 2.4).7 Before it appeared as a finally established structure (18th to first half of 19th century), North Caucasus costume had been influenced by the Polovets, Mongols and Tartars, Seljuk Turks, Ottoman Turks and Persians. Meanwhile, the basic cut of the dress often remained unchanged and passed from generation to generation. Nomads repeatedly swept through the territory of the Northern Caucasus. Contact between the indigenous population and the folk from the steppes brought changes in the basic form of female dress. The fact that age-old dress patterns – the cut, structure and silhouette – were rejected overnight cannot be explained away as a mere fad. Evidently, more complex causes underlie such a change. They can be connected both to the shift of women’s position in society, and the change in female activities, since folk costume invariably responds to daily needs. Regrettably, the data available to date hold no clues to explain this

The synchronous level of cultural evolution also has a territorial aspect. The central part of the Northern Caucasus which comprises modern Chechnya, Ingushetia, Osetia, Kabarda-Balkar and Karachay-Cherkessia, borders upon Kalmykia to the north and the peoples of Daghestan to the east. To the south, beyond the Greater Caucasus ridge, lie Georgia, Armenia and Azerbaijan. Never in all its historical epochs was the North Caucasus isolated but, to the contrary, it maintained constant contact with surrounding states and peoples, which naturally influenced the evolving culture within the region. Accordingly, it is likely that the territories of the western and central parts of the Northern Caucasus should be viewed as possessing a shared cultural and historical landscape, due to their shared territory, climate and environment which determined the similarities in farming and economic development in the region, as well as the shared basic underlying culture and the innovative processes affecting it.

Fig. 2.4: The apodictic reconstruction of Alan men’s and children’s clothing (7th – 10th century). Men’s caftan of linen textile tabby weave, decorated around the neck and sleeve edge with Sogdian silk cloth (National History Museum of Karachay-Cherkessia. Cherkessk museum depository, inventory # 9644/ 2; published by Dode 2001: 14. Fig. 3). Head-dress of linen cloth helmet form (National History Museum of Karachay-Cherkessia. Cherkessk museum depository, inventory # 9107/98). The trousers of homespun fabric, made from three squares (National History Museum of Karachay-Cherkessia. Cherkessk museum depository, inventory # 9107/310; published by Dode 2001: 10, Fig. 17). Linen stockings on a leather strap (National History Museum of Karachay-Cherkessia. Cherkessk museum depository, inventory # 9107/390, 9107/391; published by Dode 2001: 18, Fig. 11). Leather belt with metal suite (find by S.A. Varchenko in Nizhny Arkhyz. Not published). Child headpiece of silk fabric with bands of dark and light fur (National History Museum of Karachay-Cherkessia.Cherkessk museum depository, inventory # 3969/90; published by Dode 2001: 33, Fig. 27). Shirt – jackets made of silk fabric (find by V.N. Kaminsky, I.V. Kaminska in the Lower Arkhyz. National History Museum of Karachay-Cherkessia. Cherkessk museum depository, inventory # 9107/107; published by Dode 2001: 29. Fig. 23). Lower silk shirt (find by V.N. Kaminsky, I.V. Kaminska in the Lower Arkhyz; National History Museum of Karachay-Cherkessia. Cherkessk museum depository, inventory # 9107/109; published by Dode, 2001: 29, Fig. 23). Reconstruction of the cingulate clothing made by analogy with the male specimens. Children’s wooden dagger (find S.A. Varchenko in Nizhny Arkhyz. Not published). 7

As with geographic zones, so historic cultural zones suggest that areas of transition should exist between interacting regions. Given the North Caucasus on the one hand and, for example, Central Asia on the other, as cultural and historical landscapes in contact, then Daghestan and Chechnya will appear as the transition zone for their cultural interaction. Here in a single assembly, details of North Caucasus and Central Asian costumes can be found. The narrow-waist cut, the distinctly ‘triangular’ silhouette, the restricted colour spectrum and the restrained well-designed and scanty decoration are all distinctive elements of North Caucasus

13

Dress and Identity

women’s attire. Central Asian costume is typified by its T-shaped cut, the ‘rectangular’ silhouette, loud motley colours, and an abundance of jingling decorations. Chechen assemblages often include a dress of traditional North Caucasus cut, with a breast piece and a belt. Yet there is one difference: they are made of parti-coloured cloth, something not typical in the clothes of the tribes in the western North Caucasus where the expression ‘a loud dress’ used to mean ‘tasteless’. In this case Chechen and Daghestani women share a predilection in their choice of printed cloth. It points to eastern trends in colour preferences in Chechnya and Daghestan. Their everyday costume was dominated by a combination of T-cut shirt-gown, pantaloons, and a headscarf. The T-cut, as the most ancient form among all North Caucasus tribes, was preserved in the undershirt used as underwear and requiring over-clothes. For their daily dress, Chechen and Daghestani women normally wore a shirt-gown and pantaloons (without any over-clothes). A preference for scarves, kerchiefs and shawls rather than caps brings Chechen and Daghestani dress closer to those of their eastern neighbours. The predominance of North Caucasus features in Chechen assemblages, and Central Asian features in Daghestani assemblages can be explained by the ripples of cultural influence growing weaker from the centre outwards.

socio-political bonds and ideological exchange. Analysis of the historical levels will help to identify consistent links between historical events and cultural phenomena. However, typological phenomena have to be identified, and the conditions that allow this cause to have its effect. Individual events that result from typological series are also of research interest, if such events bring new components to cultural combinations. Contact and acquisition of material culture resulted from diverse kinds of cultural historical interplay, from peaceful trade relations and an autocrat’s remuneration for barbarian nobility, to military conquest. As an illustration, silks and silk clothes entered the Caucasus as trade imports, as remuneration, and as means of bribery: various sources trace this back from to the earlier Middle Ages. Byzantine and Sogdian garments triggered local imitation of the originals (Ierusalimskaya 1992: 8) similar to the effect produced by Iranian and Turkish dress in later times. Contact invariably generates semantic and symbolic conjunctions. As they appeared together with commodities or spoils of war, ornamental motifs, images of fantastic creatures, attractive or mysterious fruits could not remain unnoticed. However, one interpretation in the literature insists that the wealth of new artistic textile images had little influence on local arts, and that the textiles were cut and divided with no attention to the pattern design (Ierusalimskaya 1992: 8). There is no knowledge of the principle that regulated division of tributes paid in silk. Silk was probably first divided into a required number of pieces which made it impossible to preserve complete patterns intact. Therefore, silk decorations on canvas clothes appeared as a patchwork of fragments of many colours and ornaments. For example, to produce a caftan of Byzantine silk with a pheasant pattern, the seamstress would select a fragment of the same cloth to fashion the bird’s legs, which were missing on the main piece of cloth. But apparently she might only have a mirror image of a fragment with the ornamental motif available (Ierusalimskaya 1992: 8). Obviously, the chaotic mix of silks in Alan clothes was not so much the result of indifference to textile pattern and palette, as of the quantity and quality of fragments available, that is, the rules of distribution of silks among the community members. This issue has not yet received the coverage it deserves in the historical literature.

Examination of costume within the territories of cultural evolution helps to resolve the following issues: the establishment of vectors of cultural interaction and the causes of their shifts in direction; diagnosis of the mechanisms of territorial and cultural contact between highlanders, farmers of the plains and nomads; and ascertaining the extent of influence from key civilisation centres (Classical, Christian, Muslim, nomadic) on the evolution of the culture of the North Caucasus. Historical context Historical information means that all information from a text can only be understood and correctly appreciated if considered against its specific historical environment and connections. Most obviously, in certain cases it is historical events which produce an impulse for cultural change. For example, as the result of the Iran-Byzantium wars (502629), Iran made it impossible for Byzantium to use any known eastern routes for importing raw silk from the Orient. The Byzantine embassy then found a new caravan route over the Caucasus ridge passes in order to bypass Iran. As a result, the local Alan tribes gained access to silk, which they used to make clothes. The implications of this event are of great significance for the historian. Alan economic development in the 7th-10th centuries was such that most other items being equal, only the amount of silk (a luxury) used in a costume could reveal the existence of any social stratification.

Meanwhile, there is evidence that the local inhabitants were far from unfamiliar with the images that arrived, not only as textile samples but also in oral tradition, brought by the same merchants who transported silks, jades, exotic fruit and spices through the Alan territories. So for example a set of items forming an Alan woman’s necklace included a peach stone (Dode 2001: 25-6). No peaches grew in the Alan lands in the Middle Ages. Evidently peaches found their way in with silk caravans and merchants who brought exotic goods not only from faraway China, but also from Samarkand, and told the locals legends of the properties of this fruit. In China, where the peach was first cultivated some five thousand years ago, its fruit and the tree itself

Not only can events or phenomena be captured in costume, but also cultural and historical interactions can identify material values, semantic and symbolic conjunctions,

14

Zvezdana Dode: Costume as Text

were said to possess purifying qualities to exorcise evil spirits and disease. The Chinese used peach stones as medicine (Shefer 1981: 243). The inclusion of a peach stone in a necklace with other objects valued for their magic rules out its random choice.

these informative levels helps viewer/reader to understand it in the context of key categories of cultural environments. Bibliography Anna Comnena. 1996. Aleksiada. St-Petersburg: ‘Alteyia’. Комнина, А. Алексиада / А. Комнина; пер. с греч. – СПб.: Алетейя, 1996. – 703 с. Armianskie istochniki ob alanah. 1985. Ed. I. Yerevan. Армянские источники об аланах. / Составитель и автор комментариев Р.А. Габриелян – Ереван: Изд-во АН Арм. СССР., 1985. Artamonov, M.I. (1962): Артамонов, М. И. История хазар / М. И. Артамонов. – Л.: Изд-во Гос. Berestenev V.A., Flekset L.A., Lukyanova L.M. (1982): Makrostnuktura volokon i elementarnyh nitei i osobennosti ih razrusheniia. Moscow. Берестенев, В. А. Макроструктура волокон и элементарных нитей и особенности их разрушения / В. А. Берестенев, Л. А. Флексер, Л. М. Лукьянова. – М.: Лег. и пищ. пром-ть, 1982. – 247 с. Dalgat U.V. (1972): Geroicheskii epos chechencev I ingushei. Moscow. Далгат, У. . Героический эпос чеченцев и ингушей / У. Б. Далгат. – М.: Наука,1972. - 467 с. Dode, Z.V. (1991): Доде, З. В. Аланский женский головной убор из могильника Подорванная Балка в Нижнем Архызе / З. В. Доде // Вопросы археологии и истории Карачаево-Черкесии: сб. научн. тр. – Черкесск: КЧНИИИФЭ, 1991. - С.124-131. Ierusalimskaja, A. A . Von China nach Byzanz / А. А. Ierusalimskaja, В. Borkopp. – Munchen, 1996. - 108 p. Dode Z. V. (2001): Srednevekovyi kostium narodov Severnogo Kavkaza. Moscow: ‘Vostochnaia literatura’. Доде, З. В. Средневековый костюм народов Северного Кавказа: Очерки истории / З. В. Доде.- М.: “Восточная литература”, 2001. – 136 с. Feofilact Simokatta. (1957): Istoriya. Moscow. Феофилакт Симокатта. История: пер. с греч. – М.: Издво Акад. наук СССР, 1957. – 224 с. Hairedinova E. A. (2000): ‘Zhensky kostium s yuzhnokrymskimi orlinogolovymi priazhkami’. Materials on Archaeology, History and Ethnography of Tavria, Volume VII. Simferopol. Хайрединова, Э. А. Женский костюм с южно-крымскими орлиными пряжками / Э. А. Хайрединова // Материалы по археологии, истории и этнографии Таврии. – Симферополь,2000,.– Вып.7. - С.91-133. Hairedinova E. A. (2002): ‘Bosporsky zhensky kostium pervoi poloviny V v. (po materialam nekropolei)’. The Bosporus Phenomenon. Burial Memorials and Sanctuaries. St. Petersburg. Хайрединова, Э.А., Боспорский женский костюм 1 половины V века (по материалам некрополей) / Э. А. Хайрединова // Боспорский феномен. Погребальные памятники и святилища. – СПб.: Изд. Гос. Эрмитажа, 2002. - С.242-245. Ierusahmskaia A.A. (1996): Die Graber der

The image of Senmurv (see Fig. 2.2) can be seen in the same context: a winged dog whose original image entered the Alan world on Persian silks, it remained as an image copied onto a wooden pen case (binat-i-hitsau) which, by Alan superstition, housed the protector of the household and the hearth. Iranian mythology perceived Senmurv as the protector of plants in spite of its ferocious appearance. The combination of the formidable look and kindness corresponded with the principle idea of Iranian Zoroastrianism: the interaction of good and evil. It is hardly an accident to find the image of this fantastic animal, foreign to local ideology, on an object endowed with clear sacred significance. The likeness of the winged dog on a wooden pen case is primitive, but the craftsman probably did not seek to copy the look of the animal he had seen on silk cloth – he tried to convey its core essence, of which he evidently had some understanding. Besides, the existing sophisticated cult of the tree in Alan beliefs prepared the ground to assimilate the image of a creature-protector of plants from the donor culture. The choice of borrowed elements through contacts between two cultures allows us to determine value priorities and particularities of mentality of the borrowing people. Cultural historical interactions were never unilateral. Evidence exists of how components of medieval barbarian costume were further developed in a different cultural environment, and reflected the value scale of that environment. Read in this way, the costume as text generates new meanings and new texts. For example, the Mameluke armed forces in Egypt included troops of North Caucasus tribesmen, and by the 1390s Circassian Mamelukes even established a dynasty of their own. During Mameluke rule in Egypt (1250-1517) the Tartar caftan and the Circassian skull-cap gained popularity. If it was previously considered bad taste to go out without the turban, then after the Mamelukes’ arrival people found nothing wrong with wearing the skull-cap when going to the market or the mosque. In this example the adoption of headgear shape from another culture was connected with unusual circumstances. Sodomy, common among the Mamelukes, spread among the locals. The sultans marked this phenomenon as a novelty introduced into the country by the Tartars who brought with them youths they treated as women. Wearing the Circassian cap that covered the forehead and the neck, these youths resembled maidens. The Egyptian historian Al-Makrizi (1364-1442) noted that, seeing their husbands’ inclination for boys, some wives also began wearing such caps to win back their husband’s love. Besides, such hats were much less expensive than other styles. This is an example of an item of Caucasus costume acting as a cultural phenomenon within a different ethnic environment (Dode 2001: 82-4). Costume read as text at

15

Dress and Identity

Moscevaya Balka Fruhmittelalterliche Funde an der Nordkaukasischen Seidenstrasse. Munchen. Ierusalimskaya A.A. (1992): Kavkaz na shelkovom puti. St-Petersburg: ‘Ermitag’. Иерусалимская, А. А. Кавказ на Шелковом пути / А. А. Иерусалимская. – СПб.: Государственный Эрмитаж, 1992. - 70 с. Kavkazskie gorci. (1992): Sbornik svedenii - Highlanders of Caucasus. Ed. I. Tbilisi 1886, reprinted edition. Moscow: ‘Adir’. Сборник сведений о кавказских горцах, издаваемый с соизволения его Императорского Высочества Главнокомандующего Кавказскою Армиею при Кавказском горском управлении. – М.: МНТПО «Адир», 1992. – Вып. 1. – 387 с. (Репринт. Воспроизведение изд 1868 г., Тифлис). Krylasova N.B. (2001): ‘Zhensky kostium po materialam Kanevskogo mogilnika’. The cultures of the Eurasian Steppes in the First Half of the Ist millennium AD [From the History of the Costume] Vol. 1. Samara. Крыласова, Н. Б., Женский костюм по материалам Каневского могильника / Н. Б.Крыласова // Культура евразийских степей II пол. I тысячелетия н.э. Из истории костюма: В 2 т. – Самара: ПО «Самвен», 2001.– Т.1. - С. 226-241. Lotman Yu, M. (2000): Semiosfera. St-Petersburg: ‘Iskusstvo- SPb’. Лотман, Ю. М.,. Семиосфера / Ю. М. Лотман. – СПб.: Искусство-СПб, 2001. - 703 с. Markovin V.I. (1983): ‘Vizantiiskaia tkan s zolotym shitiem iz Sentinskogo hrama’. Problems of Archeology and Ethnography of Karachai-Circassia. Cherkessk. Марковин, В. И. Византийская ткань с золотым шитьем из Сентинского храма / В. И. Марковинн // Проблемы археологии и этнографии Карачаево-Черкесии. – Черкесск: КНИИИФЭ, 1983. - С.67-76. Mastykova A.V. (2002): ‘Kostium alan verhney Kubani v epohu velikogo pereseleniia narodov (po materialam mogilnika Baital-Chapkan)’. XXII Krupnov Studies in the Archeology of the North Caucasus, Report Summary. Yessentuki – Kislovodsk. Мастыкова, А. В. Костюм алан верхней Кубани в эпоху великого переселения народов (по материалам могильника Байтал-Чапкан) / А. В. Мастыкова // XXII Крупновские чтения по археологии Северного Кавказа: тез. докл. – Ессентуки-Кисловодск, 2002. - С.87-90. Minorski V. F. (1963): Istoriia Shirvana I Derbenta v X-XI stoletiiah. Moscow. Минорский, В. Ф. История Ширвана и Дербента X-XI веков / В. Ф. Минорский. - М.: Изд. Вост. лит., 1963. - 265 с. Narty (1957): Kabardinsky epos. Moscow.

Нарты: Кабардинский эпос. - М.: Гослитиздат, 1957. 527 с. Narty (1974): Adygeiskii geroicheskii epos. V.2. Moscow. Нарты: Адыгский героический эпос. - М.: Наука, 1974. - 415 с. Narty (1989): Osetinskii geroicheskii epos. V.2. Moscow: ‘Vostochnaia Literatura’. Нарты: Осетинский героический эпос. - М.: Наука, 1989. - Т.2. - 496 с. Pletneva S.A. (1967): Ot kochevii k gorodam. Moscow. Плетнева, С. А. От кочевий к городам. Салтово-маяцкая культура. / С. А. Плетнева // МИА. М. «Наука», 1967. – 198 с. Prohorova T.A., Guguev V.K. (1992): ‘Bogatoe sarmatskoe pogrebenie v kurgane 10 kobiakovskogo mogilnika’. Soviet Archaeology Vol. 1. Прохорова, Т. А. Богатое сарматское погребение в кургане 10 Кобяковского могильника / Т. А. Прохорова, В. К. Гугуев // Сов. археология. – 1992. – № 1. - С. 142 - 161. Rudnev V. P. (1997): Slovar kulturi XX veka. Moscow: ‘Agraf’. Руднев, В. П. Словарь культуры XX века / В. П. Руднев. – М.: Аграф, 1997. – 381 с. Runich, (1971): Рунич, А. П. Скальные захоронения в окрестностях Кисловодска / А. П. Рунич // Сов. археология. – 1971. – №2. – С.167-178. Shefer E. (1981): The Golden Peaches of Samarkand (in Russian). Moscow. Шефер, Э. Золотые персики Самарканда. Книга о чужеземных диковинах в империи Тан / Э. Шефер. М.: Восточная литература, 1981. – 607 с. Simonenko A.M. (1992): ‘Farzoi I Inismei — aorsy ili alany? Ancient History Observer Jfs 3. Симоненко, А. В. Фарзой и Инисмей – аорсы или аланы? / А. В. Симоненко // Вестник древней истории. – 1992. – № 3. – С.148-162. Studenetskaia E.N. (1989): Odezhda narodov Severnogo Kavkaza v XVIII -XX vv. Moscow: ‘Nauka’. Студенецкая, Е. Н. Одежда народов Северного Кавказа XVIII – XX вв. / Е. Н. Студенецкая. - М.: Наука, 1989. – 286 с. Vartovski M. (1988): Modeli: reprezentaciia i nauchnoe ponimanie. Moscow: ‘Progress’. Вартофский, М. Модели. Репрезентация и научное понимание / М. Вартофский; пер. с англ. – М.: “Прогресс”, 1988. – 506 с. Zweig Stefan. (1994): Maria Stuart. Stavropol: ‘Fond culturi’. Цвейг Стефан. Мария Стюарт. – Ставрополь: «Фонд культуры».

16

Chapter 3

Veiling the Spartan Woman Lloyd Llewellyn-Jones1

That Spartan women are problematic is undeniable. Although they have been championed by several notable contemporary feminist historians, it is incontestable that Spartan women remain, to most scholars, the proverbial fly in the ointment of ancient Greek gender studies. Of course, the notion of the ‘Spartan mirage’ has shaped and influenced our perception of Spartan women for several generations, and many scholars continue to portray them as unusually progressive, independent and powerful, at least in relation to women from other Greek societies, most notably Attica. Spartan women come across as very modern, certainly when judged by the standards of Greek antiquity. Thus Sarah Pomeroy’s latest study on the role of Spartan women is marketed by its publisher as:

undoubtedly ‘different’ (in some ways) from their Athenian sisters, and dress may have been an indicator of this to some extent. Yet in many other ways the women of Sparta shared a common life-experience with women in other Greek city-states. Most Greek women, to all intents and purposes, participated in and operated under a similar maleconstructed gender ideology and, in fact, led comparable daily lives (Millender 1999: 372-3; Foxhall & Stears 2000; Cartledge 2001). This article does not aim to give yet another ‘reading’ of the familiar but puzzling evidence for Spartan women’s lives per se. Rather, it seeks to add to our corpus of knowledge of Spartan women by bringing into focus a group of images of women that come from Sparta and the wider Peloponnese and that have remained oddly understudied, despite the important impact they have on our understanding of how Spartan women were envisaged within their own society. In addition to these material representations, I want to focus on several well-known texts in order to reassess what they tell us about female clothing codes in Spartan society.

A unique and fascinating glimpse of the women who inhabited ancient Sparta. Sparta … was renowned in the ancient world as a stoic and martial city-state, and most of what we know about Sparta concerns its military history and male-dominated social structure. Yet Spartan women were in many ways among the most liberated of the ancient world, receiving formal instruction in poetry, music, dance, and physical education. … Spartan Women seeks to reconstruct the lives and the world of Sparta’s women. (Dust jacket of Pomeroy 2002)

In analysing these artworks and texts, I want to suggest that Spartan women, like the women of the rest of the Greek world, in fact, shared in an ideology that required them to cover their heads and – when occasion demanded – their faces too, with a garment that should be classified as a veil.4 Veiling, I suggest, was part of the Spartan system of gender identity.

The book’s cover illustration shows a very familiar image: a little bronze figurine of a Spartan girl who hitches up the skirt of her chitōniskos, a short, breast-exposing tunic in order that her manly stride should be unimpeded.2 Spartan Women’s Lib, it seems, is expressed through Spartan dress – or undress, as the case may be.3

If that is indeed the case, then why has ‘Spartan veiling’ been missing from histories of Spartan women up until now? There are perhaps two reasons. First, despite the overwhelming evidence for the use of the veil in ancient Greece as a whole, there is a general disregard for, or deliberate avoidance of, the subject of the importance of veiling in ancient Greek societies in classical scholarship in general.5 However, more recent work undertaken on the subject is concerned with emphasizing the point that ancient Greece was a veil-society, and that it shared in a ‘veil ideology’ still seen in contemporary societies, especially

I do not want to dismiss this idea totally. There is much of value in this type of ‘reading’: Spartan women were Drawings by L. Llewellyn-Jones. London, British Museum 208, from Prisrend (?), Albania (ancient Epirus), c. 580 BC. For details see Scanlon (2002: 101-2). 3 Thus, see Fantham, Foley, Kampen, Pomeroy & Shapiro (1994: 60): ‘a young female athlete [wears]… a short sleeveless dress, not unlike that of an Amazon. Such an image is unusual in the context of Ancient Greece, where female figures (other than prostitutes) are normally shown in layered garments that cover the whole body.’ On Spartan dress see Ephraim (1989: esp. 7). 1

2

On the use of the word ‘veil’ and ancient Greek terminology for ‘veil’ see Llewellyn-Jones (2003: 3-39). 5 For reasons behind the lack of scholarly interest in Greek veils see the discussion in Llewellyn-Jones (2003: 3-9). 4

17

Dress and Identity

with understanding the make-up of Laconian society from the inside, then it is logical to begin by looking at evidence from within Sparta itself. That evidence is, of necessity, material.7 Yet images of women made by Spartans within and for their society have too often been neglected.8

Fig. 3.1: The contemporary Iranian chador resembles the ancient Greek pharos-veil.

Not that there is an over-abundance of female imagery to work with. Representations of women in Spartan art are rare and we simply do not find the rich profusion of female images familiar to us from Attic art, which allow us a glimpse into the world of women or, more accurately, the male Athenian ideology of womanhood.9 Some scholars, however, have interpreted the lack of female images as a further nod towards the empowerment of Spartan women: ‘Judging from the visual propaganda directed at women in the rest of the Greek world, a modern feminist might consider the absence of art as a positive and creative force for Spartan women’ (Pomeroy 2002. 170). Yet that view undoubtedly takes female empowerment too far; certainly, historians of other ancient world cultures who deal with a similar paucity of female images do not equate the lack of representations with any kind of female autonomy.10 Those scholars who have turned their attention towards Spartan imagery, however, have concentrated on a familiar (but tiny) sample of artworks that represent women, most notably the bronze statuettes of female athletes, the bronze nude caryatid mirror handles, and the painted pot fragments depicting women (hetairai?) at symposia.11 Much of this material evidence stems from the archaic period and may be considered to be of limited value if scholarly interest (as is often the case) resides in classical Sparta. But it should be remembered that the archaic visual evidence is still much nearer in time to the classical Spartans than the majority of non-Laconian texts that probe classical Spartan society, and should be privileged accordingly.

in the Muslim world.6 In fact, evidence for the daily use of the veil by women throughout the ancient Greek world is strong.

Much of the literary evidence for Spartan women comes from decidedly Athenocentric sources, hostile to Sparta. This familiar literary topos has recently been well explored by Ellen Millender, who notes that the Athenians frequently stereotyped their enemies by their odd sexual and genderrelated practices (Millender 1999). In Athenian literature ‘otherness’ is reflected in the idea that Spartan women enjoyed a particular sexual freedom often represented in their habit of ‘going Dorian’ – that is to say, appearing

By ‘veil’, I refer to an unstitched garment, like a mantle or cloak, which has the capacity to be pulled up onto the head and, if required, across the lower face, like the modern Iranian chador (Fig. 3.1). This was a routine way of wearing an ‘uncut garment’ in ancient Greek societies, and pan-Hellenic literary and material evidence from several successive centuries strongly supports the notion that the veiling of the female head or lower face was routine for many types of women in all Greek societies (Cairns 2002; Llewellyn-Jones 2003: 49-61, 104-113, 162-165).

Powell (1998: 119) attests to the validity of this: ‘scenes of human activity [in Laconian art] give a… revealing, as well as a more sympathetic view [of Spartan life]’. 8 There are exceptions. See, for example, Fitzhardinge (1980); Pipili (1987); Powell (1998); Pomeroy (2002: 161-70 for an overview). 9 On women in Attic art and ideology see Lewis (2002) and LlewellynJones (2002). 10 See Albenda (1983; 1987) on the lack of representations of women in Assyrian art. 11 Figures of athletes: (Scanlon 1988 & 2002: 101-5); Bonfante (1989); Pomeroy (2002: 164-65). Caryatid mirrors: Scanlon (1988 & 2002: 127-37); Stewart (1997: 29-34, 108-19); Rolley (1977); Pomeroy (2002: 164-65). See also Richter (1938) who argues that the caryatids represent Corinthian prostitutes and not Spartan athletes at all. Vase paintings (especially the so-called ‘Mitra Vase’): Pipili (1998); Smith (1998); Powell (1998); Pomeroy (2002: 165-9). 7

The second reason that Spartan veiling is so conspicuous by its absence in scholarship is perhaps more perturbing. There is, among most historians of Sparta, a tendency to privilege the literary evidence over the visual, an understandable approach if the aim is to understand the narrative rudiments of Spartan history. If the aim, however, is truly concerned 6 See, most importantly, Blundell (1998, 2002); Cairns (1996a, 1996b, 2001, 2002); Llewellyn-Jones (2003).

18

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.2: Figure following Larry Gonick (1990) The Cartoon History of the Universe

naked or at the very least in a state of dishabille. This is still a popular image. Larry Gonick’s perceptive cartoon history of ancient Greece plays up the stereotype (Fig. 3.2): a few Spartan beauties are actively engaged in naked sports, while behind the boundary wall of the exercise yard, two panting non-Spartiate males ‘admire’ their efforts (Gonick 1990: 234. Cf. his vision of Athenian women: 1990: 324).

In the play the Athenian women admire Lampito’s brazen Spartan beauty. This peculiar kind of womanly, yet virile, Spartan beauty so neatly embodied in Lampito is also noted by Plutarch, who famously claimed that Spartan girls were ‘freed from all delicacy and effeminacy… since they partook no less than the men in bravery and ambition’ (Lycurgus 14.2). However, this belies some of the Spartan iconography in which, as we will see, feminine daintiness is frequently highlighted. Plutarch also notes that ‘there was nothing disreputable about the girls’ nudity’ (Lycurgus 14.4), although Thomas Scanlon has correctly emphasised that ‘nudity’ in this context probably means ‘scantily clad’ or with less clothing than was normal for women outside Spartan society.16 The short Doric chitōn, the chitōniskos, or chitōn exomis, pinned on one shoulder, open at one side and pulled up over the knee, became regarded as Spartan dress par excellence.17 This rather scandalous tunic famously earned Spartan women the epithet ‘thigh flashers’ (Ibycus fr. 339) and in a scene in Euripides’ tragedy Andromache, Peleus, the caricature of a vapid Athenian patriot, accuses Spartan women of dressing immodestly – by flashing too much thigh – and Spartan men of allowing their wives

Alcman’s Partheneion famously recalls the ritualised athletic activities of the beautiful girls of Sparta, perhaps in relation to the cult of the most beautiful of all Spartan women – Helen.12 Later Aristophanes lampooned the same erotic athletic prowess of these Spartan lovelies in the character of Lampito, the female Spartan delegate to Athens, in his comedy Lysistrata of 411 BC. Lampito is the perfect product of Athens’ myth of the empowered Spartan women, for, while all Spartan women were beautiful (the epithet ‘Sparta of the beautiful women’ is first found in Homer, Odyssey 13. 412),13 Lampito has a strikingly virile beauty, and is endowed with a strength that can throttle an ox.14 Her breasts are large and firm, her buttocks are taut, and her semi-nude body is displayed in all its glory to her Athenian sisters whom, we learn from a later scene, are themselves circumspectly veiled (Lysistrata 532-4).15 On the Partheneion as evidence for Spartan female exercise see Hesychius s.v. en Drionas. See further Scanlon (1988: 214 n. 63) and Millender (1999: 383 n. 54). 13 See further discussion in Pomeroy (2002: 132-3). 14 On Lampito as a caricature see Millender (1999: 358). 15 See further Llewellyn-Jones (2003: 270-1). Interestingly, even though the Athenian women are circumspectly veiled, Aristophanes portrays them as far more sexually adventurous than the unveiled Lampito who is merely the passive recipient of their sexual attention. 12

Scanlon (2002: 125; see also 101-5). See further Ludwig (2002: 264) sv LSJ gymnos. On female nudity at Sparta, Critias (DK(6) 88, fr. 32) explicitly refers to female athletic nudity in his admiration of the Laconian methods of producing healthy children. See further Plut. Mor. 227e and his discussion of Spartan ‘incitements towards marriage’ among which, at Lyc. 15.1, he includes Spartan girls ‘undressing’ (apoduseis). 17 Xen. Anab. 5.4.13; Xen. Mem. 2.7.5; Aelian VH 9.34; Arist. Birds 946. See further discussions in Parisinou (2002). 16

19

Dress and Identity

and daughters too much licence (Andromache 592-604).18 Whether Spartan women really dressed as immodestly as Euripides and Aristophanes propose is contestable, and it is more probable that these scenes play up the Spartan sexual stereotypes.19

in Sparta naked or semi-naked exercise or ritual was part of a state-endorsed system of education and initiation that placed high value on the fertility and childbearing capacities of young, unmarried girls. But as we shall see below, the evidence makes it clear that outside of an athletic, ritualistic, or state-condoned setting, female nudity was not the norm and, in fact the non-Spartan sources that speak about married women’s revealing dress are all very late (Pollux 2.187, 7. 54-5; Clement of Alexandria 2.10, 114.1). In fact, there is no mention at all of an older married woman wearing le minimum in the classical sources.

We should not necessarily see all the literary references to nudity or revealing Spartan female dress as being representative of what all Spartan women wore on a daily basis. So what then was the reality? Archaic Spartan art does contain images of scantily dressed young female athletes, and I think it only fair to concede that the Doric chitōn or the chitōniskos were worn by some young women, but not necessarily day after day and certainly not by all women. The images of pubescent girls on bronze mirrors suggest that nudity was a Spartan practice, and, although the blatant eroticism of these figures should not be underplayed, we need to hypothesise that nudity and semi-nudity was not practiced on a routine basis outside of a specific context, no more than male nudity was seen in public on a daily basis on the streets of Athens. In general, ancient Greek society certainly expected male nudity at specific occasions, permitted it at other events, and even appears to have tolerated it at a more pedestrian level. 20 Female nudity, however, was generally not acceptable, at least not if a woman wanted to maintain her good name, give an impression of a morally upright character, or uphold the respectability of the men of her family. We will see below how this applies to Spartan women too.

Scanlon neatly sums up my own conclusions about Spartan nudity or semi-nudity. He has persuasively suggested that in their athletic contests, the young women’s short revealing tunics were worn: …to emphasise the essentially non-athletic aspects of the [sport] which was more initiatory ritual than contest. To compete in the nude would have… looked like mimicry of a men’s activity. So dress was chosen which was unmistakably typical of the “everyday” activity of men, allowing for freedom of movement and also socially “proper” in not being too revealing… The exomis is therefore an ambiguous costume that well suits the status of the maidens, alluding to the masculine yet revealing the feminine, maintaining normal sartorial modesty yet unveiling the body as an erotic object. (Scanlon 2002: 108-9)

Female nudity at Sparta must have been situational. Girls appeared naked or more likely semi-nude at certain times (perhaps even on a daily basis), but in specific places. Young unmarried women no doubt had licence to appear semi-nude in the eugenic context of athletic competitions associated especially with rites of passage, as recounted by Plutarch (Lycurgus. 14. 4-6; cf. Moralia. 227e, Lycurgus 15.1). In this respect, Spartan custom does not differ that much from other Greek rites observed at, say, Brauron, where young girls performed naked or semi-nude in honour of Artemis (Scanlon 2002: 126). The difference was that

With this in mind, how then do we go about reconstructing the daily-wear of Spartan women? Unfortunately, the imbalance in the use scholars make of the material evidence from Sparta is a matter of grave concern and only clouds our understanding of daily habits. The twenty-six examples of the mirror handles depicting nude bronze caryatids have been well published and have attracted much criticism, most recently by Andrew Stewart, Scanlon, and Pomeroy (Scanlon 2002: 127-37; Pomeroy 2002: 164-65). In contrast, the images we have of fully dressed and, indeed, veiled Spartan women remain, on the whole and at best, unknown and unanalysed. At worst, however, is the possibility that although known and published (and indeed all the artefacts I will examine here have been formally studied, albeit often in isolation), the images have been deliberately ignored since they do not conform with scholarship’s expectations of how Spartan women should appear. The imbalance in our acknowledgement and use of the iconographic source material is unsettling.

See also Soph. fr. 872 (describing Hermione); Cartledge (1981: 91 ff.); Scanlon (2002: 125). 19 On ancient and modern linkages between nudity and licentiousness see Millender (1999: 383 n. 51). 20 The literature on male nudity is enormous, but see in particular Spivey (1996: 112-13); Stewart (1997: 26 ff.); Bonfante (1989); Osborne (1998: 80-104). I follow Osborne’s argument that male nudity in Greek art is a symbolic construct that represents the heroic and mature male and works alongside the image of the beardless youth. I do not suppose that men generally walked around naked in daily life, but suggest that nudity was an accepted part of athletic society at the gymnasium and other sporting events and no doubt was a central element of the public ritual of looking at boys’ genitals which so delighted Philokleon (Arist. Wasps 578). It is highly likely that people even turned a blind eye to an open robe or carelessly wrapped himation at a night-time revel, but I find it hard to believe that naked Athenian males paraded alongside vulnerable maidens up to the temple of Athene during the Great Panatheneia. Male nudity was acceptable in controlled contexts, where the Athenian council inspected the bodies of young men, or at places where women were banned from entering. Moreover, the Greek word aidoia used for sexual organs (like the Latin pudenda) testifies that male nudity was not totally accepted. See further Ferrari (1990: 189). See most recently the interesting observations on civic nudity by Ludwig (2002: 261-318). 18

So what did Spartan women habitually wear – at least when off the sports ground? Archaic evidence, especially from the sanctuary of Artemis Orthia at Sparta, suggests that Spartan women loved their finery, and certainly in the late Archaic period they wore heavily patterned woollen Doric gowns that fully covered and concealed their bodies (Hodkinson 2000: 228-30). Moreover, as Karen Stears and Lin Foxhall

20

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.3: Terracotta head. Sparta Museum 3100.

have shown, much of Spartan women’s clothing was woven at home, and some textiles – the women’s finest work no doubt – were later proudly dedicated to Artemis Orthia together with the loom weights that helped create the garments (Foxhall & Stears 2000). In this respect, too, Spartan women shared a common tradition with the women of the rest of Greece who regularly offered textiles of their own making to goddesses primarily connected to marriage and childbirth (Linders 1972).

arranged curls are visible at the hairline.21 The patterned stripes are similar to those recalled by Sappho (fr. 98) who recollects that it was a custom in her mother’s generation for a woman to have ‘her locks bound in a purple [headcovering]’, whilst from her own time, she refers to ‘a decorated head-covering from Sardis.’22 It has recently been suggested that the Oriental-looking archaic korai from an Isthmian perirrhanterion (a large shallow lustral bowl resting on a stand composed of four standing women), dated c. 670-50, have similar veils draped over their heads: what at first look like plain, thick bands of hair appear on close examination to contain elements of bright pigmentation (red and blue), suggesting the kind of coloured head-covering worn by the Spartan terracotta head.23 A similar veil is also worn by the female heads appearing on the rim of the Chian ‘Aphrodite Bowl’ discovered at Naukratis (c. 600 BC), where patterned bands

The veil is certainly the type of garment likely to have been created by Spartan women at home. Literary and artistic evidence attests to the antiquity of the Spartan veil. According to Homer’s Odyssey (4.623) at a banquet in Sparta held in honour of Telemakhos, the women who serve the guests are the wives of the Spartan nobles and appear ‘beautifully veiled’ (kallikrademnos) as they carry out their duties. A terracotta head found on the Acropolis at Sparta bears out Homer’s portrait of veiled Spartan women, as the female wears a head-veil ornamented with delicate patterned bands (Fig. 3.3). Dated to c. 680-70 BC, the head’s glazed decoration is carried out with great attention to detail: the veil is divided into bands that are decorated with woven patterns and floral rosettes; a vertical striped border surrounds the forehead. Beneath the veil, carefully

The terracotta head is one of the earliest mould-made figures to be found in Greece, and moreover, one that clearly attests the shift from the Geometric to the Daedalic style of art, see Higgins (1967: 51-2). 22 If Sappho’s floruit was c. 612-08 BC then her mother could have been referring to a headdress worn in the mid 7th century. 23 See Sturgeon 1987: 44; Ridgway (1993: 126-8). A similar style is found on a number of terracotta heads from Rhodes, see Pinney & Ridgway (1979: 235. No. 114, fig. A); Jacopi (1931-1939: Pl. IV). See most recently Boardman (2000: 95 ff.); Llewellyn-Jones (2003: 45, fig. 18). 21

21

Dress and Identity

It is generally accepted that the mainland-style korai found on the perirrhanteria are derived from Near Eastern originals, via Samian influence; this Eastern bias is felt in the dress of the korai who, in addition to the short head-veil, can alternatively wear longer veils which (probably) hang down their backs before they are pulled around the front of the torso and tucked into a belt. Such a style is the hallmark of Near Eastern dress and was adopted by the women of East Greece too, if the depictions of the korai are anything to go by.27 The fact that highly decorated Near Eastern fashions were entering Greece during the archaic period is well documented in sculpture, vase painting, and in poetry. Although material evidence from Sparta is lacking, according to Alkman (1. 67-8) Spartan girls wore rich and elaborate Eastern-style gowns for their religious duties and were familiar with the styles of Lydia, particularly in headgear. Finds of miniature statuary of clothed females from Sparta, particularly from the sanctuary of Artemis Orthia, testify to the influx of foreign textiles which are rendered in minute detail on the otherwise crude statuettes (Foxhall & Stears 2000: 304-5). Short head-veils appear to have disappeared from fashion by the early 6th century.28 They were replaced instead by garment known as the pharos, a term that is usually translated as ‘mantle’ or ‘robe’, although by derivation pharos simply means ‘a piece of clothing’.29 It was easily made from a huge oblong of fabric and was therefore open to considerable variety in its wearing and there is a clear development in the use of this versatile length of fabric over the centuries. From the Archaic period to the Hellenistic age, it tends to be worn by women as a cloak covering the shoulders and arms, and it is frequently pulled over the head as a veil. Given the absence, outside Homer, of any satisfactory Greek word for an archaic-style mantle, we can use the word pharos when discussing this style of garment in its 7th and 6th century context and, where appropriate, for later examples too.30 Consequently when it is drawn over the head or across the face, we can reasonably employ the term ‘pharos-veil.’ 31

Fig. 3.4: Four sided marble slab. Scene of Helen and Menelaos(?). Sparta Museum 1.

are once again prominent (Boardman 1998: 145; LlewellynJones 2003: 45, 46, fig. 19). It should be borne in mind that perirrhanteria were also made and used in Sparta and that they share the same artistic motif of standing veiled women supporting a basin, although no complete example has been found.24 Following the observation that all the known perirrhanteria (nine in total) share identical features, Fitzhardinge argues that the perirrhanteria found scattered throughout the Greek world were all the product of a single Laconian workshop based at Caryae.25 If this is the case, then it is feasible to suggest that the vogue for (Eastern-style?) short patterned head veils so frequently attested in the art works of this period was a result of a fashionable Spartan influence permeating outwards.26 This is not to be considered unique, for as John Boardman notes, although Sparta may seem a surprising home for fashionable, foreign-looking, objects found elsewhere in Greece, Laconia’s special links with the East and with some Ionians is also reflected in Sparta’s creation and use of ivory objects (Boardman 1978: 25).

Judging from the iconographic evidence, the most popular Laconian and Peloponnesian veil style of the archaic and See Ridgway (1993: 127). Cf. Sturgeon (1987: 42) who calls it a shawl. For the development of veil styles in this period see Llewellyn-Jones (2003). 29 Pharos worn by a man: Il. 2. 43, 8. 221, Od. 15. 61, Xenoph. 3. 3, Pherecyd. Syr. Fr. 68 Schibli (DK B2) col. 1.14, Hdt. 9. 109, E. El. 1221; worn by women: Od. 5. 230, Hesiod. Works &Days 198, A. Ch. 11 etc. Losfeld (1991: 338) defines it as ‘long manteau de lin féminin et quelquefois masculin’. See also pp. 82-3, 179, 313, 321. 30 Although the pharos is the only Homeric garment shared by both men and women, it is only members of the nobility who are represented as wearing it. For the pharos, no brooches or pins are mentioned nor does it seem to be used as a blanket at night; it was probably made of linen since the epithet euplunēs – ‘well-washed’ (Od. 8. 392, 425; 13. 67; 16. 173) – seems better suited to linen than to a woollen cloak and it does not seem to have been worn so much for warmth as for a conspicuous display of wealth and status. 31 In his study of the classical himation, Repond did not use the term pharos. Instead he classified seven main types of drapery-styles for the himation (although I am not convinced that they fall into such neat categories). My definition pharos-veil best fits his category of himationtype 1. See Repond (1931: 40-59). See also Pollux, Onom. 7. 42, 48. 27 28

A fragmentary base from Sparta is discussed by Hamdorf (1964). The whole series of perirrhanteria, except those from Sparta and Amyclae, is shown by Richter (1968: figs 31-65). 25 Fitzhardinge (1980: 72). It is also suggested that the post-Daedalic head of Hera found at Olympia is also of Spartan design and construction. 26 For the spread of the short head-veil in Greece see Llewellyn-Jones (2003: 44-46, 56-59). 24

22

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.5: Male and female figures from Tegea. Tegea Museum 1605

classical periods was a pharos-type.32 It is well represented on a marble pyramidal slab from Magoula in Sparta (dated c. 600-570 BC) with relief decoration on all sides, one of which possibly shows the meeting of Helen and Menelaos at the sack of Troy. Here Helen wears the veil over her head and draws a fold around her body (Fig. 3.4).33 The idea that the veiled figure is Helen is plausible not only because the relief is Laconian (and may once have stood on a heroon of Helen or of Helen and Menelaos), but also because the confrontation between Helen and her estranged husband was a very popular one in archaic Greek art generally. The Spartan artist, despite the limited space of the stele, attempts to show Helen raising a portion of her veil with her left hand, a gesture frequently adopted by Helen on archaic pottery.34 The sculptor thus shows Helen’s head and body circumspectly wrapped in the folds of the pharos-veil

as if to suggest she is a good and modest wife (Pomeroy 2002: 169). However, Pomeroy suggests that “Helen meets [Menelaos] without cowering and faces him boldly, looking directly into his eyes”. But the presence of the veil here, as in archaic painted examples, strongly suggests otherwise. The veil is commonly found in Greek art being utilised by its wearer in the mode adopted by Helen. A female frequently raises part of it with one arm (usually the left arm) that apparently extends in front of her so that the veil forms a large and distinctive flap of cloth which frames her face. It is important to note that this ‘veil-gesture’35 is first encountered in the Greek artistic canon in the art of the Peloponnese in the late 7th century BC on a relief from a sanctuary of Athena Soteria and Poseidon in the region of Asea. It shows either a nuptial scene or a procession of gods – which is more likely, or both combined (Fig. 3.5).36 The stone slab is poorly preserved although it is possible to make out, depicted in high relief, a male (god?) figure with long braided hair, a section of which falls down his back while another is draped forward over his shoulder.37 The smaller figure of the woman (goddess?) holds a segment of her veil aloft in a typical veil-gesture encountered so frequently in examples from Corinth and Athens during this period. The veil forms an arch which covers the woman’s head and face so completely that only a hint of a hairline on her right-facing profile is visible. The gesture performed here actually indicates that the female figure is in the act of

On the pharos-veil see further Llewellyn-Jones (2003: 49-53). See Pipili (1987: 31). If the female figure is Helen then the veil that she wears has a crucial role to play in her re-captivation of her husband. The cutting of the relief and its preservation is poor, but we are able to discern that Helen’s pharos is used to veil the back of her head where it hangs down over her shoulders and upper arms. No indications of the colour or pattern of the veil that were originally picked out in paint now survive. See further Llewellyn-Jones (2003: 51 & fig. 31). 34 See Lexicon Iconographicum Mythologicae Classicae (LIMC) sv. Helene nos. 157-161, 211-214, 217, 219, 221 (where the veil falls off the head), 222, 224, 291, 293, 295, 297, 299, 300, 301, 303-306, 315, 321, 322, 327, 328, 331, 333, 335, 337, 338, 342, 343, 344, 347, 349, 353. The motif goes back at least as far as the famous relief pithos from Mykonos (c. 675) which affords an early depiction of a pharos-veil as worn by Helen. See metope 7 at LIMC sv. Helene no. 225. The Mykonos Pithos must rank as the single most important source for the Ilioupersis myth and this vignette is the earliest known depiction of the recovery of Helen by Menelaos. See Ahlberg-Cornell (1992: 78-80). See also Anderson (1997: 182-91). The scene continues in popularity well into the third century. Helen’s image alters according to taste and, interestingly, she is always depicted in the height of fashion. See, for example, Austin (1994: 75, fig. 3) for Helen wearing the short, shoulder-length shaal-veil, popular c. 500. Helen is just one of many women depicted on the Mykonos vase, but she is the only veiled figure. 32 33

For the use of the term ‘veil-gesture’ see Llewellyn-Jones (2003: 98110). 36 Tegea Museum 1605. See for example the procession of gods in the wedding scene on the François Vase by Sophilos. See Boardman (1974: fig. 24 and 25.1). 37 This figure was originally identified as female. This is certainly wrong. See Raftopoulou (1993: 1). 35

23

Dress and Identity

covering or (depending on the context) uncovering her face – an artistic motif frequently found throughout the Greek world throughout the archaic and classical periods (see Llewellyn-Jones 2003: 98-110). However, it is interesting to note that in fact the Spartan examples of the gesture are some of the earliest encountered in Greece, which suggests that not only was the veil a facet of archaic Laconian society but also that the artistic motif may have had its origins in Spartan (or at least Peloponnesian) tradition (Raftopoulou 1993). It is, of course, a two-dimensional rendition of a threedimensional subject and consequently can be difficult to ‘read’ correctly, but the logic of the two-dimensional veilgesture becomes a little easier to interpret when dealing with frontal-facing subjects, such as a well-known metope from Mycenae, dating to c. 630 BC (Fig. 3.6). It is clear that here we have a woman wearing her pharos over her head and that she pulls a large fold of it across her bust, in a three-dimensional rendition of the veil-gesture.38 The woman in the metope does not cover her face, but none of the veil-gesture scenes explicitly show that, although I would suggest face-veiling is actually intended if we follow the literary evidence at least. Greek art, even at this early stage and even in the Peloponnese, privileged an unimpeded view of the human face.39 Sometimes this penchant goes to an extreme, as in a remarkable image on a bronze mitra from Olympia (dated to around 650 BC), which probably depicts Orestes’ murder of Klytemnestra (Fig. 3.7).40 The queen is covered by an arch of elaborately worked cloth, and she holds her veil aloft with one hand, a gesture that supposes her to be veiling her face (and figure). The fact that she is shown underneath the arched veil is probably an artistic device that allows the viewer to still see the queen despite the fact that her body is intended to be entirely shrouded within her veil. The concentration is on the figure underneath the clothing.41

Fig. 3.6: Metope from Mycenae. Athens National Museum 2869.

not a matter of concern since both categories of women utilise the veil gesture in Greek art of this period. However, the veiled woman on a grave relief from Tegea in the Peloponnese (dated to around 510 BC) is definitely mortal. She sits in the presence of (probably) her deceased son (shown naked) and her seated husband (Fig. 3.9).43 Several scenes like the Tegea stele show a seated woman adopting the veil-gesture; most examples are categorised by the term ‘hero-relief’, of which about thirty examples (mainly in a fragmentary state) are known. Most originate at Yeraki and Khrysafa, although clay reproductions have been discovered near Sparta at Amyclae (upstream on the Eurotas river) at the Sanctuary of Cassandra. Less than a quarter of the reliefs can be dated to the Hellenistic era (although they are crafted in an archaic style), and the majority date to the latter part of the 6th century and the early fifth century BC. Most follow a regular iconographic scheme, depicting a seated god and goddess (or hero and heroine) holding various attributes, such as a kantharos or a cup, and frequently attended by miniature worshippers offering gifts, such as cockerels and flowers (Andronikos 1972). Moreover, these reliefs, like the perirrhanteria before them, betray a Spartan ‘adaptation of Ionian models for the hair and clothing into the local style.’ (Charbonneaux, Martin & Villard 1971: 148).

A delicate ivory plaque dating to the opening decades of the 6th century BC and found at the Artemis Orthia sanctuary at Sparta itself shows how the wearer has pulled her voluminous pharos around her shoulders in order to conceal much of her body and uses the excess fabric to cover her face too (Fig. 3.8).42 The ivory image obviously lacks a context and it is difficult to know if the woman represented here is a goddess or a mortal worshipper, although that is Boardman (1993: 39, fig. 35) describes it thus: ‘a woman draws [a] cloak over head, a gesture of modesty or rank’. See further, Sweeney, Curry & Tzedakis (1987: 108-9, no. 29); Llewellyn-Jones (2003: 51, 52, fig. 30). 39 For the privileging of the face in art see Llewellyn-Jones (2003: 1067). 40 Ahlberg-Cornell (1992: 92-3) believes that it represents Clytemnestra and Orestes. 41 Given the context of the scene, Ahlberg-Cornell (1992: 92) has interpreted Clytemnestra as hiding beneath her veil as Orestes is about to deliver the death blow. In this way, I think that the Queen’s imprisonment within her veil at the moment of her death parallels the death of Agamemnon caught within Clytemnestra’s woven web at the time of his own slaughter. See Aeschylus Ag. 1383-5. 42 Athens National Museum 15847. See Dawkins (1906-7: 98): “[a woman] in a cloak, which she holds in front of her face.” See further Raftopoulou (1993: 4). 38

In a well-known 6th-century example from Khrysafa near Sparta (Fig. 3.10), a long-haired god turns towards the viewer, but the goddess seated next to him – wearing ‘Oriental’ slippers with turned-up toes – is only viewed in profile and, accordingly, adopts the standard veil-gesture to 43

24

Athens National Museum 55. See Raftopoulou (1993: 1 and fig. 3).

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.7: Bronze mitra from Olympia. Olympia Museum.

shoulders to form a veil. The back of the pharos hangs low down her back (Bonfante 2003: 186-7).

indicate that she discreetly covers herself.44 On first sight it is difficult to see that the female’s head is veiled, since the sculptor is keen to depict the uninterrupted view of her delicate ringlets and curls. The veil is only represented being held aloft behind the woman’s profiled features. In fact, the veil, which was probably originally brightly painted, is used to highlight her profile but is not represented covering the part of her profile visible to the viewer.45 Later the complex design is simplified and becomes increasingly standardised and by the fifth century the god too is rendered in profile, although his consort’s veil-gesture remains unchanged – as do her eastern slippers (Fig. 3.11).

Sometimes, but more rarely for Sparta and the Peloponnese, the veil-gesture found on the votive plaques, stelai, and ivory figurines is reduced to a mere delicate touching of the veil, particularly in early fifth-century examples, such as a stone relief from the Maina (Fig. 3.13) depicting a young woman in the act of pouring a libation. Her very light and diaphanous mantle has slipped off her head (showing her short hair bound beneath a filet) and onto her shoulders, although she takes a section of the robe and holds it in front of her face.46 This delicate rendition of the veil-gesture is paralleled many times in contemporary Athenian examples wherein women pluck sections of their clothes – veils, mantles, or sleeves (Llewellyn-Jones 2003: 100, figs 106-110). It is clear that this Laconian artist relished the opportunity that the gesture gave him to experiment with both the depiction of the hands and fingers and the range of

Coincidentally, a three-dimensional representation of the costume worn by the goddesses of the votive plaques comes in the form of a beautiful bronze figurine of a mature woman from the rim of the Vix crater, dated to c. 530 BC (Fig. 3.12) and originating, in all probability, from Tarentum or Rhegium, although clearly based on Laconian models (Charbonneaux, Martin & Villard 1971: 148). She wears a long peplos and oriental-style slippers and her oblong pharos is drawn up, by its shorter end, over her head and

Pomeroy (2002: 42) suggests that young women wore their hair long and uncovered – although this is not supported by the Maina relief. Conversely, Pomeroy argues that married women wore their hair cut short and covered by a veil. She is correct to highlight the veil, but it is impossible to judge from the material evidence if their hair really was short. Horace Odes 2.11.24, for instance, recalls a woman with her long hair caught up in a twist or knot ‘Laconian style’. On Spartan hair see David (1992: esp. 17). 46

For a discussion of the reliefs see Fitzhardinge (1980: 80-82); Hafner (1968: 98-9); Boardman (1993: 165). 45 On the difficulty or reading this gesture see Llewellyn-Jones (2003: 97-98). 44

25

Dress and Identity

action of veiling. As a convention, veiling is as necessary in a display of aidōs as is careful avoidance of the gaze of strangers and the lowering of the head, a notion that Euripides’ Hecuba makes clear: ‘Custom forbids that a woman look into the eyes of men’ (Euripides Hec. 974-75). The ‘veil-gesture’ found in Greek iconography (including that of Sparta) is strongly related to the concept of aidōs (Cairns 1996a). Veiling is a distinctive response to (and display of) aidōs, and the covering of the head and face is a natural reaction to express, contain, or rebuff sexual shame and to preserve a sense of modest detachment. For it is the conscientious movement of raising the head-veil to cover the face that enables the viewer to recognise that a woman knows the rules of the honour-shame code and plays by them accordingly. As a result, the viewer too participates in the game (Llewellyn-Jones 2003: 169-173). It has often been noted how women in Greek art are laid open to the common gaze and that viewers want to see the female body, but wish to be assured of the respectability of the model. The idea is given a further twist in images of women carved into grave stelai like those found in Sparta and Attica where women are frequently shown holding onto their veils as they are exposed to the public gaze – even though the scenes often attempt to be domestic vignettes. Grave stelai frequently invite the passer-by to stop, read, and remember, but the compulsion towards female modesty remains, even for the dead (Fig. 3.14). In a society where women were not supposed to have a public identity, the existence of public sculptures of women seems anomalous; therefore, the sculpted bodies of women on the grave stelai invite the gaze, but simultaneously rebuff it and they deny the viewer the privilege of looking at them by adopting the veil-pose. The veil-gesture is therefore a defensive signal of blockage and a barrier to unadulterated viewing.49

Fig. 3.8: Ivory figure of a veiled woman from the sanctuary of Artemis Orthia at Sparta. Athens National Museum 15847.

effects that could be created by the veil falling in a variety of folds around the face, head, and shoulders.

Properly regulated, women could be put on display as statues and reliefs and painted images and even celebrated in coin-portraits and could be open to the public view: modesty expressed through dress and pose was one way of demonstrating that women were conforming to the established social nomos and that they were under control. The veil-gesture in art, like the physical motion of veiling in real-life, displayed a woman’s self-awareness of aidōs and in return demanded the aidōs of others.

So, judging by the material evidence, Spartan women veiled. The question is why? What message is conveyed by female veiling? Of course, there are multiple messages sent out when a woman veils her head or face. She might be making a bid for status; she might be making a move that attracts the attention of others – to show off her sexuality or to express anger or grief. All of these motives are found in Greek texts from Homer onwards.47 But the most common use of the veil expresses female awareness of modesty, shame, honour and social custom or nomos. Veiling as a social nomos is a convention in tune with society, to borrow a pun from Plato.48

The fact that the gesture is found in Spartan art strongly suggests that Spartan women also participated in the honour-shame code understood and acted on by Greek women from other societies. That they participated in the code in daily life and in the imagined artistic ideology of gender representation suggests that Spartan women were not as emancipated as is often thought and certainly not too out of step with women from other poleis. In this respect

In particular the concept of aidōs – modesty, reserve, knowing one’s place – is best displayed by the self-aware For a full discussion see Llewellyn-Jones (2003). Nomos is also used for ‘tune’ and ‘melody’ and is a favourite theme for word-play for Plato. See Laws 700b, 722e, 734e, 775b, 799e. A good definition of nomos can be found in Cartledge, Millett & Todd (1990: 231-32). Cf. LSJ s.v. nomos.

Even when true portrait statues of women became more common in the Hellenistic period, the appearance of the veiled head was an essential part of the acceptable iconography even though the portrait features could not be masked by a veil. See Smith (1991: 83-6, figs. 111-116). For public images of women in the Hellenistic period see van Bremen (1996).

47

49

48

26

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.9: Funerary stele from Tegea. Athens National Museum 55.

The image of Aidos, some thirty stades distant from the city, they say was dedicated by Ikarius, the following being the reason for making it: when Ikarius first gave Penelope in marriage to Odysseus, he tried to make Odysseus settle in Lacedaemon, but failing in the attempt, he next pleaded with his daughter to remain behind, and when she was setting out towards Ithaka he followed the chariot, begging her to stay. Odysseus endured it for a time, but at last he told Penelope either to accompany him willingly, or else, if she preferred her father, to go back to Lacedaemon. They say that she made no answer, but covered herself with a veil in reply to the question, so that Ikarius, realising that she wished to leave with Odysseus, let her go, and dedicated an image of Aidos; for Penelope, they say, had reached this point in the road when she veiled herself (Pausanius 3. 20. 10-11).

it is best to follow Paul Cartledge who has stressed that ‘Spartan women did enjoy certain freedoms… that were denied to their Athenian counterparts, but they were not, to put it mildly, as liberated as all that’ (Cartledge 2001: 106). Interestingly, the aition for the veil-gesture itself actually occurs in, of all places, Sparta, when Penelope abandoned her father for the sake of Odysseus and veiled her head in shame. According to this story, the veiling of the heads of all modest women can be associated with Penelope’s sense of shame50 and, according to Pausanias (3.20.10-11), a statue of Aidos which he saw in Sparta relates directly to the tale: The Romans also had a tradition that the bride’s veil was associated with an act of betrayal. Commenting on the custom of veiling the bride’s head, the lexicographer Festus (s.v. nuptias) noted that the act of covering the head (and the verb obnubere) was identical to the act of covering the heads of parricides: ‘Aelius and Cincius [on the derivation of ‘nuptias’]: since the head of the bride is veiled [obnuvatur] with a flame-coloured veil, [an act] that the ancients termed ‘obnubere’. For this reason [they say] that the law also bids veil [obnubere] the head of him who has slain a parent, that is to say, ‘obvolvere’.’ Both the bride and the parricide are guilty of breaking the bond with the father. 50

There is a strong supposition that the statue noted but not described by Pausanias showed a woman veiling and

27

Dress and Identity

Fig. 3.10: Spartan hero-relief. Berlin Staat. Mus. 731.

and Torelli have suggested that the aition provided by the ancient travel-writer may relate to a popular tradition attached to a statue of a veiled woman set up, perhaps, on a Spartan tomb. Alternatively, or in addition, his story could well have been inspired by the Laconian hero-reliefs which, as we have seen, frequently show a female adopting the veil-gesture.52 It is apparent that women at Sparta were veiled and that the existence of the influence of aidōs, expressed in the hero-reliefs and other monuments, was just as prevalent in Laconian society as it was in other parts of the Greek world (see also Odyssey 4. 623).

perhaps performing the veil-gesture, and a speculative identification has been made with a fifth century BC seated statue-type of Penelope (Fig. 3.15).51 I am not convinced, however, that aidōs is the emotion being portrayed in the statue: the seated statue-type, which can be compared to the figure of Penelope in the famous Chiusi Vase (Fig. 3.16), shows a veiled woman, but the veil is not held in the conventional veil-gesture. Instead the vase painting and the sculpture show her hand touching her chin in a gesture that is usually thought to indicate grief (Neumann 1965: 139 and fig. 69) or possibly pensive thought (BuitronOliver & Cohen 1995: 46). It is more difficult to assign Penelope’s imagery the standard veil-gesture pose, despite the suitability of the subject and the relevance of the motif to the Penelope story.

In a routine observation of Spartan customs, Plutarch also recalls that the women of Sparta were veiled; he draws attention to the fact because the tradition of Spartan

However, in their commentary on Pausanias Book 3, Musti

See Musti & Torelli (1991). It is argued that the female figure of Aidos is linked to the cult of Hades/Aidoneus and that the statue of Aidōs was a vestige of a pre-existing cult of Hades. There may well have been some confusion in fact about the reading and writing of the word AIDEUS (Hades) inscribed on a statue. For details see Richer (1999). 52

LIMC s.v. Aidos. The name of the seated woman in the statue type is certain, see Boardman (1985: 51); see also Cairns (1996a: 155 and n. 37); Buitron-Oliver & Cohen (1995: 43-48). 51

28

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.11: Spartan hero-relief. Sparta Museum 4.

and aidōs.54 Secondly, Plutarch’s Kharillos specifically states that the reason why Spartan wives are veiled is in order that they stay loyally and modestly faithful to their husbands, so the reasoning behind veiling the wives of Sparta is fundamentally no different from any other veilsociety, ancient or modern: female modesty and chastity is emphasised by veiling. Plutarch’s shock-factor comes into play with the mention of the Spartan girls (men’s daughters and sisters) who are taken into public areas unveiled in a deliberate bid for them to be seen by men and subsequently married off.55 The public display of the faces (and figures?) of young women is in accord with Plutarch’s larger agenda of envisioning Laconian society as being emphatically and essentially different from other Greek communities, a divergence that is, as we have noted, well emphasised

veiling included a rather avant-garde custom that was only practised in Laconia. Plutarch’s account is set at the time of King Kharillos, a contemporary of Lykurgos and in the Sayings of the Spartans he recalls, When someone asked why they [the Spartans] took their girls into public areas unveiled, but their married women veiled, he [Kharillos] said, ‘Because the girls have to find husbands, and the married women have to keep those they have!’ 53 There are several points to note about Plutarch’s observation: firstly, we need to stress the idea that Plutarch does not think that Spartan veiling is necessarily out of the ordinary but instead he assumes that Sparta is part of a long pan-Hellenic tradition whereby women are habitually veiled in accordance with established notions of sophrosyne

On Spartan notions of sophrosyne and aidōs see Humble (1999) and Richer (1999). 55 For a discussion of the age of veiling girls see Llewellyn-Jones (2003: 215-19). 54

Mor. 232 C. For the problems of using Plutarch as a source for Spartan women see Powell (1999). 53

29

Dress and Identity

Fig. 3.13: Fragment of a Spartan hero-relief. Copenhagen. Ny Carlsberg Glyptoket 423.

practices. In Attica we have proof that young women were veiled even before marriage (Llewellyn-Jones 2003: 216 ff). Expanding the evidence into the wider Peloponnese, Athenaeus (quoting Pythaenetus) notes the startling effect that unveiling had on (male) observers of Peloponnesian girls in general (not specifically Spartans) who appeared uncovered in public: Pythaenetus in the third book of his On Aegina says that Periander saw in Epidauros the daughter of Prokles, Melissa, dressed in Peloponnesian fashion – that is she wore no veil – but was dressed in a simple chitōn while she acted as wine-pourer for the workmen in the fields. And falling in love with her, he married her (Atheneus Deip. 589F).

Fig. 3.12: Mature woman in pharos veil. Rim of Vix crater, Chatillon-sur-Siene.

by the Spartan treatment of young girls. Since, Plutarch states, the Spartans saw nothing disreputable about female nudity among the young, and actually revered the cult of the female body, the unveiled appearance of young women of marriageable age, we can assume, would not be thought shameful or compromising.56 The Spartan men of Plutarch’s account are just being resourceful. They take their daughters out in public unveiled so that they will quickly attract good husbands and providers. His non-Spartan readers were no doubt more astonished by his revelations of Spartan feminine immodesty and the abuse of regular veiling

It is particularly interesting to note that Melissa’s public appearance in Epidauros without a concealing veil is justified and qualified by Athenaeus’ (or Pythaenetus’) statement that she was dressed ‘in Peloponnesian fashion’, in other words, that she merely accommodated to typical Peloponnesian (and perhaps Spartan) customs for young women. Therefore, if Plutarch and Athenaeus can be believed (and there is no reason to dismiss their evidence summarily), this perhaps suggests that the artistic representations we have examined tend to show only married mortal women or married goddesses. Perhaps the young maiden’s exposed head in the Maina relief (Fig. 3.13)

56 For Plutarch’s comments on Spartan women see especially Lycurgus 14-15.

30

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.14: Grave stele of the sisters Demetria and Pamphile c. 320 BC Keramikos Museum, Athens.

is an accurate depiction of a Spartan custom. Certainly Melissa’s appearance in male society without her veil is condoned by Peloponnesian society: the girl appears in close proximity to strange men as she pours them wine (the act carried out by the Maina maiden too), and it is clear that she is ‘on parade’ with the hoped-for result that marriage to one of the men she serves will follow, as indeed is the case, albeit that she does not marry a local man, but a Corinthian

stranger. In fact, there is an Athenian endorsement for this kind of veiling trait found in Euripides’ Andromache. While Pelius may propound the familiar cliché about Spartan moral laxity and female corporal display, his view is not completely endorsed by Euripides, certainly not in his construction of the Spartan Hermione, the proud daughter of Menelaos. When, in the play, the decorous Hermione loses all self-control and lets down her virtuous preserve of

31

Dress and Identity

Fig. 3.15: Seated sculpture of a veiled Penelope. Roman marble copy of a Greek original c. 460 BC, Vatican Museum 754.

aidōs, her jealous anguish over the concubine Andromache and the anxiety she feels about her own status as mistress of Neoptelemos’s household is reflected through reference to her dress.57 At line 830 she unveils her head (which had obviously been so decently and richly covered with a pharos-veil at her first entrance at line 147) as part of an agonised display of her loss of sophrosyne and cries aloud, ‘Woe! Woe! Away from my locks fine-spun veil, into the air!’ (Euripides Andr. 829-31).

her body but the Spartan princess replies, with a resignation that contrasts vividly with her first entrance when her gorgeous costume expressed and heightened her sense of her own status and dignity, ‘What is the point of veiling my breast with robes? What I have done to my husband is plain and evident and unhidden’ (Euripides Andr. 833-35). In many contemporary veil-societies a woman like Hermione, who appears unveiled in public – but especially before a man, even her husband – is considered naked (Mernissi 1975: 144; Chowdhry 1994: 283). If she appears unveiled before a stranger her good name is seriously compromised and she is generally regarded as inherently immodest and vulgar (Papanek 1973: 283-325, esp. 297). In northern India, for example, an unveiled woman is said

Hermione’s unreserved lack of shame is intensified by the fact that she not only exposes her head but also her breasts (Euripides Andr. 832). Her dazed nurse begs her to re-veil For comments see Allen (2000) and Lloyd (1994). See also comments by Millender (1999: 359-61). 57

32

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Fig. 3.16: Penelope at her loom. Skyphos from Chiusi by the Penelope Painter, c. 440 BC. Mus. Chiusi 1831.

to ‘go about exposing her breasts together with her face’. At the heart of the contemporary ideology of veiling, and the Greek ideology too – including that of Sparta – pounds the idea that a woman should only unveil for those men most closely related to her, and that unveiling in front of any strange men without the sanction of her male guardian is regarded as a violation of the familial bond (Llewellyn‑Jones 2011). If we follow Plutarch’s reasoning, Spartan girls unveiled themselves in public when accompanied by their fathers, and with their fathers’ permission, in order to attract a husband. Once married, he suggests, Spartan women did not allow themselves to be viewed by strangers – therefore following similar codes found in many ancient and contemporary veil-societies.

suggests that Spartan women were participants in the same honour-shame code experienced by other Greek women. While young unmarried girls could, on occasion, participate in state sponsored nudity, for much of the time they, together with married women, kept their bodies hidden behind multiple layers of clothing, and their heads and sometimes even their faces shrouded beneath veils. This does not negate the notion that some Spartan women had more autonomy in some areas of life than their Athenian counterparts and that ritualistic athletic contests with seminudity might have formed part of their formative existence, but the fact remains, in daily life the Spartan woman, like the Athenian woman, was a veiled woman. Bibliography

The Spartan mirage of female liberation expressed through revealing clothes or even full nudity has to be tempered. A closer look at Spartan iconography and a handful of neglected texts reveals quite a different picture and

Ahlberg-Cornell, G. (1992): Myths and Epic in Early Greek Art: Representation and Interpretation. Studies in Mediterranean Archaeology. Vol. C. Jonsered.

33

Dress and Identity

stages’, in M. McDonald & L. Hurcombe, (eds), Gender and Material Culture. London 300-14. Gonick, L. (1990): The Cartoon History of the Universe, volumes 1-7. From the Big Bang to Alexander the Great. New York. Hafner, G. (1968): Kreta und Hellas. Baden-Baden. Hamdorf, F.W. (1964): ‘Lakonische perirrhanteria’, Mitteilungen des deutschen archäologischen Instituts 87. 47-64. Higgins, R.A. (1967): Greek Terracottas. London. Hodkinson, S. (2000): Property and Wealth in Classical Sparta. London & Swansea. Humble, N. (1999): ‘Sophrosyne and the Spartans in Xenophon’ in A. Powell & S. Hodkinson, (eds). Sparta. New Perspectives, 339-353. London. Jacopi, G. (1931-1939): Clara Rhodos, IV, Esplorazione archeologiche di Camiro I. Rhodes. Lewis, S. (2002): The Athenian Woman. An Iconographic Handbook. London. Linders, T. (1972): Studies in the Treasure Records of Artemis Brauronia found in Athens. Stockholm. Llewellyn-Jones, L. (2003): Aphrodite’s Tortoise. The Veiled Woman of Ancient Greece. Swansea. Llewellyn-Jones, L. (2011): ‘Domestic Abuse and Violence against women in Ancient Greece’ in S. Lambert (ed.) Sociable Man. Essays in Honour of Nick Fisher. Swansea 231-66. Lloyd, M. (1994): Euripides’ Andromache, with an introduction and commentary. Warminster. Ludwig, P.W. (2002): Eros and Polis. Desire and Community in Greek Political Theory. Cambridge. Mernissi, F. (1975): Beyond the Veil: Male-female Dynamics in Muslim Society. Cambridge, Massachusetts. Millender, E. (1999): ‘Athenian Ideology and the Empowered Spartan Woman’, in Hodkinson, S. & Powell, A. (eds), Sparta: New Perspectives, 355-91. London & Swansea. Musti, D. & Torelli, M. (1991): Pausania. Guida della Grecia. Libro III. La Laconia. Milan. Neumann, G. (1965): Gesten und Gebärden in der griechischen Kunst. Berlin. Osborne, R. (1998): ‘Men without clothes: Heroic nakedness and Greek art’, in Wyke, M. (ed.), Gender and the Body in the Ancient Mediterranean, 80-104. Oxford. Papanek, H. (1973): ‘Purdah: Separate worlds and symbolic shelter’, Comparative Studies in Society and History 15: 283-325. Parisinou, E. (2002): ‘The ‘Language’ of Female Hunting Outfit in Ancient Greece’, in L. Llewellyn-Jones (ed.), Women’s Dress in the Ancient Greek World, 55-72. London & Swansea. Pinney, G. & Ridgway, B.S. (1979): Aspects of Archaic Greece. Allentown. Pipili, M. (1987): Laconian Iconography of the Sixth Century BC. Oxford. Pomeroy, S.B. (2002): Spartan Women. Oxford Powell, A. (1998): ‘Sixth-Century Lakonian Vase-Painting. Continuities and discontinuities with the ‘Lykourgian’ ethos’, in N. Fisher & H. van Wees (eds), Archaic

Albenda, P. (1983): ‘Western Asiatic women in the Iron Age: Their image revealed’, Biblical Archaeologist 46 (2): 82-8. Albenda, P. (1987):‘Woman, child, and family: Their imagery in Assyrian art’, in J-M. Durand, (ed), La femme dans le proche-orient antique, 17-21. Paris. Allen, W. (2000): The Andromache and Euripidean Tragedy. Oxford. Andronikos, M. (1972): ‘Laconian Reliefs’, Peloponnesiaka I: 254-314. Blundell, S. (1998): Women in Classical Athens. Bristol. Boardman, J. (1985): Greek Sculpture. The Classical Period. London. Boardman, J. (1993): Greek Sculpture: The Archaic Period. London. Boardman, J. (2000): Persia and the West. An Archaeological Investigation of the Genesis of Achaemenid Art. London. Bonfante, L. (1989): ‘Nudity as a costume in Classical art’, American Journal of Archaeology 93: 543-70. Buitron-Oliver, D. & Cohen, B. (1995): ‘Between Skylla and Penelope: Female characters of the Odyssey in Archaic and Classical Greek art’ in B. Cohen, (ed.), The Distaff Side. Representing the Female in Homer’s Odyssey, 29-58. Oxford. Cairns, D.L. (1990): ‘Mixing with men and Nausicaa’s Nemesis’, Classical Quarterly 40: 263-6. Cairns, D.L. (1993): Aidos. The Psychology and Ethics of Honour and Shame in Ancient Greek Literature. Oxford. Cairns, D.L. (1996a):‘Veiling, aÓδωs, and a red-figure amphora by Phinitas’, Journal of Hellenic Studies 116: 152-58. Cairns, D.L. (1996b): ‘Off with her aÓδωs: Herodotus 1.8.34’, Classical Quarterly 46.1: 78-83. Cairns, D.L. (2001): ‘Anger and the veil in ancient Greek culture’, Greece &Rome 48.1: 18-32. Cairns, D.L. (2002): ‘The meaning of the veil in ancient Greek culture’, in L. Llewellyn-Jones (ed.), Women’s Dress in the Ancient Greek world, 73-94. London & Swansea. Cartledge, P. (1981): ‘Spartan wives: Liberation or licence?’, Classical Quarterly 31.1: 84-105. Cartledge, P. (2001): Spartan Reflections. London. Cartledge, P, Millett, P. & Todd, S., (eds), (1990): Nomos. Essays in Athenian Law, Politics and Society. Cambridge. Chowdhry, P. (1994): The Veiled Women. Shifting Gender Equations in Rural Haryana 1880-1990. Oxford. Ephraim, D. (1989): ‘Dress in Spartan Society’, Ancient World 19: 3-13. Ephraim, D. (1992): ‘Sparta’s Social Hair’, Eranos 90: 11-21. Fantham, E, Foley, H.P, Kampen, N.B, Pomeroy, S.B. & Shapiro, A. (1994): Women in the Classical World. Oxford. Ferrari, G. (1990): ‘Figures of speech: the picture of Aidos’, METIS 5 (1-2): 185-204. Fitzhardinge, L.F. (1980): The Spartans. London. Foxhall, L & Stears, K. (2000): ‘Redressing the Balance: dedications of clothing to Artemis and the order of life

34

Lloyd Llewellyn-Jones: Veiling the Spartan Woman

Greece. New Approaches and New Evidence, 119-46. London & Swansea. Raftopoulou, E.G. (1993): ‘Sur certains archétypes de thèmes iconographiques provenant du centre du Péloponnèse’, in O. Palagia & W. Coulson, (eds), Sculpture from Arcadia and Laconia. Proceedings of an international conference held at the American School of Classical Studies at Athens, April 10-14, 1992, 1-12. Oxford. Richer, N. (1999): ‘Aidōs at Sparta’ in A. Powell & S. Hodkinson (eds), Sparta. New Perspectives, 91-115. London & Swansea. Richter, G. (1938): ‘An Archaic Greek Mirror’, American Journal of Archaeology 42: 337-44. Richter, G. (1968): Korai. Archaic Greek maidens. A study of the development of the Kore type in Greek sculpture. London. Ridgway, B.S. (1993): The Archaic style in Greek sculpture. Chicago. Rolley, C. (1977): ‘Le problème de l’art laconien’, Ktèma 2: 125-40.

Scanlon, T.F. (2002): Eros and Greek Athletics. Oxford. Smith, R.R.R. (1991): Hellenistic Sculpture, a Handbook. London. Smith, T.J. (1998): ‘Dances, Drinks and Dedications: Archaic Komos in Laconia’ in W.G. Cavanagh & S.E.C Walker (eds), Sparta in Laconia: Proceedings of the 19th British Museum Classical Colloquium. British School at Athens, vol. 4: 75-81. London. Spivey, N.J. (1996): Understanding Greek Sculpture. Ancient Meanings, Modern Readings. London. Sturgeon, M.C. (1987): Isthmia. Excavations by the University of Chicago under the auspices of the American School of Classical Studies at Athens. Volume IV. Sculpture I: 1952-1967. Princeton. Stewart, A. (1997): Art, Desire and the Body in Ancient Greece. Cambridge. Sweeney, J., Curry, T. & Tzedakis, V. (eds), (1987): The Human Figure in Early Greek Art. Athens. van Bremen, R. (1996): The Limits of Participation. Women and Civic Life in the Greek East in the Hellenistic and Roman periods. Amsterdam.

35

Chapter 4

Dressing to Please Themselves: Clothing Choices for Roman Women Mary Harlow

In the modern western world the clothed body is integral to self-representation and identity. For us, this expression of identity through dress implies both agency and a range of available options. An individual can use a dress style that reinforces or resists social stereotypes and, even in societies which have strong cultural rules surrounding dress and the body, there is often the facility to subvert social mores or expectations (Twigg 2007: 286). Here, I want to explore how far Roman women could make choices about how they dressed and whether dress could give them access to a mode of self-representation that transcended the dominant literary and visual discourses – in short, can we uncover a level of female agency in the surviving source material. In examining Roman dress we are constrained by the surviving sources, and in discussing Roman female dress we are further hindered by the lack of a female voice in the relevant literature and, arguably, also in the surviving visual and material culture. The question of agency tends to be side-stepped by discussing Roman female dress in terms of how it reflects, reinforces or rejects social and cultural meanings (see for example Harlow 2004b; Olsen 2002). This chapter will be confined to two aspects: the role of sculpture in providing images to emulate and the act of choice in the question of colour.

Looking at shops and market stalls in some of the cities it was striking to see on one side of a shop the traditional dark clothing that was visible on women in the streets while on the other side of the same shop there would be a wealth of colour, style and textiles that seemed completely out of keeping with the public image. Brightly coloured, long, short and quite revealing styles are not uncommon. In response to the question of when women might have the opportunity to wear such fashions, the answer was simply at home with their husbands or at social events where normal practice dictated sexual segregation. The implication is that women could dress to please themselves when among other women of the family. Other glimpses of such ‘choice’ were revealed as we travelled through the country: a woman opening a door to another woman – the one on the street covered with the chador, the one inside the house wearing a brightly coloured shirt and carefully arranged hair; younger women in jeans and T-shirts getting out of cars and quickly ensuring the chador covered all but a glimpse of dress beneath. As a long garment with no fastenings the chador requires at least one hand to keep it closed so restricts movement. However, women manage to hold it closed while carrying bags of shopping, small children – and sometimes both, but in this case the chador was held tight in the teeth. In the cities more than the countryside we also saw blatant subversions of this public female image. These came in the form of very obvious uses of cosmetics or elaborately plaited or dyed fringes just visible under the head covering – and in one case a very elegant turban which covered the head but left the neck exposed. I was merely an observer of this phenomenon, standing outside the social mores of Iranian society and, with no understanding of the spoken or written language, it was hard to interpret these different images. An outsider has little sense of where such dress choices sit in the moral spectrum that dictates female dress in modern Iran. 1 This, it seems to me, is very much the position historians are in when viewing the sources for Roman female dress.

The impetus for the chapter came partly from a trip to Iran made in 2006 organised by Dr Lloyd Llewellyn-Jones. This was my first real close encounter with a society where female dress is prescribed in a way very different from western Europe. The chador, the long black cloak which covers the head and body, is typical dress in both the cities and the countryside (see Fig 3.1 Llewellyn-Jones, this volume). It is not the only form of female dress, many women wear long dark trousers or skirts topped with long sleeved tunics and headscarves but the effect is similar – to hide the shape and any individuality of the female body. Head coverings also vary from the simple head scarf tied under the chin, to the hijab, to niqāb, and the chador. One cannot do justice to the nuances of current Iranian dress from the experience of two weeks travelling in the country but the impressions, particularly of the relationship between public and private dress, raised the issue of how such dress codes might allow us to reconsider approaches to Roman female dress.

1 Acknowledgements: I would like to thank my colleagues Elena Theodorakopoulos, Niall McKeown and Gillian Shepherd for their comments on this paper – it is undoubtedly better for them. For studies of modern Middle Eastern Dress see: Vogelsang-Eastwood 2010: 288-307; Stillman and Stillman (2003); Sabbagh (1996); El Guindi (1999); Arthur (2000).

37

Dress and Identity

The other impetus for this paper is also a side effect of the trip to Iran. Even in hotels that were clearly aimed at the western traveller there was a distinct shortage of full length mirrors. Often the only mirror was a small face glass in the bathroom. As western women we were preoccupied with how we looked, essentially to ensure that we were not transgressing any major rules in terms of general covering and colour (which we did often, despite our best efforts) and the lack of a full length mirror to check appearance was a disadvantage for those of us used to giving ourselves the once-over before we leave the house. The implications for the study of Roman women and female cultus in general are apparent. Roman women had face mirrors of which the reflective ability was variable but nothing like the range or quality we have available now. Today it is possible to view oneself in full length mirrors, in the reflections of windows and in various shiny surfaces; but the same degree of self evaluation was not available to Roman women so how did they assess their overall image? Did they trust the opinions of their slave maids if they had them? How far were they influenced by styles or ways of wearing clothes they saw on other women? Did the visual images of female statuary and women in wall paintings which surrounded them in their daily lives influence a choice of style? Were women constrained by the prescriptive tone of male literary voices, and did they police each other? How far could they exercise choice and still remain within the realms of respectability – if in fact that was their aim? Could women such as prostitutes and courtesans who stood outside the notion of moral society achieve the veneer of respectability through dress? These are all matters that deserve consideration.

to read dress at any interpretive level we are faced with multiple layers of meaning created from sources (visual, literary and material culture) that do not speak the same language. They are self-referential and have connections but none can be elided with the others and there is no easy way of reading them as a single or homogenised system (Warwick & Cavallero 1998: 62). For Roman dress evidence is remarkably varied: visual evidence comes from sculpture, wall paintings, mosaics and some plastic arts such as gold glass, silverware etc. Written descriptions of dress can be found in almost all genres of literature. Material remains are few and fragmentary and mostly come from eastern contexts (see Harris this volume). I have argued elsewhere that the media from which we garner information about Roman dress control what we can say about it (Harlow 2004b and 2007a). However, this approach is arguably too prescriptive and closed in terms of allowing any sense of female expression through dress and adornment. It works on a series of presuppositions based on the source of information and makes genre the ruling component. In this reading the demands of the genre supersede or confound any attempt at accessing even a tentative social reality. For example: women in elegy – not representative of ‘real’ women; women in satire – not representative of ‘real’ women; women in moralising texts – not representative of ‘real’ women; statues and paintings of mythological/divine women – not representative of ‘real’ women; honorific statues/portraits of actual women – still not representative of ‘real’ women because they were controlled by strict demands of public sculpture as a reflection of accepted social morality. Each genre of evidence – visual or literary – will produce a particular image of the dressed or undressed woman that is shared by the audience, who will recognise subtle nuances depending on their own experience, education and gender. For the rest of this paper I want to re-examine some of this material, keeping in mind the issues raised by genre (but not simply surrendering to them) in order to search for areas where Roman women may have shown agency through dress, rather than assuming them to be simply one-dimensional symbols. This is not to suggest a simplistic recreation of ‘real’ women, or to blur realism with social reality, but to allow a more flexible approach to the evidence imagining how women may have made choices about what to wear.

Sources for Roman female dress are essentially of three types: literary images; visual images and some very limited fragments of extant textiles. These elements cannot be jigsaw-ed together to create a unified image, each has to be taken in context. For modern dress, a similar set of interpretive issues was articulated by Roland Barthes in The Fashion System (1967).2 He identified three types of garment: image-clothing, written clothing and real clothing (1985: 3-4). In his reading image-clothing is that represented in photographs and fashion drawings; written clothing is the literary description of such an image; real clothing is the garment itself. He makes the point that these three types of garment do not speak the same language. Not only that, they have internal rules and agenda that control the messages they put across: fashion photography, for instance is a genre in its own right, it follows its own trends and presents an image that conforms to the language of that genre. Written descriptions of clothing likewise conform to the genre in which they are written (Barthes 1985: 12-20, Taylor 2002: 90ff). Actual garments themselves are yet another dimension. While the item of dress is the material reality that image and written clothing mean to represent, it is different again. It is the starting point, model and frame of reference for the first two but it is unique in structure and signals yet more layers of meaning to the viewer. In looking

Roman society was a very visual culture in which, to paraphrase Mary Douglas, the social body controlled the physical body – outward appearance was important and considered to reflect both the social position and the moral character of the individual.3 The dress of both men and women was subject to scrutiny. Male rhetoric on the subject of female dress and toilette, whether in prose or poetry, almost always took a negative view, and any female selfrepresentation that deviated from that considered ‘natural’ Mary Douglas (1970: 65). On the use of dress as a tool in describing character see for example articles by Bender, Heskel and Edwards in Sebesta and Bonfante (eds) (1994); Harlow (2005: 145-9); Langlands (2006: 68-73). 3

Barthes’ Système de la Mode first published in 1967 (trans. M. Ward & R. Howard The Fashion System London 1985).

2

38

Mary Harlow: Dressing to Please Themselves

was deemed deceptive, and thus undermining of the social order. There is a tension between the dominating literary discourses and the material remains: the archaeological and iconographic material suggests that women spent time and care cultivating their appearance, and that a refined elegance was very much part of the identity of the good Roman wife.4

Catia, her long robe conceals all else’ (Sat. 1.2.94-5) – Catia was a notorious adulteress who showed off her lovely legs by wearing a short tunic). The wearing of the stola would have added weight and warmth to the standard dress and may not have been ideal in Mediterranean climate, or for working women. The impracticalities of the stola seem to have been acted upon by Roman matronae as by the late 2nd century Tertullian complained that women no longer routinely wore it (de Pall. 4.9; Croom 2002: 74); but despite this it remained the most symbolic of female garments.

A brief outline of the ‘uniform’ of the upper class Roman women as expressed in the sources and the moral values ascribed to particular items of clothing follows to provide context.5 The basic dress of all Roman women was essentially made up of some form of tunic and mantle. The tunic was in fact the basis of all dress in Rome, male and female; it cut across class, age and gender. It could be as simple as two rectangles of cloth with a gap for arms and head or it could have more complex shaping. Even the basic rectangular shape could allow for sleeves which appear to be gap–fastened from elbow to neck. Later versions appear to have inset sleeves. The essential differences in the basic garment were in terms of the textiles it was made from and its length. It could be plain, coloured or decorated, depending on status and class, and perhaps age of the wearer (for full details see Pausch 2003). For women it was usually bound under the bust, although this must have been tricky to maintain even for the usual activities of a restricted upper class life style. Length also varied, again, primarily dictated by class and rank. Wealthy and upper class women wore tunics that came to the ground and showed very little of the foot. Working women appear to have worn more practical lengths, but the short tunic – above the knee – would indicate slavery or extreme poverty.

Over the tunic (and stola if appropriate) a woman would wear a mantle, a wrap that could cover the head if required and wrap the body for both warmth and hide it for modesty. For a matrona this mantle was the palla. This was made of a large rectangular piece of cloth and could be worn in a number of ways but is most commonly shown as worn over the left shoulder, taken around the back and then either pulled over or under the right shoulder and worn across the body. It reached usually at least to mid-calf, although it could come in different sizes, and required one hand at least to hold it in place if the wearer was in movement. The palla further wrapped the body and kept it hidden from view re-enforcing notions of modesty and submissiveness (Horace Sat 1.2.99; Martial 11.104.7; Aulus Gellius NA 6.12). Moreover it could also be pulled up and used as a veil to cover the head, and if required also pulled across the face (for general descriptions of the mantle see Croom 2002: 87-9; Olsen 2008a: 33-6).6 An upper class Roman woman then was ideally swathed in relatively long, loose drapery which wrapped around, rather than clung to the body. How far the physical dress would restrict body movement, and the ways could women manipulate this relatively restricted wardrobe are issues that need consideration. Wearing the tunic, stola and palla the Roman matron would have had relatively restricted movement. Hands and arms would have been needed to keep material in place and physical movement constrained by the length and relative bulkiness of the layers. Dress would have been made up of perhaps one light layer of

A married woman (matrona) could wear a garment over the tunic known as the stola. This resembled a pinafore with shoulder straps and reached to the ground, again belted under the bust (Olsen 2008a: 27-33; Croom 2002: 73-6; Wilson 1938: 155-62). Upper class women wore it long, perhaps with a band around the lower edge (instita – see discussion, Olsen 2008a: 30; Sebesta 1994: 49; contra – instita as straps - see Beiber 1977: 23; Croom 2002: 74). The stola (perhaps also synonymous with the phrase longa vestis – see Olsen 2008a: 28-30; Sebesta 1994: 49) could only be worn by a woman who was in iustum matrimonium (legitimate marriage) so it immediately signalled a woman who was a citizen, and of a certain status. It therefore embodied all the virtues of the ideal wife: modesty, chastity, faithfulness and honourable behaviour, and was used as shorthand for such by some authors (e.g Ovid Ars Am. 1.31-2; Martial 1.35.). The length of the stola and the fact it was worn over another garment effectively hid the body from view as Horace and other writers complained (‘…in a matrona one can only see her face, for unless she be a

Veiling is a problematic issue for historians of Roman dress. The few references to veiling which imply that it was the norm for women to cover their heads when out in public are at odds with the many statues and portrait busts displaying unveiled women and illustrating elaborate hair styles. Wall paintings of outdoor scenes also show women with uncovered heads but the status of such women is harder to discern. A piece of evidence often quoted to support the notion of female veiling is a story told by Valerius Maximus (1st century AD) about C. Sulpicius Gallus, consul in 166BC. Gallus apparently divorced his wife for going outside with her head uncovered (capite aperto foris). His reason for this was that she had exposed to everyone what only his eyes should see (Val Max. 6.3.10). It should be noted that this anecdote is part of a section on severity and is followed by two more short stories of wives being divorced: one for talking to a freedwoman (libertina vulgari, so perhaps a prostitute) in public and the other for attending the games without her husband’s permission. The moral here is: ‘while women were thus checked in the old days, their minds stayed away from wrongdoings’ (Val Max. 6.3.11-12). The idea that morals were better and women were better behaved in ‘the good old days’ is a topos of Latin literature. There are other mentions of veiled women but they are few and we should be careful of creating social norms from such a paucity of evidence. For further discussion see Macmullen (1980); Sebesta (1994: 48-9); Olsen (2008a: 33-4); D’Angelo (1995: 131-64). 6

On toilette in general see D’Ambra (2007: 111-27). For deceptive nature of female toilette see Richlin (1995: 213) and her references; on more positive reading through iconography see Shumka (2008: 173-91). 5 Detailed discussion of particular items of female dress can be found in a number of recent publications: see for instance, Harlow (2004b), Olsen (2002; 2008a: 25-36); Croom (2000: 73-116); Sebesta (1994: 46-53); Cleland, Davies & Llewellyn-Jones (2007). 4

39

Dress and Identity

underwear, with a wool or linen tunic over that, belted somehow to contain the width and retain modesty if not fully sewn up both sides; a matrona might have worn a wool stola over the top. These layers would not have sat easily on top of each other, there would have been bunching of fabric and shifting of layers. The palla added yet another layer. Hands had to be occupied holding dress in place when in movement, and steps needed to be short and controlled in order to keep the whole ensemble in some sort of containment, and avoid tripping up. Roman women must no doubt have learned how to wear such layers with ease in the same way that Iranian women wear the chador but similarly such dress also meant that when in public women needed to concentrate literally on ‘holding themselves together’. When outside domestic confines then, the respectable female body was physically covered and metaphorically wrapped in a series of social codes expressed in her dress and her movement.

body language and dress, and appears to be embodied in these representations. These statue types used the wrapped, draped and usually veiled female body as their starting point. The pudicitia pose usually has one arm wrapped in the mantle bent across the front of the body. The other arm, bent at the elbow, is raised and the hand, with index finger raised, touches the mantle as it passes along the side of the head or face. This gesture implies a readiness to pull the mantle across the face (in the act of full veiling) if required. In these images the head, or certainly the eyes, have a downward cast. The so-called Herculaneum women types are likewise heavily draped, usually with one hand held loose down the side of the body, often hidden by the palla, while the other, bent at the elbow and wrapped inside the mantle, pulls the cloth across the body. These figures are usually, but not invariably, veiled. The only parts of the body visible in these models are the face, the hands and the feet. Scholars have noted several things about these figure types: the inclined head and lowered gaze suggest modesty – the figure is not making eye contact with the viewer. This modesty is re-enforced by the body language of the figures: the arms are set in defensive, postures and the force and fall of the drapery presents a closed body image, again suggesting vulnerability; but also unassailability and sexual integrity. (Davies 1997: 102-4; Trimble 2000: 65-6).7

Respectable women out in public had an image to maintain and they had various ways of assessing that image, in the absence of adequate mirrors. The most obvious way would no doubt be viewing other clothed women but they would also have seen visual images of clothed (and unclothed) female bodies wearing dress in very particular ways, mainly in statuary. The question of how far ‘real women’ would have internalised the public images they saw around themselves is debatable. Jane Fejfer is in no doubt:

The images are ambiguous: they embody the paradox of modesty on display. Modesty, of course, is a virtue that should eschew display. The figures appear at first glance to conform to the ideal of the modest matrona but they are not quite the visual version of Seneca’s model wife who if she ‘wants to be safe from the lust of a seducer must go out dressed up only so far as to avoid unkemptness’. Any communication, even with relatives, should be with eyes cast down and cause a blush (Contr. 2.7.3). In Seneca, the contrast is with the woman who is the physical embodiment of impudicitia – one who goes out with a ‘face made up to look utterly seductive, naked hardly less obvious than if you had taken off your clothes’ and giving advance warning of her shamelessness in ‘the way she dresses, the way she walks, the ways she talks, her appearance’ (Contr. 2.7.4). For Seneca, it seems, a Roman husband could have a dull wife or an adulterous one, with little middle ground.

However, common to all the different statuary representations of Roman women is the fact that they were not ‘real’. This is because the costumes and the way in which they were draped were characteristic of images of goddesses rather than normal women. The Roman viewer may not have associated a particular statuary type and its variants with the goddess whom it originally represented, but its costume was probably clearly distinguishable from the everyday dress which a woman wore during her appearance in public (Fejfer 2008: 344-5). While Fejfer accepts that the tunic, stola and palla were the central garments in the Roman woman’s wardrobe, she argues that the way they are depicted in sculpture would have borne little relation to the garments actually worn by women. Female statuary at Rome is divided into six main types: the ‘Ceres’ type, ‘Pudicitia’ type, ‘Small’ and ‘Large Herculaneum’ women types and the so-called ‘shoulder-bundle’ and ‘hip-bundle’ types (Fejfer: 2008: 335). Three of these styles embody an ideal of the virtuous and virtuously dressed woman: the so-called ‘Pudicitia’ type and the ‘Small and Large Herculaneum women’ (see also Beiber 1977: 148-62; Davies 2002; Trimble 2000: 5665 on the role of such statues as visual representations of the Augustan moral codes). Pudicitia is a complex concept. For women it meant modesty, sexual purity, chasteness within marriage, and repelling the advances of anyone, especially males, from outside the family (see Langlands 2006). It was a virtue that needed to be visible in manner,

The statue types reflect an ambiguity that is not present in Seneca’s declamation. The heavy but elegant drapery on the statues suggests wealth, while the modelling on some of the figures accentuates breasts and pubic area, so while the wrapped body suggests sexual integrity the modelling still manages to stress sexuality. Several modern authors have commented on the fact that the apparent veiling gesture could also be read as unveiling and inviting the viewer to gaze further with the promise of more (see for example, Davies 2002: 236; Smith 1991; 84-5; Trimble 2000: 66). In these statues the shape of the body is often revealed in surprising ways given at least two and often three layers of clothing are apparently worn. Both the tunic and the stola There is extensive literature on these statue types see also: Beiber (1962, 1977); Smith (1991, 1998). 7

40

Mary Harlow: Dressing to Please Themselves

can be depicted of a very light material, revealing breasts and navels, thus in theory contradicting the ideals of moral uprightness associated with it (Fejfer 2008: 335, 345).

and, even within the relatively limited repertoire of styles, Roman women had access to a range of choice of material and colour. A text that gives some insight into the choices and reasons for those choices is Ovid’s Ars Amatoria 3. This is a piece of didactic writing which seeks to instruct, in the voice of the praeceptor (teacher), puellae (girls, young women) on how to appear in order to attract lovers. Ovid says the poem is a response to a demand from the girls to write a book for them to parallel the earlier two books addresses to men (Ars Am. 2. 745-6). Where the first two books taught the rules of the elegiac game to men, this was to teach them to women.9 As in more traditional didactic literature there are plenty of references to mythological figures but Ovid writes not in epic metre, the usual form for didactic verse, but in elegiac (Gibson 2003: 8). Gibson places the Ars Amatoria in the ‘erotodidactic’ tradition where the author assumes the role of a woman, often in the persona of a lena (procuress), instructing other women with the audience placed in the role of eavesdroppers (Gibson 2003: 18 - 21). It is a carefully constructed and manipulating literary work in which cultus is expounded in positive terms.10

However, Roman women, at least in urban contexts, were surrounded by such images and the statue types were replicated across the empire. Using idealised sculpted images as references for clothing is problematic and while I am not suggesting that statues act in the same way as fashion plates, such standardised images must have created an idea of the overall impression the dressed woman was meant to make in draped and wrapped garments, even if the version of the sculpted ideal could not be achieved in reality (Beiber 1977: 153 and figs 713-5 on the difficulty of replicating the styles). It may be, however, that the poses, particularly of the Herculaneum women types, did reflect some of the realities of wearing such garments: it was necessary that both hands and upper arms were held tight to the body to hold it all together, focussing the wearer inward. Images and social reality, in both literature and visual culture, tend to shape each other (see Trimble 2000: 64-67; D’Ambra 1996; Stewart 2008: 97-101). Sculptors followed artistic precedents but they also created images that spoke to a contemporary audience, they too may have looked at women wearing the tunic, stola and palla and used reality as a model. This is, of course, a circular argument. However, women did wear loose and draped clothing and it had to be held together in particular ways to make it function, and there are only so many ways of doing this and achieving an acceptable public look. We cannot know how the female viewer related to the images that surrounded them, if they laughed at them as preposterous male fantasies of the clothed (and unclothed) female bodies or if they sought to emulate them, but they are images that may have given women who lived without the benefit of full length mirrors an idea of what the dressed body looked like. We should allow women some room for agency rather imagining them conforming to the stereotypes. Ancient writers were clearly anxious about female intentions and the potentially subverting nature of manipulating drapery. As noted above, the wrapped body could also be unwrapped. The very basic nature of female dress with its fundamentally simple shaping allowed women a certain freedom about what they chose to outline, expose or hide. Women had the power to manipulate their drapery within the relatively limited wardrobe repertoire. Roman satirists and elegists talk about artfully arranged clothing that could hide flaws or show off more attractive body parts (Tib. 1.6.18; Propertius 2.22A 8; Martial 8.68; Juvenal 6.260). The idea that clothing not held together will easily come adrift is also a motif used to characterise mental disarray or lack of control (cf. Catullus 64. 63-70 on abandoned Ariadne), suggesting that it was easy to lose clothing if care was not taken.8

The relevant section starts with an introduction (101-34) which claims that while in the old days young women did not cultivate their looks, now in the new (Augustan) Rome a different approach to cultivated beauty is allowable, in opposition to the rusticity (rusticias) of the past. Ovid is taking a very particular view here; instead of engaging in the traditional Latin trope of the good old days he actively welcomes the artistry and artifice of new Augustan Rome: ‘Let ancient times delight others, I congratulate myself I was not born until now’ (l. 121-2). In Book 3 Ovid suggests that women can now cultivate attractiveness; they should create themselves as a work of art that looks natural.11 However, even at the start the praeceptor warns against excess: while the ‘vast wealth of the conquered world’ (l.114) is available puellae should not wear exotic gems or dresses weighed down with gold thread (l.129-33). Ovid compares excess which ‘often repels’ with elegance which attracts (munditiis capimur l.133). In this area he is in line with fellow elegists (Richlin 1995: 187; Watson 1982: 239). Munditia carries implications of refined neatness, restraint and cleanliness and it is this tone that underlies all the subsequent advice coupled with the general principle that any choice of hairstyle or colour of dress should suit the There is an extensive bibliography on Ovid and the Ars Amatoria (see Richlin, Watson and Downing on bibliography) and for this paper I have made great use of Roy Gibson’s 2003 introduction and commentary on Ars Amatoria 3. 10 Ovid’s notion of cultus is at variance with that of Propertius and Tibullus. Where Ovid plays up elements of constructed beauty and emphasises artistry and technique in the maintenance of good looks, the others praise natural beauty and decry artificiality. See Watson (1982: 238-9). 11 The whole section of ll. 109-130 is an unpicking of ideas about the Augustan Golden Age, here stressing the advantages of the modern (Watson 1982, particularly p. 241). The idea of cultus is stressed: ‘Care will give good looks: looks neglected go to waste (l.105). This is direct contrast to Propertius who denigrates any form of artificial aid (Watson 1982: 238). See also Downing (1999); Gibson (2003: 139-48); Richlin (1995: 188). 9

The ability to enact choice is one of the markers of agency For a similar contrast in art see example of half-dressed maenad paired with dressed matron from capitals from House of the Figured Capitals, Pompeii in Clarke (2003: 248-51). I thank Elena Theodorakopoulos for the Catullus reference. 8

41

Dress and Identity

shape and complexion of the individual. Ovid starts with hairstyles. Hair should be controlled but as no single style will suit everyone, women should choose one to suit the shape of the face, for example, an oval face suits a centre parting, a round face the nodus style (l.135-40).12 Ovid lists a range of styles: loose, braided, waved, tightly bound, and an artfully neglected look, but privileges none. Some styles have mythological associations. For instance braided hair is like that of Diana the huntress (l.143) – this reference would assume that women might actually look at visual and sculptural images and seek to emulate them in some fashion in the same way that we see the nodus style of Livia emulated by other women in sculpture (for examples see Bartman 1999).13

this advice gives weight to Ovid’s mock-serious intention of writing a ‘handbook’ and gives a clear impression that deliberately choosing to enhance physical appearance could be viewed as positive.15 It also suggests women could exercise choice not only in whether to engage with cultus or not but also to make minute decisions about all aspects of their appearance depending on their budget and, presumably, social status. Textiles of many colours would have been available on the Roman market and images of women dressed in coloured clothing were visual as well as literary. Some of the surviving Herculaneum women statues show traces of colour (Trimble 2000: 43), wall paintings also survive with coloured clothing but the best examples are in Egyptian mummy portraits from the Roman period. Both time and environment of survival may mean that the original brightness, tone and even hue are not accessible to us – but the point is that women could make choices. In the world outside Ovid’s poem women of all classes could choose to wear only colours that suited their complexions (although even this is also a choice defined by cultural codes and different tastes) or they could have chosen to demonstrate wealth and status with the use of expensive dyes and gold, or they may have chosen a colour that pleased them regardless of its moral connotations in literary texts. Likewise, while Ovid might stress the use of wool, plenty of other textiles were available on the Roman market. Silk comes in for a particularly bad press from the moralisers – it was expensive, exotic and eastern and could be very sheer and thus revealing – all elements that were easily exploited in satire and rhetorical exaggeration (e.g. Seneca De Beneficiis 7.9.5). Coan silk, a particularly transparent weave originating in the island of Kos, came in for particular diatribes and inferences (Seneca ad Helviam 16.4; Propertius 2.1.5-6; Horace Satires 1.2.101-2; Olsen 2008a: 14). Linen was also used for clothing and a silk and linen mix was known. Cotton was less often used for clothing. Women living in Rome would also have seen the mixed checked weaves common in Gaul, light linens from Greece and Egypt, decorated and brocaded silks from the east. We can tell little about the clothing shopping habits of any class of Roman woman but we should not imagine that in their private lives at least they were constrained by the literary proscriptions of male authored literature. Women may indeed have chosen fabrics that suited their body shape regardless of comment (cf. Ovid Ars Am. 3. 263-74).

The poem next moves on to give advice on dress and in particular on colours. The topic is again introduced by a rejection of excess succinctly phrased in terms of gold and purple – both of which were synonymous with luxuria (l. 169-73). Next follows a catalogue of colours including various shades of blue (colours of the sky); grey, yellow, grey-green or sea-green, amethyst purple, rose-white, dark brown of chestnuts and lighter brown of almonds, ending with ‘as many colours as there are spring flowers’ (l.17489). The list highlights that a range of choice was available but the context of the poem which Ovid’s immediate audience would have recognised, suggests they might only be appropriate for some women. Ovid’s catalogue has similarities with the one given by Plautus: azure, saffron, sea blue, buttercup, walnut, wax (Epidicus 229-35) which is also about courtesan preferences. Both lists make the point that colour was possible and desirable in female dress (Sebesta 1984: 65-76; Olsen 2008a: 11-13). Ovid subtly stresses the idea of moderation, or at least restraint, in the use of colour by making the point that it is wool which holds such dyes (l.187).14 Wool is very much the textile of the virtuous woman in contrast to expensive silk or exotic more diaphanous, figure hugging or transparent materials. In Ovid’s own time Augustus was said only to have worn homespun wool worked by women of his family (Suetonius, Aug. 73) – and even if this is a highly disingenuous statement the inference of the association of the good wife with wool working is clear. As with hairstyles, Ovid’s praeceptor suggests that women should choose colours that complement their skin tones: pale skinned women suit darker colours and those with darker skins should wear white (l.189-92). There is no sense here that women should wear a rainbow of available colours, but rather chose one that suits. The conservative nature of

While excerpts of text should not be considered outside the whole or taken beyond the bounds of genre sections of Ars Amatoria give a voice to advice that is not beyond the realms of common sense and we might accept as an insight into the range of choices a woman might have available and the criteria by which she might decide. This is a rare

There is a large bibliography on Roman hairstyles but for an introduction see: Olsen (2008a: 70-76); Bartman (2001); Croom (2002: 98-107). Giordano and Casale (1991, 2nd ed rev. Garcia y Garcia). See Myerowitz Levine (1995) on gendered meaning of hair in ancient Mediterranean. 13 Cf Properius 1.3 where the sleeping Cynthia is likened to a series of known paintings and sculptures. Downing (1999: 235-51) argues that in Ars Am. 3 Ovid’s preceptor is transforming a woman into a statue through cultus, in contrast to Metamorphoses where Pygmalion turns a statue into reality. 14 On the range of ancient dyes available see Cardon (2007); Sebesta (1984: 65-76); Forbes (1965); Wilson (1938: 6-10). 12

Bradley (2009: 182-7) argues that the colours used by Ovid in Ars Am. are yet another layer of literary games: colour allows women yet another layer of deception. The names of colours recall a series of associations, some mythical, which create yet more imaginative outlets for the informed reader. Ovid makes a case for colour but warns his audience that this could still be misleading. 15

42

Mary Harlow: Dressing to Please Themselves

moment in Latin literature and while it can hardly be judged neutral, it is at least not overtly stressing the negatives in the way women dress. The Ars Amatoria pretends not to be addressed to respectable upper class women, and indeed, explicitly warns them off: ‘keep away you slender fillets, emblems of modesty, and the long skirt (instita longa) that hides the feet’ (Ars Am. 1. 31-2) but it is not too much to imagine that the advice it contains might reflect the conversations of upper class salons or between mistress and slave. If the puellae to whom the advice is addressed followed it they would not have looked like the ‘painted trollop’ of satire, on the contrary given the conservative nature of the advice they would have much in common, in outward appearance at least, with the Roman matron. Presumably this eliding of female imagery was part of the subtlety of Ovid’s writing in contrast to the Amores or other sections of Ars Amatoria (see Gibson 2003: 25-35 on the relationship between Ars Amatoria and lex Julia). Upper class men expected their wives to look elegant, despite rhetorical comments to the contrary. In order to maintain their appearances and conform to social mores upper class women must have mastered the ability to walk the fine line between excess and moderation in their appearance and this involved time for work on their dress, their hairstyles and their complexions. They had the economic power to choose finest wools and a range of dyes and if they could avoid the pitfalls of over-adornment, they might by default have made the choices reflected in Ovid’s writings. Conversely, of course, Ovid’s advice could have allowed the puella to cultivate the look of the refined and elegant matrona thus empowering her to move in upper class circles. This would confirm all traditional Roman prejudices that any cultivation of the female self was duplicitous and counterfeit.

immoral behaviour, the archaeological record suggests that the production of cosmetics and instruments to apply them were commonplace (Stewart 2007 and this volume; Olsen 2009). Roman women do not appear to have felt constrained by the rhetoric of male authored literature, on the contrary they seem to have actively engaged in grooming activities which gave them control over their appearances and allowed them to cultivate self presentation. As observers in Iran we saw the public image of women, with only glimpses of dress habits in the private domain; for Rome we lack even those glimpses unless we are prepared to allow some sense of social reality to reach us through the dominant male ideologies in the literary and visual culture. To return to the methodological issue exemplified by Barthes: there is a tension between the written and visual sources and a question of how far we can exploit them to access a sense of homogenised social reality. However, if we cannot find ways to do this we are in danger of turning the dressed women of Rome into entirely fictional personae. Roman social codes did define dress in particular ways and produced a very powerful idea in literature and art of what the upright moral woman should wear but even within this idea there is no single narrative and no single over-arching idea of identity. Evidence demonstrates that within a relatively limited repertoire of styles a range of choice existed in terms of colour and textiles, and also that colour was very much part of the Roman visual world. The constant refrain against female adornment throughout the time span of the Roman empire suggests that women certainly were exploiting the market that was available to them despite any disquiet it might cause their menfolk. Roman writers were adept at manipulating the image of the dressed (and undressed) woman to suit their agenda and presumably women were equally as adept at manipulating their own draped clothing to suit their agenda, or at least give them power over their immediate social space.

The archaeological and iconographic record offers a picture that also differs from the dominantly pejorative literature. Material remains and depictions of women’s toilet items (mundus muliebris) demonstrate that arranging appearances was very much part of a woman’s life and presumably far more part of the lives of wealthy and upper classes who had time to expend on it (D’Ambra 2007: 111-28 includes toilette in a chapter on ‘Women’s work’; Shumka 2008; Swift and Stewart this volume). Crafting a suitable appearance must have been an integral part of female identity in the Roman world, even if that identity was closely associated with that of their male kin. While the standard female garments of tunic and mantle could be worn in such a way as to establish very generalised identities: woman as wife, mother, widow etc., a woman could personalise her identity by refining this look – in making choices about which elements to adorn (or not). The cultivation of an elegant appearance took time and money and this cultivation was celebrated on funerary monuments (see Swift this volume; Shumka 2008). A well groomed look was something for women to exhibit and for husbands to be proud of, but the trick was, as Ovid recommends, to make it look as natural as possible. While satirists may have depicted such cultivation as nothing more than trickery that concealed not only physical deformity but also encouraged

Bibliography Arthur, L. (2000): Undressing Religion: Commitment and Conversion from a Cross- cultural Perspective. Oxford & New York. Barthes, R. (1985): The Fashion System. Trans. M. Ward & R. Howard. London. Bartman, E. (1999): Portraits of Livia: Imaging the Imperial Woman in Augustan Rome. Cambridge. Bartman, E. (2001) ‘Hair and Artifice of Roman Female Adornment’, American Journal of Archaeology 105: 1-25. Bieber, M. (1962): ‘The copies of the Herculaneum women’, Proceedings of the American Philosophical Society 106: 111-34. Bieber, M. (1977): Ancient Copies. Contributions to the History of Greek and Roman Art. New York. Bradley, M. (2009): Colour and Meaning in Ancient Rome. Cambridge. Cardon, D. (2007): Natural Dyes. London. Cavallaro, D. and Warwick, A. (2001): Fashioning the

43

Dress and Identity

Frame. Boundaries, Dress and the Body. Oxford & New York. Clarke, J. R. (2006): Art in the Lives of Ordinary Romans. Berkeley. Cleland, L., Davies, G. & Llewellyn-Jones, L. (2007): Greek and Roman Dress from A-Z. London. Cokayne, K. (2003): Experiencing Old Age in Ancient Rome. London. Corbeill, A. (2004): Nature Embodied: Gesture in Ancient Rome. Princeton & Oxford. Croom, A. T. (2000): Roman Clothing and Fashion. Stroud. D’Angelo, M. R. (1995): ‘Veils, Virgins, and the Tongues of Men and Angels: Womens’ Heads in Early Christianity’, in H. Eilberg-Schwartz & W. Doniger (eds), Off With Her Head: The Denial of Women’s Identity in Myth, Religion and Culture, 131-64. Berkeley & London. D’Ambra, E. (1996): ‘The calculus of Venus: nude portraits of Roman matrons’, in Natalie Boymel Kampen (ed.) Sexuality in Ancient Art. Cambridge. 219-32. D’Ambra, E. (2007): Roman Women. Cambridge. Davies, G. (1997): ‘Gender and body language in Roman art’, in T. Cornell and K. Lomas (eds). Gender and Ethnicity in Ancient Italy, 97-108. London. Davies, G. (2002): ‘Clothes as sign: the case of the large and small Herculaneum women’ in L.J. Llewellyn-Jones (ed.) Women’s Dress in the Ancient Greek World. 227-42. London & Swansea. Douglas, M. (1970): Natural Symbols. London. Downing, E. (1999):’Anti-Pygmalion: The Praeceptor in Ars Amatoria, Book 3’, in J. Porter (ed.) Constructions of the Classical Body, 235-51. Ann Arbor. Edmondson, J. & Keith, A. (eds) (2008): Roman Dress and the Fabrics of Roman Culture. Toronto. Eilberg-Schwartz, H. & Doniger, W. (eds), (1995): Off With Her Head: The Denial of Women’s Identity in Myth, Religion and Culture. Berkeley & London. El Guindi, F. (1999): Veil. Modesty, Privacy and Resistance. Oxford & New York. Fejfer, J. (2008): Roman Portraits in Context. Berlin & New York. Forbes, R. J. (1965): Studies in Ancient Technology vol. III. Leiden. Gibson, R. (2003): Ovid: Ars Amatoria Book 3. Cambridge. Giordano, C. & Casale, A. (1991): (2nd ed. rev. Garcia y Garcia). Perfumes, Unguents and Hairstyles in Pompeii (Profumi, Unguenti e Acconciature in Pompei Antica). Rome. Harlow, M. (2004a): ‘Clothes maketh man: power dressing and elite masculinity in the late Roman world’, in L. Brubaker & J. Smith (eds) Gender in the Early Medieval World. East and West 300-900. 44-69. Cambridge. Harlow, M. (2004b): ‘Female dress, 3rd – 6th centuries: the messages in the media’. Antiquité Tardive 12: 203-15. Harlow, M. (2005): ‘Dress in the Historia Augusta: the role of dress in historical narrative’, in L. Cleland, M. Harlow & L.J. Llewellyn-Jones (eds) The Clothed Body in the Ancient World. 143-53. Oxford. Harlow, M. (2007a): ‘The impossible art of dressing to please: Jerome and the rhetoric of dress’, in L. Lavan,

E. Swift & T. Putzeys (eds) Objects in Context, Objects in Space: Material Spatiality in Late Antiquity. 531-47. Leiden. Harlow, M. (2007b): ‘Blurred visions: male perceptions of the female life course – the case of Aemilia Pudentilla’, in M. Harlow & R. Laurence (eds) Age and Ageing in the Roman Empire. Journal of Roman Archaeology Supplementary Series 65: 195-208. Langlands, R. (2006): Sexual Morality in Ancient Rome. Cambridge. Macmullen, R. (1980): ‘Women in public in the Roman empire’, Historia 29: 208-18. Myerwitz Levine, (1995): ‘The gendered grammar of ancient Mediterranean hair,’ in H. Eilberg-Schwartz & W. Doniger (eds), Off With Her Head: The Denial of Women’s Identity in Myth, Religion and Culture, 76-130. Berkeley & London. Olsen, K. (2002): ‘Matrona and whore: clothing of women in Roman antiquity’. Fashion Theory 6.4: 387-420 Olsen, K. (2008a): Dress and the Roman Woman. London. Olsen, K. (2008b): ‘The Appearance of the Young Roman Girl’, in J. Edmondson & A. Keith (eds), Roman Dress and the Fabrics of Roman Culture, 139-57. Toronto, Buffalo & London. Olsen, K. (2009): ‘Cosmetics in Roman antiquity: substance, remedy and poison’, Classical World 102.3: 291-310. Pausch, M. (2003): Die römische Tunika. Verlag. Richlin, A. (1995): ‘Making up a Woman. The Face of Roman Gender’, in H. Eilberg-Schwartz & W. Doniger (eds), Off With Her Head: The Denial of Women’s Identity in Myth, Religion and Culture, 185-213. Berkeley & London. Sabbagh, S. (1996): Arab Women: Between Defiance and Restraint. Northampton MA. Sebesta, J. & Bonfante, L. (eds) (1994): The World of Roman Costume. Wisconsin. Shumka, L. (2008): ‘Designing women: the representation of women’s toiletries on funerary monuments in Roman Italy’, in J. Edmondson & A. Keith (eds) (2008): Roman Dress and the Fabrics of Roman Culture, 172-91. Toronto. Smith, R. R. (1991): Hellenistic Sculpture. London. Smith, R. R. (1998): ‘Cultural choice and political identity in honorific portrait statues in the Greek East in the second century AD’, Journal of Roman Studies 88: 5693. Stewart, P. (2008): The Social History of Roman Art. Cambridge. Stillman Y. & Stillman, N. (2003): Arab Dress: A Short History from the Dawn of Islam to Modern Times. Leiden. Taylor, L. (2002): The Study of Dress History. Manchester. Trimble, J. (2000): ‘Replicating the body politic: the Herculaneum women statue types in Early Imperial Italy’. Journal of Roman Archaeology. 13: 41-68. Twigg, J. (2007): ‘Clothing, age and the body: a critical review’. Ageing and Society 27: 285-305. Vogelsang-Eastwood, G. (2010): ‘Iranian regional dress’

44

Mary Harlow: Dressing to Please Themselves

Watson, P. (1982): ‘Ovid and Cultus: Ars Amatoria 3 11328’. TAPA 112: 237-44. Wilson, L. (1938): The Clothing of the Ancient Romans. Baltimore.

and ‘Iranian urban dress’ in G. Vogelsang-Eastwood (ed.) Berg Encyclopedia of World Dress and Fashion vol 5: Central and Southwest Asia 288-307, Oxford. Warwick, A. & Cavallaro, D. (2001): Fashioning the Frame: Boundaries, Dress and the Body. Oxford & New York.

45

Chapter 5

The Archaeology of Adornment and the Toilet in Roman Britain and Gaul Ellen Swift

Written sources from the early Roman period confirm that, in Roman culture, an elite woman would normally spend part of the morning in an elaborate toilet, including bathing, dressing the hair, and applying perfume, makeup, and jewellery (Balsdon 1962: 261; Fleming 1997: 15, 2831). Maria Wyke (1994) examines this practice in more detail, showing that body care and the use of cosmetics and jewellery were important in defining an elite woman and delineating her role as a consumer and exhibitor of the wealth of her husband. Only elite women could afford the time and expense necessary to an elaborate toilet; though men were also expected to be groomed to some extent, spending too much time on one’s toilet, and using cosmetics, was perceived as effeminate and undesirably oriental in a man; toilet and adornment was therefore both a marker of wealth, and of womanhood and femininity.

of toilet and adornment appropriate to Projecta’s role as a married woman, for instance a scene of Projecta attended by servants, and looking in a mirror while a hair-pin is placed in the hair. The self-referential decoration points to the use of the casket as a receptacle for toilet items and jewellery (see Elsner (2003); Swift (2009: 125-129) for a further interpretation of the casket). Further instances of similar iconography can be cited, for instance the 4th century palace wall paintings decorating the Imperial palace at Trier, in which a woman’s adornment through jewellery exemplifies her elite feminine status (see Warland (1994) for a further discussion of these and other scenes). Though it is important to stress that this type of presentation of a woman’s elite role did not necessarily extend to the more westerly parts of the Empire such as Britain and Gaul beyond its Imperial capital, it can certainly be suggested that a minority of women in these provinces had an elite feminine identity constructed for them at death in this way. A number of grave monuments from major Roman towns in Britain and Gaul, similar to those found in Italy, may be used as a clear illustration of this. Though the themes of the toilet and adornment expressed in the iconography of these objects do not occur with great frequency, they suggest a representation of ideal status for some women very similar to that described above.

This representation of feminine identity through the process of the toilet is clearly shown on a number of grave monuments from Italy showing toilet scenes. They include a 1st century funerary altar from Aquileia in the northern part of Italy (Hope 2001: 111 and plate 4) showing a woman holding a mirror, and two sarcophagi (dating to the second half of the 3rd century AD) from Arezzo and Ostia respectively, that show a woman, presumably the deceased, who sits in a chair attended by servants, who are doing her hair and bringing to her objects that are to be used in the toilet (Amedick 1991: 108). Natalie Kampen discusses the theme briefly in more general terms, using evidence from the grave monuments of Asia Minor, where mirrors are included alongside other objects signifying feminine roles such as spindle-whorls and cosmetics. She also makes the association with women’s self-presentation and their appropriate roles in relation to men (Kampen 1996: 22). The persistence of the theme through several centuries is evident in a representation of a woman attended by servants, and engaged in her toilet, on the 4th century Projecta casket, found on the Esquiline Hill in Rome. The iconography of the Projecta casket in particular has been studied by many previous commentators, and it has been suggested that it is a marriage gift, showing the married couple, Projecta and her husband. It also includes the scenes

Funerary relief sculpture with toilet and adornment scenes There are several variations of the basic theme in 2nd and 3rd century relief sculpture from Britain and Gaul (see Fig. 5.2 and Figs 5.3 and 5.5 for details). Several show the iconic scene of a woman looking in a mirror attended by servants. Representations of women holding mirrors and other toilet items, but without servants, are slightly more common; the other objects held include caskets (perhaps jewellery caskets), a long-necked flask, a basket, and other indeterminate objects (representations also occur of women holding mirrors without other objects, which could have a different symbolism). Given the more explicit scenes of the toilet that include servants, it seems likely

47

Dress and Identity

that representations of women holding more than one item associated with adornment and the toilet (especially those that include a toilet casket) are also intended to represent the process of the toilet, i.e. the process of construction of a woman’s elite feminine identity. The dating of this type of scene appears to be concentrated in the 2nd and 3rd centuries AD, also evident from the two sarcophagi lids from Aquileia mentioned above.

documented (Philpott 1991). However, beyond occasional considerations of ‘mirror’ burials (see for example Philpott 1991: 182) toilet assemblages in graves have not received much attention. Crummy, Crummy and Crossan (1993: 273) make the important connection between texts which refer to a woman’s casket of toilet articles and jewellery, and examples found in cemeteries at Colchester of caskets containing jewellery. Analogous examples of caskets containing jewellery occur in grave contexts in Italy, as well as grave assemblages (where sex is identifiable all female) containing both mirrors and jewellery, with a date range from the 1st to the 3rd century (Martin-Kilcher 2000: 64-67 and table 7.1). This suggests that Crummy et al. are correct to link the burials at Colchester to Roman cultural practices.

Grave assemblages Given the existence of this type of iconography in funerary monuments, it is perhaps likely that objects in grave assemblages might be used in a similar way. Pearce (2000) gives a useful overview of current approaches to burial evidence. General questions of social status and religious belief as visible in burial practice have been studied (e.g. Jones 1984; Alcock 1980), and the occurrence of particular items of material culture in graves has been well-

Current approaches stress that burial ritual is linked to the construction of an identity at death for the deceased that may show the ideals of a society rather than reality (Pearce 2000: 5). An integrated approach which includes the study

Fig. 5.1: Relief sculptures depicting female toilet from the western Roman provinces. Scene Woman sitting in chair, looking in a mirror, attended by servants Woman holding a mirror, attended by servant/s Woman holding a mirror, attended by servant/s Woman holding a mirror, attended by servant/s Woman holding mirror and casket Woman holding mirror and basket Woman holding mirror and long-necked flask Woman holding mirror and unidentified object Woman holding mirror and unidentified object

Example

Date

Reference

Relief sculpture, Neumagen

2nd or 3rd century (Hatt 1986: 10)

Espérandieu 1908: vol. 2 no. 5142

Relief sculpture on tombstone, Chester

2nd half of 3rd century (Henig 2004: 25-6)

Wright & Richmond 1955: cat no. 120 plate XXXI Henig 2004: cat. no. 74

Relief sculpture on sarcophagus lid, Agen

Espérandieu 1908: vol 2 no. 1253

Relief sculpture, Musee de Luxembourg (unprovenanced)

Espérandieu 1913: vol. 5 no. 4156

Relief sculpture, probably from tombstone, Vanvey-sur-Ource Relief sculpture on tombstone, Bordeaux Fragment of relief sculpture, Autun museum (locally provenanced stone)

Espérandieu 1911: vol. 4 no.3407 Espérandieu 1913: vol. 5 no. 1157

Antonine period Braemer (1959:45)

Espérandieu 1910: vol. 3 no. 1952

Relief sculpture, Bordeaux Antonine period. Braemer (1959: 43)

Relief sculpture, Bordeaux

Woman holding mirror only

Relief sculpture, Bordeaux

Woman holding casket and finger-ring

Relief sculpture on funerary monument, Arlon

Second century, Lefebvre (1975, 48-50

Relief sculpture, Lyon

Sas & Thoen 2003, 2nd-3rd century

Woman holding casket, wearing necklace and earrings Woman holding casket and unidentified object Woman with casket at feet

Relief sculpture, Musee de Picardie (unprovenanced) Relief sculpture on tombstone, South Shields

48

Espérandieu 1913: vol 5 no. 1167 Espérandieu 1913: vol. 5 no. 1171 Espérandieu 1913 vol 5 no.1160 Espérandieu 1913 vol 5 no. 4156 Lefebvre no.23 Espérandieu 1910 vol 3 no. 1787 Espérandieu 1913 vol 5 no. 3948 Toynbee 1962, No. 87, plate 85. Also Arch Ael. ser 4 XXXVII 1959, 203-7 & plates 32-33

Ellen Swift: The Archaeology of Adornment and the Toilet

Fig. 5.2: Detail of relief sculpture on sarcophagus lid from Agen.

Fig. 5.3: Tombstone from Autun, © Brigitte Maurice-Chabard, Musée Rolin, Autun.

49

Dress and Identity

cosmetics, and indeed Cool links it to the Bacchic cult, Allason-Jones (2005: 125) observes that rouge is known to have been made from the lees of red wine in the Roman period (see also Stewart this volume: s.v. faex). Phials from Roman Britain have also been found containing white or blue-grey powders or paint-like contents (Wright 1872: 165-7; Rudd & Daines 1968: 15-18) and these seem likely to have been for cosmetic use. In a number of cases, phials are accompanied by a rod or ligula (see details in Fig. 5.4). From the evidence of a burial from Eynesford, where a rod was actually found inside a phial along with a blue-grey powder, it seems very likely that these rods were used to access the contents of the phial (they often have quite narrow necks). A glass bird-shaped flask whose contents have been analysed and suggested to be cosmetic is also known from a site in Spain (Perez-Arantegui 1996). Philpott (1991: 117) and Price (1978: 102) also associate glass phials with perfume, oil or cosmetic, as does AllasonJones (2005: 124-7).

of funerary monuments alongside evidence from cemeteries enables the archaeologist to look at these ideal constructions of identity in two different media (although unfortunately the opportunity to study a grave monument alongside the contents of the actual grave to which it relates is unlikely to arise for Romano-British material). Several particular objects appear to be significant in representing the toilet and adornment in Roman material culture: mirrors and caskets particularly, and also glass flasks or phials that may have held perfume and cosmetics, and jewellery; cosmetic palettes are a rarer find (see Riha (1986), who also includes other toilet utensils deliberately omitted here as they are more ambiguous). Mirrors, caskets, phials, jewellery, and palettes could have a variety of meanings, contingent on the particular way in which they are assembled in a burial, and the cultural value attached to the items by the mourners. Philpott (1991: 182), for example, notes that mirrors are found in burials in the later pre-Roman Iron Age. This makes it clear that a mirror by itself in a burial, especially in an early burial which could represent pre-Roman cultural trends, does not necessarily index the specifically Roman cultural practice of the feminine toilet routine (this is discussed further below). The correlation between toilet items and graves sexed as female in Roman period burials, together with the appearance of the same objects on female grave monuments held by women or servants, is however suggestive of their sometime use to construct a Roman-style feminine identity. This is worth exploring particularly in the light of the grave monuments and other evidence discussed above.

Though items such as rings and brooches are worn by both sexes in the Roman period, some types of jewellery, such as bracelets, necklaces and earrings, are also widely acknowledged to be feminine objects and are found overwhelmingly in female graves (though occasionally bracelets are associated with a military practice of donatives). Caskets, too, have a strong association with female burials (for example, all where sex was identifiable came from female graves at Butt Road in Colchester: Crummy, Crummy & Crossan 1993: table 8.3) and examples have been found containing feminine jewellery and toilet articles (e.g. graves 69 and 519 at Butt Road, Colchester Crummy, Crummy & Crossan (1993: 274); see also further examples in Fig. 5.4 below and those cited above for Italy).

Roman Britain Firstly, the evidence of the association of various objects with the female toilet needs to be considered. From grave monuments, mirrors, phials and caskets seem to be significant objects. Grave assemblages for Roman Britain excavated up to the end of the 1980s are usefully summarised by R. Philpott. Of nine burials with mirrors, seven can be identified as female, with two of unidentifiable sex (Philpott 1991: 355, table A32). Crummy, Crummy and Crossan (1993: 273) also note the female association of mirrors at the Colchester Butt Road cemeteries, published since Philpott’s compilation of material.

Having established the relationship of these types of objects to the toilet and/or a feminine identity, it is next important to examine the details of their actual occurrence in grave assemblages from Roman Britain. These are given in Figure 5.4 below (from Philpott 1991 plus more recent examples), and show that a minority of burials in Roman Britain contained objects that could be collectively interpreted as a Roman-style assemblage of equipment to be used during the routine of toilet and adornment. Given that one item alone could have varying significance, I have chosen to examine assemblages which contain two or more of the following items: toilet casket, mirror, glass phial/s; and, optionally, jewellery. Sometimes one or more of the objects is found inside the casket, in an additional confirmation of its association with the toilet and adornment in this particular context.

Glass phials are also more likely to be found in female than male graves (Philpott 1991, table A29), a conclusion also reached by Cool in a study of pipette-shaped glass phials from burial contexts. Opinions vary as to the contents of these phials, and they could of course have had varying contents and functions, especially since they occur as part of quite diverse assemblages (see Cool 2002 for an interpretation relating to the practice of mystery cults in Roman Britain, which also summarises grave assemblages that include the pipette-shaped form of phial). Cool (2002: 137, 143) notes that this shape appears to be associated with rich burials. The analysis of the contents of one phial suggested resinated wine (Alcock 1980: 62; Cool 2006: 159). Though this does not seem immediately related to

Items identified as ‘toilet sets’ comprising metal nailcleaners, tweezers, cosmetic spoons, and the like, are also related to the toilet and one example is mentioned below as it occurs with several other toilet-related items. However, these ‘toilet sets’ of metal tools will not be used as diagnostic objects here, since they are not represented

50

Ellen Swift: The Archaeology of Adornment and the Toilet

Fig. 5.4: Grave assemblages with toilet items

sex from skeleton

relevant items within grave assemblage

date

not known

bracelet, glass phials, mirror

AD 50-100

not known

mirror, bone box, glass phial,[ligula]

1st century AD

female

shale or jet ‘rings’ (probably bracelets), mirror in case, casket remains [also toilet-set]

AD 100-150

No grave number, (Wroxeter)

not known

beads, mirror, glass phials containing white substance, bone needle (probably used as dipping rod); all contained within wooden casket/box

No dating

Grave 45, (Ospringe)

female

bracelets, mirror, glass phial

AD 190-260

Grave 94, (Ospringe)

?female

beads, bracelets, glass mirror, glass phial

AD 200-250

Grave e vi, (The Mount, York)

female

bracelets, glass phial and ringshaped flask, casket fittings

3rd-4th century AD

Grave c vi 9 (Railway cemetery, York)

female

mirror handle, glass vessels and jewellery, all in box/casket

female

glass phial and dipping rod, jet box and hair accessory

AD 350-410

not known

2 mirrors, glass phial, fingerring, glass ring

AD 120-50

not known

mirror, glass phial

no date given

female

casket, mirror, bead

Not known

female

casket, mirror, necklaces, hairpin, 2 brooches, ?finger-ring, needle, spatula

Probably 2nd-3rd century

Grave no. and location Grave 44 (Colchester) Joslin collection No grave number, (Colchester) Grave 81/81b (Colchester) Joslin collection

Sarcophagus burial, (London) Grave 1092 1 (West Tenter Street, London) a. (Stanstead, Essex). Grave 25 (Colchester) Grave 302 (Colchester)

reference Philpott 1991: table A11; May 1930: 265-6 Philpott 1991: table A11; Smith 1922, 97 Philpott 1991: table A11; May 1930: 275-6; Crummy, Crummy & Crossan 1993: table 8.3 Philpott 1991: table A11; Wright 1872: 165-7 Philpott 1991: table A32; Whiting 1926: 125-31 Philpott 1991: table A32; Whiting 1926: 125-31 Philpott 1991: table A29; RCHM 1962: 140 and fig.59 Philpott 1991: table A32; RCHM 1962: 82; 140 Thomas 2004: 20-25 Philpott 1991: table A11; Whytehead 1986: 32, 89-91, fig. 38 Philpott 1991: table A11; Anon 1989: ix Crummy, Crummy & Crossan 1993: table 8.3 Crummy, Crummy & Crossan 1993: table 8.3 & p265

(Bel & Feugère 2002: 146), including cosmetic palettes, spatulas, a pyxis, and a bronze phial. Graves 243 (AD 100-200) and 280 (AD 70-150) in particular contained a mirror and a glass phial; parallels are drawn with similar burials at Rustrel and Maubrec. Bel associates the presence of these kinds of objects with Roman influence, an interpretation also suggested by Bertrand (2003) who brings together evidence of dress accessories and toilet equipment (including toilet-sets as well as mirrors, phials, cosmetic palettes, combs, etc.) from the tribal area of Picton (Deux-Sèvres, Vienne). According to Bertrand, some graves at Dune contained more than one toilet item, but specific details are not given. She does however cite a grave at Ronsenac (Charente) dating to the end of the late 1st or early 2nd century AD which contained glass phials, a mirror, and a rock-crystal finger-ring (Bertrand 2003: 135).

on funerary relief sculpture and, in Roman Britain at least, seem to be associated with the indigenous rather than Mediterranean Roman culture in Britain (see Eckardt 2005); it is also not clear that they are feminine objects as they have been found with both sexes. The relationship between these and other toilet items is discussed further in the concluding section. Since the evidence relating to tombstones with depictions of the female routine of the toilet is more frequent in Gaul than in Britain, it is also useful to examine some cemetery evidence from Gaul, although there is no single compilation of burial practices for the province. Evidence from the middle of Roman Gaul is summarised by Bel and Feugère (2002), who examine particularly the cemetery at Valladas, (1st to the mid-3rd century), and compare this to evidence from other Roman sites in Gaul. Some graves at Valladas that contained mirrors also included other toilet items

It is interesting to note the provenance and dating of these

51

Dress and Identity

Fig. 5.5: Tombstone from Chester, © Grosvenor Museum, Chester.

burials across Britain and Gaul. The date-range spans the whole of the Roman period, with several burials of 2nd3rd century date analogous to the dates of the funerary monuments described above. More significant, perhaps, are the locations of the burials – most of the Romano-British examples are from major Roman towns where exposure to Mediterranean Roman culture would have been greatest (in support of this, Philpott (1991: 117) notes that burials with glass phials first appear in the earliest phase of Roman conquest at military sites and towns with early foundations). There is also a wide distribution, with examples from the north-east, north-west, and south-east of Britain, suggesting that the deposition of this type of toilet assemblage in a grave – with its typical objects mirror, glass phial, casket – was not a regional practice. The association of some of these graves with other status-signifiers such as stone sarcophagi (e.g. the examples above from London, and York c vi – see Fig. 5.4) is also notable. The evidence suggests

that assemblages of grave-goods related to the toilet and adornment routine, as well as funerary iconography, are being used to construct an elite feminine identity at death related to the norms of Roman culture. If the evidence can be interpreted as the construction of a ‘Roman-style’ elite feminine identity, however, it must be observed that there are also many female burials without this assemblage of objects. It remains a minority practice. DNA and isotope analysis of the skeleton from the London sarcophagus burial noted in Figure 5.4 has suggested that the woman originated in south-east Europe, and it might be proposed that perhaps some of the other women buried with a similar toilet assemblage documented above may also have originated outside Roman Britain. It is important to distinguish the burial practice described above from other, superficially similar, burials, LPRIAvery early Roman in date, containing mirrors and/or

52

Ellen Swift: The Archaeology of Adornment and the Toilet

indigenous toilet articles such as cosmetic grinders (see Jackson 1985), which, along with other phenomena such as a massive increase in the use of brooches, appear to relate to an earlier ‘wave’ of cultural change precipitated by Roman contact with Britain before the Claudian conquest (See Hill 1997; Jundi & Hill 1998). Hill (1997: 100-101) suggests that the appearance of this toilet equipment relates to fundamental changes in the idea of the body and conscious body care which is not itself gendered. Such material culture is sometimes found in prominent display, for example toilet sets found hanging from Romano-British brooches (e.g. Hattatt 2000: fig. 210 no. 603) and cosmetic grinders doubling as pendants (see Jackson 1985 for some examples). This display of the equipment supports Hill’s argument (1997: 101) that the use of such material culture is not solely to do with personal body care but is also being used to project a particular identity related to its ostentatious use.

Fig. 5.6: Bone hair-pin from London, © Museum of London.

What we are looking at here is apparently successive ‘waves’ of the effects of Roman contact. The first relates to indigenous adoption and appropriation of the process of the toilet into the expression of elite local identities; the second relates to the later presentation of a specifically ‘Roman’ cultural identity brought about through the wider Roman impact of the conquest, including the presence of high-status women from other parts of the Roman Empire. Decorated objects associated with adornment and the toilet This evidence of funerary monuments and grave assemblages relates to the construction of a feminine, elite status at death for particular women who died in Roman Britain and Gaul. Toilet and adornment viewed as an appropriate way for a woman to construct her elite status in life can be traced through decorated jewellery and equipment (see also Swift (2009: 150-154) from which some of the following material is amplified). An illustrative example is the series of pins which has been found in both Britain and other parts of the Empire, decorated with representations of women and goddesses (Fig. 5.6), or with hands holding various objects (Figure 5.7). A number of these pins are made from semi-precious materials such as ivory or silver; some are finely carved bone (there are also less finely worked bone copies, which I see as slightly different and which cannot necessarily be included in the same interpretation). The high-status material, subjectmatter, and provenance of the silver and ivory pins (from major Roman towns, e.g. London and Colchester) and those more finely carved in bone suggest that they would probably have been owned by women who were to some extent familiar with Roman culture. Grave-contexts and research on the relationship between pin size and hairstyle confirm that Roman period pins from Roman Britain were invariably used in women’s hairstyles (Cool 1990). Sometimes pins will also have had other uses, for example in ritual deposits, but these do not concern us here. The pins

occur in several different types, most of them within Cool’s groups 17 and 18: 1. Hair-pins with representations of women with an elaborate hairstyle 2. Hair-pins with representations of Venus (sometimes engaged in toilet activities, e.g. putting on a sandal or looking in a mirror or with hands holding object associated with her, e.g. shells) 3. Hair-pins with hands holding objects associated with the toilet e.g. combs 4. Hair-pins with representations of spherical objects, sometimes with leaves (interpreted as pomegranates) or without (often interpreted as eggs) 5. Hair-pins with representations of other goddesses associated with fertility, e.g. a pin showing a hand holding a bust of Isis Clearly all of these different subjects are related and there are two broad themes that emerge, both with a strong gender-specific dimension. First, the theme of toilet and adornment and secondly, the theme of fertility. Interpreting the iconography of the hair-pins Henig (1977) notes that in the Roman world busts often had funerary connotations and views it as significant that gods

53

Dress and Identity

Fig. 5.7: Copper alloy hair-pin from late 1st-early 2nd century context, Silchester Insula IX, © University of Reading.

and goddesses found on hair-pins are often those associated with other worlds, such as Minerva, Mercury and Isis. Johns (1996) suggests that the iconography of the Isis pin from London (a hand holding a bust of Isis) relates to the fertility aspect of the goddess and also, more generally, as a symbol of good fortune. These pins are least relevant to the current study, though the theme of fertility will recur.

that the spherical or egg-shaped object could be seen as a fertility symbol. For a woman to wear a pin with symbolism relating to fertility would obviously be appropriate in a Roman context in which the bearing of children was seen as an important, perhaps the most important, role that a woman could have. The spherical object can, though, be linked to Venus in her role as exemplar of beauty, the subject of some other figurative hair-pins. Venus is shown holding a fruit in some Roman iconography (for example the Arles Venus in the Musée de Louvre, or, to use a more provincial example, the mosaic from Rudston villa, depicting Venus, illustrated in Dunbabin 1999: fig. 99). These representations refer to the myth in which Venus is given an apple by Paris as a prize for her supreme beauty. This connection thus links the pin-types with spherical objects (the apple) to those with more explicit iconography related to Venus, which include a hand holding a shell (Sas & Thoen 2002: cat. no. 220) and representations of Venus, for instance, the well-known example from London in which she puts on a sandal (Sas & Thoen 2002: cat. no. 219). The use of themes relating to Venus to decorate dress accessories is clearly appropriate to the way that she personifies feminine beauty in Roman culture.

Several previous researchers focus on the interpretation of the most ambiguous pins in terms of iconography: the pins which show a hand holding a spherical object (Fig. 5.7). Von Gonzenbach (1950-51) listed a large number of examples of figurative hair-pins in the categories described above, in a wide range of materials and found at sites across the Empire. Basing her interpretation on the example of a hairpin with a hand holding a comb, she suggested that pins with hands holding spherical objects might also be related to adornment, and that the spherical objects themselves might be hair ornaments. One at least of these objects has a hole through it, as if it is a bead. Arthur (1977) also focused on the hair-pins with spherical objects, constructing an elaborate argument. Using a hairpin from Aquileia showing a hand holding a pomegranate, a symbol of immortality in Roman culture, he suggested that hair-pins of hands holding an egg-shaped object with leaves are showing a combined egg-pomegranate that can be interpreted as relating to the mystery cult of Orpheus. The Orphic interpretation rests partly on the representation of a snake encircling the wrist of one hand-shaped pin, and upon the idea of the Orphic egg. Johns (1996: 117) however, casts doubt on this by identifying the snake as a known form of bracelet. Arthur’s other suggestion was

Hair-pins which show a woman with an elaborate hairstyle, sometimes including a representation of rows of pins among the hair (Fig. 5.6), can also therefore best be interpreted as relating to adornment, a conclusion also reached by Hall and Wardle (2005) in their examination of some examples from Roman London. The theme of the toilet and adornment dominates the iconography of these figurative hair-pins, in a kind of self-

54

Ellen Swift: The Archaeology of Adornment and the Toilet

and in addition, both the burials and funerary monuments, and the figurative hair-pins in luxury materials, are strongly associated with find-spots in major Roman towns where exposure to Mediterranean Roman cultural norms would have been greatest. For a relatively small number of women living and/or dying in Britain and Gaul, the construction of their ideal identity through material culture is suggestive of exposure to the norms of Mediterranean Roman culture and the stereotypical Graeco-Roman ideal of the elite woman as the consumer and exhibitor of wealth and leisure, displaying the elite status of her husband and family.

referential decoration that suggests their appropriate use. It is also important to notice that many of the hair-pins actually show the action of the toilet: not just Venus, but Venus explicitly engaged in her toilet, for example, putting on a sandal; not just a comb, but a hand holding a comb, and therefore in the process of dressing the hair (see Swift 2009: 123-136 & 150-154 for a further discussion of selfreferential decoration on these and other items). Of course, interpretation becomes a good deal more complex when we examine pins with similar subject matter made from more widely available materials such as bone, or bronze; Von Gonzenbach (1950-1) catalogues a good number of these, mostly from Vindonissa; many others are known, for example from London. Some of these are finely carved, and perhaps comparable to the examples in high status materials; others, though, are rudimentary copies and it cannot be assumed that those who made and used them necessarily had the same exposure to Roman culture that would allow them to understand the iconography in the way described above. Interpretation of these examples depends to some extent on the aspirations of those who bought them. If the iconography was understood, then the intent might be aspirational; by wearing such a pin, a woman associates herself with elite Roman women. However, they may simply have been purchased as ‘fashionable’ items, desirable because they emulated Roman culture in general; the intention in this case may have been to indicate participation within Roman culture without more specific connotations. Thirdly, of course, those making and using the pins may have reinterpreted the iconography to give it a different meaning.

Bibliography Alcock, J. (1980): ‘Classical religious belief and burial practice in Roman Britain’, Archaeological Journal 137: 50-80. Allason-Jones, L. (2005): Women in Roman Britain. Council for British Archaeology, York. Amedick, R. (1991): Die Antiken Sarkophagreliefs, Bd. 1 T. 4 Die Sarkophage mit Darstellungen aus dem Menschenleben: Vita Privata. Gebr., Berlin. Anon (1989): ‘Chasing after dinosaurs at Stansted Airport, ix’, Essex Archaeology no 6 (Essex County Council Supplement). Arthur, P. (1977): ‘Eggs and pomegranates: an example of symbolism in Roman Britain’ in T. Munby & M. Henig (eds), Roman Life and Art in Britain, 367-374 BAR 41. Oxford. Balsdon, J. (1962): Roman Women: their History and Habits. London. Bel, V. (2002): Pratiques Funéraires du Haut-Empire dans le midi de la Gaule: la nécropole Gallo-Romaine du Valladas à Saint-Paul-trois-châteaux (Drôme). Lattes. Bel, V. & Feugère, M. (2002): ‘La Toilette’ in Bel, V. Pratiques Funéraires du Haut-Empire dans le midi de la Gaule: la nécropole Gallo-Romaine du Valladas à Saint-Paul-trois-châteaux (Drôme), 146-149. Lattes. Bertrand, I. (2003): Objets de parure et de soins du corps d’époque romaine dans l’est picton (Deux-Sèvres, Vienne). Association des publications Chauvinoises. Braemer, F. (1959): Les stèles funéraires à personnages de Bordeaux, I-III siècle. Paris. Cool, H. (1990): ‘Roman metal hair pins from southern Britain’, The Archaeological Journal 147: 148-82. Cool, H. (2002): ‘Bottles for Bacchus?’ in M. AldhouseGreen and P. Webster (eds) Artefacts and Archaeology: Aspects of the Celtic and Roman World, 132-15. Cardiff. Cool, H. (2006): ‘Pipette flask, 159’ in E. Hartley, J.Hawkes, M. Henig & F. Mee (eds) Constantine the Great: York’s Roman Emperor. London &York. Crummy, N. Crummy, P. & Crossan, C. (1993): Excavations of Roman and Later Cemeteries, Churches and Monastic Buildings in Colchester, 1971-88, 257-275. Colchester. Dunbabin, K. (1999): Mosaics of the Greek & Roman World, Cambridge. Eckardt, H. (2005): ‘The social distribution of Roman artefacts: the case of nail cleaners and brooches in Britain’, Journal of Roman Archaeology 18: 139-160.

Let us revisit the recently excavated sarcophagus of a woman from Roman London to link together the evidence from grave assemblages with that represented by decorated objects. The stone sarcophagus contained a lead coffin decorated with shells (Thomas 2004: 20). Marine iconography in a funerary context is often interpreted as referring to a belief in life after death, and the soul’s journey over the sea to the Isles of the Blessed. Given the grave assemblage within the sarcophagus, however, which includes a hair accessory, jet box and phial with dipping rod, it can perhaps be suggested instead that the shells are used to recall Venus, born in the sea, and therefore that the decoration of the lead coffin reinforces the message that is implicit in the grave goods included within: the construction of a woman’s elite feminine status at death. We can recognise in a wide range of different types of evidence a dominant discourse which apparently existed largely unchanged for much of the Roman period, in which the representation of the process of toilet and adornment was a recognised marker of elite status and identity for some women not only in Italy, but also in provincial contexts like Britain and Gaul (perhaps in some instances because the women themselves did not originate from these provinces). Each aspect of the material record that has been examined here relates to high-status culture (tombstones; sarcophagi; luxury materials such as ivory and silver, etc.)

55

Dress and Identity

Lefèbvre, L. (1975): Les sculptures gallo-Romaines de Musées d’Arlon. Arlon. Martin-Kilcher, S. (2000): ‘Mors immatura in the Roman world: a mirror of society and tradition’ in J. Pearce, M. Struck & M. Millett (eds) Burial, Society and Context in the Roman World, 63-77. Oxford. May, T. (1930): Catalogue of the Roman pottery in the Colchester and Essex Museum. Cambridge. Pearce, J. (2000): ‘Burial, society and context in the provincial Roman world’ in J. Pearce, M. Struck & M. Millett (eds), Burial, Society and Context in the Roman world, 1-12. Oxford. Perez-Arantegui, J. (1996): ‘Analysis of the products contained in two Roman glass ungentaria from the colony of Celsa, Spain’, Journal of Archaeological Science 23: 649-655. Philpott, R. (1991): Burial Practices in Roman Britain. BAR 219. Oxford. Price, J. (1978): ‘The glass flask’ in J. Collis (ed.) Winchester Excavations vol II: 1949-1960, excavations in the suburbs and western part of the town, 102. Winchester. Riha, E. (1986): Römisches toilettgerät und medizinische Instrumente aus Augst und Kaiseraugst, Forschungen in Augst 6. Augst. Royal Commission on Historic Monuments (1962): An Inventory of the Historical Monuments in the City of York, vol. 1 Eburacum Roman York. Leicester. Rudd, T. & Daines, C. (1968): ‘A Romano-British settlement site at Brickhills estate, Eynesbury, Hunts’, Proceedings of the Cambridgeshire Archaeological Society 61: 15-18. K. Sas & H. Thoen (eds) (2003): Schone Schijn: Romeinse juweelkunst in West-Europa/Brillance et Prestige: La joaillerie romaine en Europe occidentale. Leuven. Smith, C. (1922): A Guide to the Antiquities of Roman Britain in the Department of British and Medieval Antiquities. London. Swift, E. (2009) Style and Function in Roman Decoration: living with objects and interiors. Farnham. Thomas, C. (2004): Life and Death in London’s East End: 2000 years at Spitalfields. London. Toynbee, J. (1962): Art in Roman Britain. London. Von Gonzenbach, V. (1950-51): ‘Zwei typen figurlich verzierter Haarpfeile’, Jahresbericht der Gesellschaft pro Vindonissa 1950-1: 3-19. Warland, R. (1994): ‘Status und formular in der Repräsentation der spätantiken Führungsschicht’, Mitteilungen des Deutschen Archaeologischen Iinstituts: römische Abteilung 101: 174-202. Whiting, W. (1926): ‘The Roman cemeteries at Ospringe, 113-122’, Archaeologia Cantiana 38: 123-151. Whytehead, R. (1986): ‘The excavation of an area within a Roman cemetery at West Tenter Street, London E1’, Transactions of the London and Middlesex Archaeological Society 37: 23-124. Wright, T. (1872): Uriconium: A Historical account of the Ancient Roman City. London/Shrewsbury. Wright, R. and Richmond, I. (1955): Roman Inscribed and

Elsner, J. (2003): ‘Visualising women in late Antique Rome: the Projecta casket’, in C. Entwistle, C. (ed.) Through a Glass Brightly: Studies in Byzantine and Medieval Art and Archaeology Presented to David Buckton, 22-36. Oxford. Espérandieu, E. (1908): Recueil Général des Bas-reliefs, statues et bustes de la Gaule Romaine, Tome deuxième, Aquitaine. Paris. Espérandieu, E. (1910): Recueil Général des Bas-reliefs, statues et bustes de la Gaule Romaine, Tome troisième, Lyonnaise première partie. Paris. Espérandieu, E. (1911): Recueil Général des Bas-reliefs, statues et bustes de la Gaule Romaine, Tome quatrième, Lyonnaise deuxième partie. Paris. Espérandieu, E. (1913): Recueil Général des Bas-reliefs, statues et bustes de la Gaule Romaine, Tome cinquième, Belgique première partie. Paris. Fleming, S. (1997): Roman Glass: Reflections of Everyday Life. Philadelphia. Hall, J. & Wardle, A. (2005): ‘Dedicated followers of fashion? Decorative bone hair-pins from Roman London’ in N. Crummy (ed.): Image, Craft and the Classical World, Essays in honour of Donald Bailey and Catherine Johns, 173-179. Monographies Instrumentum 29. Montagnac. Hatt, J. (1986): La Tombe Gallo-Romaine. Paris. Hattatt, R. (2000): A Visual Catalogue of Richard Hattatt’s Ancient Brooches. Oxford. Henig, M. (1977): ‘Roman gemstones: figure type and adaptation’ in J. Munby & M. Henig (eds), Roman Life and Art in Britain. 341-346. BAR 41. Oxford. Henig, M. (2004): Roman Sculpture from the North-west Midlands. CSIR Vol.1 Fasc.9, Oxford. Hill, J. (1997): ‘The end of one kind of body and the beginning of another kind of body? Toilet instruments and ‘Romanization’ in A. Gwilt & C. Haselgrove (eds), Reconstructing Iron Age Societies, 96-107. Oxford. Hope, V. (2001): Constructing Identity: the Roman Funerary Monuments of Aquileia, Mainz and Nîmes. BAR International Series 960. Oxford. Jackson, R. (1985): ‘Cosmetic sets from late Iron Age and Roman Britain’, Britannia 16: 165-92. Johns, C. (1996): ‘Isis, not Cybele: a bone hairpin from London’ in J. Bird, M. Hassall & H. Sheldon (eds), Interpreting Roman London, 115-118. Oxford. Jones, R. (1984): ‘Death and distinction’ in T. Blagg & A. King (eds), Military and Civilian in Roman Britain. BAR 136, 219-226. Oxford. Jundi, S. & Hill, J. (1998): ‘Brooches and identities in First century A.D. Britain: more than meets the eye?’ in C. Forcey, J. Hawthorne and R. Witcher (eds), TRAC 97: Proceedings of the Seventh Annual Theoretical Roman Archaeology Conference, University of Nottingham 1997, 125-137. Oxford. Kampen, N. (1996): ‘Gender theory in Roman art’ in D. Kleiner & S. Matheson (eds), I Claudia: Women in Ancient Rome, 14-26. Yale University Art Gallery, Austin.

56

Ellen Swift: The Archaeology of Adornment and the Toilet

Sculptured Stone in the Grosvenor Museum, Chester. Chester. Wyke, M. (1994): ‘Woman in the mirror: the rhetoric of

adornment in the Roman world’ in L. Archer, S. Fischler & M. Wyke (eds) Women in Ancient society, 134-151. London.

57

Chapter 6

Dress and Cultural Identity in the Roman Empire Ursula Rothe

The study of Roman culture in the provinces: a situation report

their meanings.1 Through this, it was argued, one could gain access to the processes at work among the non-elite population who were not represented in written sources. More problematically, it was often argued that continuity of pre-Roman material culture could be interpreted as covert resistance to Roman cultural imperialism.2 This approach was not spared criticism, however, and the points raised at the time are still of relevance today.

In recent decades the traditional view of Roman cultural expansion and its effect on the people Rome conquered has undergone a fundamental reappraisal. Not only have scholars begun to construct a more detailed image of the forms and extent of cultural change, but it has also emerged just how complex and multi-layered cultural interaction in the Roman provinces really was. This has led to an increasing realisation (in part due to the influence of trends in the social sciences) that identity is one of the basic concepts at the heart of cultural processes.

First, profound psychological reasons for patterns in the distribution of material culture are sometimes automatically assumed when more pragmatic considerations appear at least as likely to have played a role. The consumption of different types of pottery, for instance, was determined not only by personal taste or cultural orientation, but perhaps above all by the financial capacity of the consumer as well as the market availability and utility of the vessel. Second, the persistence of pre-Roman traditions is likely to have often been the result of passive continuity of habitus when there was no good reason for change. Third, as items of material culture were often put to different uses by different people, one cannot presuppose a certain lifestyle, and as such a specific social meaning, from consumption of a particular item. Fourth, the pace of psychological reorientation and the embracing of new ideas and values often occur at a different pace to that of the adoption of new items of material culture (Woolf 1998: 16).

It is a commonplace that equating items of material culture to ethnic identity is fraught with problems. Siân Jones (1997) has demonstrated that this dilemma has been a leitmotif of archaeological research from its inception and has argued for a more reflective approach to interpreting material culture in terms of ethnicity. However, perhaps as a result of a certain euphoria at having at last found a concept to replace the troublesome ‘Romanisation,’ archaeologists and cultural historians have nonetheless begun to attach the concept of identity to all manner of material culture with less caution than should perhaps be exercised. Identity has become the new catch-phrase for themes which a decade ago would have been formulated in terms of degrees of Romanisation. While the concept of differing cultural identities is vastly superior to the ‘Romanisation’ model due to its ability to integrate complex and multi-directional processes and to incorporate cultural constructs that are not ‘pure’ in character, the old question must again be raised as to whether it is justified to see profound meaning in the distribution of all items of material culture, regardless of their form.

At the same time, in order for archaeological research to contain meaning beyond the mere visual description of past societies, evidence must indeed be analysed and interpreted. The answer lies in appreciating that the ‘hidden The term ‘hidden transcript’ was first coined by Scott (1990). For the use of the concept by Roman archaeologists, see for example Hingley (1997: 81ff.) and Mattingly (1997: 15). 2 See for both examples and criticism, inter alia, Benabou (1976); a number of contributions to Pippidi (1976) and Raaflaub et al. (1987); Drinkwater & Vertet (1992); Galliou (1992); Hingley (1996; 1997); Webster (1996; 1997). This idea was, however, not entirely new. In 1966 Lambrechts published a small monograph entitled De geestelijke weerstand van de westelijke provincies tegen Rome in which he argued that the non-adoption of items of Roman culture represented resistance to the ‘eenheidspolitiek’ of the Romans. 1

In some ways the present problem promises to replicate a discussion that took place from the 1970s onward, and in particular in the 1990s amongst protagonists of the Romanisation debate. A number of scholars argued that ‘hidden transcripts’ lay embedded in distributions of material culture, waiting for archaeologists to decode

59

Dress and Identity

Dress and cultural identity

transcripts’ contained in material evidence differ from type to type, and that separate interpretational frameworks must be developed that correspond to the specific meanings and uses of each source type.3

The term ‘cultural identity’ is preferable to ‘ethnicity’ when approaching material culture in the provinces, as ethnic identity is generally associated with continuity from the past and a certain lack of capacity for evolution. While ethnicity, thus defined, is a useful term to describe the continuity of pre-Roman cultural elements among some communities in the provinces of the Roman Empire, it is perhaps less accurate when applied to ‘Roman culture’ as such (Rothe 2009: 5ff.). In the light of recent studies, some of which have been inspired by current theories on our own ‘global culture’, Roman culture must now be seen as a dynamic and flexible entity that was susceptible to change, geographically widespread, and attainable by different groups of people.8 As such, it existed on a different level to continuing tribal or regional identities and, importantly, could co-exist with them. Moreover, the Roman Empire possessed various hybrid cultures and inspired new regional identities that cannot easily be described as ‘ethnicity’. In order to be able to include various types and levels of culture, then, the more general term ‘cultural identity’ is perhaps more appropriate, and taken to mean the identity that pertains to membership of a culture or cultures of whatever type.

It is particularly against this background that dress comes into its own, because it has a number of unique features which set it apart from other types of material culture and make it an ideal base from which to investigate developments in social and cultural identity. These will be discussed more fully below. Following on from this it is somewhat surprising that very little research has been devoted to exploring and analysing the cultural meaning of dress behaviour in the Roman provinces.4 That is not to say that the dress of the Roman Empire has been generally neglected. It has, however, traditionally been the preserve of either dress or textile historians. As a result we have excellent descriptive studies of both Roman clothing and native dress in various regions of the Roman Empire, in particular that of Palmyra and the north-western and Danube provinces, as well as detailed knowledge of textile technology and fabric types.5 By and large, however, the aim of research in these fields has been to reconstruct and describe, not to analyse the meaning of, dress types and patterns in their distribution. Recently, a number of scholars, some of them contributors to this volume, have begun to use clothing and adornment to investigate different aspects of social identity such as status, gender and power relations.6 But in terms of pursuing cultural identity in the Roman provinces, dress is as yet an underused resource.7

The substantial body of theories and case studies surrounding dress in sociology and cultural anthropology demonstrate that dress presents an excellent medium to observe cultural identity and changes therein. First, dress is worn on the body. With hairstyle, adornment and bodily alterations, dress is what visually marks a person out as a socio-cultural, as opposed to a biological, entity. Dress communicates in an immediate manner to those viewing the person the place the wearer occupies in his or her social and cultural context, and has for this reason been characterised by one scholar as ‘a visual metaphor for identity’ (Davis 1992: 25). The element of communication is crucial, as it is also the main factor that defines identity: unlike the ‘self’ it cannot exist without an audience.9

A point already made briefly by Hanson (1994: 157ff.). Cf. a recent critique of the identity paradigm by Pitts (2007) in which he argues that not the items of material culture as such, but their social contexts and use, where archaeologically visible, should be used as indicators of cultural change, as they reflect changes in social behaviour. His argument follows a similar logic to mine in that the mere existence of garments and accessories cannot be used to map cultural change, while patterns in where and by whom they are worn can. However convincing Pitts’ study of consumption in Britain is, I am not persuaded that changes in social practice necessarily reflect changes in identity. I would suggest that, like spheres of material culture, different spheres of social practice had different meanings. 4 Cf. cultural anthropology, where dress has been a major new focus, particularly since the 1990s; see studies such as Kuper and Hendrickson in southern Africa (Kuper 1973; Hendrickson 1996a; 1996b), the work of mainly Dutch anthropologists in Indonesia (e.g. Schulte Nordholt 1997) and a recent study by Tarlo (1996) in India. 5 For descriptive studies see: Blümner (1911); Bieber (1931; 1977); Wilson (1924; 1938); Goette (1990); Croom (2002); for Palmyra: Colledge (1976:145-149); Taha (1982); Goldman (1994); for N. Western & Danube provinces see: Langlois (1959-1962); Wild (1968a; 1968b; 1985); Böhme (1985); Böhme-Schönberger (1997); Láng (1919); Mautner and Geramb (1932); Fitz (1957); Čremošnik (1964); Garbsch (1965; 1985); Bíró (2003); Kovrig (1937); Patek (1942); Pochmarski (2003); on textiles see: Roche-Bernard (1993); Wild (1970a; 1970b; 1977), Pfister (1934-1940); al-As’ad, Chehade and Schmidt-Colinet (1995); Schmidt-Colinet, Stauffer & al-As’ad (2000); Schmidt-Colinet (2000); Stauffer (1995; 1996). 6 E.g. contributions in parts I and II of Sebesta & Bonfante (1994); Freigang (1997b); Chausson/Inglebert (2003); Pausch (2003); Harlow (2004); most contributions to Cleland, Harlow & Llewellyn-Jones (2005); Olson (2008); Colburn/Heyn (2008) and Edmondson & Keith (2008). 7 Exceptions include brief discussions of the meaning of dress on gravestones, for example in Wild (1985: 409-413), Freigang (1997a: northern Gaul/Rhineland), Goldman (1994); Schmidt-Colinet (2004: Palmyra) or Čremošnik (1964), Mosser (2003), Pochmarksi (2004) and Faber and Jilek (2006: Danube provinces). The author’s PhD thesis 3

As a result, identity is something which must be conveyed using signs that can be read by others. While other forms of material culture can also convey such messages, what is worn on the body is, with language, the most direct and personal medium of expression.10 In fact, some social scientists see dress as a type of non-verbal language,11 (Rothe 2009) has attempted to make inroads in this regard by conducting a detailed study of dress behaviour in the Rhine-Moselle region. 8 For example, Woolf (1998); Hingley (2005); Wallace-Hadrill (2008). 9 Stone (1962: 93): “‘Identity’ is not a substitute word for ‘self’. Instead, when one has identity, he is situated … One’s ‘identity’ is established when others place him as a social object by assigning him the same words of identity that he appropriates for himself or announces”. (Stone’s italics). 10 In view of this it is interesting that J.P. Wild also took linguistic groupings into account in his influential work on dress in the north-western provinces (i.e. Wild 1968b; 1985). For the latter see Kuper (1973: 366): ‘a person’s relationship to his clothing is at once different from and more intimate than his relationship to all other material objects.’ 11 The original idea is attributed to structuralists like Lévi-Strauss and Barthes (for example Barthes 1967), who had compared clothing communication to linguistic models such as that of Saussure (Davis 1992:

60

Ursula Rothe: Dress and cultural identity in the Roman Empire

and Stone argued that not only verbal communication and gestures, but also dress, should serve as a basis for research on identity when formulating his theory of symbolic interactionism.12

Anthropologists, traditionally concerned with ‘pure’ or ‘authentic’ dress behaviour, have experienced somewhat of a breakthrough in the past two decades as they have begun to realise the significance of ‘the mixture of clothes people normally wear’ (Schulte Nordholt 1997: 2). Humans often quite consciously negotiate their place in a new cultural or political context by combining aspects from both, or many, cultures. In this way, for instance, wealthy Parsi men in Bombay under British rule adopted selected elements of British dress into their otherwise Indian ensemble to symbolise their eloquence in both the local and European cultural worlds (Dar 1969: 75-78; Tarlo 1996: 48ff.) and modern Indian women often wear western accessories such as high-heeled shoes and handbags to symbolise their participation in global fashion, while their otherwise Indian dress assures the viewer of their continued local cultural loyalty (Tarlo 1996: 46). Similar considerations have inspired people in various historical situations to wear different ensembles for different occasions depending on which cultural symbols are seen to be relevant in that particular sphere.17

Stone also raised a second important characteristic of dress: it is not only directly communicative, but can also be involved in dialogue: In appearances, then, selves are established and mobilised. As the self is dressed, it is simultaneously addressed, for, whenever we clothe ourselves, we dress ‘toward’ or address some audience whose validating responses are essential to the establishment of our self. Such responses may, of course, also be challenges, in which case a new program is aroused (Stone 1962: 101-2).13 As such, clothing should not be seen merely as a passive reflection of a person’s place in his or her world. Dress choices are both public and personal and as such are dictated not only by social constraints but also individual choice.14 Clothing can thus also be actively manipulated and used as a medium of asserting a position in a non-verbal dialogue relating to social and cultural processes going on in a person’s surroundings.15 The work of Kuper and Hendrickson on sub-Saharan Africa has demonstrated how clothing played an important role in interaction between members of the colonial powers and native people, the vast majority of which was interpersonal but plagued by language problems, rendering visual communication of utmost importance (Kuper 1973: 347 ff.; Hendrickson 1996a: 15). Anthropologists cite examples in which clothing was used deliberately to express conformity to the colonial culture, often linked with the occupation of the wearer and the type and extent of his or her contact with members of the colonial power (Renne 1996; Schoss 1996). Conversely, many cases show how clothing can be used, implicitly or explicitly, to demonstrate resistance to colonial culture.16

Fourth, dress is highly symbolic, for while climate and other practicalities play a role in choice of dress, both history and the modern world abound in examples of impractical clothing, from the notoriously uncomfortable Roman toga, the insistence on tight and weighty Western dress among representatives of the European powers in their often hot and humid colonies. As such, clothing choice is determined by both rational and irrational considerations and can provide a gateway into the latter in the context of cultural exchange. Finally, and related to this, is the fact that clothing, perhaps more than any other form of material culture, can reflect ideas and values, and therefore has the potential to inform us of the more subtle and indirect consequences of cultural dialogue.18 This applies most obviously to women’s dress, as ideas concerning female modesty are both very important in many human societies and often closely related to how women cover their bodies. Moreover, female propriety has often played a major role in the negotiation of cultural identity, especially in cases where an adoption of new clothing is considered symbolic of an abandonment of that culture’s ideal of female respectability. However, other, non-gender-related values and ideas can also be expressed in dress. Tarlo has said of her research on dress in rural northern India:

Third, dress is flexible and multidimensional. Consequently, it has the ability to reflect complex and multiple identities, both social and cultural. This means that dress not only simultaneously expresses, say, gender, class and cultural identity, but can also reflect a combination of cultural allegiances in the same person when garments are mixed. 5 note 2). Comments made by Hiler and Hiler in 1939 show, however, that the idea was not entirely new: ‘[M]any hidden characteristics of the individual are written for those capable of reading them, in the subtle and unapprehended language of the attempt at conformity or adaptation shown on his back.’ (Hiler & Hiler (1939: xi). But it was only in the context of the ‘burgeoning field of semiotics’ (Davis 1992: 5) that dress has recently been treated directly by sociologists as a ‘language’ with its own different types of ‘words’ (for example Lurie 1992), or, in the case of Fred Davis’ research, as a type of code (picking up on ideas from Umberto Eco’s Theory of Semiotics 1979). See also Dode, this volume. 12 Stone (1962: 87). Cf. Ruesch & Kees (1956) and Rose (1962). 13 Cf. Hartmann (1949): dress as both stimulus and response. 14 Kuper (1973: 365 ff.); Tarlo (1996: 318); Hendrickson (1996a: 2). 15 See Kuper (1973: 347 ff.); Roach & Eicher (1965: 3); Hendrickson (1996a: 2). 16 For this reason, Hendrickson (1996a: 15) refers to the clothed body as ‘perhaps the quintessential subversive object sign.’

…dress codes are often at the centre of a number of wider issues concerning modesty, honour and respect, and… a clash between different styles of clothing is often symbolic of a wider conflict For India see Dar (1969: 78); Ramanujan (1984: 32); Tarlo (1996: 52-56), for Indonesia see Gelman Taylor (1997: 101), for Swaziland see Kuper (1973). 18 See for example Gelman Taylor (1997: 112) and Nandy (1983: xi): ‘…colonialism colonizes minds in addition to bodies and it releases forces within the colonised societies to alter their cultural priorities once and for all.’ 17

61

Dress and Identity

between different cultural and social values and norms.19

although often also depicted, are underrepresented. Dress historians have been able to typologise dress ensembles from these sources, but only in those parts of the empire where they are plentiful. As a result, it will never be possible to write a detailed history of dress behaviour for all regions of the Roman Empire. Nevertheless, we would be missing an opportunity if we ignored the dress behaviour of those for which the sources do speak.

In Tarlo and Dar’s studies, the men who adopted some elements of British dress did so to symbolise commitment to the idea of ‘progress’ and ‘enlightenment’ that was propagated in their (Westernised) education. For this reason, the dress of a colonising power can symbolise a number of things quite apart from ethnicity, as Kees van Dijk has shown for Indonesia:

A second source problem arises when endeavouring to determine the social and cultural meanings of the various garments and ensembles. Ancient literary sources are only of limited value: while providing vital information on the meaning of various Roman garments, such as the toga, to people in the imperial centre, such sources give only brief and often inaccurate details of dress behaviour in the provinces.

Depending on the epoch, ‘dressing like a European’ could reflect a variety of motives. It could be a sign of accepting European culture as a term of reference, or of a desire to be part of a sophisticated fashionable world. Following a European style might indicate acceptance of or accommodation to Dutch rule. It could also be a signal that one did not consider oneself inferior to Europeans, or that one was part of a modern world of progress and therefore rejected the traditional mores of local society and of traditional Islam (van Dijk 1997: 54).

There are, however, a number of ways in which one can discover something of the meaning of garments. First, it is usually possible to determine, at least to some extent and occasionally not without controversy, the origins of different types of dress, such that garments and ensembles can be linked to cultural groups. Dress which originated within a cultural group will have symbolised something different to dress which was imported into that sphere. In other words, in some cases the Roman-native paradigm is justified and useful. A 1st-century grave stele from Koblenz on the Rhine (Fig. 6.1),21 for example, shows a couple in the pre-Roman dress of the region (hooded Gallic cape and scarf for the man, and, for the woman, a tight, long-sleeved bodice, loose, sleeveless overtunic held by fibulae at the shoulders and neck, cloak, disc pendant and bonnet).22 To their right sits a younger man in an elaborate chair with a footstool wearing a tunica and pallium and holding a scroll. In a separate gallery below, five young women are depicted in the Roman tunicae and pallae, but also wearing around their necks the torques and disc pendants that formed part of the native ensemble.23 Although the inscription is only partially legible, it would seem that the younger man and women are the couple’s children, and, given the dating of the stone, one can interpret the differing dress behaviour of the two generations as symbolising a new cultural orientation within that family, even without explicit information to that effect.

Similarly, the idea of shame propagated by Christian missionaries led many native inhabitants of European colonies to begin to cover their bodies more.20 What they covered their bodies with is a separate, if nonetheless related, matter. The implications of these observations are clear: dress behaviour can express not only allegiance to a particular cultural identity but also to its values and ideas. This presents perhaps the biggest challenge to those wishing to use dress to investigate cultural processes, but it also provides a rare opportunity to tease out the various layers and knock-on effects of cultural interaction. Dress and identity in the Roman provinces It has been shown above how dress has played a major role in both the expression and negotiation of cultural identity in diverse societies and historical situations, but how does this relate to the Roman Empire? Most of the theory cited above stems from anthropology and sociology, and these disciplines base their theoretical frameworks on fieldwork conducted in the modern era, and have therefore had a far superior source base than any scholar of antiquity can hope for.

Correlation can also be applied to data to reveal what certain dress styles meant to their wearers. Particularly in the case of grave monuments, which frequently include further information in inscriptions and other scenes about the people depicted, it is often possible to group people according to occupation, wealth or political status, gender

The source base for dress in the Roman Empire is by comparison less satisfactory in two ways: because textiles rarely survive in archaeological contexts, and due to the difficulties of reconstructing garments from metal accessories, we are largely dependent on pictorial evidence such as frescoes, statues and reliefs. These, however, are not found everywhere in the empire throughout its history and tend to depict the more wealthy, such that the lower classes,

Espérandieu (1907-1981: VII 5770; VIII 6184; CIL XIII 7627); Krüger (1938); Bauchhenß (1975: 88-91); Boppert (1992: 27ff. & Fig. 6). 22 This is commonly called ‘Menimane’s ensemble’ after a woman portrayed with her husband, the nauta (sailor) Blussus, on a gravestone from Mainz (Fig. 7.2); CIL XIII 7067; Espérandieu (1907-1981:VII 5815); Boppert (1992 no. 2). The dress worn by the ‘father’ in the background of the stone from Koblenz is also identical to that worn by Blussus. 23 The second, fourth and fifth wear a torque, the second and fifth a disc pendant. 21

Tarlo (1996: 13). Gelman Taylor (1997: 112) also made a point, in her study of colonial Java, of the fact that dress change in that region revealed not only the spread of ‘trade items’ but also of ‘ideas’. 20 For example in India: Tarlo (1996: 35 nt. 8). 19

62

Ursula Rothe: Dress and cultural identity in the Roman Empire

Fig. 6.1> Family Gravestone, Koblenz, 1st century AD (Photograph Mittelrheinmuseum Koblenz).

and the region in which they lived. When certain dress choices correlate strongly with these categories, what is already known about these particular groups and their place in society can be combined with their dress choice to formulate theories as to the meaning of the dress type in that group. In the civitas Treverorum in northern Gaul, for example, the toga is comparatively rare on funerary portraits, and is restricted almost entirely to the more wealthy merchants and landowners. This cannot be wholly attributed to legal status: although the toga could, at least officially, only be worn by Roman citizens, inscriptions point to a high number of people with this status in the

region.24 Even if this were not so, one would expect a dramatic increase in toga depictions after the enactment of the constitutio Antoniniana in AD 212 and this is not the case, although many portraits date to this later period. The disparity between the epigraphic evidence and the frequency of the toga shows that many Roman citizens must have chosen not to wear the toga in their portraits. Indeed, stones from neighbouring districts show this did occur: several men bear the tria nomina but wear Gallic dress.25 Not the legal, but the wealth and occupational 24 25

63

Raepsaet-Charlier (2001b: 465; 2001a: 349ff.). From Cologne: CIL XIII 8344; Galsterer & Galsterer (1975 no. 314

Dress and Identity

status of these men was responsible for their choice of the toga for their portraits. As they are the most likely to have held public office and/or to have travelled widely in the empire on business (see Krier 1981), their choice of the toga reflects an identification with the wider cultural sphere of the Roman Empire, and stands in marked contrast to the regional Gallic dress worn by the majority of their Treveran contemporaries on their gravestones (Rothe 2009).

circumstances, but represent a more general reaction of an autochthonous population to incorporation into an empire with all the cultural pressures, socio-structural changes and new geographical perspectives that this brought. Such forces were, to an extent, also at work among provincial populations of the Roman Empire. Consequently, it can be argued that the corresponding ancient phenomenon was a result of similar considerations.28

A third possibility for interpretation arises from comparative work. In cases where both the cultural circumstances and the dress behaviour are similar to those in more recent and better documented case studies, theories can be applied to the ancient examples that stem from the scholarly interpretation of their more recent counterparts. In northern Gaul and the Rhineland, for example, changes in the distribution of brooches in the archaeological record in the late 1st century AD correspond to changes in women’s dress on grave portraits, and together show that a new panregional, Gallic ensemble began to be worn in place of the former, more localised dress such as Menimane’s ensemble (Fig. 6.2). The fact that this new dress originated among the provincial population and was native in character taken with its regional distribution, seems to point to the development of a new regional cultural identity that stood apart from both the new Roman and former smaller native cultural spheres (Rothe 2009: 53-58).

Comparative work can, however, be a risky affair (Golden 1992: 324). There are obvious dangers in using material from other disciplines without having a wider knowledge of discourse in that field. Nonetheless, if we accept that the basis of academic progress is dialogue, a one-way flow of information from the social sciences is detrimental to our research. It is virtually impossible for one person to master several disciplines at once, but it is possible is to increase communication between members of various disciplines working on similar themes, such that comparative research can be scrutinised and refined, and new ideas and models exchanged. Conclusion: towards an anthropology of dress in the Roman Empire In such a discipline as Roman provincial studies, with its less than perfect source base, it is essential that all available sources are used to their full advantage. This means making use of the wealth of information we have in some regions for dress. But in order to do this it is essential that we work towards a theoretical framework that allows us to understand the garments, ensembles and patterns in dress behaviour over time and space. We need to continue to extend the study of dress from the mere description of ‘pure’ dress types to an integrated investigation of patterns in dress behaviour and the information it imparts about cultural and social processes in past societies. At this level, dress could form an ideal centre around which to conduct truly interdisciplinary research, in which scholars of dress behaviour from different historical periods and disciplines could communicate their findings and comment on each others’ research.29 The study of dress holds enormous potential for better understanding provincial societies, we just need to know how to find what we are looking for.

Why this occurred and what this new regional identity signified is difficult to obtain from the ancient evidence alone. A survey of anthropological literature, however, reveals that such a phenomenon has been a relatively common occurrence among the native populations of both European colonies and modern states that are under the strong influence of Western culture. Examples of the formation of new, pan-regional identities with corresponding dress styles include the Tswana of southern Africa (Comaroff & Comaroff 1992: 235-45), the Kayapo of Brazil (Turner 1991) , the Australian Aborigines26 and India.27 When studying the first-hand accounts of the people involved, it is evident that the motivation behind these processes was similar in all the various cases, and were not peculiar to certain groups of people under specific & pl. 69); from near Bonn: CIL XIII 8105; Espérandieu (1907-1981: VIII 6276; Bauchhenß (1979 no. 26). From the civitas Mediomatricorum: Freigang (1997a cat. nos: Med. 167, 173, 189, 192, 198). See Schlippschuh (1974: 162) for the decreasing significance of Roman citizenship among provincials. 26 Jones & Hill-Burnett (1982); Tonkinson (1990). Cf. also the clothing behaviour of native North Americans who have been adopting an increasingly ‘pan-Indian’ style of dress that is distinctly native but borrows elements from various specific tribal dress ensembles (see Thompson Miller 1979: 329). 27 For example, Tarlo (1996: 46; 128; 145; 147; 249ff.; 256). See also Mehta (1990: 124) who describes the adoption of the sari by Oswal women as ‘the symbol of … entry into a new world’. The causes of this new perspective are cited by Bhandari as being the spread of education, the breaking-down of traditional occupations (with their work-specific dress), urbanisation, migration, the effects of the media and social mobility (Bhandari 2005: 24). Cf. also the villagers of colonial Java at the turn of the century who migrated in increasing numbers to cities and whose ‘increasingly neutral outfit’ in the sarong gave them ‘access to a strange new space’ (Mrázek 1997: 119).

Bibliography al-As’ad, K., Chehade, J. & Schmidt-Colinet, A. (1995): ‘Die Textilien aus Palmyra. Ein internationales und interdisziplinäres Projekt’ in A. Schmidt-Colinet (ed.),

See also Jones (1997: 96): ‘Each example has its own particularities and European colonialism as a whole is arguably characterised by certain specific concrete historical conditions in contrast to other periods. However, despite variations in the particular conditions in which ethnic identity is constructed, and in the form that ethnicity takes, it can be argued that similar, although in some cases less radical, processes of objectification are involved in the construction of a consciousness of ethnicity within diverse socio-historical contexts.’ 29 See, for example, recent undertakings along these lines with regard to empires and economic globalisation (Alcock et al. 2001; Schreiber 2000). 28

64

Ursula Rothe: Dress and cultural identity in the Roman Empire

Fig. 6.2> Gravestone for Blussus and Menimane, Mainz, early-mid 1st century AD (Photograph Landesmuseum Mainz: ursula. [email protected]).

Palmyra: Kulturbegegnung im Grenzbereich, 54-56. Mainz. Alcock, S.E. et al. (eds) (2001): Empires: Perspectives from Archaeology and History. Cambridge. Barthes, R. (1967): Système de la mode. Paris. Bauchhenß, G. (1975): ‘Römische Grabmäler aus den Randgebieten des Neuwieder Beckens’ Jahrbuch des Römisch-Germanischen Zentralmuseums 22: 81-95. Bauchhenß, G. (ed.) (1979): CSIR Deutschland III,2:

Germania Inferior: Bonn und Umgebung: Zivile Grabdenkmäler. Bonn. Benabou, M. (1976): La résistance africaine à la romanisation. Paris. Bhandari, V. (2005): Costume, Textiles and Jewellery of India: Traditions in Rajasthan. London. Bieber, M. (1931): s.v. stola in: Pauly-Wissowa 7: 56-62. Bieber, M. (1977): Ancient Copies. New York. Bíró, M.T. (2003): ‘Ein neuerer Beitrag zur pannonischen einheimischen Frauentracht’ Pannonica: 89-102.

65

Dress and Identity

Research on the Romanisation of the Western Provinces. BAR 571: 29-33. Oxford. Galsterer, B. & Galsterer, H. (1975): Die römischen Steininschriften aus Köln. Cologne. Garbsch, J. (1965): Die norisch-pannonische Frauentracht im 1. und 2. Jahrhundert (Münchener Beiträge zur Vorund Frühgeschichte 11). Munich. Garbsch, J. (1985): ‘Die norisch-pannonische Tracht’, ANRW II.12.3: 546-577. Gelman Taylor, J. (1997): ‘Costume and gender in colonial Java, 1800-1940’ in H. Schulte Nordholt (ed.), Outward Appearances. Dressing State and Society in Indonesia. 85-116. Leiden. Goette, H.R. (1990): Studien zu römischen Togadarstellungen (Beiträge zur Erschliessung hellenistischer und frühkaiserzeitlicher Skulptur und Architektur 10). Mainz. Golden, M. (1992): ‘The uses of cross-cultural comparison in ancient social history’, Echos de Monde Classique/ Classical Views 11: 309-331. Goldman, B. (1994): ‘Graeco-Roman dress in SyroMesopotamia’ in J.L. Sebesta & L. Bonfante (eds), The World of Roman Costume, 163-181. Madison. Hanson, W.S. (1994): ‘Dealing with the barbarians. The Romanization of Britain’ in B. Vyner (ed.), Building on the Past. Papers Celebrating 150 Years of the Royal Archaeological Institute, 149-163. London. Harlow, M. (2004): ‘Clothes maketh man: Power dressing and the elite male in the later Roman world’ in L. Brubaker & J. Smith (eds), Gender in the Early Medieval World, 44-69. Cambridge. Hartman, G.W. (1949): ‘Clothing: personal problem and social issue’, Journal of Home Economics 41: 295-298. Hendrickson, H. (1996a): ‘Introduction’ in H. Hendrickson (ed.), Clothing and Difference. Embodied Identities in Colonial and Post-Colonial Africa, 1-16. Durham & London. Hendrickson, H. (1996b): ‘Bodies and flags: The representation of Herero identity in colonial Namibia’ in H. Hendrickson (ed.), Clothing and Difference. Embodied Identities in Colonial and Post-Colonial Africa, 213-244. Durham & London. Hiler, H. & Hiler, M. (1939): Bibliography of Costume. New York. Hingley, R. (1996): ‘The ‘legacy’ of Rome: the rise, decline, and fall of the theory of Romanization’ in J. Webster & N. Cooper (eds), Roman Imperialism: Post-colonial Perspectives, 35-48. Leicester. Hingley, R. (1997): ‘Resistance and domination: social change in Roman Britain’ in D.J. Mattingly (ed.), Dialogues in Roman Imperialism: Power, Discourse, and Discrepant Experience in the Roman Empire. Journal of Roman Archaeology Suppl. Series 23: 81-102. Portsmouth (USA). Hingley, R. (2005): Globalizing Roman Culture. Unity, Diversity and Empire. Oxford & New York. Jones, S. (1997): The Archaeology of Ethnicity. Constructing Identities in Past and Present. London & New York. Jones, D. & Hill-Burnett, J. (1982): ‘The political context

Blümner, H. (1911): ‘Die römischen Privataltertümer’, Handbuch der klassischen Altertumswissenschaft IV.2.II. Munich. Böhme, A. (1985): ‘Tracht- und Bestattungssitten in den germanischen Provinzen und der Belgica’ ANRW II.12.3: 423-455. Böhme-Schönberger, A. (1997): Kleidung und Schmuck in Rom und den Provinzen. Stuttgart. Boppert, W. (ed.) (1992): CSIR Deutschland II,6: Germania Superior: Zivile Grabsteine aus Mainz und Umgebung. Mainz. Chausson, F. & Inglebert, H. (eds) (2003): Costume et sociéte dans l’antiquité et le haut moyen age. Paris Cleland, L., Harlow, M. & Llewellyn-Jones, L. (eds) (2005): The Clothed Body in the Ancient World. Oxford. Colburn, C.S. & Heyn, M.K. (eds) (2008): Reading a Dynamic Canvas. Adornment in the Ancient Mediterranean World. Newcastle. Colledge, M.A.R. (1976): The Art of Palmyra. London. Comaroff, J.L. & Comaroff, J. (1992): ‘Bodily reform as historical practice’ in J.L. Comaroff & J. Comaroff (eds), Ethnography and the Historical Imagination, 69-91. Boulder. Čremošnik, I. (1964): ‘Die einheimische Tracht Noricums, Pannoniens und Illyricums und ihre Vorbilder’, Latomus 23: 760-773. Croom, A. (2002): Roman Clothing and Fashion. Stroud. Dar, S.N. (1969): Costumes of India and Pakistan: A Historical and Cultural Study. Bombay. Davis, F. (1992): Fashion, Culture and Identity. Chicago. Drinkwater, J. & Vertet, H. (1992): ‘“Opportunity’ or ‘Opposition” in Roman Gaul?’ in M. Wood & F. Queiroga (eds), Current Research on the Romanisation of the Western Provinces. BAR 571: 25-28. Eco, U. (1979): A Theory of Semiotics. Bloomington. Edmondson, J. & Keith, A. (eds) (2008): Roman Dress and the Fabrics of Roman Culture. Toronto. Espérandieu, E. (1907-1981): Recueil Général des BasReliefs, Statues et Bustes de la Gaule Romaine. Paris. Faber, A. & Jilek, S. (2006): ‘Das Rollenbild norischpannonischer Frauen und seine Darstellung in den Gräbern: drunter und drüber’, Bayerische Vorgeschichtsblätter 71: 149-157. Fitz, J. (1957): ‘Az Eraviskusz női viselt / Die Tracht der Eravisaerinnern,’ Archaeologiai Entesitő 84: 133-154. Freigang, Y. (1997a): ‘Die Grabmäler der gallo-römischen Kultur im Moselland. Studien zur Selbstdarstellung einer Gesellschaft’, Jahrbuch des Römisch-Germanischen Zentralmuseums 44.1: 277-440. Freigang, Y. (1997b): ‘Die Bedeutung der Kleidung und der Attribute auf Grabmälern im nordöstlichen Teil der Provinz Gallia Belgica’ in B. Djuric & I. Lazar (eds), Akten des 4. internationalen Kolloquiums über Probleme des provinzialrömischen Kunstschaffens, 107-117. Ljubljana. Galliou, P. (1992): ‘L’Armorique romaine: mutations et resistances’ in M. Wood & F. Queiroga (eds), Current

66

Ursula Rothe: Dress and cultural identity in the Roman Empire

of ethnogenesis: an Australian example’ in M.C. Howard (ed.), Aboriginal Power in Australian Society, 214-246. St. Lucia. Kovrig, J. (1937): Die Haupttypen der kaiserzeitlichen Fibeln in Pannonien. Budapest. Krier, J. (1981): Die Treverer ausserhalb ihrer civitas. Mobilität und Aufstieg (Trierer Zeitschrift Beiheft 5). Trier. Krüger, E. (1938): ‘Ein römischer Familien-Grabstein in Koblenz’ in Festschrift A. Oxé, 128-134. Darmstadt. Kuper, H. (1973): ‘Costume and identity’, Comparative Studies in Society and History 15.3: 348-367. Lambrechts, P. (1966): De geestelijke weerstand van de westelijke provincies tegen Rome. Brussels. Láng, M. (1919): ‘Die pannonische Frauentracht’, Österreichische Jahreshefte 19-20: 209-260. Beiblatt. Langlois, S. (1959-1962): ‘La vêtement gallo-romain d’après les scènes figureés sur des reliefs du Museé Archéologique de Dijon’ Mémoires de la Commission des Antiquités du Département de la Côte-d’Or 25: 195208. Lurie, A. (1992): The Language of Clothes. London. Mattingly, D.J. (1997): ‘Dialogues of power and experience in the Roman Empire’ in D.J. Mattingly (ed.), Dialogues in Roman Imperialism: Power, Discourse, and Discrepant Experience in the Roman Empire. Journal of Roman Archaeology Suppl. Series 23: 7-26. Portsmouth (USA). Mautner, K. & von Geramb, V. (1932): Steirisches Trachtenbuch. Graz. Mehta, R. (1990): ‘From Purdah to Modernity’ in B.R. Nanda (ed.), Indian Women: from Purdah to Modernity, 113-128. London. Mosser, M. (2003): ‘Die Bevölkerung von Vindobona im Spiegel ihrer Denkmäler’ in P. Noelke (ed.), Romanisation und Resistenz in Plastik, Architektur und Inschriften der Provinzen des Imperium Romanum. Neue Funde und Forschungen, 363-384. Mainz. Mrázek, R. (1997): ‘Indonesian dandy. The politics of clothes in the late colonial period, 1893-1942’ in H. Schulte Nordholt (ed.), Outward Appearances. Dressing State and Society in Indonesia, 117-150. Leiden. Nandy, A. (1983): The Intimate Enemy: Loss and Recovery of Self under Colonialism. Delhi & Oxford. Olson, K. (2008): Dress and the Roman Woman. Selfpresentation and Society. London/New York. Patek, E. von (1942): A Pannoniai fibulatipusok elterjedése és eredete. Verbreitung und Herkunft der römischen Fibeltypen in Pannonien. Budapest. Pausch (2003): Die römische Tunika. Ein Beitrag zur Peregrinisierung der antiken Kleidung. Augsburg. Pfister, R. (1934-1940): Textiles de Palmyre I-III. Paris. Pippidi, D.M. (ed.) (1976): Assimilation et résistance à la culture gréco-romaine dans le monde ancien. Travaux du VI congrés international d’Études classiques 1974. Paris. Pitts, M. (2007): ‘The emperor’s new clothes? The utility of identity in Roman archaeology’, American Journal of Archaeology 111: 693-713.

Pochmarski, E. (2003): ‘Zur Typologie und Chronologie der sog. norischen Mädchen’, Anodos 3: 181-193. Pochmarksi, E. (2004): ‘Das sogenannte norische Mädchen. Ein Beispiel für den Ausdruck lokaler Identität in der provinzialrömischen Plastik’ in A. Schmidt-Colinet (ed.), Lokale Identitäten in Randgebieten des Römischen Reiches. Akten des Internationalen Symposiums in Wiener Neustadt, 24. - 26. April 2003 (Wiener Forschungen zur Archäologie 7), 161-173. Vienna. Raaflaub, K.A. et al. (eds) (1987): Opposition et résistance à l’empire d’Auguste à Trajan: 9 exposés suivis de discussions; Vandoeuvres-Genève 25-30 août 1986 (Entretiens sur l’antiquité classique 33). Geneva. Raepsaet-Charlier, M.-T. (2001a): ‘Charactéristiques et particularités de l’onomastique trévire’ in M. DondinPayre & M.-T. Raepsaet-Charlier (eds), Noms, identités culturelles et romanisation sous le Haut-Empire, 343398. Brussels. Raepsaet-Charlier, M.-T. (2001b): ‘Onomastique et romanisation: éléments d’une comparaison entre les provinces de Gaule Belgique et de Germanie inférieure’ in M. Dondin-Payre & M.-T. Raepsaet-Charlier (eds), Noms, identités culturelles et romanisation sous le HautEmpire, 399-470. Brussels. Ramanujan, M. (1984): ‘The language of clothes: an Indian perspective’, Media Development 31(4): 30-33. Renne, E.P. (1996): ‘Virginity cloths and vaginal coverings’, in Ekiti, Nigeria’ in H. Hendrickson (ed.), Clothing and Difference. Embodied Identities in Colonial and PostColonial Africa, 19-33. Durham & London. Roach, M.E. & Eicher, J.B. (eds) (1965): Dress, Adornment and the Social Order. New York. Roche-Bernard, G. (1993): Costumes et textiles en Gaule romaine. Paris. Rose, A.M. (1962): ‘A systematic summary of symbolic interaction theory’ in A.M. Rose (ed.), Human Behaviour and Social Processes, 3-19. Boston. Rothe, U. (2009): Dress and Cultural Identity in the RhineMoselle Region of the Roman Empire. BAR S2038. Oxford. Ruesch, J. & Kees, W. (1956): Nonverbal Communication: Notes on the Visual Perception of Human Relations. Berkeley & Los Angeles. Schlippschuh, O. (1974): Die Händler im römischen Kaiserreich in Gallien, Germanien und den Donauprovinzen. Amsterdam. Schmidt-Colinet, A. (2000): ‘“Best wishes from China”. Ancient textiles from Palmyra. A glimpse on globalization in antiquity’ in S. Düll, O. Neumaier & G. Zecha (eds), Das Spiel mit der Antike zwischen Antikensehnsucht und Alltagsrealität. Festschrift zum 85. Geburtstag von Rupprecht Düll, 281-289. Möhnesee. Schmidt-Colinet, A. (2004): ‘Palmyrenische Grabkunst als Ausdruck lokaler Identität(en): Fallbeispiele’ in A. Schmidt-Colinet (ed.), Lokale Identitäten in Randgebieten des Römischen Reiches. Akten des Internationalen Symposiums in Wiener Neustadt, 24. 26. April 2003 (Wiener Forschungen zur Archäologie 7), 189-198.Vienna.

67

Dress and Identity

Schmidt-Colinet, A., Stauffer, A. & al-As’ad, K. (2000): Die Textilien aus Palmyra. Neue und alte Funde. Mit Beiträgen von H. Böhmer, L. von Falkenhausen, R. Karadag und A. Rinuy (Damaszener Forschungen 8). Mainz. Schoss, J. (1996): ‘Dressed to “shine”: Work, leisure and style in Malindi, Kenya’ in H. Hendrickson (ed.), Clothing and Difference. Embodied Identities in Colonial and Post-Colonial Africa, 157-188. Durham & London. Schreiber, W. (ed.) (2000): Vom Imperium Romanum zum Global Village. ‘Globalisierungen’ im Spiegel der Geschichte. Neuried. Schulte Nordholt, H. (1997): ‘Introduction’ in H. Schulte Nordholt (ed.), Outward Appearances. Dressing State and Society in Indonesia, 1-37. Leiden. Scott, J.C. (1990): Domination and the Arts of Resistance: Hidden Transcripts. New Haven. Sebesta, J.L. & Bonfante, L. (eds) (1994): The World of Roman Costume. Madison. Stauffer, A. (1995): ‘Kleider, Kissen, bunte Tücher’ in A. Schmidt-Colinet (ed.), Palmyra: Kulturbegegnung im Grenzbereich, 57-71. Mainz. Stauffer, A. (1996): ‘Textiles from Palmyra: Local production and the import and imitation of Chinese silk weavings’ in Palmyra and the Silk Road: Special Issue Documenting the Activities of the International Colloquium, Damascus 1996 = Les annales archéologiques arabes syriennes 42: 425-430. Stone, G.P. (1962): ‘Appearance and the self’ in A.M. Rose (ed.), Human Behaviour and Social Processes. Boston. Taha, A. (1982): ‘Men’s costume in Palmyra’, Les annales archéologiques arabes syriennes 32: 117-132. Tarlo, E. (1996): Clothing Matters: Dress and Identity in India. Chicago & London. Thompson Miller, M. (1979): ‘Sexual differentiation and acculturation in Potawatomi costume’ in J.M. Cordwell & R.A. Schwartz (eds), The Fabrics of Culture: The Anthropology of Clothing and Adornment, 313-330. Den Haag. Tonkinson, M.E. (1990): ‘Is it in the blood? Australian

Aboriginal identity’ in J. Linnekin & L. Poyer (eds), Cultural Identity and Ethnicity in the Pacific, 191-309. Honolulu. Turner, T. (1991): ‘Representing, resisting, rethinking: historical transformations of Kayapo culture and anthropological consciousness’ in G.W. Stocking (ed.), Colonial Situations: Essays on the Contextualizations of Ethnographic Knowledge, 285-313. Madison. Van Dijk, K. (1997): ‘Sarong, jubbah and trousers. Appearance as a means of distinction and discrimination’ in H. Schulte Nordholt (ed.), Outward Appearances. Dressing State and Society in Indonesia, 39-83. Leiden. Wallace-Hadrill, A. (2008): Rome’s Cultural Revolution. Cambridge. Webster, J. (1996): ‘Roman imperialism and the ‘post imperial age’ in J. Webster & N. Cooper (eds), Roman Imperialism: Post-colonial Perspectives, 1-17. Leicester. Webster, J. (1997): ‘A negotiated syncretism: readings on the development of Romano-Celtic religion’ in D.J. Mattingly (ed.), Dialogues in Roman Imperialism. Power, Discourse and Discrepant Experience in the Roman Empire. Journal of Roman Archaeology Suppl. Series 23: 165-184. Portsmouth (USA). Wild, J.P. (1968a): ‘Die Frauentracht der Ubier’, Germania 46(1): 67-73. Wild, J.P. (1968b): ‘Clothing in the north-west provinces of the Roman Empire’, Bonner Jahrbuch 168: 166-239. Wild, J.P. (1970a): Textile Manufacture in the Northern Roman Provinces. Cambridge. Wild, J.P. (1970b): ‘Textilfunde aus der Memoria IIK in Xanten’, Bonner Jahrbuch 170, 267-270. Wild, J.P. (1977): The Textiles, Vindolanda III. Haltwhistle. Wild, J.P. (1985): ‘The clothing of Britannia, Gallia Belgica and Germania Inferior’, ANRW II.12.3: 362-423. Wilson, L.M. (1924): The Roman Toga. Baltimore. Wilson, L.M. (1938): The Clothing of the Ancient Romans. Baltimore. Woolf, G. (1998): Becoming Roman. The Origins of Provincial Civilization in Gaul. Cambridge.

68

Chapter 7

Investigating the Emperor’s Toga: Privileging Images on Roman Coins Ray Laurence

All clothing is men’s or children’s or women’s or that which can be worn by either sex or that worn by slaves. Men’s clothing is provided for the benefit of the pater familias, such as togas, tunics, cloaks, bedspreads, coverlets, blankets and the like. Children’s garments are clothes used only for this purpose, such as togae praetextae, coats, chlamydes, and cloaks we provide for our sons. Women’s clothes are those acquired for the materfamilias which a man cannot easily use without incurring censure, such as robes, wraps, undergarments, head coverings, belts, turbans, which have been acquired more with a view to covering the head than for decorative effect, coverlets and mantles. Clothes adapted to the use of either sex are those which a woman shares in common with her husband, for instance, where a mantle or cloak is of the type that a man or his wife can wear it without criticism, and other garments of this nature. Slaves’ clothing is that acquired for dressing the household, such as blankets, tunics, mantles, bed-linen and the like (Digest 34.2.23).

we might describe as accessories (Swift 2000, 2003, 2004 and this volume), and of surviving textiles (see Rogers, Jørgenson and Rast-Eicher 2001), but the dominant mode of analysis depends on the use of a selective corpus of visual images from which we may begin to reconstruct or discuss the manner in which persons were dressed (Goldman 1994; Stone 1994). The interest in reconstruction of dress from sculpture goes back to at least the 19th century and Thomas Hope’s 1812 Costumes of the Greeks and Romans which drew on material in the new museums of Napoleonic Paris full of material looted from Italy, and research undertaken at the re-organised L’École Française de Rome (Hope 1812: xvi; compare Wilson 1924; Houston 1931; Wilson 1938; Laver 1964; Richardson and Richardson 1964 and 1966; Symons 1987). Wilson (1938: v) suggests that there are three sources for the study of clothing: literature, Roman sculpture and painting, of which she regards texts as ‘incidental’, the paintings ‘scanty’ and the sculpture as ‘by far the most definite’. In the tradition of Roman dress history we find a prioritisation of sculpture over all other forms evidence. Another factor should not be underplayed in the reception of antiquity: place a set of people into dress that is labelled ancient and you will find that viewers will regard them as having the characteristics, gestures and features of the films of Hollywood epic (Wyke 1996). However, what they are actually representing is a moving image of an antique statue that should be seen to have a tradition going back at least to Emma Hamilton’s famous “Attitudes” through which she created a moving image of goddesses, performed in the courts of Europe (Jenkins & Sloan 1996: 252-61). After one of Emma’s performance, Goethe was to comment that Hamilton had found in Emma ‘all the antiquities, all the profiles of Sicilian coins, even the Apollo Belvedere’ (Goethe 1962: 208) and Emma was to appear in numerous paintings as a classical goddess, a Bacchante, or a Sibyl (Jenkins & Sloan 1996: 263-75). Emma Hamilton’s inspiration and repertoire drew on sculpture as did Hope’s 1812 volume on Greek and Roman costume. Indeed, we should see Hope’s volume as engaged with this tradition of late 18th century re-enactment, and the imagining of an inhabited past with dressed figures. Just as today, we can find a close relationship between historical

[Augustus] desired also to revive the ancient fashion of dress and once when he saw in a contio a throng of men in dark cloaks, he cried indignantly, ‘behold them “Romans, lords of the world, the nation clad in the toga”’ and he directed the aediles never again to allow anyone to appear in the Forum or its neighbourhood except in the toga and without a cloak’ (Suetonius Augustus 29.5). The study of Roman dress, including its reconstruction, is dominated by the study of statuary, relief sculpture, and figures in wall-paintings and on mosaics (Candilio 2004; Pisani Sartorio & Liberati Silverio 2000; Croom 2000; Roche-Bernard & Ferdiere 1993; papers in Llewellyn-Jones 2002 and Sebasta and Bonfante 1994; interestingly those in Cleland, Harlow & Llewellyn-Jones 2005 and Chausson & Inglebert 2005 have a different focus with a greater emphasis on texts). It needs to be noted that there has also been close attention paid to the material remains of what

69

Dress and Identity

re-enactment, televised or not, and the development of dress history, that leads the re-enactor towards a sculptural pose.

occurring under the new regime of Nerva and Trajan. Yet as Uzzi (2005) demonstrates numerous 2nd century images have their origins in the 1st century AD.

The privileging of sculpture over other forms of representation creates the pattern of Roman dress history, but this phenomenon is not unique to historians of Roman dress. Two recent publications on dress and imperial imagery have drawn on statuary, texts and coinage of the 2nd century AD (Hales 2005; Rawson 2001, 2003). The first, by Shelley Hales, dwells at length on the toga of emperors in texts and in sculpture, but pays little attention to the representation of dress on coins. The second by Beryl Rawson examines the representation of children in 2nd century imagery. Her focus, in contrast to that of Hales, is on statuary and coinage in the 2nd century AD and interestingly she does not relate the imagery from the 2nd century to earlier images found mostly on coins in the mid to late 1st century, notably under the condemned Nero and Domitian. Perhaps taking a lead from Pliny the Younger’s works, notably his Panegyricus, there is an impression of a new beginning and a new imagery in the 2nd century and this keys these findings into a long-established pattern of viewing a disjuncture in imperial art between that of the ‘Spanish’ emperors and their predecessors (see Breckenbridge 1981: 497-8). The problem with these different analyses is that they are in part determined by the survival of a number of key images – most notably the Arch at Beneventum at the beginning of the Via Traiana, and an absence or non-survival of the sculptural images of condemned emperors (see papers in Varner 2001 for discussion of survival of images and inscriptions of Nero and Domitian). The Trajanic Arch at Beneventum is an iconic monument in the study of art history: for Bianchi Bandinelli (1970: 235), this monument was the first time that ‘the lower classes appear on an official monument’. The inequality between ruler and his subjects is beautifully expressed via the medium of dress and gesture of the participants engaging with their emperor (on gesture as a cultural system see Corbeill 2004: 1-11). However, to say it is the first of such monument is problematic. Varner (2004) has demonstrated that following the damnatio memoriae of Domitian’s statues, referred to by Pliny (Paneg.52.4-5; see Stewart 2003: 267-90 on statue riots), many were transformed into the image of Nerva. In fact 82% of all surviving statues of Nerva were re-cut images of Domitian. A similar phenomenon is identified with reference to the statues of Caligula and Nero. The head is the only item re-cut to represent an older man on a young man’s body. Damnatio could extend to the words of the deposed emperor: a good saying of Domitian, ‘No one believes there are plots until the emperor is dead’, was deliberately and consciously re-allocated in antiquity to the ‘good’ emperor Hadrian (Laurence & Paterson 1999). The killing of the emperor Domitian created a disjuncture in our evidence and is reflected in the writings of Pliny the Younger in both his Panegyric and his Letters. However, Bartsch (1994: 162-5) demonstrates the continuities of the rhetoric of power under Domitian and Trajan. Reading this fracture within our body of evidence, it is perfectly possible to see within it a difference in attitude and representation

This paper takes the analysis of Hales and Rawson a little further to examine the development of the images of the dressed emperor and his subjects (adult males, adult females and children) through the medium of coinage and to privilege coinage as a body evidence for the study of dress history. There are coins representing every reign, whether the emperor was ‘good’, ‘bad’ or ‘indifferent’, and they can be seen as a system of representation with a precise dating system that can be unpicked and analysed and, unlike sculpture or texts, coins survive from all periods and all reigns. The paper seeks to challenge a disjuncture in the representation of the emperor and his subjects over the period from c. AD 40 to AD 120, and resists a dichotomy between the forms of representation of emperors in the 2nd century and those of the Flavian and Julio-Claudian emperors in the 1st century. The images on coins, although they might be dismissed by Roman dress historians as lacking detail, had a resonance for those living across the empire. The images were created not just in Rome but also in the provinces and, unlike sculpture, coins circulated across the empire and were produced as an image of home or of a shared ideology of Rome’s homeland (a factor in the distribution of togate sculpture in the North-West provinces: see Wild 1968: 188-92). These are images of what it meant to be Roman (Uzzi 2005: 16) and allow us to view the role of dress in the creation of an image of Romaness and of masculinity.1 By focussing on images of dressed emperors and their subjects on coins, the paper maintains a distance from the very idea that the clothing represented can be remade – since the lack of actual detail prevents recreation – and aligns the paper with recent scholarship on the role of dress as a narrative device within texts (Linderski 2002: 342-64; Harlow 2004, 2005; Callu 2004 on dress in later texts). As Harlow (2005: 152) has suggested: ‘Using dress to track change over time is as complex a business as tracking changes in dress’ and there is a danger that we develop a circularity of argument: for example, dress changes because the emperor has no children, and the emperor has no children and hence the nature of dress on coins changes. However, maybe just maybe, coinage (unlike texts) has biases that point in a similar direction rather than in numerous different directions according to time and the authorial voice (see Kleiner 2001: 56 on difference in representation of women in art according to medium). It is this factor that gives the images on coins a greater relevance for the understanding of the meaning of dress and the ability of dress to become part of the construction of both specific events and abstract concepts under the Roman emperors.

See opening quotation from the Digest 34.2.23; Dyck (2001) on Cicero’s use of the imagery of dress to disfranchise or even emasculate his opponents; Heskell (1994); Vout (1996); Richardson & Richardson (1966) on the imagery of the toga. 1

70

Ray Laurence: Investigating the Emperor’s Toga

Methodology: learning from New Dress History

theme) has stressed that in terms of Roman dress, it was acceptable for a Roman to go Greek and wear a pallium in his hortus, but for public occasions the toga was the dress of the Roman man (Dyck 2001: 125). It was acceptable to code-switch from one form of dress to another dependant on the context. Just as an emperor speaking to troops as imperator would not wear a toga, the emperor as princeps (first citizen) wears a toga to address soldiers not directly under his command. However, at moments of crisis or if the state was imperilled, the emperor could switch from a toga into military dress. As habitus changed so could the dress of the emperor. In this paper we will examine togate figures, and emperors in standing position or sitting on a rostra in front of inferiors. It is easy as the viewer to make the conceptual jump to assume that the emperor is addressing his subjects. Glenys Davies (2005: 121) reminds us that such images of toga-clad Romans are representations, rather than images of how people wore a toga or other forms of dress. Just as the togate statue is designed to express masculine power and control, the toga-clad emperors and subjects represent a similar vision (Davies 2005: 126). A feature that was aped by children represented wearing the toga praetexta (see images in Gabelmann 1985; also Richardson and Richardson 1966: 253-5, 265-8 on the toga and tirocinium) and thus via the images asserting their citizenship and superiority to those who were not citizens – slaves or foreigners (Uzzi 2005: 29-30) The appearance of children in association with the emperor on coinage has an important ideological value and Pliny the Younger makes it clear that the presence of children was a feature of scenes of imperial largesse under both Trajan and his predecessors:

The entire field of dress history or dress studies has burst across old boundaries and flourishes in a more open-minded multi-disciplinary atmosphere (Taylor 2004: 279; compare Breward 1998; Ribeiro 1998). Dress history is a new subject, with a shorter history than that of archaeology (including New Archaeology). It is a contested arena, in which a ‘New Dress History’ has emerged, in part from a greater interest in dress within gender studies and New Art History (Breward 1998; Ribeiro 1998; Taylor 1998; Styles 1998). At its most sophisticated it can link a change in footwear to the nature of the built environment and produce a startling insight into the gendered nature of walking in the 18th century city (McNeil & Riello 2005). However, since dress historians do not really look at antiquity, archaeologists and historians will have to pursue research in this new field (but see Sichell 1983, Black, Garland & Kennett 1980 for two page summaries on Roman dress). There are a number of principles for the following analysis of dress on Roman coins drawn from recent publications by dress historians that need to be stated at the outset: • ‘there are three basic concepts which inform consideration about human appearance: form (the body), function and fashion’ (Cumming 1998: 18): fashion signifies the human mind’s adaptation of the function of clothing • distinction in dress presents a system whereby each class is actively distinguished (Taylor 2002: 95): in the case of the Roman toga distinction could be created through different styles and quality of fabric • Dress historians have rejected the use of ‘high literature’ as opposed to diaries to normalise the celebrity or elite practices of dress (Taylor 2002: 90) • character, likeness and costume are fused visually (Taylor 2002: 95): on Roman coins dress actively contributes to our understanding of the characteristic of the emperor portrayed via the obverse • fashion is part of the cultural construction of an embodied identity (Cumming 2004): we should note that the coins featuring an emperor (always dressed) are themselves subject to the fashion of representation in this medium • dress is not just about ‘clothing’ the body, but should be understood as part of the formation of a normative understanding of status and gender (Breward 1998: 303) • the visual image or representation of a dressed person(s), for example on Roman imperial coinage, is not reporting an event but part of the creation of the event itself (drawing on Ribeiro 1998), for example the distribution of coins at a congiarium (see below).

On the day of a congiarium it had been the custom for swarms of children, the populus of the future, to watch the emperor’s public appearance and line his path. Every parent’s concern was to show his little ones mounted on his shoulders, to teach them flattering words and fawning phrases…’ (Pliny, Panegyric. 26) The images of dressed citizens and children on coins maintain a concept of stability and continuity in the institution of imperial rule between the past and the present and for the future generations to come, even if the emperor represented on the reverse may have changed (see Pani 1991). Coins and images There is a picture-language on coins that develops over the course of the latter part of 1st century and early decades of the 2nd century AD, identified by Jocelyn Toynbee (1956). The scenes of a toga-clad emperor with his subjects under the legends of Annona (corn supply personified), Liberalitas (liberty), Alimenta (food/money distribution) replace, or can be replaced by, visual personifications of these attributes of government. The pose and position of the standing or seated emperor on coins may owe much to the posture and gesture of gods and goddesses on earlier or contemporary coins.

The dressed figures on coinage are a representation of a particular form of cultural habitus (Sebasta 1994: 46). Andrew Wallace-Hadrill (1998; 2008 expands on this

71

Dress and Identity

Fig. 7.1: Orich sestertius of Nero, unknown mint. Obverse. NEROCLAVDCAESARAVGGERPMTRPIMPPP Laureate head of Nero l., globe at point of bust Reverse. ANNONA AVGVSTI CERES Ceres seated l.; before her Annona; between them an altar with modius; in background prow Weight: 28.67g The Barber Institute Coin Collection, R0967; G. Haines Collection.

Fig. 7.2: Silver denarius of Trajan, mint of Rome AD112-114 Obverse. IMPTRAIANOAVGGERDACPMTRPCOSVIPP Bust of Trajan r., laureate, draped on l. shoulder Reverse. ALIMITAL(in exergue); S•P•Q•R•OPTIMOPRINCIPI Abundantia standing l. holding ears of corn and cornucopiae; at her feet a child standing and holding a roll Weight: 2.87g The Barber Institute Coin Collection, R1086; G. Haines Collection.

The paper by Toynbee includes an analysis of the creation of the personification of Annona. Visual personifications, such as Annona, on coins may have a direct relationship to statuary and friezes erected in Rome and across the empire (Stewart 2003: 208-14 on statues on coins). Cornelius Vermeule (1987: 12) observes that Alimenta or Tutela may be seen as personifications of a similar nature to Annona, or Ops, and that there is a problem in the identification of fragmentary statuary of these deities, since they share so many of the characteristics of the goddess Fortuna and, hence, any fragment will be confidently identified as the goddess Fortuna in art historical literature (Vermeule 1987: 20). The picture language of the emperor has three primary guises determined by his dress: general/victor in military dress, as the serving princeps in toga, or as priest with toga drawn over his head (Harlow 2005: 145). From the mid 1st century AD into the mid 2nd century AD there is a gradual shift from visual representation of the specific event such as a distribution of money, a congiarium, to the use of this same imagery to represent a more generic concept, such as Liberalitas. The image of dressed figures on coinage shifts from being part of a specific event to become part of the process of government. It then articulates the abstract concept of government by utilising images of specific events which had become embedded as the images associated with coinage. As we shall see, it was in the mid 1st century that a new set images of dressed emperors, their children, and their citizens were created and became part of a repertoire of coin images that would be appropriated or re-invented in the 2nd century to represent abstract concepts of good government. These concepts coincide with many of those found in Pliny’s Panegyricus which presents Trajan as the exceptional “good” emperor.

an individual but the collective body of Roman citizens or their actions (Crawford 1974, nos 301/1, 372, 413/1, 419/2; 351/1 is difficult to interpret) or persons from the past (Crawford 1974 no. 433/1). Under the triumvirate this changed: Q. Cornuficius presented himself on his coins not as an imperator but as a toga-clad augur (priest) (Crawford 1974 nos 509/1-5, for comparison see Buttrey 1956). We also find an historical representation of a toga-clad M. Claudius Marcellus from 222 BC depositing a trophy (the spolia opima) in a temple during his fifth consulship (BMCRR 4206-8). The toga-clad emperor did not appear on coinage until after the secular games of 17 BC, and remained a rarity. There is an appearance of a generous toga-clad Augustus at the games in 16 BC (BMCRR 4487), which was an innovation that Augustus also applied in the realm of statuary from the same date – 17 BC (Zanker 1988: 127). The imagery of a toga-clad emperor in association with toga-clad citizens comes to symbolise the notion that the emperor was also a citizen. This can be seen from the coins minted under Augustus that include toga-clad images of Agrippa and Augustus on the obverse (BMCRR 4646, 4647, 4657-59, 4660; BMC Augustus 103, 115) symbolising not only their equality but also their status as Roman citizens, superior to provincials yet equal to other Romans. Augustus’ image as a toga-clad princeps is even maintained in military contexts (BMC Augustus 443) and when receiving children as hostages from barbarians (BMC Augustus 492). However, the image on the obverse should not be read in isolation and turning to the reverse, the viewer would find the image of the emperor’s head in profile – the greatest indication possible of social difference and without precedent in the Republic prior to the dictatorship of Caesar (Wallace-Hadrill 1986: 68). Turning back to the obverse, words were included to explain the scene – a further innovation and quite different from the earlier Roman coins of the Republic (Levick 1999 examines the debates surrounding the impact of these words). I do not want to enter the debate over whether these novel images persuaded

The image of a real person dressed in a toga, displaying rhetorical gestures was part of the new iconography of the emperors on coins. During the Republic, toga-clad figures seldom appeared, and where they did they represented not

72

Ray Laurence: Investigating the Emperor’s Toga

the exception of Nero in Alexandria (RPC 1. 5203, 5242, 5253), but as togate Romans: Britannicus appears on a coin from Tralles (RPC 1. 2654) in a toga with his sisters Octavia and Antonia on a coin minted from Caesarea in Cappadocia (RPC 1.3627) and another from Caesarea Paneas in Syria (RPC 1. 4842). The sons of Vespasian appear in togas on the obverse of coins with the head of Vespasian on the reverse in the province of Asia (RPC 2. 805, 811, 819). This is a contrast to the representation of the demoi (people) of their own cities, who are dressed and posed quite differently (e.g. RPC 1. 2928, 2.1369). Roman men were represented in togas, whereas Greeks were represented in Greek dress. Unsurprisingly we find victorious commanders represented in military dress (RPC 1. 3629). Women in the imperial family were represented dressed, but with the attributes of goddesses with the exception of the coin representing Octavia performing a ritual on the obverse and the head of Agrippina on the reverse. Here the iconography follows a pattern of male children of an emperor in togas performing a ritual with the head of their father on the reverse of the coin. The Romaness of these images was created by their dress, clearly differentiated from that of the Greek provincials who were minting the coins.

Fig. 7.3: Orich sestertius of Hadrian, mint of Rome AD 118 Obverse. IMPCAESARTRAIANVS HADRIANVSAVG Bust of Hadrian laureate r., with drapery on l. shoulder Reverse. PONTMAXTRPOTCOSIISC LIBERALITASAVG (in exergue) Hadrian seated l. on platform; in background, Liberalitas standing l., holding tesserae Weight: 21.08 The Barber Institute Coin Collection, R1146; G. Haines Collection.

or not, but instead I want to observe their development and recognise that the moneyers at the mint had developed a visual language of imperial rule that focussed on a dressed emperor interacting with his subjects by the end of the 1st century AD (Wallace-Hadrill 1986 offers a balanced view of the debate over propaganda). The full articulation of this visual language is investigated in the four case studies below.

Returning to the coinage of the imperial mints and the representation of the arrangement of a standing togate emperor on coinage, we need to consider the meaning of the image that is presented by the dressed figure. The toga of the emperor can be viewed as a symbolic act by the moneyer at the mint to recall a vital moment: the point in a speech or at an event that is depicted. The toga was part of what Cicero called the sermo corporis (language of the body) or the eloquentia corporis (eloquence of the body – De Oratore 3.222; see Davies 2005: 125, and Graf 1991 for further analysis of posture and gesture) that enables the delivery of a speech. As Quintilian (Inst.11.3.137-49;) makes clear, the speaker at the commencement of his speech was expected to have his toga well positioned; whereas by the end of a speech, the toga could be dishevelled through the exertion of speaking and this transformation represented the man of action. The linkage between gesture and dress should be maintained, since Cicero (Off.1.130) makes a connection between the gendering of both gestures and dress: ‘one must remove from one’s appearance all dress unworthy of a man, and one should beware of similar mistakes in gestures and movements’ (Graf 1991: 45). The figure giving a speech needed to be standing erect, a sign of nobility, and also had to avoid any accusations of arrogance (Graf 1991: 46) and to appear as an equal to his audience of social inferiors. The toga was a means to create an environment of habitus of social equality, whilst at the same time limiting the movement and gestures available to the orator. However, Graf (1991: 51) concludes that the gestures of oratory in the middle of the 1st century A.D. were both more violent and livelier than those of Cicero’s time, quite different to those taught by Greek rhetoricians, and distinguished from the gestures of actors on the stage. In toga-clad emperors on coins there is a coincidence with the deportment and posture found in statuary that can be seen as distinctly Roman and

The cities of the empire adopted the sculptural representations of emperors, so we should expect to see a similar adoption of the images of togate emperors on the coins minted by cities in the east. However, relatively few coins minted by these provincial cities in the 1st century AD include images of toga-clad figures on the obverse. The mints that do adopt the emperor and his family as an image for their coinage reveal the importance of the toga as marking an individual out as a Roman. We can identify a coin depicting toga clad images of Gaius and Lucius from Pergamon (RPC 1.2362; on statuary of Gaius and Lucius see Pollini 1987). On the coinage of Magnesia, we can identify imperial figures such as Tiberius raising Tyche of Magnesia following the earthquake of AD 17 (RPC 1.2451); Agrippina the Elder dressed with attributes of Demeter and Germanicus with toga arranged for sacrifice (RPC 1.2454); and Agrippina the Younger dressed with attributes of Demeter and a togate Nero (RPC 1.2457). The tradition of representing the empress as Demeter goes back to the time of Augustus with Livia’s appearance with these attributes on a coin from Tralles (RPC 1.2647-8). At Caesarea Paneas in Syria, Agrippina is depicted on the reverse of a coin dressed and seated in the pose of the goddess Fortuna with the standing figure of her daughter-inlaw Octavia making sacrifice with a patera on the obverse (RPC 1. 4845/1, 4845/2). Other coins from other cities would seem to depict Agrippina the Younger as Tyche or Fortuna (RPC 1.4859-60). However, the male members of the imperial family are not represented as gods, with

73

Dress and Identity

as developing a conception of an identity of the emperor as a citizen and as a Roman superior to non-Romans.

message conveyed via the language of dress that the Annona Augusta provided food to the young children of Italy. Choices are being made here: the emperor does not appear – instead we have a personification of the process of food supply to those living in the capital. After Domitian, this image disappears from the coinage of Nerva and Trajan and the mint reverts to the original Neronian imagery of Ceres plus Annona (BMC Nerva 101). The Domitianic image then reappears in the same format under Hadrian (BMC Hadrian 1346, 1347), alongside coins representing the Annona as a single figure holding the rudder of a ship (BMC Hadrian 1574-78) and through the image of just the modius (BMC Hadrian 596-601). This establishes the iconography of the child with undressed lower body.

Case study I: from Annona to Alimenta The coin type with the legend Annona would appear to have developed only after the construction of the new harbour at Portus in the 1st century AD and its use by moneyers was frequent, with revivals and adaptations over the course of the following century or so. These developments illustrate the dynamics of the imagery of coinage into which the figure of a toga-clad emperor or an emperor in military dress was to be inserted. On the early Neronian coins, the figure of Annona is represented as a fully dressed standing female figure holding a cornucopia, adjacent to a seated figure of Ceres with the legend Annona Aug. Ceres SC. In the background is the prow of a ship and we find the addition of a representation of the dry measure known as a modius (BMC Nero 127, 128, 129, 305, 306, 307. See also Fig. 7.1). Later 2nd century developments of the simple image include the further addition of a lighthouse, a rudder and a tablet or list of recipients (BMC Antoninus Pius 1655). Significantly, these are developments from a Claudian Ceres coin (BMC Claudius 136, 138). Ceres had been the traditional format in which empresses had been represented. Annona can appear on her own without Ceres. Annona and Ceres appear semi-dressed from Nero to Domitian, but from Nerva onwards we find them fully dressed. However, it is in the interaction of two figures that we find developments. Under Vitellius (AD 69, BMC 47) the mint at Rome issued a coin with Annona and the emperor in military dress to depict the soldier emperor as the guarantor of the Annona. Vespasian, like Galba, issued coins depicting a seated Annona in the manner of Ceres, but with a bag on her lap (BMC Galba 141, Vespasian 730) and we find also the development of an association between Annona and Aequitas under the Flavians (BMC Titus 152, 153, 154). Under Domitian, the figure of a seated Annona is now holding a bag open on her lap. Behind her is the prow of the ship, but more importantly and innovatively in front of her is a small figure who would seem to be naked from the waist up or from the top of the legs down (BMC Domitian 304, 347, 415). This figure is seen placing or taking something out of the bag depending on the coin type and the viewer’s interpretation. The dress of this small person on the Domitianic coin has few parallels in coin iconography however, famously on the Ara Pacis two halfnaked children are visible – often interpreted as barbarian children within the court of Augustus. In relation to this image on the coin of Domitian, Mattingly (BMC II: xcvii) rejects the notion of this being an actual child and sees the half-naked figure as that of an Italian farmer, which would provide a link to Suetonius’ view of Domitian attempting to revive Italian farming (Dom.7). Such a view does not sit well with the association of Annona with a ship’s prow in the images (very similar to the iconography of Venus as the caretaker of sailors). If we view the image on the coin as that of a half dressed child in a similar form to that of children on the Ara Pacis, we are viewing an ideological

A variation on the simple standing figure of Annona Augusta occurs under Trajan alongside the development of the imagery of the Alimenta Italia and Trajan as Optimus Princeps. Here, Annona appears draped with a crown of corn ears with her left hand holding a cornucopia and ears of corn outstretched in her right to a small person clad in a toga (BMC Trajan 468, 469, 470, 471, 472, 869, 918, 959*, 973, 996). The cornucopia provides a link back to Annona, but the situation of a fully dressed Alimenta provides a means of distinction. There is the additional variant of the child holding a roll of papyrus (BMC Trajan 469; see also Fig. 7.2) to represent the privilege of admission to the Alimenta Italia. The toga-clad child can be identified as representing the citizen and, in association with the figure normally seen on the Annona Augusta coinage, as demonstrating an extension of imperial largesse from the capital Rome into Italy. This would appear to be an extension of the existing image of Annona to represent the new institution of largesse – the Alimenta. The association with Trajan is made via the written message: SPQR Optimo Principi Alim Ital. Other Trajanic coins replace the figure of Annona with that of the emperor. He is dressed in a toga and extends the range of recipients from that of the male child to include female children as well. Trajan appears in such imagery clad in a toga, holding a scroll to represent the lists of the recipients of the Alimenta Italia in his left hand and extending his right hand to two children. The boy is togate and stretches up with only his right hand as befits a toga wearer, whereas the girl is wearing a robe and stretches up with both hands (BMC Trajan 378). This gender distinction is not just created through dress but is extended by gesture and the fact that the male is shown as taller than the female. Texts refer to the release of children from swaddling beginning with the right hand to create a generic right-handed Roman adult (in addition, the left hand was used to control the drapery of the toga, leaving the right free for rhetorical gestures). In the coinage, we see this feature represented as a gender distinction and might consider whether the right handed male can be related to the act of writing and literacy as well as simply being a feature of toga wearing. The legend on this coin has far more to say about the princeps: Alim Ital Cos V PP SPQR Optimo Princ (reference to Trajan’s role in the Alimenta along with holding his 5th consulship, being Pater Patriae,

74

Ray Laurence: Investigating the Emperor’s Toga

alimenta coinage as an anthropomorphic representation of Italia. However, the format utilised is an abbreviation of that found on the alimenta scene from the arch at Benevento, which features more than one female figure. Hence, we might suggest that these figures might instead be Italian mothers, represented by their dress as human rather than as an anthropomorphic representation. After all, no scholar has seen the toga-clad male as a personification of a god or anthropomorphic representation of civitas, hence perhaps we should not be too eager to regard the female clothed figure with children as a goddess.

with the authority of the Senate and People of Rome and being Optimus Princeps). Another Trajanic alimenta coin develops this theme. Trajan is now seated in a curule chair wearing a toga and holding a sceptre in his left hand, he extends his right hand to a female figure carrying a child in her left hand and leading a standing child with her right (BMC Trajan 870, 918, 959). This iconography can be related back to the Domitianic image of Annona Augusta seated before the child discussed earlier in this paper. It might also owe rather more to the imagery associated with Tutela on coinage. Tutela is a familiar term related to guardianship (Buckland 1963: 142-61 is fundamental). It can also be found in the naming of deities as related to that of the Genius of the male head of a Roman family (CIL 2.4092, 3. 445, 13.939; Stat. Silv. 4.6.32). It is almost as though Tutela could be seen as a substitute for the Genius. Horace in Odes 4.14.43 also creates a link with Italia and Tutela presiding over all of Italy. The reverse of a rare coin type of Vespasian minted in AD 70, and others issued in AD 71, shows a seated woman with her right hand on the head of a child standing before her and her left hand on the shoulder of a second child (BMC Vespasian 527, 595, 808*; there are examples Tutela Italiae from the reigns of Nerva, but these are considered to be fakes). However, like many of the Flavian coins the type was taken over from the extant coin types of Vitellius (Krupp & Krupp 1961). The very concept of Flavians as tutors to the newly founded state ties in with the construction of the new dynasty’s legitimacy. Vespasian is emperor, but he is performing the public munera in the position of princeps (Buckland 1963: 142-61 for discussion of Tutela). Significantly, the coin type was issued by Vespasian a hundred years after the battle of Actium (Krupp & Krupp 1961). Vespasian makes an inventory of the estate or the state, he provides for the maintenance of the estate’s children or the citizens of the state, and he manages the patrimonium of the citizens. It is these coins that are the origin of the 2nd century coinage featuring two children and a seated toga clad emperor. The Trajanic coin with the alimenta legend has a togate emperor seated as Optimus Princeps in a central position, just as Tutela had taken that position on the Vespasianic coins. This female figure is often identified as Italia and is integrated into the Trajanic programme through the coinage of restoration with a kneeling female figure, probably in a stola, being hauled to her feet by a toga-clad Trajan, as can be seen from a coin in the same format with the legend Rest(ituta) Italia (BMC Trajan 404). The image could be read as the alien Spanish emperor restoring Italia (Lo Cascio 1998 for discussion). These coin types are modelled on two coins of Vespasian featuring the legends Libertas Restituta and Roma Resurgens with the emperor in a similar position before a kneeling Libertas or Roma (BMC Vespasian 549, 565) and go back to the coinage of Augustus that proclaimed a restored Republic at the beginning of the Empire. The Trajanic format of Restituta Italia presents a more plausible female figure that is also seen to have the same body language and drapery in the alimenta coinage. This has led scholars to view the female figure in the

The image of the togate emperor on these coins can be related back to the original meaning of the togate figures on coins of the republic or even of the first emperor, Augustus. Trajan has been substituted for the goddess Annona/Alimenta and appears as Optimus Princeps, in his toga rather than military dress, enabling the restoration of Italy, care for children and the future of the state. He is simply one of us, but also a superior being who through duty can care for us all as the best man in the state. This leads us to the fact that the togate emperor could replace the allegorical figure of a deity called Alimenta or Annona. There is a deliberate avoidance of his power derived from the military, unlike the civil war imagery of Vitellius assuring the Annona in his breastplate and greaves. The Trajanic toga veils power, whilst at the same time the position of the togate emperor is that of a deity. Case study II: from Congiarium to Liberalitas The figure of a toga-clad emperor engaging with togaclad citizens, adults and children, developed as an image on coins from the middle of the 1st century AD and was consistently utilised by emperors afterwards to represent their relationship with the urban plebs at Rome. The Neronian image drew on that of an Augustan coin (BMCRR 4487), but developed the image further and added a new legend. The coin commemorates, or perhaps was given out, at the Congiarium of Nero in AD 57 (for full discussion of congiaria, see Spinola 1990; Carlson 1975; Brilliant 1963: 77; Hannestad 1986: 111-14; BMC Nero 136, 137, 138 for examples). Nero is seated on a curule chair, wearing a toga, bare headed with his hands on his knees – he is inactive. A tunic wearing attendant, sometimes seated on an ordinary stool, hands a tessera (a token) or some coins to a toga clad citizen, who holds out the fold in his toga to receive it, behind him stands a small boy (Pliny Pan.26 on presence of children at congiaria). In the background, there are representations of the goddess Minerva and the god Liberalitas. Vermeule (1987: 84-5) suggests that the statue was brought out for the occasion and may have resided in the Basilica Julia or later in the Basilica Ulpia (see Mattingly 1967: 148, 162 Millar 1977: 136). A variation strips down this image to that of just Nero on the platform, with the attendant and the citizen placed on the ground with only Minerva presiding over the scene in the background without Liberalitas (BMC Nero 139, 140). This second Neronian type is repeated under Vespasian in AD 72 with

75

Dress and Identity

Titus in the role of togate benefactor and the head of his father on the reverse (BMC Vespasian 629). The congiarium is next shown on the coinage under Nerva. Here, we find a significant innovation, the introduction of a listing or an account board as an accessory. Nerva is seated figure in a curule chair wearing a toga, seated in front of him is a tunic clad attendant with a toga clad citizen standing on the steps to the platform.

made clear in a coin issued by Antoninus Pius, under the legend Cong. Aug. VIII, Cos.III, Liberalitas holds up an abacus and a cornucopia (BMC Ant. Pius 1004, 1005) – the cornucopia also places this figure in close juxtaposition with that of Annona and returns us to the imagery of the goddess Fortuna. We even see under Marcus Aurelius, coins dropping from Liberalitas’ ‘abacus’ into the folds of the toga of the citizen, whilst the togate figures of Marcus Aurelius and Lucius Verus look on (BMC Marcus Aurelius 1277, 1278, 1317).

A striking aside on the imagery is that it coincides in layout with that of a ‘bread distribution’ or ‘bread seller’ from a wall-painting from Pompeii. Looking at the image (see Clarke 2003: 259-61 for image and full description), we can see that the man in the white garment is seated on a platform, as the emperor is on the congiarium coinage. He is distinguished from the other persons by the colour of his garment. The three other figures conform in gesture to those found on the congiarium coins. This includes a smaller younger person. There is evidence at Pompeii for the distribution of honey alongside the munera or spectacle of a beast hunt (CIL 4.1186; Tumolesi 1980 no.25) and on another occasion, Tiberius Claudius Verus, the giver of a beast hunt in honour of Nero, distributed 373 [somethings] (the word is undecipherable in the announcement of the beast hunt) in AD 62 (CIL 4.7989a; Tumolesi 1980: no.18). Usually, it is suggested that the 373 refers to sestertii, but could equally refer to food of some description. Hence, I would suggest that the image of the ‘bread seller’ is in fact a piece of high status art, rather than art of the ordinary Romans (Clarke 2003: 259-61; Fröhlich 1991: 236-41 cannot locate a precise iconographic parallel for the scene) that commemorates a giver of games distributing bread as part of the celebration. Is the man wearing a white toga? The location in the tablinum of an atrium house (7.3.30) points to a prestige event and the painting has been seen to be an image of an aedile distributing bread (Fiorelli 2001: 87-8). The absence of togas on the part of the recipients might give substance to Juvenal’s (Sat.3, 171-2) claim that outside the capital the toga was seldom worn.

These innovations should not distract us from the consistency of toga-clad emperor’s benevolence to togaclad citizens. The toga denotes the citizens’ right to benefit from the emperor’s benevolence, marking out the urban plebs as having a very unique relationship with the emperor (Woolf 1990 for discussion). The toga denotes difference between the emperor as giver, the citizens as recipients from the servant or administrator who actually distributed the coins. The role of the toga as a signifier of status is not the same here as in the alimenta imagery, where the emperor in his toga was found to be replacing the image of a deity. The togate figures of the early congiaria of Nero remain consistently posed as emperor and people joined by the benevolence of the former. What changes in the second century is the replacement of the emperor by a personification of Liberalitas distributing the cash to toga clad citizens. Case study III: from Adlocutio to Exercitus The development of scenes of an emperor speaking to his soldiers provides a means to access the meaning of dress in this context. Soldiers are signified as a quite different group from citizens by their military equipment and dress, whilst the emperor could be represented either as an imperator in military dress or a toga. Both can be found in sculpture (Kleiner 1992: 63-4 on Augustus in toga and the Prima Porta statue; Rose 1997 for examples of JulioClaudian dynastic statuary utilising toga and military dress), but in the images placed on coins by the mint we find an expression of differentiation via dress. The format would seem to have its origins in Caligula’s reign with a toga-clad emperor speaking to a group of soldiers with the legend Adlocut Coh (BMC Caligula 33; Brilliant 1963: 67; Hannestad 1986: 98). Usually interpreted as the emperor speaking to a cohort of praetorians rather than any other soldiers, the image breaks new ground and represents for the first time the emperor standing and speaking to a group of his subjects. The image is repeated under Nero (BMC Nero 122, 124, 125, 126, 304; Hannestad 1986: 115), again the emperor is clad in a toga. The interaction of a civilian princeps with the military is expressed through dress. Unlike the emperor on Trajan’s Column, the emperor is demilitarised and represented as a civilian via his toga just as the triumviral general Q. Cornuficius chose his image for his coins as a togate augur, rather than as a cuirassed imperator. Interestingly, the Flavians issued a posthumous coin featuring Galba in military dress with a bare head and

To return to the imagery on the coinage, there is consistency in terms of dress on the coins, but we should note some innovations following from the murder of Domitian. Liberalitas appears in the background with an account board on the coin of Nerva (BMC Nerva 87, 97; Brilliant 1963: 106). This image of Liberalitas takes on greater prominence in the coinage of Trajan to celebrate the second and third Congiaria (BMC Trajan 767, 768, 769). The coin type appears in almost identical form but with Liberalitas holding an abacus or account board. This can be seen to not just represent the generosity of the emperor but also the control of expenditure. Liberalitas is also depicted with the legend SPQR Optimo Principi under Trajan holding a cornucopia, which was often seen to be inverted from which coins flowed (BMC Trajan 174). This image is later represented under Hadrian with the legend Liberalitas (BMC Hadrian 523, 556-60, 1041, 1432; see also Fig. 7.3). The interrelationship of this coin with that of the earlier Congiarium and the Liberalitas types is

76

Ray Laurence: Investigating the Emperor’s Toga

no helmet in the same format as the coin of Nero with a simple legend: Adlocutio (BMC Galba 249; Brilliant 1963: 87) allowing for a reference to the conflict of AD 69 and also to Galba’s wearing armour or at least a linen cuirass within the city of Rome (Suet.Galba 19). The next issue in which we find Adlocutio is that of Nerva (BMC 85; Brilliant 1963: 105-6, now in the Louvre) here the emperor returns to a toga, a format also found in the Cancelleria Reliefs – although reworked from an original with Domitian as the emperor (Varner 2001: 13). The legend creates an association between the act of Adlocutio, the senate and the figure of the emperor as Augustus. The scene and legend creates an interaction between the military and the emperor to legitimate the tripartite creation of the new emperor, after the murder of Domitian, by Senate, soldiers and the people. The emperor can be seen to represent not just the commander issuing orders or parleying with his troops, but he is also the representative of a civilian world of toga-clad citizens and interacting on their behalf with the soldiers. The choice of dress maintains the emperor as a toga-clad citizen, rather than as a soldier whose power is derived from the weapons of his army. Thus, dress maintains the fiction of the emperor as civilian or as a citizen and distant from military power, particularly in Italy.

figure as female. The two children she has with her or is creating hold a bunch of grapes and corn ears – the symbols of the transformation of the province from barbarism to agriculture and production. By the following reign the representation of the province of Dacia relies solely on the figure itself seated on its rock (BMC Hadrian 1735-46) and the Antonine issues tend to feature a Dacia standing with a military standard and the rock is no longer depicted (BMC Ant. Pius 1187, 1188). An important parallel might be the representation of Judaea on coins minted under Hadrian. The type features a togate Hadrian sacrificing over an altar, facing Judaea holding a patera. In front of the altar are two toga-clad children (BMC Hadrian 1757). This coin is usually interpreted as a commemoration of the founding of the colony at Jerusalem that would in turn produce togaclad worshippers of the Roman gods in due course. Case study IV: The children of emperors The representation of the biological children of emperors comes to the fore in AD 69 with the coin of Vitellius representing his son and daughter and that of Vespasian showing his two sons (BMC Vitellius 12, 27; Vespasian 45, 46, 338, 430, 443, 455). These coins make it clear that the conflict of AD 69 had become a dynastic battle. Vespasian’s sons, who were older than Vitellius’ children, appear in togas. Like the emperor (Corbeill 2004: 14166), the children of emperors had become spectacles in their own right to be viewed and gazed upon. The Flavian imagery has its origins in the coinage of Claudius that represents the appearance of Nero in AD 51 in an adult toga, probably only aged twelve (since his birthday was on the 15 December) and the senate marked out the future in their desire that Nero would be consul before he was twenty. We know that he gained the title Princeps Iuventutis, along with proconsular imperium outside Rome, and can locate him in the coinage as the primary figure in a congiarium in the forum (see Spinola 1990 on congiaria; Brilliant 1963: 77). These instances we find were occasions for the assertion of the adultness of Nero and, as we learn from texts, just as at the Troy Games in the Circus, he rode as a magistrate holding imperium, in complete contrast to his step-brother wearing a toga praetexta and thus seen by all as a child (Tac. Ann.12.41). Dress marked out Nero as the key attraction for the Roman gaze at this event. In addition, he drilled the praetorians with shield in hand (Suet. Nero 7), further asserting his adulthood. When the civil wars of AD 69 were over, Titus and Domitian could appear on the coinage as Principes Iuventutis in both togate and the military dress of the Troy games (BMC Vespasian 392, 393, 394, 395). These would appear to copy civil war coins minted in Illyricum (BMC Vespasian 426, 426). The conception of these two young men as Principes Iuventutes owes its precedent to the Claudius-Nero relationship of AD 51, in which power was conveyed via dress in the toga of a magistrate and in the military dress of the Troy Games. The legend Princeps Iuventutis also appears on coins minted in the reign of Vespasian with the image of Titus or Domitian on the obverse, with a variety of goddesses on the reverse:

The distinction of dress between military emperor and togate citizens was subject to some alteration in the 2nd century through an innovation under Hadrian. This created an expectation that the emperor would appear talking to soldiers in military dress under the legend Exercitus and the relevant name of the army, for example Exercitus Syriacus (BMC Hadrian 1689, 1690). These coin types should be seen alongside the Adventus coinage that is connected to the journeys of Hadrian to those provinces that did not contain armies (e.g. BMC Hadrian 1670 or 1671). Here, a togaclad princeps faces a personification of the province clad in civilian rather than military dress. The original meaning of adlocutio depended on the differentiation by dress of a civilian princeps, now we have a new meaning that emphasises the emperor as soldier rather than togate civilian – when speaking to soldiers, but can return to the civilian realm when speaking to civilians. This pushes us back to the imagery of Galba produced in AD 69 and the context of civil war. The emperor by the 2nd century is the general and no longer appears in the role of a civilian princeps when speaking to soldiers. Christopher Howego (2005: 10-11) argues that the format of dress, military/togate, was the means to represent a fundamental geographical division of the Roman provinces into military/non-military and reflects the dress code of governors of imperial and senatorial provinces (Dio 53.13). Within the imagery of anthropomorphic representations of the provinces, it needs to be highlighted that two images of newly conquered regions, produced in the 2nd century, included the figures of children. Trajan’s conquest of Dacia was marked by an issue of coinage under the legend Dacia Augusta Provincia. Dacia appears in military dress seated on a rock, rather than a chair to symbolise the barbarian origins of the province (BMC Trajan 960, 990). Coin cataloguers see the

77

Dress and Identity

Conclusions: understanding the Emperor’s toga

Spes, Vesta, Salus etcetera (BMC 154-156, 260-69, 346, 481, 686, 699, 713, 724*). This theme continues once Titus is emperor with coins featuring Domitian on the obverse (BMC Titus 83-91). Even though Titus and Domitian were holders of office as consul or praetor, their role as Princeps Iuventutis continued to be paraded. These coins represent the power of these imperial adult children of the time, but also express a hope for the future. In contrast, on Vespasian’s coinage we see the old man hunched in his toga with his son Titus upright in the togate position of the orator holding a globe over a rudder, with the legend Providentia Augusta.

There are certain key observations to be made from the images of emperors on coins. The first is that, unlike in sculpture, the emperor is not represented as a god but his mother, child or wife may appear with the attributes of a female deity. The second is that in this medium, the emperor and his children are always dressed. When he does appear, the preferred mode of dress is the toga. The emperor is quite simply the toga clad Roman, who ruled the world (Verg.Aen.1.282). If that world was being torn apart by civil war, his dress may change into that of the general guaranteeing the world order and the annona. The representation of his subjects on coins tended to be as toga-clad beneficiaries, perhaps differentiated by age, or as soldiers. What is innovative, in terms of representation, is that on the emperor’s coinage we find not only allegorical images but also the incorporation of dressed Romans into the repertoire of allegorical images with words to explain the visual representation.

The lack of direct male heirs to Domitian, Nerva, Trajan and Hadrian led to a renewal of creativity in the representation of what we might term substitute children on the coinage. Many of these images are derived from the reign of Nero or from the civil wars of AD 69. Dress is instrumental to their creation as children of Romans, rather than miniaturised slave attendants. These were to become the images of ‘good government’, but actually owe their inception to the ‘bad government’ of Nero or Domitian and the chaos of civil war. This new set of images developed within the repertoire of coin types during the mid to late 1st century and into the early 2nd century AD. These revolved around the representation of the emperor interacting with his subjects. Rather than seeing the arch of Trajan at Beneventum as expressing a new form of representation, we should see this as the fully developed form of attempts to define allegories of good government taking precedents from the coinage of earlier emperors (D’Ambra 1998: 85 on arch and good government). Particularly notable is the development of a language of dress to represent soldiers and citizens – male, female, adult, child – within the emperor’s gaze. Alongside the visual representation, there is a written explanation of the scene that is represented through brief statements such as ‘Annona Aug’ or ‘Alim Ital’. These images of dressed leaders are often shown alongside children of the state rather than of themselves. It is the emperors without legitimate children or heirs who utilised the iconography of a generous Pater Patriae or Optimus Princeps caring for his clothed children. Unlike statuary, in which we find emperors in a state of dress and undress, coins have an internal language of representation that utilises dress strictly to define the emperor’s civil and military role. He either appears in a toga or equipped with armour and weaponry, but without a helmet. It is male gods who appear in a state of undress, for example Jupiter carrying thunderbolts crowning a toga clad emperor and legend Conservatori Patris Patriae (BMC Trajan 493, 494, 495, 496). Just as the emperor is defined by dress, so are those who are seen to interact with him – the toga-clad citizens or soldiers with weapons and standards. Here, we see the representation of the exceptional Emperor interacting with his ordinary subjects. The effect is to create a composition that allows for the individuals represented to become examples of the beneficiaries of the emperor’s generosity or management of the state.

In terms of dress history the coinage from the reigns of Caligula and Nero shows representations of the emperor in a toga interacting with other toga-clad individuals. These innovative coin issues were to be copied or adapted under later emperors. This evidence demonstrates the importance of dress in the representation of the deeds of Roman emperors and its greater ideological value under the empire compared to the toga’s invisibility on the coinage of the Republic. The emperor was to be thought of in his toga in these contexts, as were members of his family and future successors. Yet they were also commanders of the army, successful in war, and it seems surprising that the gloria associated with the military and represented by military equipment and dress is all but absent. Only under Hadrian do we see a consistently militarised emperor and even then only in the context of the representation of the army of a particular province. Once the context of an image shifted to that of a province with no army, we see Hadrian switch back to his toga. In this, the mints in Rome and in the provinces extended the traditional meaning of the toga-clad person as the power of Rome to the person of the princeps. This extension of meaning was accompanied with a far greater number of togate figures on the coinage under the empire from the reign of Nero onwards than under the Republic or under Augustus. It seems ironic that Nero, as a ‘bad’ emperor, was criticised for his inappropriate dress (Plin. N.H.8.196), while his coinage reveals him as the toga-clad Roman fulfilling many of the roles of a later Optimus Princeps – Trajan. This is not to suggest that either image of Nero is more correct, but instead to flag up the importance of the language of dress for the emperors and their subjects. The focus on the emperor as ruler, as much in the biographies of Suetonius as on the coins, causes a widening in the number of signifiers of power. Whereas the toga might have been seen to represent some form of national dress (Stone 1994), under the empire it takes on a new importance as a symbol of the emperor’s power as well as Rome’s domination of the world. Alongside this, we

78

Ray Laurence: Investigating the Emperor’s Toga

Acknowledgements

might point to a change in fashion, not only were the people of Rome the gens togata from Augustus onwards (Verg. Aen.1.282; Zanker 1988: 162-5), they were also dressed in garments of a better quality from this time onwards made from a smoother type of cloth (Plin.N.H.8.195; compare Varro Modius 8; Ovid Ars.Am.3.445; Diodorus Frag. 37.3.4). Following the battle of Actium (31 BC) we see a re-working of the toga and a re-invention of a tradition in which the toga was understood as a form of traditional dress dating back to the Etruscan kings of Rome. Interestingly, the Roman toga could be sat alongside the highly Hellenised repertoire of images that represented imperial rule in sculpture. In these cases, dress in a toga could be utilised to re-establish Roman identity for individuals represented as half-dressed deities (discussed by Hales 2005). What these images of toga clad emperors and subjects indicate is that the toga is now of greater symbolic relevance than before: it is the dress for representing the people and their emperor. Moreover, it was a garment thought to have been utilised to denote power by the kings of Rome and was understood in these terms during the 1st century AD (Pliny N.H.8.197, 9.136; Dionysus of Halicarnassus 3.61; Livy 1.8.3). We lack is any information on the weave, colour or type of cloth worn by the emperor and his subjects at a congiarium; instead the emperor appears to be dressed in the same way as his subjects and only differentiated by his position on a raised platform. Perhaps a Roman viewer would have imagined things differently, with his emperor wearing a magnificent toga that delineated further his superiority over his grateful toga clad subjects receiving their gift from him (Festus 237L; Dio 49.16, 58.11-12). The Romans’ knowledge of the toga as a garment worn by the kings of Rome may have shifted the meaning of these scenes on the coins from images of equality to the embodiment of kingship and monarchy. The toga was the emperor’s garment, but it was also a Roman garment (Harlow 2005 for later sources and tradition of toga wearing emperors) that was monarchical. Wearing a toga confirmed figure as Roman emperor, what differentiated the emperor’s toga from that of a citizen’s was its quality. Hence a form of dress, the toga, could have rather more meanings than first meets the eye and we might even suggest that the dress of the emperor re-invented the tradition dress of the kings of Rome. Augustus’ attempt to revive toga wearing at Rome (opening quotation) should perhaps be viewed as creating a new dress code for the principate, in which the emperor could express ideas not just of his equality with his people as a citizen of Rome, but also his separation from them as their monarch (Hales 2005: 133-4 offers a senatorial perspective on the dress code). The toga is not the same garment under the empire as under the republic, it has a new set of meanings and connotations and had been re-invented for a new purpose: to delineate the power of Rome’s citizens and its emperor (Suetonius Aug.60) and its first emperor, Augustus, was remembered as a master of this new dress code (Suetonius Aug.73).

I would like to thank Mary Harlow for first running the seminar series in which this paper first was articulated. In its delivery to this final form, I have discovered much about Dress History particularly from my colleagues in the Institute of Archaeology and Antiquity. I have to also thank the editors for patience and good humour in bringing the piece in line with IAA Series formatting. The images of coins that accompany this paper are from the Barber Institute (University of Birmingham). Bibliography Bartsch, S. (1994): Actors in the Audience: Theatricality and Doublespeak from Nero to Hadrian. Cambridge Mass. Bianchi Bandinelli, R. (1970): Rome: The Centre of Power. London. Black, J.A., Garland, M. & Kennett, F. (1980): A History of Fashion. London. Breckenbridge, J.D. (1981): ‘Roman Imperial Portraiture from Augustus to Gallienus’, ANRW II.12.2: 479-512. Breward, C. (1998): ‘Cultures, Identities, Histories: Fashioning a Cultural Approach to Dress’, Fashion Theory 2: 201-14. Brilliant, R. (1963): Gesture and Rank in Roman Art. The Use of Gesture to Denote Status in Roman Sculpture and Coinage, (Memoirs of the Connecticut Academy of Arts and Sciences 14). New Haven. Buckland, W.W. (1963): A Textbook of Roman Law from Augustus to Justinian. Cambridge. Buttrey, T.V. (1956): ‘The Triumviral Portrait Gold of the Quattuorviri Monetales of 42 B.C. (American Numismatic Society: Numisamtic Notes and Monographs 137). Callu, J.-P. (2004): ‘L’habit et l’ordre social: Le témoignage de l’Histoire Auguste’, Antiquité Tardive 12: 187-94. Carlson, C.W. (1975) ‘Cogiaria and Liberalitas’, Journal of the Society for Ancient Numismatics 6: 59-63. Candilio, D. (2004): Moda costume e bellezza nella Roma antica. Rome. Chausson, F. and Inglebert, H. (eds) (2003): Costume et societé dans l’Antiquité et le haut Moyen Age. Paris. Clarke, J. (2003): Art in the Lives of Ordinary Romans: Visual representation and non-elite viewers in Italy 100 B.C. – A.D.315. Berkeley. Cleland, L., Harlow, M. and Llewellyn-Jones, L. (eds) (2005): The Clothed Body in the Ancient World. Oxford. Corbeill, A. (2004): Nature Embodied: Gesture in Ancient Rome. Princeton N.J. Cumming, V. (1998): The Visual History of Costume Accessories. London. Cumming, V. (2004): Understanding Fashion History. London. Croom, A.T. (2000): Roman Clothing and Fashion. Stroud. D’Ambra, E. (1998): Art and Identity in the Roman World. London.

79

Dress and Identity

Dandre Bardon, M.F. (1784-6): Costume des anciens peoples à l’usage des artistes. Paris. Delmaire, R. (2004): ‘Le vêtement dans les sources juridiques du bas-empire’ Antiquité Tardive 12: 195-202. Dyck, A.R. (2001): ‘Dressing to kill: Attire as a proof and means of characterization in Cicero’s speeches’, Arethusa 34: 119-30. Fiorelli, G. (1875): La descrizione di Pompeii. Naples. Fröhlich, T. (1991): Lararien – und Fassadenbilder in den Vesuvstädten: Untersuchungen zur ‘volksümlichen’ pompejanischen Malerei. Mainz. Gabelmann, H. (1985): ‘Römische Kinder in Toga Praetexta’, Jahrbuch des Deutschen Archäologischen Instituts 100: 497-41. Goethe, J.W. von (1962): Italian Journey, 1786-1788. Trans. W.H. Auden & E. Mayer. London. Goldman, N. (1994): ‘Reconstructing Roman Clothing’ in Sebasta and Bonfante (eds): 213-37. Graf, F. (1991): ‘Gestures and conventions: the gestures of Roman actors and orators’, in J. Bremmer and H. Roodenburg (eds) A History of Gesture. From Antiquity to the Present Day. Oxford. Hannestad, N. (1986): Roman Art and Imperial Policy. Aarhus. Harlow, M. (2004): ‘Female dress, third – sixth century: The message of the media?’ Antiquité Tardive 12: 203215. Harlow, M. (2005): ‘Dress in the Historia Augusta: The role of dress in historical narrative’, in Cleland, Harlow, and Llewellyn-Jones 2005: 143-53. Heskell, J. (1994): ‘Cicero as evidence for attitudes to dress in the Late Republic’ in Sebasta and Bonfante 133-45. Hope, T. (1812): Costumes of the Greeks and Romans. New York (re-issued 1962). Houston, M.G. (1931): Ancient Greek, Roman and Byzantine Costume and Decoration. London. Howego, C. (2005): ‘Coinage and Identity in the Roman Provinces’, in C. Howego, V. Heuchert and A. Burnett eds) Coinage and Identity in the Roman Provinces: 1-17. Oxford. Jenkins, I. & Sloan, K. (1996): Vases and Volcanoes: Sir William Hamilton and His Collection. London. Kleiner, D.E.E. (2001): ‘Now you see them, now you don’t: the presence and absence of women in Roman art’, in E. Varner (ed.) From Caligula to Constantine: Tyranny and Transformation in Roman Portraiture: 45-57. New York. Krupp, E. & Krupp, F. (1961): ‘The Tutela Type of Vitellius’, Numistatic Chronicle Ser. 7. Vol. 1: 129-30. Laver, J. (1964): Costume in Antiquity. London. Levick, B. (1999): ‘Messages on the Roman Coinage: Types and Inscriptions’, in G.M. Paul and M.I. Ierardi (eds) Roman Coins and Public Life under the Empire: 41-60. Ann Arbor. Linderski, J. (2002): ‘The pontiff and the tribune: The death of Tiberius Gracchus’, Athenaeum 90: 339-66. Llewellyn-Jones, L. (ed.) (2002): Women’s Dress in the Greek and Roman World. London. Lo Cascio, E. (1998): ‘Alimenta Italiae’ in J. González (ed.) Trajano: Emperador de Roma: 287-312. Rome.

Mattingly, H. (1967): Roman Coins from Earliest Times to the Fall of the Western Empire. London. McNeil, P. & Riello, G. (2005): ‘The Art and Science of Walking: Gender, space and the fashionable body in the long eighteenth century’, Fashion Theory 9: 175-204. Millar, F. (1977): The Emperor in the Roman World. London. Pani, M. (1991): ‘Il senso del nuovo fra costume e politica’ in M. Pani (ed.) Continuitä e trasformazioni fra repubblica e principato: Instituzioni, politica, societä: 109-32. Bari. Pisani Sartorio, G. and Liberati Silverio, A. 2000 L’abbigliamento (Museo della Civilità Romana 22). Rome. Pollini, J. (1987): The Portraiture of Gaius and Lucius Caesar. New York. Ribeiro, A. (1998): ‘Refashioning Art: Some Visual Approaches to the Study of Dress History’, Fashion Theory 2: 315-26. Richardson, L. and Richardson, E.H. (1964): ‘Ad cohibendum brachium toga’, American Journal of Archaeology 68: 199-200. Richardson, E.H. and Richardson, L. (1966): ‘Ad cohibendum brachium toga: An archaeological examination of Cicero, Pro Caelio 5.11’, Yale Classical Studies 19: 251-68. Roche-Bernard, G. and Ferdiere, A. (1993): Costumes et Textiles en Gaule Romaine. Paris. Rogers, P.W., Jørgenson, L.B. and Rast-Eicher, A. (eds) (2001): The Roman Textile Industry and Its Influence. Oxford. Rose, C.B. (1997): Dynastic Commemoration and Imperial Portraiture in the Julio-Claudian Period. Cambridge. Sebasta, J.L. (1994): ‘Symbolism in the Costume of the Roman Woman’, in Sebasta and Bonfante 1994: 46-53. Sebasta, J.L. and Bonfante, L. (eds) (1994): The World of Roman Costume. Madison. Sichell, M. (1983): History of Children’s Costume. London. Spinola, G. (1990): Il congiarium in età imperiale: Aspetti iconografici e topografici. Rome. Stewart, P.C.N. (2003): Statues in Roman Society. Oxford. Stone, S. (1994): ‘The Toga: From National Dress to Ceremonial Costume’ in Sebasta and Bonfante (eds) 13-45 Styles, J. (1998): ‘Dress in History: reflection on a contested terrain’, Fashion Theory 2: 383-90. Swift, E. (2000): Regionality in Dress Accessories in the late Roman West. Montagnac. Swift, E. (2003): Roman Dress Accessories. Princes Risborough. Swift, E. (2004): ‘Dress Accessories, Culture and Identity in the late Roman period’, Antiquité Tardive 12: 217-22. Symons, D.J. (1987): Costume of Ancient Rome. London Taylor, L. (1998): ‘Doing the Laundry? A reassessment of object-based dress history’, Fashion Theory 2: 337-58. Taylor, L. (2002): The Study of Dress History. Manchester. Taylor, L. (2004): Establishing Dress History. Manchester. Toynbee, J.M.C. (1956): ‘Picture-Language in Roman Art and Coinage’, in R.A.G. Carson and C.H.V. Sutherland

80

Ray Laurence: Investigating the Emperor’s Toga

(eds) Essays in Roman Coinage Presented to Harold Mattingly: 205-26. Oxford. Uzzi, J.D. (2005): Children in the Visual Arts in Imperial Rome. Cambridge. Varner, E. (2001): ‘Tyranny and the Transformation of the Roman Visual Landscape’ From Caligula to Constantine: Tyranny and Transformation in Roman Portraiture: 9-26. New York. Varner, E. (2004): Mutilation and Transformation: Damnatio Memoriae and Roman Imperial Portraiture. Leiden. Vermeule, C. (1987): The Cult Images of Imperial Rome. Rome. Vout, C. (1996): ‘The myth of the toga: understanding the history of Roman dress’, Greece and Rome 43: 204-20. Wallace-Hadrill, A. (1986): ‘Image and Authority in the Coinage of Augustus’, Journal of Roman Studies 76: 66-87.

Wallace-Hadrill, A. (1998): ‘To be Roman, go Greek: Thoughts on Hellenization at Rome’, Bulletin of the Institute of Classical Studies Supplement 71: 79-92. Wallace-Hadrill, A. (2008): Rome’s Cultural Revolution. Cambridge. Wild, J.P. (1968): ‘Clothing in the North-West Provinces of the Roman Empire’, Bonner Jahrbucher 168: 166-240. Wilson, L.M. (1924): The Roman Toga. Baltimore. Wilson, L.M. (1938): The Clothing of the Ancient Romans. Baltimore. Woolf, G. (1990): ‘Food, Poverty, and Patronage. The Significance of the Epigraphy of the Roman Alimentary Schemes in Early Imperial Italy’, Papers of the British School at Rome, 58: 197-228. Wyke, M. (1996): Projecting the Past. London. Zanker, P. (1988): The Power of Images in the Age of Augustus. Ann Arbor.

81

Chapter 8

Anglo-Saxon Woman: Fame, Anonymity, Identity and Clothing Gale R. Owen-Crocker Famous names

11th-century queen consorts Emma3 and Edith.4 Our sources of evidence are documentary, the work of male writers, even when the women personally commissioned the texts as Emma and Edith did. These were women who grasped their opportunities, in some cases having rejected or survived their spouses. Yet in each of these cases the woman was initially given the potential for leadership because of the rank of her husband, father or other male relative; and her biography, such as it is, is presented to us through the filter of a masculine, monastic author.

Feminist scholars sometimes complain that women’s history consists largely of a few ‘famous names’: Boadicea; Cleopatra; Mary, Queen of Scots; Catherine the Great; Florence Nightingale; Marie Curie and so on. There are far fewer famous women than famous men and most creative acts of history (inventions and art) are attributed to men, as well of course as the destructive acts of war which have made men into heroes. Many of the women celebrated in history books were queens, thrust into power by the absence or inadequacy of male heirs. The non-royal minority of famous historical women carved out pioneering paths in a male world and their gender is remarkable in contexts where women are rarely mentioned.

Named women are not evenly spread, in chronological terms, since they cluster in the period recorded in Bede’s Ecclesiastical History, which was completed in 731, and the years of the continental mission of Boniface (martyred in 755), whose correspondence with both men and women survives;5 and in the 10th and 11th centuries from when most diplomas survive, whether originals or copies (or forgeries). Although these women were important, in that they were achievers who influenced the lives of others, numerically they are few, both in relation to the number of named Anglo-Saxon men in documentary sources, and in relation to the female population of England.

The accusation that the story of women is confined to ‘famous names’ is undoubtedly true of the Christian AngloSaxon period, that is from the 7th century, when Christian missions introduced literacy and humanistic art to England, up to the Norman Conquest of 1066. The best-known women of Christian Anglo-Saxon England are Bertha, the Frankish Christian consort to King Ethelbert of Kent, the monarch who received the mission of Augustine in 597, opening up Anglo-Saxon England to Christianity (Bede Historia Ecclesiastica i.25, ii.5; Colgrave and Mynors 1969: 72-5; 150-1); St Hilda, a member of the Northumbrian royal family and an influential foundress-abbess (Bede HE iii.24, iv.23; Colgrave and Mynors 1969: 292-3; 40415); St Æthelthryth, daughter of an East Anglian king who preserved her virginity through two royal marriages, eventually becoming a nun, foundress-abbess, famed ascetic and virgin saint;1 Æthelflæd, ‘Lady of the Mercians’, who ruled Mercia after the death of her husband, its ealdorman, and who co-operated with her brother King Edward ‘the Elder’ of Wessex in defence against the Vikings;2 and the

Named but intangible women The legitimisation of academic interest in the history of women and the establishment of feminist research projects have resulted in the mining of once-obscure sources in attempts to uncover more material about female AngloSaxons. Through study of non-literary texts, especially charters and letters, women have emerged as important and respected contributors to the early development of the Emma, of the Norman ducal family, was married first to King Æthelræd ‘the Unready’ of England, when she was given the English name Ælfgifu; and secondly to King Cnut, of the usurping Danish dynasty. She was the patron of a text about Cnut’s family known as the Encomium Emmae Reginae (Campbell 1979). For an assessment of her career and power see Stafford (1997, passim). 4 Eldest daughter of the powerful Earl Godwin and his royally-connected wife Gytha (she was sister-in-law to King Cnut), Edith was married to King Edward ‘the Confessor’. His successor King Harold II, who perished at the Battle of Hastings, was her brother. She commissioned the Vita Edwardi (Barlow 1992); see Stafford (1997, passim). 5 Boniface’s female correspondents were Bugga and her mother Eangyth, Eadburga, Egburga and Lioba; and he mentions communicating with Wiethburga who had retired to Rome (Emerton 1940: V, XIX). The Boniface correspondence is published in Dümmler (1892) and Tangl (1916) and translated in Emerton (1940); see also Fell (1990). 3

Bede HE iv.19-20; Colgrave and Mynors (1969: 390-401). Bede’s Latin version of her name is Etheldreda. The origin and development of the saint’s cult is discussed in Blanton (2007). 2 Recorded in the C (Abingdon) version of the Anglo-Saxon Chronicle under the years 912- 918, which incorporate information from a supposed ‘Mercian Register’; Swanton (1996: 96-105). Æthelflæd was, as Pauline Stafford (1997: ix) rightly notes, the only regnant queen of the late AngloSaxon period. See further Wainwright (1959: 53-69), reprinted in Damico and Olsen (1990: 44-55). Æthelflæd’s daughter Ælfwynn, who was chosen as her successor by certain Mercians, might have had even greater claim to the title queen regnant, but she was ‘deprived of all control in Mercia, and was led into Wessex’ according to the Anglo-Saxon Chronicle C version for 919; Swanton (1996: 105). 1

83

Dress and Identity

Christian Church in England, where influential abbesses, often royal widows or divorcees, ruled over ‘double monasteries’, communities that included monks and nuns.6 These early religious establishments were characterised by matrilineal succession, in which founder abbesses were succeeded by blood relatives, sisters, daughters and nieces: Æbbe and her daughter Mildrith at Minster-in-Thanet; Æthelthryth and her sister Seaxburg, and Seaxburg’s daughter Eormenhild at Ely; Hilda, succeeded by Queen Eanflæd and her daughter Ælfflæd at Whitby. This female, monastic rule has been seen as an ecclesiastical parallel for lay kingship. Close associations could be maintained between members of the royal family in secular and ecclesiastical life and the ownership of property, especially land, controlled (see further Dockray-Miller 2000; Yorke, 2002).

book requested by a missionary friend9 and even to make a copy of a book in gold letters;10 or that a mother might leave her son a gold cup with instructions that he was to use the metal to enlarge his ring.11 All these facts are ties to yet more (unpronounceable, Anglo-Saxon) names, abbess or testatrix-and-beneficiary, and although the woman is effectively speaking her wishes and thoughts, there is very little personality detectable. Wills, by their very nature are formal affairs, and letters were used as rhetorical exercises: amongst metaphors of storms at sea, the Boniface correspondence occasionally reveals the human misery of bereavement and isolation but few other facts of day-to-day life. Any attempt to envisage these famous women of the Christian Anglo-Saxon period as physical entities, by clothing them in garments and accessories, is frustrated by the small amount of evidence. Only two women, Wynflæd and Æthelgifu, neither of whom is definitely identified from any other context, have left wills which individualise garments (Whitelock 1930: III; Whitelock et al. 1968). Although both testaments imply ownership of a wardrobe of gowns, cloaks and headdresses for both secular and ecclesiastical occasions, and Æthelgifu’s will specifies garments of different colours,12 neither includes information about cut or shape, and practically nothing can be learned from them about dress accessories such as jewellery, belts or bags.13 Other wills simply fail to see the need to describe garments: Wulfgyth leaves ane wellene kertel (Whitelock 1930: XXXII, 86, l. 2), while Wulfwaru does not even specify individual garments by name, leaving anes wifscrudes ‘one woman’s outfit’ to her elder daughter and ealles þæs wifscrudes þe þær to lafe bið ‘all the women’s clothing that remains’ to her other daughter (Whitelock 1930: XXI, 64, ll. 11-13).

From the late Anglo-Saxon period, occasional wills of wealthy women demonstrate female ownership of a vast range of property which was evidently their own to dispose of. A woman named Æthelflæd bequeathed gold rings, cups, bowls and horses, as well as lands, the enumeration of which occupy most of her will (Whitelock 1930: XIV, 34-5). Wulfwaru bequeathed crucifixes, armlets and bedclothes (Whitelock 1930: XXI, 62-3). Husband and wife Brihtric and Ælfswith left an armlet, a head band (in two separate halves), cups, swords, horses and tack (Whitelock 1930: XI, 26-9). Many women in the late Anglo-Saxon period, particularly, but not exclusively, widows, chose to live a chaste life closely associated with a religious establishment but without formally taking the veil,7 and their bequests, such as those of Æthelgifu (Whitelock, Ker and Rennell:1968) and Wynflæd (Whitelock 1930: III), reflect a firm hold on worldly property alongside piety and generosity to favourite religious foundations. Above all, the wills mention land, disposing of estates, usually multiple estates, which are variously bequeathed to the king, the Church, and (especially) to relatives; as well as individual hides of land. They specify woodland, meadow and open land, indicating that it was normal for large areas of the English landscape to be under the control of female owners, albeit under the management of male reeves and stewards.

All the women who ruled monasteries or left wills were well connected, usually of high rank. They were rich, both in terms of material possessions, and in terms of the patronage and influences they could command through their family connections. They are therefore sociologically untypical. For every named abbess or nun, we can assume a supporting cast of anonymous sisters;14 and for every wealthy secular there must have been numerous household and estate workers, largely unnamed unless they were fortunate enough to receive a bequest, or their freedom,

These researches open slits through which we can peep at the Anglo-Saxon age, revealing surprising details like the fact that a convent might trade and own a ship;8 that a nun might be expected to acquire and send a copy of a

Boniface had requested Bugga for a book but she is unable to supply it and sends money and an altar cloth in the meantime (Emerton 1940: VII, 60-1). 10 Boniface asks Eadburga to make him a copy of the Epistles of St Peter in gold letters and sends her the materials. Possibly he expects her to commission the writing, but he may be asking her to carry out the task personally (Emerton 1940: XXVI, 64-5). 11 Wynflæd’s will: Whitelock (1930: III,12-13). 12 hire rotostan cyrtel (‘her brightest cyrtle’), mine blæwenan cyrtel (‘my blue cyrtle’) and oðera hire dunnan cyrtlas (‘others of her brownish cyrtles’);Whitelock et al. (1968: 13, ll. 37-9). 13 Wynflæd bequeaths a mentelpreon (a ‘mantle-pin’ or ‘cloak-brooch’) and hyre ealdan gewiredan preon (‘her old, wire brooch’, perhaps meaning her ‘antique, filigree brooch’); Whitelock (1930: III, 10, ll.7-8; 11-12). These bequests are discussed in Owen (1979). 14 The PASE database has 99 entries under occupation ‘nun’, of which 41 are anonymous references to one or more nuns. 9

The most famous being St Hilda, whose monastery at Whitby educated five bishops and nurtured the talent of the poet Cædmon; Bede HE iv.23, iv, 24; Colgrave and Mynors (1969: 408-9; 414-21). 7 The double monasteries ruled by powerful women, like the royal families which had sponsored them, did not survive the Viking Age. Only in Wessex did the royal family continue to be closely associated with convents, and formal monasticism for women thrive in and beyond the age of the Benedictine Reform, but with considerably restricted roles for religious women. See Foot (2003). 8 Mildrith, abbess of Minster-in-Thanet, is granted freedom from tolls in a series of documents, and a ship is specifically mentioned in Æthelbald 4, Sawyer 87 (The King’s College, London, Prosopography of Anglo-Saxon England (PASE) database: persons office; abbess; Mildrith). The PASE database may be found at: http://www.pase.ac.uk:8080/pase/index.html. 6

84

Gale R. Owen-Crocker: Anglo-Saxon Woman

in a will.15 There must have been very many who lived out their lives in family and village without their names being documented.

bereaved women in illustrations of biblical narrative.18 Two contemporary images of Queen Emma, are also, I suggest, very dependent on biblical iconography. The earlier one, in the Liber Vitae of the New Minster at Winchester (London, British Library, MS Stowe 944; Fig. 8.2) a manuscript dated by internal evidence to 1031 (Keynes 1996: 15, 38), depicts Emma and her second husband, King Cnut, presenting a cross to the New Minster. The placing of two figures beneath the cross, woman on the left of the picture, man on the right, is surely derived from the well-known image of the Virgin Mary and St John at the foot of the cross at the crucifixion.19 The other ‘portrait’, in a book commissioned by the queen in praise of Cnut and his family, known as the Encomium Emmae Reginae, is also quite closely dated since the text was written after Emma’s widowhood and during the brief reign of her son Harthacnut (1040-2). It depicts a seated queen approached by three male figures. Although one man represents the writer of the book which the queen had commissioned, and the others Emma’s sons Harthacnut and Edward (later King Edward ‘the Confessor’), who are depicted in the text as living in a harmonious triumvirate with their mother, the iconography of the picture is surely derived from the seated Virgin approached by the three Magi.

Modern books on Anglo-Saxon art rightly show some sumptuous manuscript illuminations and illustrated texts such as the Old English Hexateuch (London, BL MS Cotton Claudius B iv), which teems with illustrations, both coloured line drawings and painted images. Depictions of clothed human figures are common. However, most of these pictures are so stylised as to be of limited use to a costume historian. The fact that the majority of the female figures depicted in Anglo-Saxon manuscripts are biblical (above all, the Virgin Mary) need not in itself render this evidence irrelevant, since Anglo-Saxon artists make no attempt at historical authenticity; but neither should these be unthinkingly accepted as pictures of Anglo-Saxon people. Even when historical women are represented in art, the images should be understood as iconic rather then portraiture. The well-known miniature of St Æthelthryth in The Benedictional of St Ethelwold (London, British Library, MS Additional 49598, fol. 90v; Fig. 8.1) was painted in the 10th century, as a manifestation of the deliberate revival of the saint’s cult, about three hundred years after her lifetime. There is perhaps one concession to her legendary asceticism in the colouring of her garments in shades of brown, rather than the bright colours found elsewhere in the manuscript; but the lily she holds is purely symbolic of her virginity, not in itself associated with any event in her life. There is no representation of the posthumously healed wound on her neck, and certainly not of the red tumour which preceded it, which to the dying saint recalled the jewellery worn in the vanity of her youth, a seeming reference to the luxurious garnet bullae excavated in modern times from 7th-century contexts (Fig. 8.5).16 There is no attempt to historicise Æthelthryth’s appearance; her clothing is not dissimilar to that of other female figures in 10th- and 11th-century art, with the addition of a veil hanging from her shoulders at the back, which perhaps represents her consecration as a nun.

Nevertheless, there are some individual details in the representations of Queen Emma’s costumes in these two drawings. In the Liber Vitae she wears a decorated band round her forehead. Ribbons appear from under her hood at the back and are perhaps the ends of the fillet fastened round her head, or an independent narrow ornament like the ecclesiastical stole. She has slightly flared outer sleeves; but she has no other jewellery or dress accessories. In the Encomium she wears even wider sleeves with some stylised decoration that also appears down the front of her gown. On her head is a strange, winged crown. There is no way of knowing if the artists were representing what the 11thcentury queen genuinely wore – the decorated headband does seem to have been worn by married women and, especially when made in gold, was a treasured accessory mentioned in several wills,20 but the dangling ribbons are worn by an allegorical representation of the sin Pompa,21 and might be either symbolic of pomp or a manifestation of it.

Queen Edith, depicted at the foot of her dying husband’s bed in the Bayeux Tapestry,17 weeps into her garment in a gesture of grief made familiar by its repeated use in images of the Virgin Mary at the foot of the cross and

It would be reasonable to suppose that rich women of the late Anglo-Saxon period wore fine linen and silk decorated with the embroidery for which female hands were famous. The early 10th-century embroidered vestments associated

Wynflæd names the slaves she bequeaths to relatives and those she frees; but more wills leave them anonymous. Leofgyth’s will (Whitelock 1930: XXIX, 76-9) is typical: ‘… I desire that all my men shall be free, in the household and on the estate, for my sake and for those who begot me’ (Whitelock’s translation). 16 The neatest parallel to Æthelthryth’s memory of girlhood is the necklace from a woman’s grave at Desborough, Northamptonshire, dated to the second half of the 7th century: it is composed of variously-shaped bullae – gold and gold-framed garnets – gold wire beads and a central gold cross (Webster and Blackhouse 1991: 28-9, pl. 13). There is a similar necklace – without the central cross – from Cow Low, Derbyshire, and many other jewelled bullae, particularly from Kent. 17 Bayeux Tapestry Scene 27 (Wilson 1985: pl. 30). The figure is not identified in the text of the Tapestry, but is easily recognisable from the description of the deathbed in the Vita Edwardi, which mentions the queen sitting at the dying king’s feet: Vita Ædwardi, ii.11 (Barlow 1992: 122-3). 15

The weeping Virgin may be seen in London, British Library MS Harley 2904, fol. 3v; see Owen-Crocker (2004: 203, fig. 157). Old Testament women weep into their garments in illustrations marking the passing of the generations in the illustrated Hexateuch (Dodwell and Clemoes 1974: fols 10v, 11v and 12r) and in the Old English poetic codex, Oxford, Bodleian Library, MS Junius 11, p. 59 (Ohlgren 1992: 557). 19 See Owen-Crocker (2005b: 42). For a discussion of the role of Emma in this image see Karkov (2004: 123-33). For women’s dress in 10th- and 11th-century England in general see Owen-Crocker (2004: 202-31). 20 Women’s headbands are discussed in Owen-Crocker (2004: 224-5). 21 Prudentius Psychomachia, London, British Library MS Additional 24199, fol. 21v. Illustrated in Owen-Crocker (2004: 215, fig. 174). 18

85

Dress and Identity

Fig. 8.1: St Æthelthryth in London, British Library MS Additional 49598, fol. 90v.

with the relics of St Cuthbert, presumed by their inscription to have been the product of a royal workshop in Winchester, exquisitely worked in embroidered silk and gold on a silk base, testify to the skill of Anglo-Saxon embroiderers; however, though there is evidence of male dress sumptuously decorated with gold (Owen-Crocker 2004: 233; 237), there is no evidence that women ornamented their own clothes since the misguided nuns of the early convent at Coldingham, which came to a bad end, to Bede’s evident satisfaction (Bede HE iv.25; Colgrave and Mynors 1969: 424-7). In the late Saxon period, when piety and public generosity were the qualities praised by writers, luxurious clothing for women was not mentioned.

frustratingly limited. There are archaeological ‘stray finds’ of brooches, buckles, finger rings and strap ends but since they are without human association it is impossible to tell if they were male or female possessions unless they have a name on them, and that is fairly rare.22 Even identifying the gender of a name in Old English is tenuous: if a name is unusual or unique, establishment of gender is dependent on the assumption that this corresponds to the grammatical gender of the word which makes up the second syllable of the word. Accordingly the name Bvuredrvð and the statement Eawen mie ah (‘Eawen owns me’) on simple 9th-century finger rings are assumed to refer to female The fact that when women’s names appear on Anglo-Saxon artefacts they tend not to be on the primary faces is discussed in detail in OwenCrocker (2002). 22

Evidence of dress accessories in the later Saxon period is

86

Gale R. Owen-Crocker: Anglo-Saxon Woman

Fig. 8.2: Queen Ælfgifu (Emma) and King Cnut presenting a cross to the New Minster at Winchester, London, British Library, MS Stowe 944, fol. 6r.

owners (Wilson 1964: 178, no. 85 and Page 1964: 83; Wilson 1964: 205-6, no. 145).23 If the name is known from elsewhere or appears in meaningful context the gender is more securely established: Ælfgivv me ah, ‘Ælfgivu owns me’ appears as the decorative border on a circular silver mount from Cuxton, Kent24 and Ædvwen is named as owner in an inscription on the back of an 11th-century silver disc brooch from Sutton, Isle of Ely, Cambridgeshire (Wilson 1964: 174-7 with pls XXXI-XXXII; Page 1964: 86-8). We can deduce that these owners were women because Ælfgifu/ Ælfgivu was a popular female name in the Wessex royal

family25 and Ædvwen can be identified as a woman because the inscription refers to ‘her’ (hyo and hire); but we know nothing about the individuals named on these artefacts. The only jewellery associated with a known historical woman is a small finger ring which bears an internal inscription EAÐELSVIÐ REGNA.26 The name and title have been confidently associated with Queen Æthelswith, the wife of King Burgred of Mercia. The daughter of King Æthelwulf of Wessex and sister of King Alfred ‘the Great’, she died

PASE lists 19 items under Ælfgifu. Royal women of this name included wives of kings Eadwig, Edmund and Æthelred II (his first wife as well as Ælfgifu-Emma) and the mistress or first wife of Cnut. 26 Wilson (1964: 117-19, no. 1, pl. XI.i). The inscription appears in an unnumbered drawing on p. 118. 25

Christine Fell, Women in Anglo-Saxon England; and the impact of 1066 by Cecily Clark and Elizabeth Williams (1984), p. 94 and Fig. 32. 24 Wilson 1964: no. 14, pl. XVII. The object has a central motif of a raptor and its prey. 23

87

Dress and Identity

Fig. 8.3: Women’s dress accessories: composite drawing showing (a) girdle hangers, (b) keys, (c) reliquary box (not to scale) , (d) reconstruction of a pouch with ivory ring, and (e) toilet implements. Figs a, d were drawn by Rosalyn Smith and Figs b, c and e by Christine Wetherell. Reproduced from Crocker (2004) Figs 43, 44, 45, 50, 115.

88

Gale R. Owen-Crocker: Anglo-Saxon Woman

Gender and women’s dress

in Pavia on the way to Rome in 888.27 Æthelswith is thus a well-documented historical figure, rich and royal, but not exactly a ‘famous name’ since nothing is recorded about her personal achievements and she was a queen consort rather than a queen regnant. Furthermore it is not established whether the queen herself wore this silver ring or if it bore her name because she was the donor of it to someone who would benefit from the prestige of the gift.

Many grave-goods were clearly gendered.30 Weapons – swords, spears, long fighting knives (scramasaxes) and shields – are found virtually exclusively with male skeletons, and though utilitarian knives are found with both sexes, they come to be more characteristic of male than female burials towards the end of the period of furnished graves. Horse tack, buckets and metal vessels, musical instruments and gaming pieces are characteristically male. Tweezers and pursemounts are more often found with men, at least in the early part of the period. Men occasionally have pins and sometimes buckles; but elaborate belts (with counterplates and sometimes other decorative metal attachments) are characteristically masculine equipment. On the other hand brooches are strongly female gendered. Wrist clasps and the majority of straight pins also occur with female skeletons. While men occasionally have single beads, especially attached to the sword hilt, the elaborate swathes of beads found at the neck and elsewhere on the body of very many skeletons are characteristically female ornaments. The majority of girdle adjuncts, keys, ‘girdle hangers’, rings, remains of pouches, the large single beads which functioned as spindle whorls and the small copper alloy cylinders once thought to be threadboxes but now considered reliquaries, are associated with women (Fig. 8.3). Pottery and glass vessels are found more with women than with men. It may be suggested that the gravegoods characteristic of men represent public life – warfare, riding out, feasting and entertainment, with a few useful tools like tweezers and fire-making materials, whereas female grave-goods are predominantly garment fasteners and dress accessories. This means that the majority of instances of textile survival derive from women’s clothing rather than men’s, since textile remains are usually attached to metalwork. Female clothing of the early Anglo-Saxon period is thus more easily reconstructed than male.

Since jewellery does not appear in illustrations of women we do not know how, and in what circumstances, the archaeological finds might have been worn as feminine accessories. In fact the evidence of the illuminations and that of the artefacts appears to be contradictory: there is extant jewellery which was evidently owned by women; but in late Anglo-Saxon art it is men who wear cloak brooches and women wear no jewellery at all. Any investigation of women’s history in the Christian Anglo-Saxon period is a frustrating business for scholars with an interest in dress and accessories. It can produce the names of a few famous women who cannot be dressed or accessorised, a few items of jewellery associated with named women about whom nothing is known, and other artefacts which cannot safely be associated with women at all. Tangible but anonymous women The picture could not be more different for the early AngloSaxon period. From the Migration which began in the mid5th century, up to the early decades of the 7th century and in isolated cases into the 8th, the practice of burying the dead with grave-goods has left us detailed archaeological/ anthropological evidence for the reconstruction of social history.28 In this case there is far more evidence for women than for men, since a greater number and variety of grave-goods, especially dress accessories, have been recovered from female graves than from male. The gravegoods clearly indicate gender distinctions, and study of their positioning makes it possible to reconstruct aspects of women’s burial costume, both the garments and the accessories which held it together and were appended to it.29 Grave-goods can reflect ethnicity and they can indicate date, particularly the striking differences between 5th/6th century deposits and those of the 7th/8th centuries, the so-called ‘Final Phase’ (of the practice of burying with grave-goods). More adventurously, the researcher may use grave-goods to investigate belief system, ownership in relation to age, status in local context and role in the community.

Deductions about dress have to be made from large numbers of graves, no two of which are identical, in the light of parallels from other countries and depictions of Germanic women in Roman art. Every grave-assemblage potentially tells its own story, but certain generalisations can be made. Women were dressed in garments of wool or linen. The spinning of their clothes was even and professional, their weaving consistent, whether it consisted of plain weave, simple twill or the more skilled weaves which produced a diamond- or chevron-patterned fabric; fine weave for gowns and coarser weave for cloaks. Cuffs and neck borders were strengthened and decorated with tablet woven bands.31 In the 5th and 6th centuries in Anglian and Saxon regions and in 5th-century Kent, many women wore a tubular gown (‘peplos’) clasped at the shoulders by a pair of brooches, worn over another garment with close fitting sleeves. It was usually belted or girdled, and the girdle served to suspend various tools and trinkets. The peplos went out of fashion by the 7th century and it is not clear what replaced

Her marriage and the circumstances of her death are recorded in The Anglo-Saxon Chronicle (E for 852, A for 853; A, E for 888): Swanton (1996: 66-7; 82-3 and nt. 2 for F version). 28 I focus on inhumation graves only, since cremation does not leave evidence of gender (if there are indications of gender in the decoration on cremation urns, we have not yet ‘read’ them). 29 See Owen-Crocker (2004: 35-165), which chiefly identifies general patterns; also Walton Rogers (2007: 111-228), which particularly highlights individual styles from specific examples. 27

30 31

89

Detailed statistics are available in Stoodley (1999: 24-52). For specific evidence see Coatsworth and Owen-Crocker (2007).

Dress and Identity

Ethnicity The Germanic ethnicity of the migrants is clear from the general resemblance of grave-goods to the artefacts of north-west continental Europe, but regional distinctions and local variations are apparent in all parts of Anglo-Saxon England. Women’s dress accessories broadly follow Bede’s tripartite division of the 5th-century migrants (into Angles, Saxons and Jutes, the latter in Kent, the Isle of Wight and the opposite part of the south coast) though they also reflect some cultural mingling and 6th-century immigration. Anglian women’s dress is particularly distinctive, since it often includes a third central brooch in addition to the two regularly found at the shoulders in other regions.33 Annular (ring) and cruciform brooches were particularly popular in Anglian areas and wrist clasps, a 6th-century Norwegian import, are almost exclusively found there. Although tablet-woven bands are found as wrist bands and neck-edging in all regions, the most elaborate ones, with embroidery or brocading in the Scandinavian tradition, have been found in eastern, Anglian regions. Similarly Anglian are the somewhat enigmatic articles called ‘girdle hangers’ which are found in a position consistent with suspension at the belt, but whether for practical or symbolic purpose is not clear. Simple disc brooches and the more complex saucer brooches which have angled rims and decorated front plates are characteristic of Saxon culture, with the small ‘button brooch’ variety of saucer brooch occurring both in central southern England and in Kent. Simple long brooches, with variously shaped heads and feet occur in all regions. Large, intricate square-headed brooches (Fig. 8.4), evidently high status possessions, were worn throughout the Anglo-Saxon region in the early 6th century, later becoming characteristic of eastern, Anglian areas and disappearing before the end of the century (see Hines 1997). Fig. 8.4. 6th-century great square-headed brooch from Paglesham, Essex, with stylised human faces, geometric ornament and other decoration, now incomprehensible. Southend Museum

In Kent there was a rich culture which demonstrates a blend of material from northern Denmark (Jutland) and continental Europe: the present-day Netherlands, France and the German Rhineland. It is most dazzlingly manifested in items of jewellery. In Kent small square-headed brooches of Jutish ancestry can be found, together with garnet-ornamented square- and radiate-headed brooches of Frankish derivation. Small brooches in bird-, horseand geometric shapes also show cultural influence from continental Europe. Characteristic of Kent, and again demonstrating Frankish influence, are small silver disc brooches, usually of silver, inlaid with garnets in cast settings (see Avent 1975). In the course of the 6th century these evolved into cloisonné brooches, with tiny garnets and coloured glass set into gold cells, interspersed with filigree gold, set on gold plates on the front of silver brooches, and in the 7th century into magnificent gold brooches. Dress accessories in 6th-century Kent include multiple brooches, gold-decorated fillets and wristbands and belt accessories

it, although a wide-sleeved robe worn over the tight-sleeved garment is a possibility.32 In Kent a front fastening garment became preferred in the 6th century, perhaps covered by a front fastening jacket (Walton Rogers 2007: 189-91). Outer garments, cloaks or shawls, fastened by a single brooch were worn in some cases, and Penelope Walton Rogers has recently identified fur capes, fastened by buckles, in a few late 6th- and 7th-century graves (Walton Rogers 2007: 172). Evidence for headgear is scarce, apart from fragments of gold at the heads of certain richly-equipped Kentish women of the 6th century, but Penelope Walton Rogers argues for short veils in the 5th and early 6th centuries, longer in the later 6th and 7th centuries (Walton Rogers 2007: 157-67).

See Penelope Walton Rogers’ reconstruction of the sleeves from Sutton Hoo Mound 14 (Walton Rogers 2007: 185, fig. 5.41).

33 There are accounts of early Anglo-Saxon brooches in Owen-Crocker (2004: 37-42, 138-41) and Walton Rogers (2007: 113-22).

32

90

Gale R. Owen-Crocker: Anglo-Saxon Woman

in the form of silver spoons set with garnets accompanied by crystal balls suspended in silver slings.

is contrary to the usual practice of displaying relics, and is more of a superstitious practice than a Christian one (Crowfoot 1990).

Date

It is evident that 7th-century women were proclaiming the Christian religion on their dress accessories – though one cannot know if all the individuals who wore these things were themselves converts or were merely following fashion. It is not clear, however, whether some of the motifs of earlier jewellery were equally meaningful in terms of paganism. Much early Anglo-Saxon art is zoomorphic and perhaps has its roots in pagan belief though the birds and beasts it features may have diminished into mere decoration after centuries of development. The Roman historian Tacitus records that the horse was sacred to the Germanic people, and the many ceremonial horse burials scattered throughout north-west Europe are a testament to this. It seems likely therefore that the horse heads which decorate the terminals of cruciform brooches had some religious or amuletic purpose. Boars are well documented as protective emblems in Old English literature and it is therefore interesting to find boars’ heads on a perforated buckle contemporary with cruciform decorations. A few human faces on square-headed and button brooches could represent gods, kings or ancestors. The miniature hammers and spears found as amulets in 7th-century women’s graves in Kent are most likely to be emblems of the northern gods Thunor and Woden. In this case the conversion of some of the population to Christianity possibly made it fashionable to wear religious emblems and contemporary pagan women were responding according to their own beliefs.

Artefacts can to some extent be dated by the evolution of their ornament and techniques, though innovation need not mean the abandonment of the older-established. New artefact types, such as the Scandinavian wrist clasps and Frankish brooches which appeared in different regions of England in the 6th century, may mark the arrival of new groups of female immigrants, high status brides perhaps, bringing native fashions with them which were then copied and developed by English craftsmen. A major change in burial practice came in the 7th century. Cremation in folk cemeteries had been disappearing, though high status cremations in barrows featured in the late 6th century. The 7th century was marked by the abandonment of old ‘folk cemeteries’ and the establishment of new, relatively short-term burial grounds with few grave goods, east-west orientation, burials under mounds and with evidence of various superstitious practices (such as decapitation of the corpse or burial with a bag of teeth). Where women were buried with grave-goods other than a knife, a new style of female dress is indicated by pin suites or pairs of tiny annular or penannular brooches, perhaps securing veils; and by the disappearance of the traditional brooch types and of girdle adjuncts such as keys and girdle hangers. Regional variation in dress was discontinued. Circular brooches, annular or of the elaborate Kentish type, were now worn singly rather than in pairs at the shoulders, the peplos having apparently gone out of use. The glass and amber beads formerly so popular with women gave way to necklaces of pendant bullae or festoons of metal rings to which beads were attached, and girdle adjuncts now included variously chatelaine chains, the small bronze cylinders interpreted as relic boxes and trinkets in the shapes of miniature spears and hammers.

Age/status in local context There is great variation in the quantity of grave-goods found and in the material value of them. Objects may have been deposited according to the status that came with age. Just as males seem to have acquired weapons in the order spear, shield, sword (Härke 1989), so females may have been dressed in the distinctive peplos-type gown, usually fastened by two brooches at the shoulders, during the fertile years, and may have acquired a veil, marked by a pin, on marriage.34 Sometimes the quantity of objects deposited appears to relate to the age of the person buried, but the picture is not consistent: young children were sometimes buried with a single bead or just a few of them, but occasionally a child’s skeleton is found with many beads. Older women were sometimes equipped for the grave with considerable amounts of jewellery, both worn items and newer things, perhaps the accumulation of a lifetime. Some may have jettisoned rather than increasing their possessions, however archaeological evidence leads Penelope Walton Rogers to suggest that some older women ceased to wear the peplos with its paired brooches once they were past child bearing (Walton Rogers 2007: 178). Conversely, a well-known grave from a cemetery at Holywell Row, Suffolk, contained the remains of what, from the size of

Belief system The Christian religion is unmistakeably represented in Anglo-Saxon jewellery of the early 7th century, the period following Augustine’s mission (see also Crawford 2003). The cross appears repeatedly: there is a pendent cross incorporated in the Desborough necklace (Fig. 8.5) and there have been individual finds such as the Holderness and Ixworth crosses (Owen Crocker 2004: 145, fig. 101). Circular necklace pendants of the 7th and 8th centuries are often decorated with cruciform motifs and some Kentish buckles bear perforations in cruciform designs. Most magnificently, Kentish cloisonné disc brooches bear not only cruciform designs, but also the quincunx, arguably representing the five wounds of Christ. The little bronze cylinders which sometimes contain textile and sometimes bear cruciform ornament, have been thought to be containers for holy relics; however, as Elisabeth Crowfoot noted, the burial of such reliquaries as amulets in graves

These suggestions are made on detailed studies of selected sites; Walton Rogers (2007: 178). 34

91

Dress and Identity

the badly preserved skeleton, was thought to be a young girl, accompanied by a large number of grave-goods, some highly prestigious and at least one item quite old. She had been decked out for the grave in items which included an heirloom, in a combination of Anglian and Kentish dress accessories, and with a metal weaving sword. Though young, she evidently had considerable prestige, though the assemblage may have reflected the status she would have achieved if she had reached maturity (Lethbridge 1931: 2).

been made of organic material, may well have been present in the grave originally, but would have rotted away. The spindle whorl tells us nothing about rank, because in an era when all cloth was hand made and the ratio of spinning time to weaving time was about 10:1, it is likely that all women could spin. However the presence of a metal weaving beater, a sword-shaped implement for packing the weft on a warp-weighted loom, is rare and found with high status jewellery, so it was evidently a prestige object. It probably indicates the lady who owned the weaving equipment and controlled the servants who carried out the work, rather identifying its owner as a weaver.

Grave-goods may have been valued by quantity; they were certainly valued for rarity.35 Objects of precious metal, gold and silver, were clearly very high status possessions: Kentish gold composite brooches and gold brocaded textiles must have been possessed only by those of very high status. Gilding or silver plating would also convey prestige, and when applied to an elaborate and intricate object such as a great square-headed brooch was also clearly a sign of high rank (Fig. 8.4).Garnets and amethysts are also uncommon and obviously precious. The value of intricate and brightly coloured glass beads and imported amber ones is harder to estimate since they are not rare. Their manufacture and travel must surely have been worth something in material terms.

More problematic are the amulet collections which in the 6th century were evidently often carried in bags attached to the left side of the girdle. Some of those bags may have themselves been exotic imports since they were attached to rings of elephant ivory. Audrey Meaney notes the ‘strange collection of objects’ typically found in the remains of an amulet bag: a piece of broken glass, rings of iron or bronze, often linked together, an object at least 100 years old when buried, something of animal substance (Meaney 1981: 249-50). She suggests that the owners of these collections, which are usually unique in their cemetery, were ‘cunning women’, who employed the amulets for healing or divination (Meaney 1981: 257, 259). Tania Dickinson (1993) offers a similar interpretation for a woman who wore a collection of bucket-shaped pendants, containing spun thread, attached to a leather bib, and also carried an amulet bag at the hip containing metal rings and an antler cone. In the 7th century amulets were placed in the grave in a box. Superstition evidently continued, sometimes more strongly attested than before, into the early Christian period. Individual ornaments were also possibly prophylactic. Amber was believed to prevent and cure various ailments, and crystals, found as beads as well as mounted spheres, may also have been magical, though alternatively they may have been Christian amulets (Owen-Crocker 2004: 87, 935). We cannot know if the women wore such things only for their own benefit, or if they testify to a healing role in the community.

Wealth is relative: though a woman may be equipped with much less richness than another in a different cemetery, she may be by far the richest in her own community. All of the women we are discussing had some status since they were accompanied by any grave-goods at all – there are burials in nearly every Anglo-Saxon cemetery without grave-goods; they were also accorded the status of a place in the cemetery – there must have been members at the lowest strata of society who were not. Social role Many women were equipped with keys – simple latchlifters which could have opened chests. It seems likely that such women had charge of containers for precious things – perhaps metalwork, textiles, or wine. What is not clear is the identity of the bearer of the keys, whether the lady of the house or the chief domestic servant. Evidence from associated grave-goods is contradictory. The perforated silver spoons found with wealthy Kentish women may possibly have been for skimming wine; if so they reflect the woman’s role as provider of drink in a prosperous household (Owen-Crocker 2004: 95). The presence in a grave of undecorated pots or metal vessels may be the remains of food and drink left for the dead person; or they may testify to the role of the woman – at the humbler level of cook or the exalted level of hostess to a comitatus of warriors.

The grave-goods undeniably leave gaps – they rarely include anything organic for instance, just bone combs and occasional scraps of leather – and it is frustrating that we cannot complete the picture of female appearance since the surviving artefacts reveal nothing about women’s hairstyles or footwear. However the very presence, quantity and consistency of grave-goods demonstrates the formal acceptance of the importance of Woman in society; not just royal Woman, but prosperous middle-class Woman, the housekeeper, the spinner, the potter, the cook, the healer. It is not known whether the grave-goods accompanying a woman were her own personal property, or even if she had worn and used them during life; some or all may have been symbolic – of her gender, and her place in society – as she made her final appearance for the grave and went on to whatever afterlife her people believed in. However, the fact that costly jewellery and dress accessories were lavished

Many women are buried with a spindle whorl – a bead made of stone, glass, pottery or even amber – which acted as a flywheel on the drop spindle which was the essential textile tool of the period. The spindle itself, which would have 35

Explored in some detail in Owen-Crocker (2005a).

92

Gale R. Owen-Crocker: Anglo-Saxon Woman

on the dead body of Woman suggests she was important, symbolically as well as personally, to her family and her community and the expression of her status in preparation for her continued existence in some form was a necessary part of the funeral ritual. Thus the archaeology of the pagan and conversion period presents us with a considerable amount of evidence about Woman’s life and role in society. If the exact nature of her religious belief remains intangible, her trust in amulets, her need for supernatural reassurance is clear. Our understanding of her is imperfect, but we know this Woman. Yet there is no runic name on the back of a brooch, no inscription on a ring from the early Anglo-Saxon period. She, they, remain anonymous. *** Fig. 8.5: 7th-century necklace with cross, Desborough, Northamptonshire. Trustees of the British Museum.

As the practice of depositing grave-goods diminishes and ends in the course of the 7th and 8th centuries,36 this knowable Woman fades from view, leaving us only the names and selected details about the lives of those few individuals who impressed themselves on male authors and artists; and giving us no idea about the lifestyle, status and personality of the invisible majority.

in Anglo-Saxon graves’ in P. Walton and J.-P. Wild (eds), Textiles in Northern Archaeology, NESAT III: Textile Symposium in York 6-9 May 1987, North European Symposium for Archaeological Textiles Monograph 3, 47-56. London. Damico, H. and Olsen, A. H. (eds) (1990): New Readings on Women in Old English Literature. Bloomington Dickinson, T.M. (1993): ‘An Anglo-Saxon “cunning woman” from Bidford-on-Avon’ in M. Carver (ed.), In Search of Cult: archaeological investigations in honour of Philip Rahtz, 45-54. University of York Archaeological Papers. Woodbridge. Dockray-Miller, M. (2000): Motherhood and mothering in Anglo-Saxon England. Oxford. Dodwell, C.R and Clemoes, P. (1974): The Old English Illustrated Hexateuch. Early English Manuscripts in Facsimile 18. Copenhagen. Dümmler, E. (ed.) (1892): ‘S. Bonifatii et Lulli epistolae’, Monumenta Germaniae Historica Epistolae 3, Merovingici et Karolini Aevi, I. Berlin. Emerton, E. (1940): The Letters of Saint Boniface. New York. Fell, C. (1990): ‘Some implications of the Boniface correspondence’ in H. Damico and A.H. Olsen (eds), New Readings on Women in Old English Literature, 2943. Bloomington. Foot, S. (2003): ‘Unveiling Anglo-Saxon nuns’ in D.Wood (ed.), Women and Religion in Medieval England, 13-27. Oxford. Härke, H. (1989): ‘Early Saxon weapon burials: frequencies, distributions and weapon combinations’ in S. Chadwick Hawkes (ed.), Anglo-Saxon Weapons and Warfare, Oxford University Committee for Archaeology Monograph 21: 49-61. Hines, J. (1997): A New Corpus of Anglo-Saxon Great Square-Headed Brooches. Woodbridge (for the Society of Antiquaries of London).

Bibliography Avent, R. (1975): Anglo-Saxon Garnet and Inlaid Composite Brooches, 2 vols, BAR 11. Barlow, F. (ed. and trans.) (1992): The Life of King Edward who Rests at Westminster (2nd ed.). Oxford Medieval Texts, Oxford. Blanton, V. (2007): Signs of Devotion: the cult of St Æthelthryth in medieval England 695-1615. University PasR, PA. Campbell, A. (ed.) (1979): Encomium Emmae Reginae, Camden Soc., 3rd series, 72, London, Royal Historical Society. Coatsworth, E. and Owen-Crocker, G.R. (2007): Medieval Textiles of the British Isles AD 450-1100: An Annotated bibliography, BAR 445. Oxford. Colgrave, B. and Mynors, R.A.B. (eds) (1969): Bede’s Ecclesiastical History of the English People. Oxford. Crawford, S. (2003): ‘Anglo-Saxon women, furnished burial and the Church’ in D. Wood (ed.), Women and Religion in Medieval England, 1-12. Oxford and Oakville, CT. Crowfoot, E. (1990): ‘Textile fragments from “relic-boxes” The traditional view that grave-goods disappeared because Germanic paganism was giving way to Christianity is no longer considered the sole reason for this development, though it may account for some of the changes in burial practice in the ‘Final Phase’ cemeteries. The Church never formally forbade grave-goods and they continued in use both on the Christian Continent and in ecclesiastical burials in England. This is discussed in some detail in the opening of Owen-Crocker and Stephens (2007). Political and economic developments, and the social prestige of burial in association with a church (belief apart) were also probable factors in the change. There are very few furnished graves dating from the 8th century. The one from Boss Hall, Ipswich, Suffolk, included gold and garnet jewellery, but it had been containerised, not worn. 36

93

Dress and Identity

Karkov, C.E. (2004): The Ruler Portraits of Anglo-Saxon England. Anglo-Saxon Studies 3. Woodbridge. Keynes, S. (ed.) 1996: The Liber Vitae of the New Minster and Hyde Abbey Winchester: British Library Stowe 944: together with leaves from British Library Cotton Vespasian A. VIII and British Library Cotton Titus D. XXVII. Early English Manuscripts in Facsimile 26. Copenhagen. Lethbridge, T.C. (1931): Recent Excavations in AngloSaxon Cemeteries in Cambridgeshire and Suffolk, Cambridge Antiquarian Society Quarto Publications, 2nd series, 3. Cambridge. Meaney, A.L. (1981): Anglo-Saxon Amulets and Curing Stones, BAR, 96. Oxford. Ohlgren, T.H. (ed.) (1992): Anglo-Saxon Textual Illustration: photographs of sixteen manuscripts with descriptions and index. Kalamazoo, MI. Owen, G.R. (1979): ‘Wynflæd’s wardrobe’, Anglo-Saxon England 8: 195-222. Owen-Crocker, G.R. (2002): ‘Anglo-Saxon women: the art of concealment’, Leeds Studies in English 33: 31-51. Owen-Crocker, G.R. (2004): Dress in Anglo-Saxon England: revised and enlarged edition, Woodbridge. Owen-Crocker, G.R. (2005a): ‘Gold in the ground or just rust in the dust: measuring wealth by metalwork in Anglo-Saxon graves’ in R. Bork (ed.), De Re Metallica: The Uses of Metal in the Middle Ages, 15-29. AVISTA Studies in the History of Medieval Technology, Science and Art, 4. Aldershot. Owen-Crocker, G.R. (2005b): ‘Pomp, piety and keeping the woman in her place: the dress of Cnut and ÆlfgifuEmma’, Medieval Clothing and Textiles 1: 41-52. Owen-Crocker, G.R. and Stephens, W. (2007): ‘The Cross in the Grave: Design or Divine?’ in C. Karkov, K. Jolly and S. Keefer (eds), Cross and Culture in Anglo-Saxon England: Studies in Honor of George Hardin Brown,

The Santa Crux Halig Rod Series, Volume 1, 117-37. Morgantown, West Virginia. Page, R. (1964): ‘The Inscriptions’, Appendix A to D.M. Wilson, Anglo-Saxon Ornamental Metalwork 700-1000 in the British Museum. London. Stafford, P. (1997): Queen Emma and Queen Edith: queenship and women’s power in eleventh-century England. Oxford. Stoodley, N. (1999): The Spindle and the Spear: a critical enquiry into the construction and meaning of gender in the early Anglo-Saxon burial rite, BAR 288.Oxford. Swanton, M.J. (ed. and trans.) (1996): The Anglo-Saxon Chronicle. London. Tangl, M. (ed.) (1916): S. Bonifatii et Lulli Epistolae, Monumenta Germaniae Historica, Epistolae 4, Epistolae Selectae, 1. Berlin. Wainwright, F.T. (1959): ‘Æthelflæd Lady of the Mercians’ in P. Clemoes (ed.), The Anglo-Saxons: studies in some aspects of their history and culture, presented to Bruce Dickins, 53-69. London. Walton Rogers, P. (2007): Cloth and Clothing in Early Anglo-Saxon England AD 450-700, CBA Research Report 145. York. Webster, L. and Backhouse, J. (eds) (1991): The Making of England: Anglo-Saxon Art and Culture AD 600-900. London. Whitelock, D. (1930): Anglo-Saxon Wills. Cambridge. Whitelock, D., Ker, N. and Lord Rennell (1968): The Will of Aethelgifu. London. Wilson, D.M. (1964): Anglo-Saxon Ornamental Metalwork 700-1000 in the British Museum. London. Wilson, D.M. (1985): The Bayeux Tapestry: the complete tapestry in colour. London. Yorke, B. (2002): Nunneries and the Anglo-Saxon Royal Houses. London and New York.

94

Chapter 9

Representing Hierarchy and Homosociality: Vestments and Gender in Medieval Scotland Penelope Dransart

This paper draws together archaeological and iconographic evidence in order to examine representations which cast light on late medieval gender hierarchies. My aim is to consider certain material practices connected with the celebration of the Mass in Scotland against a background of wider developments in western Christendom. As a sacrament the Eucharist was, and is, the focal act of worship in liturgical services celebrated in Christian churches. One of the requirements for the ritual to take place is that the celebrant wears particular vestments. Sørensen (2000: 142) has discussed how dress is implicated in social relations; she considers dress to have the capacity to enable action. Vestments, when worn by an ordained priesthood, provided the outward manifestation of priestly authority.

One might infer from this terminology that there was a marked separation of priestly activity from the actions of the lay people, who stood or sat in the nave of the church. However, many authors argue that the specialised work of priests did not deny an active role for the congregants in the celebration of the Eucharist. The liturgy has been regarded as a social ritual which evolved through the Middle Ages, characteristically involving participation on the part of the clergy and the laity (Bossy 1983: 36). Duffy (1992: 220) has also considered the sharing of experience in the liturgy by social groups who might not, at first sight, have held interests in common; for example people would have had varying degrees of Latin comprehension, or none at all. Reinburg (1992: 529-30) explored differences between the Latin text of the missal used by the clergy and prayer books designed for lay men and women in medieval France. Although the priest’s role in the Eucharist was theologically prescribed, that of the laity was not and their presence was not required for the Eucharist to take place. She argued, however, that the lay prayer books assigned an important social role to the congregants and that their participation provided ‘the gestural assent’ (Reinburg 1992: 542).

It is important to take into account the spatial aspects of the ritual actions in which the robed priests projected that authority. They enacted a specific ritual gesture in a specific location in front of an altar: the elevation of the host, the wafer of unleavened bread which is transformed into the body of Christ through the words of consecration spoken by the priest. While the action of raising the host evoked complex symbolic meanings among people who partook of the ritual, other gestures were also important and require consideration. These were performed by the largely anonymous artisans who plied needle and thread and who, stitch by stitch, brought into being the vestments without which the Mass could not take place. Hence this paper has interconnecting objects: it focuses on the making, remaking and, more particularly, the wearing of vestments in a spatial context which admitted some people, but excluded others.

Nevertheless, relationships between the members of different social groups brought into play a series of inclusions and exclusions. Bossy (1983: 44) saw the distinction between the living and the dead as being particularly problematic for the medieval church. Another conceptual division to have received scholarly attention is that between head and body, in which the body of the church was perceived in a gender allegory as female. It was the bride of Christ, in a configuration which correlated the architecture of the chancel with the head, mind or soul and the nave as the female, corporeal principle (Stanbury & Raguin 2005: 6). In this paper I explore the priesthood in this male/female distinction, but I also consider the inclusion of idealisations of women in a male homosocial sphere of activity in the chancel. Corine Schleif (2005: 220) remarked that from the early medieval period and into the Middle Ages, people entered the church for Mass and separated themselves into groups which were

Several terms have been applied to the eastern end of the church where these liturgical acts were performed. They include the terms ‘presbytery’ (presbiterium) and ‘chancel’ (cancellus). Fernie (2001: 15) explained that the former accounts for the priests who officiated in the location thus designated, while the latter derives from the architectural space itself, as defined by barriers called cancelli. These two terms provide a distinction between function and structure, which Fernie (2001: 19) characterised as a set of paired concepts denoting ‘activity’ and ‘place’.

95

Dress and Identity

Fig. 9.1: St Ursula, late 15th-early 16th century embroidery mounted on a chalice veil (photograph: P. Dransart, by courtesy of Blairs Museum).

characteristically homosocial; in addition they formed groups constituted by age and/or marital status.

is an embroidered, idealised image depicting a female saint, Ursula, a virgin martyr. Embroidered in costly silk floss and couched in metal-wrapped threads, this is one of the images used in the Eucharist until recent times in the diocese of Aberdeen (Fig. 9.1). Ursula is depicted with golden hair and dark eyes; she was made to conform to a medieval standard of beauty in women evidently widespread in western European countries. Herman Pleij (2004: 50) discussed this predilection for female beauty, as voiced in literary accounts from the Low Countries, with which Scotland was engaged in a culture of intellectual and commercial exchange. The panel dates from the late 15th to early 16th century and it was cut from a worn vestment and remounted in the 19th century. It now adorns a chalice veil, while other panels from the same series were mounted as orphreys (embroidered bands) on a chasuble made of the same burgundy velvet as the chalice veil. At the time the panels were embroidered, the image of a virgin martyr represented the aestheticised ideal of womanhood in a milieu from which women were physically excluded in the restricted space in the chancel of a late medieval church or cathedral. The high altar was situated in this space and constituted the focal point of the High Mass.

The works of representation involved in the making and remaking of vestments and their use in the Eucharist provided a basis for the production of difference based on different levels of inclusivity and exclusivity. This analysis of the use of vestments in medieval Scotland builds on the work of Miri Rubin (1991: 289) who examined the social context of differential access to the actions involved in celebrating the Eucharist. Although some sentences in the writings of St Jerome at the end of the 4th century AD have been interpreted as indicating that priests donned special garments for Christian worship (Macalister 1896: 15-17), the formalisation of items of priestly dress seems to have developed gradually in the history of the church. One of the canons drawn up for the Fourth Council of Toledo (AD 633) provided for the restoration of the insignia of an unjustly defrocked cleric. The insignia mentioned included the orarium of deacons, priests and bishops, the planeta of priests and bishops, and the ring and staff of bishops (Macalister 1896: 27-8). By the Middle Ages, the orarium had developed into the stole and the planeta (or poenula) into the chasuble. The term chasuble, derived from casula, meaning ‘little house’, was a capacious cloak with an opening allowing it to be pulled over the head of the wearer (Macalister 1896: 42-5).

Another series of late 15th- to early 16th-century embroideries executed in silk floss and couched metalwrapped threads has been preserved on a chasuble at Blairs Museum, near Aberdeen. The themes depicted scenes from the Life of the Virgin Mary (Fig. 9.2). In a popular belief derived from the apocryphal Protoevangelium which was current in the 15th century, St Anne was thought to have conceived Mary at the moment when Joachim kissed her, and artists depicted the miraculous conception of

In time, the makers of vestments made increasing use of the garments as a field for visual imagery. Among the fragments of church vestments to have survived from late medieval times in Scottish cathedral and church treasuries

96

Penelope Dransart: Representing Hierarchy and Homosociality

Fig. 9.2: Virgin, child and St Anne, late 15th-early 16th century embroidery mounted on a chasuble (photograph: P. Dransart, by courtesy of Blairs Museum).

the Virgin by means of an embrace between St Anne and Joachim at the Golden Gate, Jerusalem (Stratton 1994: 21). This scene is included in one of the orphreys mounted on the chasuble (Dransart & Bogdan 2004: 464-6). The immaculatist overtones of the depiction constitute another visualisation that idealised the images of sanctified women for a predominantly male purview. Other images of the birth and childhood of the Virgin survived in orphreys stitched to a chasuble from Hartpury House, near Gloucester (now in the Victoria and Albert Museum), and a dalmatic from Whalley Abbey (now in the Burrell Collection). These embroideries are considered by King (1963: 46, Plate 21), Monnas (1994) and Johnstone (2002: Fig. 49).

of new technologies. The Black Death affected Scottish burghs in 1349 and there were many outbreaks of the plague thereafter (Dennison et al. 2002: 74). Nevertheless, an intellectual environment characterised by humanism during the period 1280-1440 in Europe was associated with what Georges Duby (1966: 11) saw as the emergence of ‘new men’ (‘des hommes nouveaux’), who were responsible for a vigorous production of representational forms. The technology of representation in the medium of embroidery should be recognised as making an important contribution to this intellectual endeavour. Duby (1966: 89) regarded the period as ceasing to be one in which Christianity was merely an affair of priests. In their religious life the laity accorded considerable importance to daily acts of piety designed to guard against the powers of evil and to attract the mercy of God. Among the eminent princes and chevaliers who, according to Duby, characterised this resurgent laity, he names two women, Isabeau or Elizabeth of Bavaria, Queen of France, and Christine de Pisan, woman of letters. However, in his comment on the representations of women carved in stone during this period, Duby (1966: 149) was of the opinion that the sculptors produced a vision of earthly women who did not mix easily in the company of saints. He contended that the radiant beauty of earthly women carved in effigies of the 14th and 15th centuries constituted a triumph of lay values in that it challenged the too-charming images of the Mother of God. The irony here is that men’s images of womanhood, commissioned by the laity, seem to vindicate those very values enunciated earlier in the Middle Ages by the clergy and analysed by Le Goff (1992).

The use of scenes concerning the life of the Virgin, as given prominence in the social context in front of the altar, calls to mind another dual distinction concerning purity and impurity. Medieval ambivalence towards marriage rested on the authority of the New Testament, in which Mary remained a virgin in marriage and her son Jesus remained a bachelor (Le Goff 1992: 95). The embroidering of chasubles and dalmatics with scenes indicating the immaculate birth of the Virgin provided a visual demonstration of the celibacy in marriage of Anne, the mother of Mary. Le Goff (1992: 102) argued that marriage was ‘the property of lay men and women; virginity, celibacy, and/or continence became the property of priests, monks, and nuns’ and that the ‘church became a society of bachelors, which imprisoned lay society in marriage’. While plague and military activities brought about demographic recession in European countries, the 14th century also witnessed prodigious developments in terms

My aim in this article is to provide a contrast to studies that examine representations of lay men and women by

97

Dress and Identity

investigating representations of the priesthood in the spaces that they created for themselves when they wore the vestments in which they celebrated Mass. Acts of patronage enabled some lay men and women to establish a material presence in the chancel, but bishops and priests were also patrons. Notwithstanding the observations made by Duby on the growing ability of the laity to produce powerful representations of itself during the 14th century, certain members of the church were accruing their own powers, especially in the consolidation of episcopal power that was occurring throughout Europe (Rubin 1991: 13-14, 51). In later medieval Scotland, bishops and priests commissioned representations of themselves. The bishops were powerful lords who exerted great influence over temporal and religious matters, as will be made clear below. It is of interest that entry into the priesthood offered a career route for men that did not depend on birthright alone. A case in point is Bishop William Elphinstone, who received papal dispensation to absolve him from the sin of illegitimacy. He entered the priesthood eventually to become Bishop of Aberdeen and, in 1494, the founder of the University of Aberdeen (Macfarlane 1985:78).

spinning, weaving and embroidering as depicted by Cosimo Tura and Francesco Cossa in The Triumph of Minerva frescoes, c. 1470 (Staniland 1991: Figs 24 and 54). Priests wore splendid vestments, the products of the largely anonymous craft workers, as a means of legitimising their collective power in a spiritual realm and they converted it into religious authority. In the eyes of the church, priests did not wear vestments to express individual power. A late 20th-century view expressed by Father Jerome Bertram was that vestments are made ‘to humiliate’ the priest. He explained: ‘vested in the raiment of the church he ceases to be himself, he “puts on Christ”, speaking not in his own name but in that of the church’ (Bertram 2002: vii). The 13th-century liturgist Bishop Durandus of Mende stipulated that the deceased were to be wrapped in a shroud prior to burial, excepting priests who were permitted to be buried in sacerdotal vestments (Ariès 1983: 246), implying that their collective identity was retained in death. Archaeological evidence for priests being buried vested for Mass is available for late medieval Scotland. A fragment of a vestment executed in couched work and split stitch was detected in the tomb of Gavin Dunbar, Bishop of Glasgow, who died in 1547 (Henshall et al. 1956: 35) and silk textiles, one patterned in a fleur de lys design of a type known to have been used in chasubles, were found in an early or mid 16th-century bishop’s tomb at Fortrose (Henshall et al. 1956: 26-27). By choosing burial in a chasuble, the garment in which the celebrant consecrated the bread and wine during the Eucharist, rather than a cope, which was used for processions, or the rochet, a linen tunic which clerics wore in daily life, priests were emphasising their exclusive theological role which sanctioned them to utter the words which converted bread and wine into the body and blood of Christ.

In the next section I will consider the textile structure and the stitching of the embroideries used in medieval Mass vestments. The presentation of surviving evidence for vestments in Scotland is followed by an examination of representations of a ritual gesture enacted by the priest in front of the altar: the elevation of the host. The church specified that certain vestments be worn and it regulated the nature of the bread and wine to be used in the Eucharist (Rubin 1991: 83-98; Dilworth 1996: 128). I then explore some examples of prelates who exerted both religious and secular authority, and who, in death, chose to represent themselves as priests robed for celebrating Mass. One was bishop of the Scottish diocese of Moray, but I also present some examples from France to suggest that Scottish prelates, like their counterparts in other western countries, interacted in a homosocial sphere that involved them in a spiritual, intellectual and political network of churchmen.

In a study of vestments, Pauline Johnstone (2002: 5) quoted from a 15th-century book known as The Imitation of Christ, attributed to Thomas à Kempis, an explanation concerning the placement of orphreys in the shape of a cross on priests’ chasubles:

Eucharistic vestments

He [the priest] wears the sign of the cross both before and behind him, that he may be ever mindful of the Lord’s Passion. He wears the Cross before him on his chasuble that he may diligently observe the footsteps of Christ, and earnestly study to follow them. His shoulders also are signed with the Cross, that he may in his mercy and for the love of God hear every injury done him by others … (Kempis 1952: 195).

This section reviews some of the surviving evidence for the mass vestments from direct and indirect sources. In the medieval church of western Europe, the labour of embroiderers and making of vestments clothed a corporate group of clerics. The church claimed the products of that labour in a gender hierarchy, which achieved an elaborate expression in the form of a male priesthood. The vestments the priests wore were the product of professional male and female embroiderers, who underwent a long apprenticeship in the medieval guild system. These embroiderers are often anonymous, and images of them are rare. In a book on medieval embroidery, Kay Staniland presented an illustration of male embroiderers taken from a north Italian manuscript dated c. 1400, depicting a scene from the Book of Exodus showing Bezalehel and Ohab making fabrics for the Tabernacle, and another illustration of young women

The Imitation of Christ was a widely circulated manuscript of which many copies have survived in Latin and in Middle English. It was read in Scotland (Donaldson 1990: 40). However, the passage cited by Johnstone does not appear in many of the copies that were circulated and it appears to be an insertion. It is not clear how the laity would have received the visual message of the priests’ humility, given

98

Penelope Dransart: Representing Hierarchy and Homosociality

their participation in the factional family-based politics of late medieval Scotland.

of treatment in the bishops’ hair, which is closely cropped and controlled beneath the mitre. Both these treatments of men’s headdress express a hegemonic form of masculine powers and denote the membership of different homosocial groups. Despite the visual distinctions, clerics could belong to two of the three estates, those of the prelates and barons. The bishops in the Lewis chess sets were designed to emphasise their priestly role in celebrating the Eucharist.

Vested for celebrating Mass, a priest wore an under-tunic called an alb, a maniple on his left wrist, and a stole, which passed round the neck and crossed on the chest. A girdle held the stole in place, the ends of which hung as far as the hem of the alb. Over the alb, the priest wore a chasuble made from rich cloth. It was often adorned with embroidered bands called orphreys. The term orphrey is derived from the Latin auriphrygium, which indicated gold embroidery (Hogarth 1987: 8). Bishops wore the same vestments as a priest for Mass, but with the addition of a tunic known as a dalmatic, between the alb and chasuble, and a mitre. A fuller account of priests’ and bishops’ vestments and accoutrements is presented in Macalister (1896: 65-136).

However, the Lewis bishop pieces have narrow orphreys and they demonstrate a restraint in ornamentation which is characteristic of the 12th and 13th centuries. Through time, vestments were used as a field for presenting elaborate visual and tactile imagery. In the 13th century, English needlework termed Opus anglicanum became famous throughout Europe in vestments, and references to it occur in papal inventories (Thurman 1992: 42). Different types of couching were used in the Middle Ages, including underside couching, which was characteristic of Opus anglicanum work and which persisted until the beginning of the 15th century (King 1963: 9). Surface couching survived for a longer period of time (Figs 9.1and 9.2). Using these techniques, embroiderers created scenes including those from the life of Christ and of his mother, Mary, and they depicted the apostles and saints. They made series of embroideries for sets of vestments to be worn by bishops, priests and deacons for special festivals in the ritual cycle. These embroideries were evidently valued and they were remounted on new vestments, perhaps more than once in their ‘biographies’. R.N. Swanson (1992: 248) stressed that vestments, along with other items such as books and church plate, were in a constant state of repair and renewal.

Early in the Middle Ages chasubles were large and bellshaped, being made from cloth cut on a circle. Indicative of such vestments are the bishops, robed for Mass, among the Scandinavian chess pieces found in 1831 in a sandbank on the coast of the Isle of Lewis in the Outer Hebrides. Ninetythree pieces in total have survived and among the bishops some are enthroned, while others are standing (Taylor 1978: 7). Some of them hold a crozier, the head of which forms a simple volute, which rests against one or other cheek of the holder. Other bishops hold a book in one hand, while a few raise the right hand in blessing. All of them wear a mitre (Madden 1832: 212). Taylor (1978: 14-15) dated the chess pieces to a period between 1150 and 1170, on account of the mitres, which are worn with the points fore and aft, rather than from side-to-side as was the practice prior to c. 1150. The bishops’ chasubles are embellished with narrow orphreys.

The chasuble became less voluminous in size, but it still provided a suitable field for the display of holy images. Embroidery remained one of the main techniques for making dramatic-looking vestments appropriate for the celebration of Mass.

The detail contained in the Lewis chess pieces is instructive. The carvers conveyed the idea of different ranks of men within society, but not of women. Amongst the men there are kings, bishops, knights and foot soldiers. In contrast, queens alone represent women. By the 15th century, men in Scottish society gained distinction through the recognition of their membership of what was termed in Scotland and elsewhere (France, for example) as the three estates: prelates, barons and burgesses. In this threefold scheme, which has been discussed by Le Goff (1988: 259-60), the church or clerical class, the military class and the ‘third estate’ comprising the ‘upper part’ of the productive class achieved iconographic representation. The men portrayed in the Lewis chess pieces could be assigned to one of these three estates. Women and serfs could not easily aspire to such recognition and representation.

Such was the need for vestments, Scottish embroiders would have produced local versions, obtaining many of their materials from the Low Countries. Named embroiderers who are known to have worked for the court and for the church during the reigns of James IV and James V included men and women, some of whom were native and some foreign, including men from the Low Countries (McRoberts 1956: 85). The embroidered vestment mentioned above from the tomb of Gavin Dunbar, Bishop of Glasgow, has couched threads with a silk core, probably wrapped with silver and gilded silver. Various patterns may be produced in couching by making the couching stitches conform to different patterns. Although fragmentary, the Dunbar embroidery displays several permutations, including a diagonal arrangement (Henshall et al. 1956: Fig. 5c). This diagonal pattern resembles the diagonally aligned floats of the supplementary-warp weave in a ribbon found in midden deposits at Dairsie, one of the palaces of the archbishops

A form of homosocial coding may be detected in differences in the headdresses of the Lewis chess pieces. The queens have no hair visible; it is carefully hidden under a veil. The kings, on the other hand, have long tresses luxuriantly arranged, alternately twisted clockwise and anticlockwise, falling from below the royal crown. There is a stark contrast

99

Dress and Identity

of St Andrews. The supplementary-warp weave of the ribbon produces an effect that mimics the more widely used technique of couching in embroidered fabrics (Dransart 1997). Although it is only 16 mm long, the Dairsie fragment is important because it suggests that vestments were being made or remade at an archiepiscopal residence of St Andrews. Such ribbons typically served as a border to cover the edges of orphreys cut from old vestments when they were remounted on new fabrics.

enclosed her in a restricted place, have been drawn back to allow her a direct view of the elevation. As the priest raises the host, the body of Christ, the kneeling queen raises her hands in a similar gesture. Jessica Brantley (2002: 103-4) has discussed this visual parallelism and has related it to the linguistic modes in the text, which alternate between vernacular Anglo-Norman and Latin. Stanbury and Raguin have explored the location of women in the spaces constituted by the parish and other churches or chapels. They argue that medieval church practices afforded wealthy women the opportunity of ‘peeking into privileged spaces’, and that such strategic affordances were related to women’s status as donors (Stanbury and Raguin 2005: 8). In this case, it is an on-looking queen who is in a restricted space, the curtains of which have been opened to allow her to observe the transubstantiated body of Christ.

An analysis of metal-wrapped threads from a late 15th century woven fragment excavated from the Archbishop’s Palace in Trondheim, Norway, was found to have gold wrapped threads by energy dispersive X-ray analysis (Peacock 1994: 258 with Fig. 11). In contrast, the metal wrapped round the threads of the Dairsie ribbon is perhaps more likely to be silver, due to its tarnished lavender-grey appearance, although gilded silver is a possibility (Peacock 1994: 254, 258).

While the chasubles worn by the bishops in the Lewis chess pieces have the appearance of being light in weight, later in the Middle Ages chasubles had become rather more weighty, due in part to the heavy couched threads, and perhaps also to successive repairs and re-mountings of the embroidered panels. In elevating the host, the extra width of the chasuble meant that its sides had to be doubled back over the priest’s shoulders, and an acolyte held the bottom edge of the chasuble to counterbalance the weight of the vestment (Reinburg 1992: 533). Many visual examples demonstrate the acolyte holding the hem of the chasuble at the elevation of the host. They include the central panel of the ‘Retable of the Seven Sacraments’ by Rogier van der Weyden c. 1445 (Duby 1966: 101); the painting of the ‘Mass of Saint Gregory’ by the Master of the Holy Kindred, c. 1486 (Museum Catharijneconvent, Utrecht; Johnstone 2002: Fig. 3); the painting of the ‘Mass of Saint Giles’ by the Master of St Giles, c. 1490 (National Gallery, London; Johnstone 2002: Plate IIA); an illuminated capital letter in a Roman missal of the second half of the 14th century (BN lat 848, fol. 194r; Rubin 1991:40) and in an illumination in the Office of the Dead in MS W. 274, fol. 118, The Walters Art Gallery, Baltimore, by the Coëtivy Master c. 1460 (illustrated on the dust jacket of Binski 1996). The image was also disseminated in the form of woodcuts. Examples are found in Mass and Altar by Hans Baldung Grien from Marcus von Lindau’s Die Zehe gebot dated 1516 (Hayum 1989: 56); The Mass of St Gregory, a woodcut from Paris c. 1490 and Elevation of the Host, f. 133v in the Missale Romanum, published in Lyon in 1512 (Reinburg 1992: 534 and 538).

The threads of the Trondheim piece interlace in a ‘basket’ weave in which the warp and weft elements are tripled (Emery 1966: 77). Because the fragment only measures 10 mm by 15 mm, various specialists consulted by Peacock suggested that it might have been needle-woven in the form of a button (Peacock 1994: 259). It might well have corresponded with an item that was listed in an inventory of the treasury of Glasgow Cathedral (1432), which included ‘a stole and maniple (without other vestures), having knops of the purest gold’ (Dowden 1898-99: 310). The elevation of the host The moment during the celebration of the Eucharist when the priest holds up the transubstantiated body and blood of Christ was a theme that artists often depicted in western European paintings, illuminations and prints. At this moment, the priest faced the altar and he held up the host above his head for the people behind him to see. This practice has been traced back to its introduction in Paris between 1196 and 1208, following which Pope Honorius III approved it in 1219. It is thought that the custom was observed in Aberdeen by about 1227, from a documentary source contained in the Aberdeen Statutes, which have survived in a 15th-century version containing earlier material (McRoberts 1957: 29). Therefore the elevation of the host was, it seems, an early introduction in Scotland, but there is little surviving iconography of it. An Anglo-Norman book of hours, the Taymouth Hours contains a depiction of the elevation of the host. A royal woman is depicted on four occasions in its pages. Dated to c. 1325-40, John Harthan (1977: 48) has suggested that the book was made for Queen Joan, wife of David II of Scotland and daughter of Edward II of England, a suggestion accepted by Caldwell (1990: 47). The elevation appears on f. 7r, in an illumination in which a priest raises the consecrated host in front of the altar while the kneeling queen also raises her hands (Fig. 9.3). The curtains, which

In the London National Gallery Mass of St Giles, the artist depicted the priest’s actions as taking place behind a curtained space; the curtain is pulled back partially to allow the patron, the king, to observe the moment during which the bread becomes the body of Christ. His retainers stand behind the curtained altar and they cannot observe the event. Andrée Hayum (1989: 81) commented that the unveiling served ‘to dramatize the sacred moment of transubstantiation’. The restrictions set up by curtains

100

Penelope Dransart: Representing Hierarchy and Homosociality

Fig. 9.3: Taymouth Hours Yates Thompson*, f. 7r. British Library.

101

Dress and Identity

round the altar and the rood screen between the chancel and nave in late medieval churches meant that the king had a privileged view of the events at the altar.

as ‘books for the unlearned’. Many of the vestments mentioned by Swanson are copes, which were used in processions, rather than chasubles. In his study of the medieval inventories of Lichfield Cathedral, he noted the problems occasioned by the frequent wearing of copes, and the damage caused by dripping candle wax at the Purification (Swanson 1992: 251). It is possible that many parishioners in medieval Scotland were more familiar with copes than chasubles, given the restrictions on viewing inside the chancel.

Miri Rubin (1991: 55-63) has examined the history of the ‘emphatic gesture’ that is involved in elevating the host. She noted that the gesture was imaged in such a manner as to emphasise the exclusive character of priestly power. In her analysis, the ‘elevation of the host was perceived as the essence of clerical office, the focus of the liturgy, the epitome and justification of clerical privilege’ (Rubin 1991: 132). It is pertinent to ask who had access to the chancel as not everyone had access to such a privileged space in the church. This issue has been raised by contributors to the book edited by Raguin and Stanbury (2005). The laity, especially the women, to whom the sacrament of Ordination was denied, only had a partial visual engagement with this important moment during the Mass.

Despite the growing powers of influence exerted by members of the craft guild in the larger burghs, the insistence of the priests to retain their privileged position provoked responses on the part of lay people. Rubin argued that the sacrament of Holy Communion did not have a unitary significance. She observed ‘no single category such as class and gender can adequately capture the variety of Eucharistic meaning’ (Rubin 1991: 288). As a consequence of priests teaching the laity of the benefits to be derived from observing the Eucharist and providing representations for people to contemplate the consecrated host in Corpus Christi processions and in works of art, Rubin (1991: 289) suggested that there was a corresponding belief that they had a right to see it. At times the people insisted upon this right. To judge from the chalice veils that have survived from medieval Scotland, one of which bears the stylised embroidered image of a Eucharistic shrine, people made the ‘same demand for exposition of the Blessed Sacrament that prevailed in the Low Countries and in the Holy Roman Empire’ (McRoberts 1964: 110).

The small church of Foulis Easter, near Dundee, rebuilt in the mid 15th century, is an example of a building consisting of a chancel and a nave with no side aisles, a characteristic form of parish church in medieval Scotland. It is unusual in that it retains evidence for its rood screen. This screen formed an entrance passage between the nave and the chancel; its carved wooden doors have survived. Two small sacristies flanked this passage and there are also the remains of a newel staircase, which would have permitted access to a rood loft. In this arrangement, the high altar at the east end of the chancel would have been obscured from the view of the congregants in the nave, but Monsignor McRoberts (1983: 384) postulated the existence of subsidiary altars, one set on each side of the passageway to the chancel. Windows in the north and south walls of the church would have lit these nave altars, and an upper window on the south side would have illuminated the rood loft (Fawcett 1994: 155).

Religious authority was embedded in kinship and politics in medieval Scotland. In the early 14th century, some of its bishops had been in the forefront of defending Scottish interests in the sphere of the church and to uphold the divinely appointed kings. The Scottish church was situated north of the territorially expansive aspirations of the metropolitan sees of York in England. In the west, parts of Scotland were under the control of the See of Trondheim in Norway. The diocese of the Isles was free of control from Trondheim, and that of Galloway from York, by c. 1350 (Barrow 1989: 69-70), but the Scottish church retained its unusual situation in which its bishops responded directly to the pope until 1472, when St Andrews became an archdiocese and the first see in Scotland to receive metropolitan status (Watt 2000: 28-30). I now comment on a prelate in the cathedral of the diocese of Moray as someone who exercised his considerable authority in secular and religious arenas, and who represented his effigy as a priest vested for Mass.

Large parish churches had many altars endowed by burgesses and lairds. St Giles, the parish church of the burgh of Edinburgh, is a case in point. In the 15th and 16th centuries, it became the custom for a newly incorporated craft guild to endow an altar and a chaplain in St Giles. The guilds acquired vestments and other items to enable Mass to be celebrated at their own altar. Lynch (1981: 28) reported that St Giles had more than forty such altars by 1560, the year in which the first reformed parliament met. According to Lynch (1981: 28-9), there was an increasing tendency for the craft guilds to exercise a considerable degree of control over their chaplain, particularly after 1466, the year in which St Giles was granted collegiate status.

Elgin Cathedral

In St Giles, it was likely that parishioners would have been able to see Mass vestments more frequently than congregants in the church of Foulis Easter, with its more typical axial architectural arrangement. Swanson (1992: 248) argued that vestments differentiated those persons who enacted the theologically prescribed liturgy from the congregants and that the embroidered imagery served

In this and the following section, my attention turns to the chancel or presbytery, which constituted an important focus of patronage for bishops and priests. In Elgin Cathedral St Mary’s Aisle dates from the 13th century. It is adjacent to the presbytery, which contained the high altar and the

102

Penelope Dransart: Representing Hierarchy and Homosociality

Fig. 9.4: John of Winchester, Bishop of Moray (died 1460), Elgin Cathedral (photograph: P. Dransart).

chancel. In the easternmost bay of the aisle, backing onto the presbytery, Bishop Winchester (1435-1460) constructed a mural tomb for his remains (Fig. 9.4). This location, close to the high altar, was a privileged one. The aisle houses the monuments of other prelates and lords (Brydall 1895: 393396; Richardson 1950: 13-14). John of Winchester’s effigy lies in a recess, underneath an ogival arch. His monument was originally painted; drawings depicting angels in the act of censing have survived in black and red lines underneath the recess (Richardson 1950: 13). His coat of arms would have been painted on the now-blank shields designed for the purpose. Dressed in Eucharistic vestments, Bishop Winchester’s crozier is aligned with the left side of his body, with the crook close to his face in the tradition of the Lewis chess-pieces.

bishop’s palace; it is likely that Winchester initiated a major rebuilding project of the palace, which was completed after his death (Lewis and Pringle 2002: 173). H.B. Mackintosh explained that erecting Spynie into a Barony with full regality rights invested the Bishop with the rights that the King had over his territory, which was a ‘formidable concession’. He commented ‘the temporal influence of the Bishops of Moray was thus growing more luxuriantly than their spiritual’ (Mackintosh 1950: 23). In death, the effigy of Bishop Winchester is robed for Mass, but his tomb also made clear his earthly kinship through the display of his heraldry. A prelate and baron, Winchester was a member of two of the three estates. This enabled him to exert a hegemonic masculinity in more than one sphere. The plan that accompanies the description of Elgin Cathedral by Richardson (1950) indicates the position of the location of the rood screen. As discussed above, this screen formed a partition between the nave and the chancel, where the canons of the cathedral sat. It is likely that lay people in the nave of Elgin Cathedral would have had limited possibilities for seeing the ritual gestures that took place in the chancel during the High Mass, as in other European countries.

During his lifetime, King James I (1394-1437) gave Winchester several roles, appointing him Prebendary of Dunkeld, Provost of Lincluden and Lord Clerk-Register before his consecration as Bishop of Moray in 1435 (Mackintosh 1950: 23; Lewis and Pringle 2002: 3). In addition to his clerical duties, he also was Master of Works to the King. During his bishopric, church lands in Moray were designated the Barony of Spynie. Spynie was the site of a former cathedral of Moray and it has an imposing

103

Dress and Identity

In some churches and monastic cathedrals there were apparently bans on the presence of women, even those of high status, including St Margaret of Scotland, Queen of Malcolm Canmore, of whom it was reported that a supernatural force repulsed her when she attempted to enter the church of St Lawrence (Schulenburg 2005: 189). The monastic cathedral of Durham provided one of the most notorious examples for restricting women’s access. Victoria Tudor (1984) argued that the misogyny attributed to St Cuthbert, who was buried in the cathedral, was posthumous. She dated it as arising after 1083 as part of a strategy adopted by the newly arrived Benedictine monks to establish a celibate priesthood by discrediting their married predecessors. In addition, she argued that efforts to ban women from the main part of the cathedral were renewed when a hagiography of St Cuthbert known as the Irish Life, which included references to the Picts in Scotland as having banned all women from entering churches dedicated to the saint, arrived at Durham at some point in the 13th century (Tudor 1984: 163). In 1417, two women were punished for dressing in male garments and entering the cathedral, by which time the alleged misogyny ‘had become an institutionalised part of the conventual routine’ (Tudor 1984: 164). The women’s act of wearing male dress seems to have been particularly transgressive.

priest, and the lower one presents the bodily remains of a departed soul. Paul Binski (1996: 142-3) observes that the tableau that Lagrange projected was a ‘multiple selfrepresentation’ that ‘multiplied focuses of devotion and interest’. Lagrange’s body was buried at Amiens, his entrails in the Avignon tomb.

Representations of priestly power in France

The most striking aspect of this self-representation is that men no longer are created in the image of God, instead God is created in the image of a bishop. This figure has a regal crown, and he wears a processional cope and a stole crossed over his chest in the manner of an ordained priest to receive the soul of St Stephen.

My last example is taken from a series of tapestries, La légende de saint Etienne, commissioned by Jean Baillet, Bishop of Auxerre (1477 – 1513) for the cathedral church of Saint-Etienne d’Auxerre. Sophie Lagabrielle commented on the self-absorption of clerical homosociality that is evoked through the exclusive character of the vestments represented in the hangings: In offering this hanging to members of the chapter, Jean Baillet wishes to honour the patron of the cathedral [St Stephen], as well as his servants, the canons, whom he introduces in the image. He takes care to reserve for the figure of the bishop the role of direct and efficacious defender of the sacred memory but proposes, besides, the vision of a clerical society providing for and tied back on itself, unconscious of contemporary changes taking place (Lagabrielle 1997: 55, my translation, emphasis in the original).

My final section cites two examples of representations of vested French bishops to demonstrate that the insistence on clerical privilege was not only a Scottish prejudice in late medieval Europe. Cardinal Jean de Lagrange, Bishop of Amiens, commissioned a tomb for himself in 1388-1389 for the choir of the collegiate Benedictine church of SaintMartial d’Avignon, at a time when the papacy was based at Avignon. The monument was over 15m high; it was badly damaged during the French Revolution but a large number of fragments and a drawing have survived (Mognetti & de Loye 1988: 9). Its iconographic programme included King Charles V and his sons Charles VI and Louis d’Orléans. Lagrange was Chancellor and Minister of Finances to Charles V as well as being tutor to the royal children. On the death of Charles V he joined the papal court of Pope Clément VII in Avignon, whose image was also incorporated in Lagrange’s monument. This iconography conveys Lagrange’s loyalty to both the French crown and the Avignon papacy. It seems that the monument constituted a grand political statement, but political events were to overtake the scheme; from 1395 Charles VI negotiated with Rome to end the schism between Rome and Avignon. The monument was finished in a hurry by a second, less skilled sculptor, and Lagrange died in 1402 (Mognetti & de Loye 1988: 9).

In Scotland, despite their vested interests, the prelates were unable to check the movement for reform. On the Twelfth Night of the year 1540, in Linlithgow Palace, King James V his court watched a performance of Sir David Lindsay’s Satire of the Three Estates. A character called Poor Man challenged Spirituality (the prelates) with these words: Our bishops, with their lusty rockets [linen / lace tunics] white, They flow in riches royally and delight; Like paradise been their palaces and places, And wants na pleasure from the fairest faces! But doubt I wald think it a pleasant life, Ay on, when I list, to part with my wife, Syne tak another of far greater beauty! But ever alas, my lords, that may not be, For I am bund, alas, in marriage, But they like ramis rudely in their rage Unpizzlet [rampant] rins amang the silly ewes, Sa lang as kind of nature in them grows (Lindsay 1967: 137-139).

The tomb incorporates three representations of the Bishop of Amiens. In one version he is robed as a priest wearing lavishly embroidered vestments (Fig. 9.5). Beneath this effigy another image depicted him as a mortal corpse in the corruption of death. In the higher image he is a sleeping

Lindsay’s unflattering characterisation of the bishops follows a markedly anticlerical vein in late medieval satire. However, it is worth noting that this allegation of their lascivious actions does not present them in the attire

104

Penelope Dransart: Representing Hierarchy and Homosociality

Fig. 9.5: Effigy of Cardinal Jean de Lagrange, Bishop of Amiens, commissioned 1388-1389. Musée du Petit Palais, Avignon (photograph: P. Dransart).

of Mass vestments. Instead, their fine linen tunics, the rochet or rocket, attracted the poet’s attention. An example of a finely pleated rochet is depicted on the mid 16thcentury effigy of Sir Donald MacDuffie, Conventual Prior of the Augustinian priory on the island of Oronsay (No 386(23) in RCAHMS 1984: 250). Carol Edington (1995: 153) argued that Lindsay used the satire to highlight the moral corruption that was occurring in a society where the leading churchmen were charged with the spiritual care of the people as well as serving as statesmen with legal and other official roles. Lindsay wrote the satire when events were leading inexorably to the meeting of Scotland’s first reformed parliament in 1560 (Dilworth 1996: 130). At the same time, the Catholic Church was holding a series of sessions known as the Council of Trent (1545-1563), which resulted in Catholic reforms including the removal of the rood screen from churches to enable the laity to see the gestures at the High Altar discussed above. However, the rood screen in Elgin Cathedral was destroyed much later, in 1640 (Fawcett 1994: 71).

used. The main focus was provided by the chasuble as a site of embroidery which enhanced an important gesture, the elevation of the host in front of the altar. This gesture requires to be understood in terms of religious practices which involved the participation of different groups of people with a homosocial character. Profound social changes occurring especially at the end of the period under consideration in the 16th century did not challenge the role of priests to administer the Eucharist. Neither Protestant nor Catholic reforms chipped the rock on which male homosocial spheres of dominance were founded, and there was as yet no representational basis for what eventually was to emerge in the form of current controversies concerning women’s ordination to the priesthood. Bibliography Ariès, P. (1983): The Hour of our Death. Harmondsworth. Barrow, G.W.S. (1989): Kingship and Unity: Scotland 1000-1306. Edinburgh. Bertram, J. (2002): ‘Foreword’, in P. Johnstone, High Fashion in the Church: the Place of Church Vestments in the History of Art from the Ninth to the Nineteenth Century, vii. Leeds.

In this article, I have attempted to examine medieval vestments in a holistic manner, comprising not only the physical space but also the social space in which they were

105

Dress and Identity

Binski, P. (1996): Medieval Death: Ritual and Representation. London. Bossy, J. (1983): ‘The mass as a social institution 12001700’, Past and Present 100: 29-61. Brantley, J. (2002): ‘Images of the vernacular in the Taymouth Book of Hours’, in A.S.G. Edwards (ed.), English Manuscript Studies 1100-1700: Vol 10 Decoration and Illustration in Medieval English Manuscripts, 83-113. London. Brydall, R. (1994-5): ‘The monumental effigies of Scotland, from the thirteenth to the fifteenth century’, Proceedings of the Society of Antiquaries of Scotland XXIX: 329-410. Caldwell, D.H. (1990): ‘In search of Scottish art: native traditions and foreign influences’, in W. Kaplan (ed.), Scotland Creates: 5000 Years of Art and Design, 45-60. London and Glasgow. Dennison, E.P., DesBrisay, G. and Diack, H.L. (2002): ‘Health in the two towns’, in E.P. Dennison, D. Ditchburn and M. Lynch (eds), Aberdeen before 1800: a New History, 70-96. Phantassie, Scotland. Dilworth, M. (1996): ‘Roman Catholic worship’, in D. Forrester and D. Murray (eds), Studies in the History of Worship in Scotland, 127-48. Edinburgh. Donaldson, G. (1990): The Faith of the Scots. London. Dowden, J. (1898-9): ‘The inventory of ornaments, jewels, relicks, vestments, service-books, etc., belonging to the Cathedral Church of Glasgow in 1432, illustrated from various sources, and more particularly from the inventories of the Cathedral of Aberdeen’, Proceedings of the Society of Antiquaries of Scotland 33: 280-329. Dransart, P. (1997): Silk ribbon with metal-wrapped threads from Dairsie. Unpublished manuscript. Dransart, P. and Bogdan, N. (2004): ‘The material culture of recusancy at Fetternear: kin and religion in postReformation Scotland’, Proceedings of the Society of Antiquaries of Scotland 134: 457-70. Duby, G. (1966): Fondements d’un nouvel humanisme 1280-1440. Geneva. Duffy, E. (1992): The Stripping of the Altars: Traditional Religion in England c.1400-c.1580. New Haven and London. Edington, C. (1995): Court and Culture in Renaissance Scotland: Sir David Lindsay of the Mount (1486-1555). Phantassie. Emery, I. (1966): The Primary Structure of Fabrics. Washington D.C. Fawcett, R. (1994): The Architectural History of Scotland: Scottish Architecture from the Accession of the Stewarts to the Reformation 1371-1560. Edinburgh. Fernie, E.C. (2001): ‘Technical terms and the understanding of English medieval architecture’, Essays in Architectural History Presented to John Newman. Architectural History 44: 13-21. Hayum, A. (1989): The Isenheim Altarpiece: God’s Medicine and the Painter’s Vision. Princeton. Henshall, A., Crowfoot, G.M., and Beckwith, J. (1956): ‘Early textiles found in Scotland: Part II – Medieval imports’, Proceedings of the Society of Antiquaries of Scotland LXXXIX: 22-39.

Hogarth, S. (1987): ‘Ecclesiastical vestments at York Minster before the Reformation’, in E. Ingram (ed.), Thread of Gold: the Embroideries and Textiles in York Minster, 8-18. Andover, Hants. Johnstone, P. (2002): High Fashion in the Church: the Place of Church Vestments in the History of Art from the Ninth to the Nineteenth Century. Leeds. Kempis, T. à (1952): Imitation of Christ, tran L. SherleyPrice. London. King, D. (1963): Opus anglicanum. English Medieval Embroidery. London. Lagabrielle, S. (1997): ‘Etienne, Gamaliel et les autres’, in S. Lagabrielle (ed.), La légende de saint Etienne et Brocarts célestes, 49-57. Avignon. Le Goff, J. (1988): Medieval Civilization 400-1500, trans. J. Barrow. Oxford and Cambridge, Mass. Le Goff, J. (1992): The Medieval Imagination, trans. A. Goldhammer. Chicago. Lewis, J. and Pringle, D. (2002): Spynie Palace and the Bishops of Moray: History, Architecture and Archaeology. Edinburgh. Lindsay, D. (1967): A Satire of the Three Estates. London. Lynch, M. (1981): Edinburgh and the Reformation. Edinburgh. Macalister, R.A.S. (1896): Ecclesiastical Vestments: their Development and History. London. Macfarlane, L.J. (1985): William Elphinstone and the Kingdom of Scotland 1431-1514: The Struggle for Order. Aberdeen. Mackintosh, H.B. (1950): ‘History’, in J.S. Richardson and H.B. Mackintosh Elgin Cathedral: The Cathedral Kirk of Moray, 18-26. Edinburgh. McRoberts, D. (1956): ‘The Fetternear Banner’, Innes Review 7: 68-86, Plates 1-11. McRoberts, D. (1964): ‘Scottish medieval chalice veils’, Innes Review15: 103-16. McRoberts, D. (1983): ‘The fifteenth-century altarpiece of Fowlis Easter church’, in A. O’Connor and D.V. Clarke (eds), From the Stone Age to the ‘Forty-Five: Studies Presented to R.B.K. Stevenson, Former Keeper National Museum of Antiquities of Scotland, 384-98. Edinburgh. Madden, F. (1832): ‘Historical remarks on the introduction of chess into Europe, and on the ancient chess-men discovered in the Isle of Lewis’, Archaeologia XXIV, 203-91. Mognetti, E. and de Loye, G. (1988): Petit guide du Musée du Petit Palais. Palais des Archevêques. Avignon. Monnas, L. (1994): ‘Opus anglicanum and renaissance velvet: the Whalley Abbey vestments’, Textile History 25, 3-27. Peacock, E. (1994): ‘SEM-EDS analysis of metal threads from Trondheim’, in Archäologische Textilfunde Textilsymposium Neümunster 4-7.5.1993 (NESAT V), 253-60. Neümunster. Pleij, H. (2004): Colors demonic and divine: shades of meaning in the middle ages and after, trans. D. Webb. New York. Raguin, V.C. and Stanbury, S. (eds) (2005): Women’s Space:

106

Penelope Dransart: Representing Hierarchy and Homosociality

Patronage, Place, and Gender in the Medieval Church. New York RCAHMS (1984): Argyll An Inventory of the Monuments: Volume 5 Islay, Jura, Colonsay and Oronsay. Edinburgh. Reinburg, V. (1992): Liturgy and the laity in late medieval and Reformation France, Sixteenth Century Journal 23(3): 526-47. Richardson, J.S. (1950): ‘Description’, in J.S. Richardson and H.B. Mackintosh, Elgin Cathedral: The Cathedral Kirk of Moray, 1-17. Edinburgh. Rubin, M. (1991): Corpus Christi: the Eucharist in Late Medieval Culture. Cambridge. Schleif, C. (2005): ‘Men on the right – women on the left: (a)symmetrical spaces and gendered places’, in V.C. Raguin and S. Stanbury (eds), Women’s Space: Patronage, Place, and Gender in the Medieval Church, 207-49. New York. Schulenburg, J.T. (2005): ‘Gender, celibacy, and proscriptions of sacred space: symbol and practice, in V.C. Raguin and S. Stanbury (eds), Women’s Space: Patronage, Place, and Gender in the Medieval Church,185-205. New York.

Sørensen, M.L.S. (2000): Gender Archaeology. Cambridge. Stanbury, S. and Raguin, V.C. (2005): ‘Introduction’, in V.C. Raguin and S. Stanbury (eds), Women’s Space: Patronage, Place, and Gender in the Medieval Church, 1-21. New York. Staniland, K. (1991): Medieval Craftsmen: Embroiderers. London. Stratton, S.L. (1994): The Immaculate Conception in Spanish Art. Cambridge. Swanson, R.N. (1992): ‘Medieval liturgy as theatre: the props’, in D. Wood (ed.), The Church and the Arts. Papers Read at the 1990 Summer Meeting and the 1991 Winter Meeting of the Ecclesiastical History Society, 239-53. Oxford. Taylor, M. (1978): The Lewis Chessmen. London. Thurman, C.C.M. (1992): Textiles in the Art Institute of Chicago. Chicago. Tudor, V. (1984): The misogyny of St Cuthbert, Archaeologia Aeliana 12 (Fifth Series): 157-67. Watt, D.E.R. (2000): Medieval Church Councils in Scotland. Edinburgh.

107

Chapter 10

Cosmetics and Perfumes in the Roman World: A Glossary Susan Stewart

Beauty products used in Roman times are only just beginning to receive the attention that has been accorded to clothing from this period in recent years. Yet, when we start to look for references to cosmetics and perfumes in contemporary literature and art we find many. Writers and artists from the Roman period were well aware of how familiar their listeners and viewers were with these products in their day-to-day lives, and how much, metaphorically speaking, the mention of cosmetics and perfumes could impart to their audience. By according beauty products the same importance allotted to them in Roman culture, we can attempt to understand the messages that a contemporary audience might have read into these references while at the same time allowing ourselves the possibility of a glimpse of what it was really like to live in the Roman world. This paper is based around the glossary that formed part of my PhD thesis.1 Here I have prefaced it with a few statements that, I hope, will show how the glossary as a whole, notwithstanding the significance of the individual entries, has some worth in terms of research into the rhetoric and the reality, relating to matters of gender, social status and identity.

principles used in these different genres in their treatment of this subject while also allowing him or her to offset the conventions of different genres, to some extent, and perhaps grasp some feeling for contemporary reality. Rhetorically speaking, cosmetics and perfumes were perceived as something used by prostitutes, those of low social standing or loose morals. This is very much the way make-up was seen in satirical verse, for example, as well as in elegiac love poetry where the female subjects would often appear to represent high-class call girls. The popularity of some of these texts has bolstered the belief that the use of cosmetics and perfumes was, in reality, confined to those involved in the sex trade (see Harlow and Swift this volume). The collected references in the glossary come from a variety of sources and often describe women in general, an indication that the balance needs to be redressed. Pliny the Elder, in particular, mentions beauty products and, more often than not, while he refers to women (as opposed to men) using them, he does not define these women as members of any particular social grouping.3 In fact, women across the social spectrum used cosmetics and perfumes. Ironically, in the case of elegiac verse and satire, the prevalence of beauty products per se may be the element of reality in a fantasy world.

In the first place, while the glossary alludes to the nonliterary material that refers to beauty products as well as to the works of art and the archaeological remains pertinent to this topic (the ampulla, speculum and ligula etc.), most of all it highlights the range of literary genres that include mention of make-up and perfumes: historical, poetic, encyclopaedic and satirical.2 This spread of material may help the scholar to build up a picture of the rhetorical

Secondly, the glossary makes the close link between beauty products and medical treatments, health and hygiene very clear. That is to say, using many of the products listed below was as much a matter of health and hygiene as it was of beautification. Individuals who did not observe the niceties of cleanliness were considered of low social class or living an immoral lifestyle that demeaned their status: so, for example, one should avoid bad breath or a strong body odour. Many of the entries in the glossary show these products as multifunctional. The association of cosmetics and perfumes with medicine and the basic need to keep clean legitimised their use and perhaps make it

Much of the material was subsequently published as a book Cosmetics and Perfumes in the Roman World, Tempus (2007). However, the glossary did not form part of that publication because I believed it had some merit on its own as a useful tool for scholars. It appears here for the first time. 2 The references given in the glossary are examples and not a fully comprehensive list. There is not space for that here. Also the point being made here in no way belittles the importance of the archaeological evidence. One should mention the particularly significant and recent find of a pot of face cream from Roman London. See Evershed, R P. et al. (November 2004) ‘Formulation of a Roman cosmetic’ Nature vol. 432: 35-6. 1

We cannot undervalue Pliny the Elder as a source of information on this topic as the glossary shows. However, I should point out that this is not the same as saying that we need to believe everything he says. 3

109

Dress and Identity

more likely that men could reasonably have used many of these products without drawing any attention or criticism.4

exceptions to this tradition, one of the most notable being woad (glastum) mentioned in Latin texts as a body paint but strictly for non-Roman use. In conclusion, the use of foundation, eye make-up, perfume, breath fresheners and other similar products was not simply a fashionable element of dress but was underpinned by concepts of beauty built on social convention, status, health, wealth and gender. A study of make-up and perfume can, I believe, reveal much about the society that used them.

Thirdly, the glossary highlights the fact that both cheap and expensive products were available. There were many products the origins and availability of which made them more exclusive and more expensive. The owners of a costly compound perfume like kyphi (the word being of Egyptian origin and meaning ‘welcome to the gods’) were probably very wealthy. Saffron (crocinum), used as a scent, and as eye make-up or an ingredient in eye make-up, was also very dear. Ownership reflected status as well as ability to pay. However, for at least some of the more expensive beauty products there was a cheap alternative that suited the purses of the less well off. Products such as lampblack (fungus lacernae) used as eye make-up or rouge (faex) made from the leftover wine were cheap and readily available to all. The climate is not much different from the fashion pages of today’s magazines where designer items are advertised alongside cheaper ready to wear alternatives.

Glossary Abrotonum: A strong smelling perfume derived from the plant southernwood. This was also used for cleaning wounds, treating asthma and cramps. Lucr. 4.125; Celsus, Med. 2.6.10, 5.5.2; Hor. Ep. 2.1.14. Absinthium: Wormwood. This plant was used as a perfume but was also mixed with rose oil to dye hair black, applied as a treatment for sore throats and used to preserve wine. Plin. HN 27.52, 15.87; Lucr. 4.125; Celsus, Med. 4.7.

Fourthly, please note the large number of anti-aging products that appear in the glossary. The importance of youth as an attribute of female beauty at least is reflected in the range of wrinkle removers and emollient creams on the market at this time. This may also be why we find so many references to hair dyes that could have covered grey hair. There was a strong desire to look young in a society where beauty and youth, as an element of beauty, could enhance one’s social status. There is little respect shown in literature or art for older women in particular.5 An elderly prostitute might not find work; there was no welfare state. Youth and beauty, which could also equate to health and therefore the ability to bear children are closely interlinked and as a result might proffer a good marriage (Soranus Gynecology 1.13).

Acacia: Gum Arabic from Egypt used as a hair dye, a scent and an eye ointment. Petron. Sat.23.5; Celsus, Med. 6.6. Acus: An ornamental pin used in the construction of often elaborate hairstyles. Pins were made of metal, wood, or bone with a sharp point at one end. The design of these pins ranged from very plain to highly decorated. Allegedly, the acus was used to punish the maidservant who did not do her job effectively. Fulvia was said to have pierced the tongue of the dead Cicero with a hairpin when his head was presented to her. Mart. 2. 66.2; Juv. Sat. 2.93; Ov. Am. 1.14.15.Cass. Dio. 47.8.1-5. Acus comatoria/Acus crinalis: A curling iron. Apul. Met. 8.13. Petron. Sat. 21.

Finally, there are interesting points to be made with regard to social status and ethnic identity within the pages of the glossary. Some entries remind us that the Romans perceived themselves of superior status to those tribes or nations whom they had subjugated. Wigs were often made of hair from captured slaves. Also, many cosmetics and perfumes originated elsewhere were imported by Rome and valued by its citizens, reflecting the Romans’ ability to absorb the practice of other cultures. There are however a few

Adeps (cygni, anseris, leonis, ursi): Animal grease from swans, ducks, lions or bears respectively. An application of animal fat was thought to improve the complexion and get rid of wrinkles. Swan’s grease (adeps cygni) was considered one of the best treatments. Bear’s grease (adeps ursi) mixed with resin from the rockrose (ladanum) and maidenhair (adiantum) was used to prevent hair loss too. Plin. HN 8.127, 28,163, 30.30.

4 There is, however, considerable scorn poured on men using hair removers (see Mart. 2.29.6, Ov. Ars. Am. 1.505 Sen. QNat. 7.32.2 for example). Men may have been wise to stick to using tweezers rather than resorting to any of the mixtures on the market, the efficacy of which is not easily understood. Many examples of tweezers have been found at Hadrian’s Wall for example (Allison-Jones (1995) ‘Sexing’ small finds’ in P.Rush (ed) Theoretical Roman Archaeology: Second conference proceedings, Aldershot, Avebury Press). The picture regarding the use of hair removers is complicated by the fact the parts of the body from which hair should be removed were prescribed by gender; This may be why there is the occasional slightly derogatory remark made even by Pliny the Elder for example. (Plin. HN. 36.154) 5 Old age in men conveyed mixed messages. For example baldness, a sign of old age, could suggest buffoonery (in comedy) and gravitas (in statuary) The senex (old man) a stock figure in Roman comedy is often bald. For statuary see for example the so-called Barbarini sculpture depicting a patrician carrying ancestral busts in the Palazzo Dei Conservatori, Rome.

Adiantum: Maidenhair fern. This was used to dye hair. When mixed with wine, celery seed and oil, maidenhair was believed to thicken and curl hair as well as prevent hair loss. It was also a treatment for stretch marks. Plin. HN 22.62, 28.163. Alabastrum: A jar or box used to store perfume or ointment. Alabaster was thought to preserve scent so this became a generic term for perfume containers. Petron. Sat. 60. 3, Frag. 18; Mart.11.8.9; Plin. HN 13.19; Gospel according to Matthew 26.7.

110

Susan Stewart: Cosmetics and Perfumes in the Roman World

Alcyoneum: A face cleanser, an ingredient in perfumery, and a treatment for ulcers and sores. The primary sources disagree on the nature and the origin of this product; it may have been obtained from bird’s nests, a kind of coral or a fungus from the sea. Plin. HN 32.86; Ov. Medic. 78; Celsus, Med. 5.6.18.

Arcula: a small box for perfume or jewels. Plaut. Mostell. 248; Apul. Met. 3.21; Cic. Att.2.1.1.

Alipilus: Person employed at the baths to remove unwanted hair. Sen. Ep.56.2; CIL 6.9141.

Asarum: A perfume made form wild nard, spikenard or hazelwort. The best was said to come from Pontus and Phrygia. Plin. HN 12.47. Gal. 6.12.4, 11.840.

Arsenicum (Arrhenicum): Arsenic or orpiment. A hairremover. See also Aurpigmentum. Plin. HN 34.178; Vitr. De Arch. 7.7.

Alum: Alum or potash. This was used as antiperspirant or deodorant and is still present in some products sold for this purpose today. Plin. HN 35.185; Scribon. 83.

Assyrium: A perfume. Its name is an indication of the area of origin. This could be used to perfume hair. Hor. Carm. 2.11.16.

Aluta: A beauty patch possibly consisting of a small piece of leather treated with alum to make it soft and pliable. See also Splenium. Mart. 7.35.1; Caes. BGall.3.13. 4; Ov. Ars. Am. 3.202.

Atriplex silvestre: an exfoliating skin cream made from turnip seed mixed with vetch, barley and lupins. Plin. HN 20.20.

Amaracinum: A strong smelling perfume or ointment made from sweet marjoram. This was also used in gynaecology and as a treatment for piles and inflammations. Lucr. 4.1179; Plin. HN 21.37.

Auripigmentum: Arsenic or orpiment. This yellow sulphurous earth was used as a hair remover. See Arsenicum. Plin. HN. 33.79; Celsus, Med. 5.5.1. Balaninum: Oil of ben nut from Syria. This expensive oil was used as a base for some perfumes as well as being applied as a treatment for dry skin. Plin. HN 23.89.

Amomum: A heavy scent from the East derived from an aromatic shrub, used as a hair perfume and funerary incense as well as being given as a cure for sleeplessness. Mart. 5.64.3; Verg. Ecl. 4.25; Plin. HN 12.49, 13.16: Pers. 3.104; Juv. Sat.4.108.

Balsamum: An expensive aromatic resin. A perfume and an ingredient in perfumes, the only source of this caustic resin was rumoured to be the king’s gardens in Judaea. Martial recommended this perfume to men. Medically it was used as a diuretic and an antidote to snake bites. Mart. 3.63.4; Plin. HN 12.118, 13.11. Juv. Sat. 2.41; Apul. Met.11.9; Tac. Germ. 45.7.

Ampelitis terra: Bituminous earth used as a colouring agent to darken eyelashes. This was also an ingredient in hair dyes and a treatment for spots. Plin. HN 35.194; Dios. Mat. Med. 5.181. Ampelos agria: A product obtained from the wild vine used to improve the complexion. In medicine it was used to treat sciatica. Plin. HN 27.44.

Bdellium: A cleanser made from the aromatic gum resin of a type of palm tree from South Arabia and Bactria. This was an ingredient in perfumes as well as in medicines, incense and magic potions. Plin. HN 12.35; Celsus, Med. 5.5.2.

Ampulla: A small round or pear shaped container for perfume or bath oil as well as wine or medicinal fluids. Suet. Dom. 21; Plin. HN 20.152; Mart. 3.82.26.

Belgicus color: A hair dye. Prop. 2.18.26. Caecilia: A product made from plant grains mixed with melted wax and applied to eradicate wrinkles. Plin. HN 26.163.

Amygdalinum: Almond oil. This was believed to improve both the tone and the colour of the skin. It was sometimes mixed with honey to remove spots and prescribed as a treatment for deafness, presumably because it cleared wax. Plin. HN 24.63, 23.85; Mart. 14.58; Plut. Mor. Frag.114.

Calamistrum: A heated instrument for curling hair. Petron. Sat.102.15; Plaut. Cur. 577; Varro. Ling. 5.129.

Anchusa: A red dye made from a root and used in perfumes. Plin HN 13.7.

Calamus: Sweet flag. This plant was used as an ingredient in perfumes and medicines. Plin. HN 12.104, 13.8.

Aphonitrum: Sodium carbonate or saltpetre. Red in colour, this purported to remove wrinkles and generally improve the complexion. Sodium carbonate when mixed with maidenhair (adiantum) was specifically applied to stretch marks. It was also used to treat skin diseases such as impetigo. Mart. 14.58.1; Plin. HN 24.63; Ov. Medic. 73.

Caliendrum: A tall hairstyle made from false hair. A wig. Hor. Sat.1.8.48. Calliblepharum: A generic term for eye make-up, this was a cosmetic for eyelids, lashes, and brows, sometimes made from rose petals and saffron or from burnt date kernels mixed with nard. The mixture could make the eyes look

111

Dress and Identity

bigger which was considered attractive. Plin. HN 35.194; 23. 97.

in the prestigious royal perfume of the Kings of Parthia (regalium). However it may have been an alternative to the more expensive cinnamomum. Plin. HN 12.135.

Capillamentum: False hair. A wig, possibly curly. Petron. Sat.110.5; Suet.Calig.11; Apul. Met. 3.16; Vitr. De Arch. 4.1.7.

Corissum: A dye extracted from St John’s Wort which coloured hair black. Plin. HN 26.164.

Caryophyllon: Dried flower buds from the clove used in perfumery, as medicine and as a food. Plin. HN 12.30.

Coryphia: A breath freshener and a treatment for wrinkles made from shellfish. Plin. HN 32.84.

Cardamomum: An aromatic perfume made from cardamom seeds. This could be used to perfume hair. Plin. HN 12.50; Celsus, Med. 3.21.7.

Corymbion: A wig, possibly curly. Petron. Sat. 110.5. Cosmianum: A perfume made by the perfumer Cosmus from cinnamon (cinnamomum) and nard (nardinum). A pomade. The term is also used to refer to pastilles made by Cosmus to freshen breath. Mart. 14.14.6; Petron. Frag. 18.

Casia: A perfume made from the flowers and the bark of this plant. It was also used in the embalming process and to treat liver complaints. Often confused with cinnamon (cinnamomum), the supply of casia is thought to have come to Rome through China via the caravan route into Arabia. Plin. HN 12.85; Tib. 1.3.61; Plaut. Cur. 101; Mart. 10.97.2.

Costus: A root used in perfumes. Mixed with honey (mel), costus was believed to improve the complexion. Plin. HN 21.76; Perip. M. Rubr. 39.

Caustica (Chattica) spuma: Hair dye. A liquid. Mart. 14.26.1.

Creta: A fine pipe-clay powder used as foundation, this made the face appear fashionably pale. It might also have served as a hair remover as well as a clothes cleaner, a letter sealant and a treatment for excessive sweating. Petron. Sat. 23.5; Ov. Ars. Am.3. 199.

Cerussa: White lead. The cosmetic was manufactured by pouring vinegar over lead shavings and applied to make the complexion fair. It was known to be poisonous. Cerussa was also used as a hair wash and an eye treatment. Exposure to the sun dried the lead out, creating potential problems. Plin. HN 28.109; Mart. 10.22.2; Celsus, Med. 3.10.2; Ov. Medic. 73.

Crocodilea: Crocodile dung. This may have been applied simply as a cleanser but it might have been used to tint the skin as well. Crocodilea might be mixed with henna (cyprinum) to give the product a more pleasant smell. Plin. HN 28.184; Hor. Epod. 12.10.

Chia terra: China clay or kaolin. Applied to the face or to the whole body, this may have been used simply as a cleanser but perhaps tinted the skin too. Plin. HN 35.194.

Crocinum: Saffron. A perfume and an eye make-up, this was an expensive product because a lot of saffron was required to make very little perfume or ointment. Ov. Ars. Am. 3.204; Mart.11.8.2; Plin. HN 21.138.

Cinnabaris: Red mercuric sulphide from Spain or India used as rouge though known to be poisonous. This product could also be used in the manufacture of perfume to add colour. Plin. HN 33.117, 29.25.

Cummium: A general term for gums or resins such as styrax or bdellium used to improve the complexion. Plin. HN 24.105.

Cinnamomum: A compound perfume made using the bark and the leaf of the cinnamon plant mixed with myrobalinum, and myrrh (murra). The price varied according to the size of leaf. Cinnamomum may have been used as a perfume by both sexes. Medicinally it was used to treat gangrene, fits and eye infections. Cinnamomum is still used today in soaps and in cosmetic preparations for men. Mart. 4.13.3; Plin. HN 12.86, 13.15.

Cyprinum: Henna oil. The scented leaves of the plant were used to produce a lilac- scented perfume. This was also mixed with crocodilea and murra or panax to make a hair remover. Mixed with vinegar, cyprinum dyed hair black while mixed with sparrow apple it turned hair red. Cyprinum was also used as a nail colour. Plin HN 13.5, 30.110, 28.183.

Cistella: A small chest in which cosmetics and perfumes could be kept. Ter. Eun. 753; Plaut. Cist. 655.

Dentifricium: A general term for tooth powder or teeth cleaner. This could be made from a variety of materials including pumice stone, powdered horn, ashes of dog’s teeth mixed with honey, murex shell, or ashes of hare’s head and orris root. Plin HN 36.153; Ov. Ars. Am. 3.216.

Collyrium: Lead sulphide used as eyeliner and a hair wash as well as having medical uses as a suppository, eye salve and a treatment for boils. Plin. HN 32.84; Juv. Sat. 6.579; Hor. Sat. 1.5.30; CIL. 7.1309.

Diapasmata: Dusting powder made from rose petals for example. To protect against sweat or bad odour this was

Comacum: Nutmeg or mace. This was an ingredient

112

Susan Stewart: Cosmetics and Perfumes in the Roman World

sprinkled over the body after a bath, allowed to dry and then washed off. Plin. HN 21.125, 13.19; Lucian Er. 39.

Galbanum: Persian or Syrian gum resin used in perfumes as a fixative and a scent. Plin. HN 12.126. CIL. 13.10021.

Dropax: Pitch. Hair-remover. Mart. 3.74.1, 10.65.8.

Galena: Eye make up. Made from malachite, a lead ore mixed with silver and applied in powdered form, it was poisonous even to the touch. Plin. HN 34.173.

Elaceomeli: Gum from the olive tree used as eye make-up to make eyes look bigger. Plin. HN 23.96.

Galericulum: A wig. Suet. Oth. 12.1.

Elate: a dark hair dye made from dates. This was also used as a breath freshener. Plin. HN 12.134.

Galerus: A wig. Juv. Sat. 6.120.

Erysisceptrum: A perfume derived from a thorny bush. Plin. HN 24.112.

Galla: Oak gallnut from which a resin was extracted to dye hair black. Verg. G. 4.267; Plin. HN 24.10.

Erythrodanum: Madder. The plant leaves were used to dye hair red. Plin. HN 24.94.

Germanae herbae: A hair dye made from unknown herbs that probably lightened hair. Ov. Ars. Am. 3.163.

Faba: Bean meal used to prevent wrinkles and stretch marks. Beans formed part of the staple diet of the poor so this was probably a cheap commodity. Ov. Medic. 72; Mart. 3.42.

Glastum: Woad. A plant dye with which Britons, Gallic and German tribes painted their whole bodies. Plin. HN. 22.2; Caes. BGall. 5.14. Hedychrum: Fragrant ointment. Cosmetic balsam. Cic. Tusc. 3.46.

Faex: Rouge made from the dregs or lees of wine. Ov. Ars Am. 3.211.

Helenium: Derived from a shrub this product was named after Helen of Troy. It was said to improve the complexion and the skin tone of the whole body leading to increased sex appeal. Plin. HN 21.159.

Favilla: Ashes used to darken brows and lashes sometimes mixed with saffron. Ov. Ars. Am.3.203. Fel irenacei: Gall of hedgehog. A hair-remover. This was sometimes mixed with other ingredients for this purpose: for example, goat’s milk. Plin. HN 30.133.

Helianthos: A body lotion made from sunflowers mixed with saffron (crocinum), bear’s fat (adeps ursi) and palm oil. Plin. HN 24.165.

Flos salis: A salt deposit used as a cleanser and a hair remover. Plin. HN 31.90.

Ichthyocella: Fish glue applied as a wrinkle remover. Plin. HN 32.84.

Folia lavi (lauri): Leaves of the laurel tree used to freshen the breath. Mart. 5.4.

Irinum: Perfume made from the root of the iris. This was often used as a deodorant and a breath freshener. Plin. HN 13.5

Folia rosarum usta: Charred rose petals. These were mixed with soot and used as eye make-up. Ov. Ars. Am. 3.203.

Kyphi: A perfume of Egyptian origin said to have contained at least sixteen and perhaps as many as fifty ingredients. Plut. Mor. 383a.

Foliatum: Nard oil. The formed the basis of a compound mixture to encourage hair growth and make hair darker. Mart. 11.27. 9; Plin. HN 13.15.

Lac: An insect dye used by women to dye their hands, feet and nails. This was also used to dye cloth. Peripl. M Rubr. 6.

Fucus: A root or lichen used as rouge. Fucus also meant tawdry or inferred a suspicious attempt to disguise. Prop. 2.18.31; Cic. De Or. 3.119; Plin. HN 31.19; Plaut. Mostell. 275.

Lacrima hederae: A hair remover made from residue from ivy. This was mixed with oil and apparently required frequent application to be effective. Plin. HN 24.79.

Fuligo: Black pigment made from soot or powdered antimony and used as eye make-up for brows or around the eyes. This was similar to our eyeliner or mascara. It may have been applied damp with a pin. Juv. Sat. 2.93.

Lacrima vitium: A gum resin from the vine mixed with oil and used as a hair remover. Plin. HN 23.3.

Fungus lacernae: Lamp black. A cheap eye make-up and alternative to kohl. Verg. G. 1.392.

Lacte asinae: Asses’ milk. A bath of asses’ milk was taken to whiten and soften skin and guard against wrinkles. Plin. HN 28.183; Dio. Cass. 62.28.

113

Dress and Identity

Ladanum: A resin from the rock rose from Southern Arabia that was mixed with soot to make an eye make-up. It was also applied with animal fat (adeps) or with myrrh (murra) to prevent hair loss or grey hair or to enhance dark hair and encourage its growth. Plin. HN 28.163.

Greek from Sicily. This required stirring before use. Plin. HN 13.13. Mel: Honey. This was used to improve the complexion. It is still widely used in modern face creams as well as bath and shower gels. Plin. HN 21.76; CIL 7.1311a.

Lapis Arabicus: An unknown stone used as teeth cleaner and a substitute for pumice stone in this respect. Plin. HN 36.153.

Melinum: A scent made from quinces. This was also a hair remover and a type of earth applied to make the face look pale. Plin. HN 35.37.Plaut. Mostell.264.

Laserpitium: Benzoin. A resin used as a fixative in perfume. This is still found in shampoos. Plin. HN 19.38.

Mendesium: A compound perfume. Plin. HN 13.5.

Ligula: A spoon with a long handle for mixing and extracting cosmetics: the end either bulbous for mixing or small for extracting contents from a container. These were made from wood, metal or bone. Plin. HN 21.84; Mart. 8.33.23.

Metopium: a compound perfume made from bitter almonds and mixed with almond oil to form a cosmetic good for dry skin. Plin. HN 13.8. Minium: red lead from Spain used as rouge. This was known to be poisonous. Plin. HN 33.111; Verg. E. 10.27.

Lixivium: A hair dye, possibly made from pressed grape must, used to lighten hair colour. Val. Max. 2.1.5.

Morum: Rouge made from mulberry juice. This was also applied as a hair dye when mixed with wine and black fig leaves. Plin. HN 23.134.

Lomentum: A skin softener, face pack, wrinkle or stretch mark remover made from bean flour and rice powder for all over body use. Mart. 14.60; Plin. HN 33.89; 33.162; CIL 4.2597.

Mundus muliebris: Toilet items belonging to a woman. Liv. 34; Dig. 34.2.25.10; Varro Ling. 5.29.129.

Lupini: Lupin seeds. These were eaten to freshen the complexion or crushed and added to face powders. Plin. HN 22.154; Ov. Medic. 69.

Murex: A cleanser and wrinkle remover made from the shell of the shellfish of that name. Plin. HN 32. 84.

Lycium: The yellow berry or buckthorn from India. Froth made from the berries was used as a cleanser. A yellow dye extracted from the roots tinted hair blonde. Plin. HN 26.164; 24.125.

Murra: Myrrh. The resin was used in hair products, in perfumes and a cleanser. It is still used in toothpastes and mouthwashes. Used as an analgesic it may have been administered on the sponge offered to Jesus on the cross. Plin. HN 28.214.

Magma: Dregs of perfume used in dusting powders. Plin. HN 13.19.

Muscis tritis: Crushed flies used to blacken eyebrows. Plin. HN 30.134.

Malabathrum: An expensive perfume made from cinnamon leaf. The leaves could be chewed to prevent bad breath. Plin. HN 23.93; Peripl. M Rubr. 56

Myrobalanum: Behen nut oil, a base for perfume. Mart. 14.57; Plin. HN 23.98. Myrothecium: A box for cosmetics and perfumes. Cic. Att. 2.1.1.

Mastiche: Mastic. A sweet smelling gum used as a skin cleanser and teeth cleaner. Plin. HN 22.45.

Narcissinum: A perfume or aromatic ointment from Lycia made from narcissus flowers mixed with myrrh (murra) and other aromatic herbs. Plin. HN 13.6.

Massaris: The grape of the wild vine used as a perfume ingredient. Plin. HN 23.9. Matticiae pilae: A blonde hair dye. This was imported from Germany and named after a German tribe. This cosmetic was sold in small cakes or balls. Mart. 14.26.27.

Nardinum: Nard or nard oil. This was an expensive compound perfume shipped around the Roman world in large and medium solid balls. Nardinum softened as well as perfumed the skin and had cleansing properties. It was believed to encourage hair growth. Plin. HN 12.42; Gal. 10.791; Peripl. M Rubr. 39.

Medicamentum: A general term for cosmetics, this also referred to a medicine, drug, poison or magic potion. Ov. Ars. Am. 3.205; Juv. Sat. 6.472.

Narthecium: A cosmetic or medicine container. Cic. Fin. 2.22.

Megalium: A perfume named after its inventor Megallos, a

114

Susan Stewart: Cosmetics and Perfumes in the Roman World

Natron: Spikenard perfume imported from Gaul. The price varied according to the size of the leaf. Natron was also used to encourage hair growth. Plin. HN 13.18.

Panchrestos: A mouthwash made from mulberries. Plin. HN 23.136. Pecten: A comb. These were made from wood or bone. Petron. Sat. 126.2; Ov. Am. 1.14.15; Apul. Met.11.9

Nuclei palmarum cremati: Burnt date stones mixed with soot or nard to make eye make-up. Plin. HN 23.97.

Pardalium: A perfume made from leopard or panther scent. Plin. HN 13.6.

Nitrum: A skin softener made from various alkalis especially soda and potash. Plin. HN 31.112. Ov. Med. 73.

Platyophthalmon: A product that made eyes look large. It gets its name from its effect. Plin. HN 33.102.

Oenanthinium: A scent made from the flower of the wild vine. This product was also used as a face cleanser. Plin. HN 13.5.

Polomemonia: A black hair dye. Plin. HN 26.164.

Oesypum: Grease extracted from the unwashed fleece of sheep or goats. A versatile product, in cosmetic terms this was used as a skin cleanser. The Attic variety was considered the best. It had a very strong smell. Ov. Ars. Am. 3.213; Plin. HN 12.74; 29.35; 30.28.

Poppaeanum: A bread dough face pack, its invention attributed to Poppaea, the wife of the Emperor Nero. This was for use at night and included asses’ milk among its ingredients. Juv. 6.462. Psilothrum: A hair remover made from arsenic and quick lime. It was green in colour and very caustic. Plin. HN 23.3; Gal. 12.450; CIL 4.2613, 4.2614; Mart. 3.74.1.

Oleum lentiscinum: Mastic oil. A hair dye and a teeth cleaner. Plin. HN 24.42. Olibanum: Frankincense. An expensive aromatic gum from the east used as a perfume and an ingredient in rejuvenating facemasks. Plin. HN 12.32; 33.102; Ov. Medic. 83ff.

Psimithium: Acetate of lead. See Cerussa. Plin. HN 34.175. Pumex: Pumice stone use as hair remover or tooth powder. Plin. HN 36.154; Ov. Ars.Am. 1.506.

Omphacium: Oil from unripe grapes or olives. This was used as a deodorant, as a cleanser, and to keep teeth looking white. Plin. HN 12.130; 23. 29; Verg. G. 2.121.

Purpurissum: Rouge made from chalk or earth dyed with purple from a shellfish called the murex. The term also meant brilliant or beautiful. Plaut. Truc 290; Plaut. Most. 261; Plin. HN 35.49.

Onyx: A perfume container. The materials onyx, marble or stalagmitic limestone were thought to preserve scent. Mart. 7.94.1.

Pyxis: A small box for storing medicines, poisons or cosmetics. Ov. Ars. Am. 3.210; Ov. Rem. Am. 353; Cic. Cael. 61. Petron. Sat. 110.2; Apul. Met. 6.16

Ophrys: A black hair dye derived from a plant belonging to the cabbage family. Plin. HN 26.164.

Quinquefolium: Cinquefoil. The root of the plant was boiled down to produce a breath freshener and a treatment for sores. Plin. HN 25.166; 25.174.

Opobalsamum: Balm of Gilead. A resin used in perfume production. Juv. Sat. 2.41; CIL 7. 1311b. Ornatrix: A hairdresser. Her job may have included fitting wigs and pulling out grey hairs as required. A skilled worker of some importance, she may have overseen the make-up as well as the dressing of her mistress’ hair. CIL 6.9732; Ov. Am. 1.14.16; Juv. Sat. 6.491.

Regalium: Royal perfume allegedly worn by the Kings of Parthia. Its ingredients included cardamomum, costus, cyprinum, panax, balaninum and murra as well as honey, wine catnip and wild grapes. Its compound nature and the origin of many of its ingredients made it an expensive luxury. Plin. HN 13.18.

Ostractae: Pieces of stone resembling potsherds used for removing hair. Plin. HN 36. 139.

Resina: Resin. A cheap hair remover used by prostitutes. Resina could also be added to perfume to preserve the scent. Mart. 12.32. 21; Plin. HN 24.32; Cels. 3.27.1.D.

Ozaentis: A strong smelling grade of nard used in perfume. Plin. HN 12.42. Panathenaicum: A scent from Athens. Plin. HN 13.6.

Rhodinum: Rose perfume, a deodorant, mouthwash or dusting powder. Plin. HN 13.17.

Panax: All-heal or ginseng used in ointments and perfumes. Plin. HN 12.127; 13.12.

Sambucus: Juice of elder berries used to dye hair black. Plin. HN 24.51. Sal: Salt, in particular a variety imported in small bricks

115

Dress and Identity

from Cappadocia, was applied to give skin a glow. Plin. HN 31.84.

shape. This may have been used, on occasion, to conceal slave brandings. Mart. 8.33.22; 2.29.9.

Saliunca: A variety of valerian or Celtic nard used in perfumes. Verg. Ecl. 5.17; Plin. HN 21.40.

Spondylium: Cow parsnip or fennel rubbed into the scalp to make hair curly. Plin. HN 12.128; 24.25.

Sampsuchinum: a perfume made from oil of marjoram (origanum) mixed with other ingredients and complicated to prepare. Plin. HN 13.10

Spuma argenti: Antimony. This was an ingredient in hair removers, hair products and skin care preparations and known to be poisonous. Plin. HN 24.58.

Sanderache: Realgar or red sulphide of arsenic used as a dye, an antiseptic and the main ingredient in a treatment for hair loss attributed to Cleopatra. Gal. 12.403-4; Plin. HN 34.177.

Spuma batava: A liquid hair dye used to tint hair red. Mart. 8.33.20.

Sanguisugae: A hair dye made from leeches decomposed in red wine used to darken hair. Plin. HN 33.68.

Stibium (Stimmi): Powdered eye make-up made from kohl or antimony sulphide. This was used for blackening brows and lashes and to make the eyes look larger by contracting the eyelids. Plin. HN 33.101.

Sanguis vestportilionum: Hair remover made from bat’s blood. Plin. HN 30.132.

Storax (Styrax): Gum resin used as a perfume and a fixative in scent. Ciris 168; Peripl. M Rubr. 28.

Sapo: A hair dye made from goat’s fat and beech tree ashes that tinted hair red-gold. The term was also used as a general term for hair dye that could be liquid or solid. Plin. HN 28.191; Tac. Hist. 13.161.

Sucinum: Amber. Ladies could hold small balls of amber in their hands when outside to counteract unpleasant odours. Mart. 3.65.5. Susinum: A compound perfume. Plin. HN 13.11.

Scrinium: A box for storing cosmetics or writing implements. Prop. 3.6.14.

Telinum: An ointment made from fenugreek used to combat body odour. Plin. HN 13.13; Plaut. Cur. 102.

Serichatum: An Arabian spice added to perfumes. Plin. HN 12.99.

Trifolium: A face cream made from small trefoil leaves or the leaves of sweet clover These leaves were also used in perfumes. Plin. HN 21.153.

Sesaminum: Sesame oil used as a base and a preservative in perfumes and a treatment for earache. Plin. HN 23.95; 13.13; Celsus, Med. 1.196.

Umor elmi: A skin care product made from the seeds of the elm tree applied to brighten the complexion. Plin. HN 24.48.

Siliginis: A bread face pack used to improve skin tone. Juv. 6.472; 5.70.

Unguentum: A general term for perfumes or ointments. Plaut. Mostell. 41; Prop. 2.4.5.

Sium: A night cream produced from a multi-flowered water plant. Plin. HN 22.84.

Viridis cortex nucis uglandis: Walnuts used to make a red hair dye. Plin. HN 15.87.

Smegma: A skin cleanser and a wash to improve hair colour made from lupins. Plin. HN 24.43; Ov. Medic. 51.

Volsellae: Tweezers for removing hair. Plaut. Cur. 577.

Speculum: A mirror. Juv. 2.99; Ov. Am. 2.17.9; Prop. 3.6.11; Gell. NA 6.12.5;Plin. HN 33.45.

Ziziphe: Jujube. A perfume. Plin. HN 21.51.

Splenium: A fake beauty spot sometimes in a crescent

116

Chapter 11

The Social Life of Museum Textiles: Some Comments on the Late Antique and Early Medieval Collection in the Ure Museum at the University of Reading Anthea Harris

Preparations for the reordering of the Ure Museum at the University of Reading in 2005 brought to light a small collection of previously largely unknown textiles. These were lightly stitched with cotton thread to nine pieces of cardboard. The back of three pieces of cardboard was labelled with ‘1889’ and the inscription ‘Ahkmîm from the collection of Rev. Greville Chester’. The labels were written in black ink in a style consistent with a late 19th-century hand. Since their closest parallels are from other cemetery sites in the Fayoum oasis area of Upper Egypt, their Ahkmîm (= ancient town of Panopolis) provenance might cautiously be accepted, especially given that the Reverend Greville John Chester (1830-1892), a well-known collector of Egyptian antiquities, was active in Egypt during this time (Dawson and Uphill 1972: 62-3). Many of the Egyptian textiles now in the collections of major British museums, including those in the British Museum, the Victoria and Albert Museum, the Ashmolean Museum and Liverpool Museum, were acquired by Chester in Egypt in the 1880s whilst the Late Antique and early medieval cemeteries there were being opened.1 It is less clear how the textiles found their way into the collections of the Ure Museum, the University of Reading not having been established at that date. However, the British Museum is recorded as having given Annie Ure (the founder of the Ure Museum) a gift of ‘unconsidered trifles’ sometime in the 1930s and it is perhaps most likely that the textiles formed part of that gift.

provenance with any precision. Thus, those pieces believed to be most ‘interesting’ or ‘important’, usually on the grounds of visual attractiveness, were detached from the main body of the textile (usually a garment or shroud) and retained, while the remainder of the textile was discarded (Thomas 2007: 241-2). Such were the quantities of textile being dug up in Egypt in the late 19th century that the garments themselves apparently became mundane even to excavators and collectors.2 The textiles in the Ure collection are fairly representative of the many thousands of so-called ‘Coptic textiles’ in museum collections across Europe and North America.3 If it is indeed the case that they were part of a gift of ‘unconsidered trifles’ from the British Museum it is easy to see why they were considered dispensable by that institution in the 1930s, for they fall into several ‘undesirable’ categories. They are either undecorated (at least in any way meaningful to a mainstream curatorial eye of that period), too small to discern or reconstruct any pattern, or have a very fragmentary or damaged decorative motif. 2005.7.1 (Cat. 1), for example, a tapestry square (tabula) cut-out with a design principally executed in russet-coloured wool on an unbleached linen background, has several comparanda in other museums, some of which are almost identical, while others are variants on the symmetrical composition of stylised canthares and vine-scroll. Yet, while the square is intact, approximately one quarter of the wool wefts are worn completely away, presumably through exposure to soil acids and the products of the process of bodily decomposition, leaving a triangular area comprised only of unbleached linen warps and wefts. These threads were surprisingly relaxed, even before conservation, and permit a detailed analysis of the object’s construction. Several other

There are 19 textiles in the collection, all of them in a fragmentary state, although sufficient monies were raised to conserve five pieces (Smith & Harris 2008). This fragmentary nature is common for Egyptian textiles of this period, most of which were retrieved from the desert sands of Egypt at a time when stratigraphical excavation techniques were not widespread and when the opportunities to acquire large quantities of exotica for private collections were more attractive (and perhaps perceived to be more important) than the impetus to record context and

For a present day perspective on the problem of quantity in excavations, see Jørgensen and Mannering (2001). 3 Estimates vary widely. Bowen (2002: 97) estimates more than 20,000 textiles in collections worldwide, yet 30 years earlier Thompson (1971: 4-5) had already estimated that there were 150,000. Many more collections have been published since the early 1970s. See Thomas (2007) for an insightful discussion of this literature. 2

The textiles now in the collections of Liverpool Museum (Museums and Galleries on Merseyside) were in the care of Norwich Castle Museum before World War Two (Gary Brown, pers comm.). 1

117

Dress and Identity

Fig. 11.1: Collection in unconserved state.

objects (for example, 2005.7.17; Cat. 17) are stained or encrusted with blood or other body fluids.

properly excavated material from Egypt, has also raised new issues. Recent AMS radiocarbon-dating results tend to indicate that a broad chronological range is possible for the textiles and increasingly suggest that the typologies linked to a very tight system of dating are no longer reliable, although where art historical dating has been expressed as a range of dates these are usually consistent with radiocarbon dates to some two centuries (van Strydonck et al. 2004: 243). As more results become available and as more 1st millennium archaeology from Egypt is published in excavation report form, it is likely that many museum collections will have to be re-examined, especially in relation to their attributed dates. Few textiles from museum collections have, as yet, been sampled for radiocarbon dating although this situation is starting to change (van Strydonck et al. 1993: 65-71; van Strydonck et al. 2004; Schrenk 2004: 476-8; Pritchard 2006: 13-25).

In the case of 2005.7.1 and a small number of other fragments (for example, 2005.7.2, 2005.7.3, 2005.7.4, 2005.7.5: Cat. 2, 3, 4, 9), the motifs represented may have been considered to be near-enough duplicates of other, perhaps less fragmentary, textiles in the British Museum’s collection and so surplus to requirements. The repertoire of colour is also more limited than is often the case. Thus, the Ure textiles include most of the usual range of motifs associated with ‘Coptic’ textiles, for example: leaves and arrows, stylised vine-scroll, acanthus leaf and canthare patterns, mythological beasts, ‘tree of life’ and geometrical designs, as well as examples of the most well-known aspects of dress decoration, such as bands (clavi), squares (tabulae) and roundels (orbiculi). In addition, there are examples of several standard weaving techniques, such as floating weft, slit tapestry, self-banding, dovetail joins and looped fringes. The ground weaves have an approximately equal proportion of warp threads to weft threads, resulting in a plain tabby cloth that is only slightly warp-faced or slightly weft-faced, if it is either. In tapestry areas the wefts are usually woven on two or three warp threads.

The realisation that the textiles were probably produced over a much longer period of time than hitherto thought likely has been instrumental in the demise of the term ‘Coptic textiles’ or ‘Coptic tapestries’. In part, too, this has been prompted by the widespread questioning of the relationship between ethnicity and material culture within archaeology and the acknowledgement that the expression of identity (including ethnicity) through material culture is socially constructed rather than fixed. In any case, the term ‘Coptic’ is a misleading one. It is a descriptor for the pre-Arab conquest population of Egypt yet is used today to refer to that sector of Egyptian society which claims to trace its origins (and language) at least as far back as the pharaonic period. It has also, of course, become inextricably linked with the native Egyptian Orthodox Church – the Coptic Orthodox Church – which split from the rest of the Byzantine Church over Christological concerns at the Council of Ephesus in AD 451, and which preserves the unique linguistic identity of this population in its liturgy (Meinardus 1999: 9). Thus, the term can be used to refer to ‘origins’, religious denomination or language, as well as both historical and contemporary populations. It does

For at least a century after they first came to the attention of the museum world, textiles from 1st millennium Egypt were analysed principally on an art-historical basis. The frequency with which given motifs occurred, where and with what variations concerned many scholars and was reflected in the way that textiles were curated and displayed in museums, as well as the content of textile catalogues. In the absence of stratigraphically retrieved exemplars, dating techniques often relied on art-historically constructed typologies. Although these still form the basis for much dating of textiles in museum collections, the use of new technologies has started to make an impact on the debate (for example, imaging technologies: Clarke et al. 2003). AMS radiocarbon-dating, coupled with the availability of

118

Anthea Harris: The Social Life of Museum Textiles

not, put simply, refer to a discrete chronological period, although this is the sense in which it has often been used.

and this has long encouraged debate about possible crosscultural interaction, as well as the transmission of textiles (and ideas) within Egypt and across its borders. This has provided a good platform for recent archaeological investigations into long-distance exchange and the role of textile evidence in elucidating this (for example, Wild 2001).

That said, there is no straightforward alternative for a group of objects which are now generally accepted as having a date range between the 3rd and 10th centuries AD, and as such span the Late Roman, Byzantine and Umayyad periods.4 The terminology used here – late antique and early medieval – is not ideal, but does give a more accurate sense of chronology and does draw the textiles into the broader debate where this terminology is applied: that surrounding the processes at work in the pan-Mediterranean world in the 1st millennium AD. This was a world of which both the producers and consumers of the textiles were an integral part.

At least one of the textiles in the Ure collection has features that make it a possible candidate for an import: 2005.7.12 (Cat. 7) is a fragment of a probable band (clavus) with a ground weave worked in a blend of blue and red wool, plied together to produce the effect of purple (seen most clearly on the reverse side of the object). This is a well-known technique, with a resultant effect often described as ‘optical purple’. Different coloured yarns were regularly combined to produce a greater variety of shades. The use of blue dyed wool in background weaves has also been associated with Persian or Persian-influenced textiles (for an early study: Kitzinger 1946).

Again, in the absence of supporting archaeological evidence, technical analyses quickly came to play a key role in the study of Egyptian textiles. A great deal of information on spinning and weaving technologies has thus been extracted, and a highly developed understanding of tapestry techniques and loom mechanics acquired. Nevertheless, it is instructive that, despite the need to examine both sides of a textile in order to assess its construction, almost without exception catalogues depict (and museums display) the front face of Egyptian textiles, rather than the reverse. Ways of ‘seeing’ these collections are seemingly entrenched and attempts to ‘see’ them from new perspectives have yet to be fully explored (Dauterman Maguire 1999: 10-11). Mindful of this consideration, the five pieces in the Ure collection selected for conservation can now be examined or displayed from any direction without removal from their casing.

Perhaps the greatest significance of the Ure collection is its usefulness for illustrating patterns of textile use, repair and re-use, as well as the processes of production itself. This is not always the case. Given that most collectors active in the late 19th and early 20th century, when the great collections of Egyptian textiles were established, preferred ‘perfect’ examples, those artefacts that provided information about the ‘life’ of the textile after its production were, ironically, selected against. Historiography which describes burial ritual as using ‘new’ or ‘best’ clothes may not always be taking this into account.

Another disadvantage of this normative approach, as Thelma Thomas (2007) has recently pointed out, is that it can lead to intransigence in the debate and a refusal to challenge data that seems not to ‘fit’ established categories of thought. She highlights the (until now) entrenched terminological distinction between ‘Byzantine silks’ and ‘Coptic tapestries’ and explores its implications for the study of textiles and of archaeology more generally. She points out that although silk as well as linen-based tapestry was been retrieved from cemeteries at Ahkmîm, there was, from the earliest publications of this material, an assumption implicit in the historiography that silk had somehow to be produced in a more ‘international’ context, whereas tapestry could be attributed to local, provincial workshops (Thomas 2007: 147). This assumption was played out in publication and museum strategies, with silk and tapestry being published and displayed separately. It is no surprise, therefore, that the Ure collection comprises only tapestry pieces and no silks, for it was put together long before this change in approach.

The collection here contains examples of darning and other forms of repair work (2005.7.11, Cat 6), and the re-use of clavi on new tunics. 2005.7.4 (Cat. 4), for example, comprises two clavi cut from (presumably) another garment, their horizontal seams turned under and stitched onto a new ground weave. The sharpness of the crease suggests the altered areas of the garment were then heat pressed. 2005.7.5 provides another example of the same technique. The Ure collection, additionally, provides an example of a ground weave with a selvedge (2005.7.19, Cat.18), permitting confirmation of the textile’s construction on the loom – without a selvedge the warps and wefts must merely be inferred. Tailoring practices in 1st millennium Egypt preferred not to use a selvedge edge, and so many textiles in museum collections cannot provide this evidence (Granger-Taylor 2000: 157-60). Interestingly, there may be an example here of left-handed sewing: the recycled clavi in 2005.7.4 (Cat. 4) are sewn onto the ground weave with a left-slanting over-stitch. All other over-stitching in this collection slants to the right.5

Yet, the advantage of adopting a normative approach to the study of spinning and weaving techniques has meant that ‘out of the ordinary’ features have been quickly identified

The very fragmentary pieces of plain weave in coloured wool threads (2005.7.19, Cat. 18), which show evidence of having once been sewn together, are more unusual still.

‘Islamic’ is avoided here because it, too, tends to obfuscate the diversity of political, religious and cultural identities.

5

Bowen (2002: 93) has identified a left-handed repair to a garment from Kellis I cemetery (Tomb 18).

4

119

Dress and Identity

Their size, coupled with collecting practices, render it unsurprising that there are few if any comparanda from museum collections – most would have been discarded on site. Yet, analogous pieces, interpreted as ‘rag amalgams’ have recently been excavated in 4th and 5th century contexts at Karanis and Berenike and, outside Egypt, at Yotata in southern Israel (Thomas 2007: 155; Wild and Wild 2000: 264-73; Shamir 2001: 96-7).6 At Berenike, where over 400 textiles have been found, it is interesting that time and energy was devoted to sewing these rag amalgams together, when access to new textiles was apparently so easy. If the Ure piece was from a burial context, it is likely that it was used as ‘packing’.

recognised, and should take place alongside other forms of textile research, in order to yield new perspectives on the role of textiles as integral to social life in late antique and early medieval Egypt.

Where evidence of physical alteration is not apparent, determining the condition in which an object was placed in the grave is not straightforward, given the lack of information on the context of its extraction and its postexcavation history. Much of the fading of colours, for example, is likely to have taken place after excavation, given the unfavourable conditions of storage. Recently excavated textiles at Fag al-Gamous, for example, have been examined for the build up of lint in order to determine whether they were placed in the grave in a pristine condition (Griggs 2005: 190). This is not possible, of course, in the case of the objects in this collection. The only object that might have been ‘new’ when deposited is 2005.7.14, (Cat. 14), interpreted as a pillow-case. This is the only piece in the collection where the threads appear to be the same brightness on both the front and reverse sides. There is no other evidence here of objects – both garments and textile accessories such as shrouds and shawls – being specially produced for burial.

220mm x 220mm; unbleached linen and dyed wool

Catalogue 1. Tapestry square (tabula) with canthares and foliate design Accession number 2005.7.1 From Akhmîm, Egypt, probably 4th – 6th century

Description Acquired in Egypt before 1889 by the Rev. Greville Chester. This is a tapestry square cut-out with a design principally executed in russet-coloured wool on an unbleached linen background. The only deviation from this is a small central circle in yellow wool, probably intended to represent gold. The design comprises a central square (84mm by 84mm) with four decorative borders. The central yellow circle (diameter 7mm) is framed by an octagon or star formed by the roots of four stylised trees. Each tree trunk runs at 45 degrees to the corners of the square and each tree has three trefoils (representing leaves), extending to the limits of the square. The roots of the trees emerge from four ornamental vases (canthares), one of which is located centrally along each side of the square.

Given what little we know about the history and provenance of the collection: it comprises objects consciously selected in the 1930s on the basis that somehow they were ‘secondrate’ or ‘surplus to requirements’, it follows that objects deposited in (and retrieved from) the grave in a pristine condition were selected against. Beyond this it is difficult to generalise. Recent work suggesting that textiles retrieved from the graves of children are, all things being equal, likely to be in a better condition than those from adult graves because smaller bodies decompose more quickly and with fewer by-products might, if we had more contextual information, permit us to conclude that most of these textiles were not part of child burials (Czaja-Szewczak 2005; Linscheid 2001: 77). There is little point in making such speculations with a collection such as this, however. Breakthroughs in our understanding of 1st millennium burial practices in Egypt are likely now to come mainly from properly-conducted (and funded) excavations, rather than museum archaeology. Yet, as I have argued here, the value of re-examining museum textile collections (including their subjection to scientific analysis) should be

The square itself is bordered by a solid russet-coloured line. This, in turn, is bordered by a four strand interlace or cable border in the flying shuttle technique. Another solid russetcoloured line frames this interlace border. Finally, the whole design is bordered by four vine scrolls with alternating trefoils and tendrils, one on each side of the square. Each vine scroll protrudes from a canthare similar to those found within the square itself. The textile is produced in finely spun linen and wool and does not appear to be shrunken, although the weave is slightly distorted in places. The square is intact, although approximately one quarter of the wool wefts are worn completely away, leaving a triangular area comprised only of unbleached linen warps and wefts. Comparisons The canthare with vine tendrils emerging from its neck is a very common motif in textiles from Egypt. Analogous tapestry squares, also from Akhmîm, are found in several museums, with a close comparison at the Textile Museum in Washington D.C. [71.119] (Trilling 1982). The same overall design can also be found superimposed on another square,

Note that Karanis, which is usually interpreted as having been abandoned in the 5th century, has recently been suggested (Keenan 2007) to have continued in use. This would cast doubt on those artefacts (including textiles) which have usually been given a 5th-century terminus post quem. 6

120

Anthea Harris: The Social Life of Museum Textiles

Cat. 1. Acc. 2005.7.1.

so as to form an eight-pointed star, as, for example, at the Cluny Museum, Paris [13162] and the Ashmolean Museum, Oxford [1888.744] (Lorquin 1992: 76, no. 11). In the latter example the canthare motif is almost identical in form and size to those forming the central part of this fragment. In other cases, a much more simplified version appears, as in an example in the Victoria and Albert Museum [2031891] (Kendrick 1922: 120, no. 281). The canthare also appears as a motif in its own right, as, for example, in the Österreichisches Museum für Angewandte Kunst in Vienna [T. 661-1883] (Noever 2005: 140, no. 80).

Dimensions: 220mm x 220mm A. Ground weave Warp: unbleached linen, S-direction spin, approximately 14 threads per cm Weft: unbleached linen, S-direction spun, approximately 15 threads per cm Weave: simple tabby (1/1)

Technical details

B. Tapestry areas

Base fabric: none

Tapestry weave

Tapestry weave

Warp: unbleached linen, S-direction spin, approximately 20 threads per cm

121

Dress and Identity

Weft: unbleached linen, S-direction spin. Russet-coloured and yellow wool, S-direction spin

120mm x 105mm; unbleached linen and dyed wool Description

Ribs per cm: 7

Acquired in Egypt before 1889 by Rev. Greville Chester.

Weave: weft-faced tabby; tapestry woven on 3 warp threads

The fragment is worked in brown-purple wool on a ground weave of unbleached linen. The design comprises a roundel containing a stylised tree with a central trunk and 16 large leaves. This is bordered by a solid circular band (7mm diameter), also worked in brown-purple wool. The tapestry is a cut-out from a larger textile and only very small fragments of the outer ground weave in unbleached linen are extant.

Special techniques: Interlace design and finer details of the canthares produced by the ‘flying shuttle’ technique in one single thread of unbleached flax. ‘Split tapestry’, dovetailing, wrapping around one warp thread and vertical weft brocading 2. Tapestry medallion (orbiculus) with ‘Tree of Life’ design Accession number 2005.7.2

The edges of the leaves are outlined and enhanced using the dovetailing technique.

From Akhmîm, Egypt, probably 4th – 6th century

The fragment is almost intact, although some of the border

Cat. 2. Acc. 2005.7.2

122

Anthea Harris: The Social Life of Museum Textiles

is missing. It is in good condition, although it appears slightly shrunken and the weave is distorted.

B. Tapestry areas Tapestry weave

Comparisons

Warp: unbleached linen, S-direction spin, approximately 18 threads per cm

‘Tree of Life’ or ‘Tree of Paradise’ motifs are relatively common in Late Roman and Byzantine textiles from Egypt and may reflect Byzantine links with the East. Several examples are preserved in museum collections, such as three in private Flemish collections and others at the Musée Historique des Tissus at Lyon [978.I.1] and in the Louvre, where the edges of the leaves are worked in the same dovetailing technique (de Moor 1993: nos. 95, 140, 143; Bourgon-Amir, 1993: 222, plate 221; du Bourguet 1964: C47/C6). The use of dovetailing to outline leaves was a common technique and can also be seen in examples from Ravenna [2473] and Bargello [599B/F], Italy (Rizzardi 1993: 47, plate 3; Peri 1996: plate 19). There is another comparison from Washington D.C. [71.49] (Trilling 1982: 72-3).

Weft: unbleached linen, S-direction spin. Purple wool, S-direction spin, Z-direction ply Weave: weft-faced tabby; tapestry woven on 2 warp threads Ribs per cm: 9 Special techniques: flying shuttle technique, vertical weft brocading, dovetailing 3. Leaf tapestry with stem Accession number 2005.7.3

Technical details

From Akhmîm, Egypt; probably 4th – 6th century

Base fabric: none

Approx. 80 mm x 204 mm; unbleached linen and dyed wool

Tapestry weave

Description

Dimensions: approximately 120mm x 105mm

Acquired in Egypt before 1889 by Rev. Greville Chester.

A. Ground weave

The fragment comprises a cut-out depiction of a fig leaf with a long stem (7mm diameter; at least 134mm in length). It is produced in brown-purple wool and unbleached linen on an unbleached linen ground weave with details in the ‘flying shuttle’ technique and vertical weft brocading. The stem features the dovetailing technique.

Warp: unbleached linen, S-direction spin, approximately 14 threads per cm (extrapolated) Weft: unbleached linen, S-direction spin, approximately 14 threads per cm (extrapolated)

The fragment is in good condition and intact, except for the truncation at the terminus of the stem and some distortion to the weave.

Weave: simple tabby (1/1)

Cat. 3. Acc. 2005.7.3

123

Dress and Identity

Comparisons

Description

Bands of tapestry terminating in a leaf motif are relatively common amongst Late Roman and Byzantine textiles from Egypt. The same basic form can be elaborated so as to make either a leaf of a tree. Although there are no precise comparisons for this particular piece, good analogies include those from the Textile Museum in Washington D.C. [71.123] as well as Paris and Worms [inv. T.513] (Trilling 1982: 69; du Bourguet 1964: D26, D28; Renner-Volbach 2002: 43-4, no. 13, plate 5).

Acquired in Egypt before 1889 by Rev. Greville Chester. The textile fragment consists of two matching tapestry bands (clavi) stitched parallel to each other on a base fabric of plain tabby weave in unbleached linen. Each band incorporates a central row of repeating vine-scroll with alternating trefoils and tendrils in brown-purple and brown-black wool. These are flanked on each side by a solid line, one of brown-purple wool and the other of brownblack wool. The solid lines are, in turn, flanked by a row of repeating small brown-purple circles, each enveloped by two brown-black scrolls and each filled with a small brown-black dot.

Technical details Base fabric: none Tapestry weave Dimensions: approximately 80 mm x 204 mm A. Ground weave Warp: unbleached linen, S-direction spin, approximately 15 threads per cm Weft: unbleached linen, S-direction spin, approximately 13 threads per cm Weave: simple tabby (1/1) B. Tapestry areas Tapestry weave Warp: unbleached linen, S-direction spin, approximately 16 threads per cm Weft: unbleached linen, S-direction spin. Brown-purple wool, S-direction spin Weave: weft-faced tapestry; tapestry woven on 2 warp threads Ribs per cm: 8 Special techniques: vertical weft brocading, flying shuttle technique, dovetailing 4. Fragment of linen with two tapestry bands (clavi) of foliate and zoomorphic design Accession number 2005.7.4 From Akhmîm, Egypt; probably 5th – 7th century 117mm x 290mm; unbleached linen and dyed wool

Cat. 4. Acc. 2005.7.4

124

Anthea Harris: The Social Life of Museum Textiles

Cat. 4. Acc. 2005.7.4 reverse side

The patterns of the two bands are almost identical to each other, suggesting that they were woven as one piece. A stylised zoomorphic motif (possibly a bird) enclosed in an oval appears on one of the bands. The oval merges at each end into the main vine scroll. This band also has two possible slits (up to 20mm in length) spaced 150mm apart and placed perpendicular to the long side of the band. These are 67mm and 217mm respectively from the hemmed edge. In other respects the two bands are identical.

The fragment is produced in extremely fine wool and linen threads. It is in good condition, although some fibres are heavily stained, possibly with body fluids. The weave is distorted. Comparisons A vine-scroll design analogous to that on the Ure piece, albeit without accompanying zoomorphic motif, is found on a fragment of band, dated to the 5th century, which is now at Bologna [inv. 487] (Ghiggini 2000: 64-5, no.30). More examples, this time with zoomorphic motifs, can be found in the Victoria and Albert Museum [T41-1936] and the Brooklyn Museum [08.480.52] (Thompson 1971: 34, no. 12). A crudely-executed design of a vine-scroll incorporating a zoomorphic motif inside an oval can be found at Bargello [inv. 596D/F] (Peri 1996).

The main design is produced in two shades of wool, now appearing brown-purple and brown-black, on a background of unbleached linen. In addition, the lips of the stylised zoomorphic motif are comprised of some three or four stitches in fine red wool. The tapestry bands are stitched to the backing in such a way that the horizontal vine-scroll pattern runs in opposite directions. The longest sides of each band have been turned under at their edges and tacked to the base fabric with a loose running stitch, before being secured with a simple left-slanting overstitch. On one short side of the fragment the ends of each band are turned under and fastened to the backcloth with overstitch, forming a simple hem. There is no indication of similar hemming on the base fabric itself.

It is possible that the tapestry band is older than the base fabric on which it is stitched and that the fragment is an example of the re-use of textiles in dress ornamentation. Such practices are known from Egypt in this period, although it is unlikely that the tapestry bands would be more than a few decades older than the base fabric to which they were stitched.

125

Dress and Identity

Technical details

trefoil. They are joined by a thin white (unbleached linen) line with two arrows at its centre. The small red medallion has a wave-crest border, also in red wool, and a leaf motif in unbleached linen in its centre.

Base fabric Dimensions: approximately 117mm x 290mm

The ground weave has areas of ‘self-banding’ (also in unbleached linen threads) on either side of the red wool band. This technique was often used in Late Roman and Byzantine weaving to draw attention to adjacent areas of decoration or to the edge of a garment without the introduction of colour. It involves multiple picks of weft being inserted without changing the position of the heddle. Ordinarily, in plain tabby weave the position of the heddle would be changed after every passage of the shuttle. In this example thicker weft threads have also been inserted to produce more variety and texture in the design.

Warp: unbleached linen, S-direction spin, approximately 13 threads per cm Weft: unbleached linen, S-direction spin, approximately 17 threads per cm Weave: simple tabby (1/1) Tapestry weave Dimensions: 55mm by 290mm; 55mm by 290mm A. Ground weave: none

The piece is in fair condition; the medallion is attached to the rest of the fragment by only a few threads.

B. Tapestry areas

Comparisons

Tapestry weave

Analogous textiles are held by the Musée de Mariemont [inv. DM96] and the Textile Museum, Washington D.C. [72.165] (Azzam and Bruwier 1997: 176, plate 58; Trilling 1982: 34, plate 5). Another comparative was excavated from grave 433 at Antinoë in the first decade of the 20th century and is now in Haute-Alsace [inv. 965.157.1-8] (Rassart-Debergh 1997: 95, 141, fig. 120).

Warp: unbleached linen, S-direction spin, approximately 16 threads per cm Weft: unbleached linen, S-direction spin. Purple and ‘brown-black’ wool, S-direction spin, Z-direction ply

Technical details

Weave: weft-faced tabby; tapestry woven on one or two warp threads

Base fabric: none

Ribs per cm: 14

Tapestry weave

Special techniques: flying shuttle technique, dovetailing

Dimensions: approximately 170mm x 35mm

5. Fragment of linen textile incorporating a small medallion motif and band

A. Ground weave

Accession number 2005.7.10

Warp: unbleached linen, S-direction spin, approximately 21 threads per cm

Provenance unknown; probably 5th – 6th century Approx. 170mm x 35mm; unbleached linen and dyed wool

Weft: unbleached linen, S-direction spin, approximately 22 threads per cm

Description

Weave: simple tabby weave (1/1)

A fragment (probably from an unbleached linen tunic) comprising a narrow band (clavus) and a medallion (possibly part of another clavus) on a ground weave of plain tabby executed in finely spun unbleached linen thread.

Other features: the clavus is flanked on one side by a narrow (3mm) area of 5-6 self-bands in unbleached linen

The band comprises a series of three elongated lozenges joined by narrow necks. These are worked in red wool against a plain unbleached linen background. At each end of the fragment the band takes the form of four red trefoil motifs, each joined to the other. The central lozenge is decorated with two small white motifs: a heart and a

Tapestry weave

B. Tapestry areas

Warp: unbleached linen, S-direction spin, approximately 18 threads per cm

126

Anthea Harris: The Social Life of Museum Textiles

Cat. 5. Acc. 2005.7.10

Cat. 5. Acc. 2005.7.10 reverse side

6. Fragment of textile with darned areas

Weft: unbleached linen, S-direction spin. Red wool, S-direction spin

Accession number 2005.7.11

Weave: tapestry weave; tapestry woven over 2 warp threads

Provenance unknown; probably 8th – 10th centuries

Ribs per cm: 9

Approx. 150mm x 110mm; dyed wool and unbleached linen

Special techniques: dovetailing, self-banding, wrapping around one warp thread

Description

127

Dress and Identity

Cat. 6. Acc. 2005.7.11

This is a fragment of weft-faced tabby comprising brown-yellow wool and blue-red wool wefts woven over unbleached linen warp threads. It is possible that the bluered wool is a form of ‘optical purple’.

larger textile and it is not possible to reconstruct the pattern, although it appears to have been comprised of a series of rectangles or squares. It is in good condition. Comparisons

The piece incorporates areas which appear to have been strengthened or repaired by darning, achieved with inlaid threads. These are thicker weft threads laid over the top of pre-existing weft threads, so that they are visible only on the front of the textile and do not appear on the reverse side except where there is a change in their direction. The inlaid threads are comprised of two-ply yellow-brown wool, spun in the S-direction and plied in the Z-direction, so as to produce an optimally stable and durable thread.

Published comparisons are rare, but this may represent the tendency of museums to accept and to publish those pieces with higher art historical than archaeological value. The appearance of fragments analogous to this may have caused them to be selected against and thus to appear missing from the published record. An exception is the child’s tunic in the Whitworth Art Gallery, which has darning in at least ten different colour threads [T.8505] (Pritchard 2005: 40, fig. 3.14).

The extant fragment appears to have been part of a much

128

Anthea Harris: The Social Life of Museum Textiles

Cat. 6. Acc. 2005.7.11 reverse side

Technical details

Weft: yellow-brown wool, S-direction spin, Z-direction ply

Base fabric: none

Special techniques: inlaid threads or darning

Tapestry weave

7. Fragment of purple tapestry with geometric design

Dimensions: 150mm x 110mm

Accession number 2005.7.12

A. Ground weave:

Provenance unknown; probably 7th – 9th century

Warp: unbleached linen, S-direction spin, approximately 9 threads per cm

Approx. 120mm x 125mm; unbleached linen and ‘bluepurple’ wool

Weft: blue-red wool, S-direction spin, Z-direction ply. Brown-yellow wool, S-direction spin, Z-direction ply

Description Fragment of a probable band (clavus). The ground weave is worked in a blend of blue and red wool, plied together to produce the effect of purple. This is a well-known technique, with a resultant effect often described as ‘optical purple’. Different coloured yarns were regularly combined to produce a greater variety of shades. The use of blue dyed wool in background weaves has also been associated with Persian or Persian-influenced textiles.

Weave: weft-faced tabby; tapestry woven over one warp thread Ribs per cm: 9 B. Darned areas: Warp: yellow-brown wool, S-direction spin, Z-direction ply

129

Dress and Identity

Cat. 7. Acc. 2005.7.12

Cat. 7. Acc. 2005.7.12 reverse side, showing blue–red wool weave

130

Anthea Harris: The Social Life of Museum Textiles

The central design on this fragment is comprised of a series of interconnecting circles and scrolls, sometimes linked by vertical lines executed in vertical weft brocading. A running wave crest motif in unbleached linen thread borders one length of the band.

Weft: unbleached linen, S-direction spin. Blue-red wool, S-direction spin, Z-direction ply Weave: weft-faced tabby; tapestry woven over one warp Ribs per cm: 8

Comparisons

Special techniques: wrapping around one warp thread

There is an analogous textile fragment in the Ashmolean Museum, Oxford [1956.702], also worked in a blend of blue and red wool that combines to produce the effect of purple. This fragment, too, is without provenance.

8. Fragment of tapestry with geometric design Accession number 2005.7.13

Technical details

Provenance unknown; probably 7th – 9th century

Base fabric: none

Approx. 35mm x 156mm; unbleached linen and dyed wool

Tapestry weave

Description

Dimensions: approximately 120mm x 125mm

A probable fragment of a wrist or neck cuff in brown-purple wool and unbleached linen.

A. Ground weave

A solid line in unbleached linen appears on three sides of the fragment. On one side of the line there is a zig-zag motif, while on the other there is a geometric design consisting of diagonal lines and small flowers or petals of red wool. The design is executed in unbleached linen threads on a brown-purple background.

Warp: Blue-red wool, S-direction spin, Z-direction ply, approximately 10 threads per cm Weft: Blue-red wool, S-direction spin, Z-direction ply, approximately 18 threads per cm Weave: weft-faced tabby

There is one selvedge edge, indicating that this is the transverse edge of the textile as woven on the loom. A zigzag motif runs alongside the selvedge, bordered by the solid line in unbleached linen.

Other features: none B. Tapestry areas

The side of the fragment running perpendicular to the selvedge has been turned under and hemmed with whiplash or overstitch. A single stitch in pink-red wool is appended to this edge, suggesting that the fragment was, at some stage, sewn to a pink-red textile of unknown form and size.

Tapestry weave Warp: unbleached linen, S-direction spin, approximately 8 threads per cm

Cat. 8. Acc. 2005.7.13

131

Dress and Identity

Comparisons Comparisons include a fragment of tapestry in the Victoria and Albert Museum, [481-1889] and another in Liverpool Museum [56.20.799] (Kendrick 1922: vol II, 85, no. 584, plate 29; Seagroatt 1965: 37, plate 18). Both of these were excavated in the Fayum area of Upper Egypt, although an origin in Syria or Anatolia has been suggested for the Victoria and Albert Museum textile. The Louvre has more comparative pieces (du Bourguet 1964: 337, F239). A fragment of tunic from Antinoë in the Museo Nazionale di Ravenna has a cuff bordered in an analogous tapestry weave, albeit with a slightly different decorative pattern [inv. 2465] (Rizzardi 1993: 108-9, pl. 44). This might suggest a similar purpose for the Ure fragment. Another comparative fragment from Antinoë is now in Haute-Alsace [Cpt.176] (Rassart-Debergh 1997: 160, fig. 253). Technical details Base fabric: none Tapestry weave Dimensions: approximately 35mm x 156mm A. Ground weave: none B. Tapestry areas Tapestry weave Warp: unbleached linen, S-direction spin. Brown-purple wool, S-direction spin, Z-direction ply. Approximately 16 threads per cm Weft: brown-purple wool, S-direction spin, Z-direction ply Weave: plain tabby; tapestry woven over 2 warp threads Ribs per cm: N/A Special techniques: ‘inlaid’ or ‘floating’ wefts 9. Fragment of linen with two tapestry bands (clavi) of foliate design Accession number 2005.7.5 Cat. 9. Acc. 2005.7.5

From Akhmîm, Egypt; probably 6th – 8th century 230 mm x 90 mm; unbleached linen and dyed wool

of unbleached linen. Each band has a design comprised of trefoils, probably representing ivy leaves, facing in the same direction. The trefoils are separated a short piece of scroll and a dot, probably representing a tendril. Each band is bordered by a solid line 6mm wide on its internal edge and 3mm wide on its external edge.

Description Acquired in Egypt before 1889 by Rev. Greville Chester. The textile fragment consists of two matching tapestry bands (clavi) stitched parallel to each other on a base fabric

The design is executed in brown-purple wool on a

132

Anthea Harris: The Social Life of Museum Textiles

Dimensions: 32mm by 170mm; 35mm by 230mm

background of unbleached linen. The tapestry bands are aligned so that the rows of trefoils run in the same direction.

A. Ground weave:

The longest sides of each band have been turned under at their edges and sewn to the base fabric with a small, neatly executed, running stitch. There is no overstitch on the upper side of the fragment, suggesting that the band was closely pinned to the base fabric whilst sewing took place and that the first side of the band was pressed with a hot iron before the second side was turned under and secured. The variation in the width of the borders is a function of one side (of each band) having been turned under more than the other.

Warp: unbleached linen, S-direction spin, approximately 10 threads per cm Weft: unbleached linen, S-direction spin, approximately 18 threads per cm Weave: slightly weft-faced tabby weave (1/1) B. Tapestry areas

There are remnants of a ground weave on one side of each band, but not the other. On the side without the ground weave it was, therefore, necessary to turn part of the border itself under in order to secure the band on the base fabric. One of the bands has been turned under slightly more than the other, so that the width of one is 32mm and the width of the other is 36mm. The space between the two bands on the base fabric is 23mm. This seems to be an example of ornamental dress textiles being re-used on new garments.

Tapestry weave Warp: unbleached linen, S-direction spin, approximately 14 threads per cm Weft: unbleached linen, S-direction spin. Purple wool, S-direction spin, Z-direction ply Weave: weft-faced tabby; tapestry woven over 2 warp threads

At one short edge of the fragment both bands are turned under and secured with overstitching in order to produce a simple hem. The base fabric itself has not been hemmed.

Ribs per cm: 7

The fragment is in good condition, although possibly slightly shrunken.

Special techniques: ‘split’ tapestry, dovetailing 10. Fragment of tapestry band (clavus) with vine-scroll

Comparisons

Accession number 2005.7.6

An analogous design from Akhmîm, was published by Forrer in 1891 (plate IV, no. 2). Another example, also comprising parallel bands, is found in the collection at Bargello and has been dated to the 5th or 6th century. Here, the parallel bands serve as a border to a geometric and foliate design [inv. 592D/F] (Peri 1996: plate 25). A more elongated version of this design, also stitched to a base fabric so as to form two parallel bands 90mm apart, can be seen in the same museum. This has been dated to the 5th century [inv. 592A/F] (Peri 1996: plate 23).

From Akhmîm, Egypt; probably 4th – 6th century 35mm x 165mm; unbleached linen and dyed wool Description Acquired in Egypt before 1889 by Rev. Greville Chester.

Dimensions: 230 mm x 90 mm

The fragment comprises a decorated band (clavus) with no ground weave or base fabric. It displays an undulating vine stem of tendrils and bunches of grapes, executed in a dark-coloured (probably originally black, dark brown or dark purple) wool on a background of (somewhat coarse) unbleached linen. The individual grapes are suggested by a judicious use of the dovetailing technique.

Warp: unbleached linen, S-direction spin, approximately 13 threads per cm

The fragment is in good condition, although encrusted with residues, possibly body fluids.

Weft: unbleached linen, S-direction spin, approximately 26 passes per cm

Comparisons

Technical details Base fabric:

A similar piece, also from Akhmîm, was published by Forrer in 1891 (plate IV, no. 13). A fragment of a roundel in the Victoria and Albert Museum [1374-1888] has an analogous grape design in dark brown wool, as does one from Bargello, which is dated to the 4th or 5th century [inv.

Weave: weft-faced tabby weave (1/1) Tapestry weave

133

Dress and Identity

Cat. 10. Acc. 2005.7.6

592C/F] (Kendrick 1922: vol. II, 102, no. 196; Peri 1996: plate 20). Bunches of similarly executed grapes border a medallion dated to the 4th century, now at Bologna [inv. 491] (Ghiggini 2000: 56-7, plate 24). The Louvre has other comparanda (du Bourguet 1964: 58, A20, A21, A22). Technical details Base fabric: none Tapestry weave Dimensions: approximately 35mm x 165mm A. Ground weave: none B. Tapestry areas Tapestry weave Warp: Unbleached linen, S-direction spin, approximately 8 threads per cm Weft: Unbleached linen, S-direction spin. Brown wool, S-direction spin, Z-direction ply Weave: weft-faced tabby; tapestry woven over 1 warp thread Ribs per cm: 8 Special techniques: dovetailing 11. Fragment of a tapestry medallion Accession number 2005.7.7 Provenance unknown; probably 5th – 6th century Approx. 143mm x 62mm; unbleached linen and dyed wool Cat. 11. Acc. 2005.7.7

Description

134

Anthea Harris: The Social Life of Museum Textiles

Ribs per cm: 8

The fragment comprises a dark-coloured (probably originally brown-purple) textile with small areas of a circular design at its opposite edges, presumed to be the border of a roundel or medallion. The border design consists of a row of pearls with their interstices filled with small dots picked out in threads of unbleached linen. There are fragments of a geometric, possibly interlace, design in the centre of the medallion, produced by the flying shuttle technique, also in unbleached linen.

Other features: Geometric pattern in centre of roundel produced by the ‘flying shuttle’ technique in one single thread of unbleached linen 12. Fragment of linen with possible foliate design Accession number 2005.7.8

The roundel with pearl border has been identified as a feature deriving from the art of the Persian Empire, and may, therefore, have entered the cultural matrix via interactions between the Sasanian and Byzantine Empires before the collapse of the former in 642 (Granger-Taylor 1993: 15-21).

Provenance unknown; probably 4th – 7th century Approx. 410mm x 250mm; unbleached linen and dyed wool Description A fragment of simple tabby weave executed in finely spun unbleached linen. Its decoration comprises four parallel tapestry bands and one looped fringe or starting border running parallel to them.

The fragment is in poor condition, with much distortion in the weave and substantial areas of residue. Comparisons

The tapestry bands are approximately 24mm wide and are separated by strips of unbleached linen tabby weave. The outermost bands are both spaced 8 weft threads apart from their neighbouring, inner-most band. The two inner-most bands are spaced 14 weft threads apart from each other.

Roundels with analogous ‘pearl’ designs are found in many museums: for example, Worms [inv. T507], Darmstadt, in Flemish collections and in the Rietz collection of the California Academy [inv. 389-2583] (Renner-Volbach 2002: 38-9, no. 7, plate 4; Renner 1985: 27-8, no. 5, plate 3; de Moor 1993; Carroll 1988: 84, no. 3). The Pfister Collection [no. 16] in the Vatican Museums has a fragment with a similar geometric design in ‘weft brocading’, as does the Musée Historique des Tissus, Lyon [907.I.84] (28 520/170)] and the Louvre [B4, B10] (Bourgon-Amir 1993: 142-3, plate 126; du Bourguet 1964: 65, 68). There are other comparanda in London and Rouen (Kendrick 1922: vol. I, 100, no. 187, plate 27; Durand and Saragoza 2002: 125, no. 87)

The tapestry is executed in wool wefts on warp threads of unbleached linen. The wool now appears as blue-purple, but is probably a 2-plied thread of red and blue yarns, the red having faded more than the blue. The design of the tapestry areas is extremely fragmentary but it appears abstract, rather than geometrical, and is possibly a foliate pattern. A repeating possible lozenge motif decorated with a simple rosette in flying shuttle technique can be discerned. At the centre of the rosette there is a small cross executed in vertical weft brocading.

Technical details Base fabric: none

The wool wefts have worn away in large areas of the fragment, leaving only the linen warps and wefts.

Tapestry weave

Comparisons

Dimensions: approximately 143mm x 62mm A. Ground weave: none

There is an analogous fringe on a plaited woollen bag from Akmîm, now in the Victoria and Albert Museum [inv. 7621886] (Kendrick 1922: vol II, 89, plate 30).

B. Tapestry areas

Technical details

Tapestry weave

Base fabric: none

Warp: unbleached linen, S-direction spin, approximately 8 threads per cm

Tapestry weave Dimensions: approximately 410mm x 250mm

Weft: Brown-purple wool, S-direction spin, Z-direction ply. Unbleached linen, S-direction spin

A. Ground weave

Weave: weft-faced tabby; tapestry woven over 1 warp thread

Warp: unbleached linen, S-direction spin, approximately 12 threads per cm

135

Dress and Identity

Cat. 12. Acc. 2005.7.8

136

Anthea Harris: The Social Life of Museum Textiles

Weft: unbleached linen, S-direction spin, approximately 12 threads per cm Weave: tabby weave Other features: one looped starting border B. Tapestry areas Tapestry weave Warp: unbleached linen, S-direction spin, approximately 24 threads per cm Weft: unbleached linen, S-direction spin. Blue-purple wool, S-direction spin Weave: weft-faced tabby weave; tapestry woven over 3 warp threads Ribs per cm: 9 Special techniques: flying shuttle technique, vertical weft brocading, wrapping around one warp thread, dovetailing 13. Fragment of linen textile with tapestry area of geometric design Accession number 2005.7.9 Provenance unknown; probably 7th – 9th centuries Approx. 178mm x 34mm; unbleached linen and dyed wool Description A fragment of tapestry area (probably a band) sewn to a base fabric of slightly weft-faced tabby weave executed in coarse unbleached linen. The tapestry areas are in white (unbleached linen) and red wool. The design is comprised of a red background with a row of repeating white circles, each filled with four red diamonds. Between each circle is a small open white square, each filled with an ‘H’ shape in white. The squares and circles are linked by diagonal lines constituting smaller white squares and forming a lattice. The tapestry area is folded under at one edge of the fragment and fastened with whip-lash stitch, forming a hem. The practice of sewing ready-made tapestry bands onto base fabrics of tabby weave is usually accepted as being a development from weaving tapestry areas as integral parts of complete or near-complete garments. Textiles produced in this way, therefore, might be given a date from the 4th to the 9th centuries. In this case, the use of a white or cream lattice design on a red background is analogous to the designs used on silks of the 7th and 8th centuries, which

Cat. 13. Acc. 2005.7.9

137

Dress and Identity

were also sometimes cut up to use as decorative features on garments (de Moor 1993: 239).

14. Fragment of unbleached linen with inlaid thread decoration

The textile is unusual insofar as the threads used in the tapestry areas are spun in the Z-direction, more commonly associated with Syria or Mesopotamia in this period. The resemblance of the fragment to silks of the same period makes an Eastern Mediterranean origin likely.

Accession number 2005.7.14

Comparisons

Description

No linen and wool comparanda are available but a silk fragment at Lyon [inv. 897.III.25], possibly produced at Antinoë, has an analogous design, albeit in blue-green and yellow (Martiniani-Reber 1986: no. 52). A fragment possibly from the same textile is in the Louvre [inv. E.29382] (Martiniani-Reber 1997: 87, no. 36).

A fragment of plain (tabby) weave executed in coarse unbleached linen threads. It has a simple, polychrome, geometric pattern comprising individual strands of bluepurple (‘optical purple’?), blue-green, light brown and dark red wool.

Provenance unknown; 5th – 9th century Approx. 270mm x 100mm

The pattern is formed of adjoining triangles, four of which (in blue-green) surround a central motif (possibly a cross) executed in light brown. The coloured wool threads were ‘inlaid’ as additional wefts during the weaving process and ‘float’ across the textile, only appearing on the surface when they are necessary to the design and only appearing on the reverse of the textile where there is a change in their direction. The inlaid thread passes over three and under a single warp thread, with at least two picks of weft being woven between each inlaid thread.

Technical details Base fabric Dimensions: approximately 114mm x 28mm Warp: unbleached linen, S-direction spin, approximately 16 threads per cm Weft: unbleached linen, S-direction spin, approximately 24 threads per cm

Although an analogous textile (in the Victoria and Albert Museum) has been dated to the 5th-6th century, others have been dated to the Islamic period, especially where a fringe is incorporated on at least one edge. The absence of a fringe on this example, together with its cross design, may suggest an earlier, Byzantine, date.

Weave: slightly weft-faced tabby Tapestry weave Dimensions: approximately 178mm x 34mm

Examination of the reverse side of the fragment reveals that there has been very little fading of the woollen threads. The woollen threads appear in the same, unworn condition on both sides of the fragments. Combined, this evidence suggests that the textile was new when deposited in the ground and that, if we may assume that it derived from a burial context, it may even have been produced or procured specifically as a burial garment.

A. Ground weave: none B. Tapestry areas Tapestry weave Warp: unbleached linen, S-direction spin, Z-direction ply, approximately 8 threads per cm

Comparisons

Weft: unbleached linen, S-direction spin. Red wool, S-direction spin, Z-direction ply

A textile from Liverpool with a comparable design was identified as a pillowcase by Seagroatt [inv. 56.21.973, see also 56.22.112] (Seagroat 1965: 36, 38, pls. 17, 19). The Ure fragment is too small to tell whether it was part of such an object, although it is produced in the same coarse linen as the Liverpool pillowcase. The Musée de Mariemont (Belgium) has pieces incorporating analogous simple geometrical designs from ‘inlaid’ wefts [inv. DM150 & DM151], as does a piece in the collections of the Victoria and Albert Museum [T.92-1922] (Azzam and Bruwier, 1997, 216-8, pl. 100, 102. The latter is also comprised of cross motifs framed by triangles. There are over 50 analogous textiles, mostly without figural motifs, in

Weave: weft-faced tabby weave; tapestry woven over 1 or 2 warp threads Ribs per cm: 11 Special techniques: ‘split’ tapestry

138

Anthea Harris: The Social Life of Museum Textiles

Cat. 14. Acc. 2005.7.14

139

Dress and Identity

Vienna (e.g. Noever 2005: 177), as well as examples with Kufic script (e.g. Rogers 1983: 14-15, 18-19, fig. 10, plate III, where a 8th-10th century textile bears the inscription ‘BISMILLAH AL RAHMAN AL’ – ‘dominion belongs to Allah’). Technical details Base fabric: none Tapestry weave Dimensions: 270mm x 100mm A. Ground weave Warp: unbleached linen, S-direction spin, approximately 9 threads per cm Weft: unbleached linen, S-direction spin, approximately 8 threads per cm Weave: simple tabby weave (1/1) Other features: none B. Tapestry areas Tapestry weave Warp: N/A

Cat. 15. Acc. 2005.7.15

Weft: blue-purple (‘optical purple’?), blue-green, light brown and dark red wool, S-direction spin, Z-direction ply

woollen strip. This is executed in unbleached linen threads. It is likely that the fragment represents a fine linen garment with a woollen cuff or neck-line, although it could be a very narrow clavus.

Weave: inlaid weft threads over 3 warp threads Ribs per cm: N/A Special techniques: ‘floating weft’ or ‘inlaid weft threads’

The piece is in good condition, albeit with distortion to the weave.

15. Fragment of fine linen cloth with narrow band of red-blue wool

Comparisons

Accession number 2005.7.15

Approx. 110mm x 60mm; unbleached linen and dyed wool

Analogous pieces were excavated in Nubia from Meroitic graves. One of these (Grave B188, number 14) yielded a fragment which, when analysed, was shown to have been dyed ‘Imperial’ purple, thus making it an important archaeological find (Thurman and Williams 1979: 60).

Description

Technical details

Fragment of a textile executed in fine unbleached linen thread in simple tabby weave. Along one side of the fragment is a red-blue (probably ‘optical purple’) strip of finely spun wool (60mm x 7mm). The woollen strip is an integral part of the textile, being woven (weft-faced) over the unbleached linen warp threads. On the opposite side of the fragment, an area of ‘self-banding’ runs parallel to the

Base fabric: 110mm x 60mm

Provenance unknown; probably 4th – 6th centuries

Warp: unbleached linen, S-direction spin, approximately 24 threads per cm Weft: unbleached linen, S-direction spin, approximately 24 threads per cm

140

Anthea Harris: The Social Life of Museum Textiles

Cat. 16. Acc. 2005.7.16

Weave: tabby weave (1/1)

Warp: unbleached linen, S-direction spin, approximately 14 threads per cm

looped twisted threads along one edge (a ‘closing cord’). The border is 10mm deep (each loop being 5mm long), with 4 loops per cm. The loops are plied in the Z-direction and then twisted in the S-direction. The fragment is comprised of cream or unbleached woollen wefts woven over warp threads in a similarly-coloured wool.

Weft: red-blue wool; S-direction spin

Comparisons

Weave: weft-faced tabby Ribs per cm: 12; woven over 2 warp threads

Comparanda may be found at Khirbet Qazone (Jordan) (Textiles 14 and 15) (Granger-Taylor 2000: 153-5, figs. 5, 9).

Other features: none

Technical details

16. Fragment of plain woollen cloth with a looped border

Base fabric

Accession number 2005.7.16

Dimensions: 105mm x 86mm

Provenance unknown; 7th-century or later?

Warp: cream or unbleached wool, S-direction spin, Z-direction ply, approximately 8 threads per cm

Tapestry area (purple stripe):

Approx. 105mm x 86mm; cream or unbleached wool

Weft: cream or unbleached wool, S-direction spin, Z-direction ply, approximately 32 threads per cm

Description

Weave: weft-faced tabby (1/1)

A rectangular fragment of weft-faced tabby, with no decoration except for a border of similarly-coloured small

141

Dress and Identity

Ribs per cm: 8 Other features: an integral border of looped twisted threads. A strand of S-spun linen protrudes from the looped border. Without examining the reverse of the fragment it is not possible to ascertain the significance of this. 17. Fragment of coarse linen cloth with decorated braid Accession number 2005.7.17 Provenance unknown; date unknown Approx. 195mm x 120mm; unbleached linen and dyed wool Description A fragment of simple tabby weave in coarsely spun unbleached linen. A fragment of tapestry ‘cord’ also in unbleached linen is crudely whip-lashed to it (along one long edge only) in similarly unbleached linen thread. This ‘cord’ is 8mm wide and 10.5 mm long. It is comprised of weft-faced tabby woven over one warp. Small heart-or lozenged-shaped gaps in the textile indicate where another thread (probably wool) has been eroded, leaving only the unbleached linen warp threads. A clump of unbleached linen threads, held together by means of a substance that may be dried body fluids, adheres to the textile. The textile itself does not appear to have been stained by the same substance (except where the linen threads adhere to it).

Cat. 17. Acc. 2005.7.17

Comparisons

Tapestry weave

None known

Warp: unbleached linen, S-direction spin, approximately 18 threads per cm

Technical details

Weft: unbleached linen, S-direction spin. Brown wool, S-direction spin

Base fabric: none Tapestry weave

Weave: weft-faced tabby; tapestry woven over 1 warp thread.

Dimensions: approximately 195mm x 120mm

Ribs per cm: 9 (extrapolated) (8 threads in total)

A. Ground weave

Special techniques: none remaining

Warp: unbleached linen, S-direction spin, approximately 12 threads per cm

18. Fragments of dark green, yellow and brown woollen textile

Weft: unbleached linen, S-direction spin, approximately 12 threads per cm

Accession numbers 2005.7.19, 2005.7.20, 2005.7.21 (accessioned separately although they probably constitute one piece)

Weave: plain tabby (1/1) Other features: several similarly-coarse threads of unbleached linen adhere to this textile by means of a large (blood?) stain.

Provenance unknown; date unknown Dark green: ranging from approx. 15mm x 11mm to approx. 95mm x 42mm; dyed wool

B. Tapestry areas

142

Anthea Harris: The Social Life of Museum Textiles

Cat. 18. Acc. 2005.7.19-21

Yellow: Up to approx. 42mm x 10mm; dyed wool

Warp: dark green wool, S-direction spin, approximately 13 threads per cm

Brown: Approx. 20mm x 15mm; dyed wool

Weft: dark green wool, S-direction spin, approximately 21 threads per cm

Description Six small pieces of simple tabby weave in finely spun dark green wool, with one small possible selvedge area where the wool is plied in the Z-direction. Four small pieces of simple tabby weave in finely spun yellow wool, with one small area of over-stitch in the same colour yarn, possibly representing a hem. One small piece of simple tabby weave in finely spun dark brown wool.

Weave: simple tabby weave (1/1)

Comparisons

Warp: yellow wool, S-direction spin, approximately 16 threads per cm

Other features: none Yellow fragment: Base fabric:

‘Rag amalgams’ from excavated contexts at Berenike, Karanis and Yotvata suggest that this roughly sewn together piece was perhaps used as packing in the grave (Thomas 2007: 155; Wild and Wild 2000: 264-73; Shamir 2001: 96-7). In non-mortuary contexts they may have served as dusters or cleaning cloths.

Weft: yellow wool, S-direction spin, approximately 21 threads per cm Weave: simple tabby weave (1/1) Other features: none

Technical details

Brown fragment:

Dark green fragment:

Base fabric:

Base fabric:

143

Dress and Identity

Cat. 19. Acc. 2005.7.18

Warp: yellow wool, S-direction spin, approximately 14 threads per cm

Dimensions: approximately 480mm x 8mm Warp: unbleached linen, S-direction spin. Orange-pink dyed linen, S-direction spin. Approximately 3 threads per cm

Weft: yellow wool, S-direction spin, approximately 14 threads per cm

Weft: unbleached linen, S-direction spin. Orange-pink dyed linen, S-direction spin. 10 threads in total (4 orange-pink linen; 6 unbleached linen).

Weave: simple tabby weave (1/1) Other features: none

Weave: weft woven over one warp thread

19. Fragment of narrow pink and white band

Ribs per cm: 3

Accession number 2005.7.18

Note on the weaves represented in this collection

Provenance unknown; date unknown (post 7th-century)

There is a vast amount of literature describing tapestry and weaving techniques used to produce textiles from Egypt in this period. Most catalogues of museum collections (including those in the bibliography below) contain a technical chapter and a glossary explaining these and other terms. In brief, tapestry is comprised of discontinuous wefts being employed to build up a design. In other words, the same weft thread does not necessarily travel from one selvedge to the other. Instead, it may be passed backward and forward over a select number of warp threads, whilst wefts of different colours are passed backward and forward over other warp threads, as necessary for the design. The wefts are packed closely together so that the warp threads are not usually visible, and the result is a weft-faced textile with areas of design. Single and multiple dovetailing, where adjoining wefts of different colours are alternately looped around the same warp thread (for example, 2005.7.1 and 2005.7.4). This enables straight or almost straight vertical lines to be woven into the design without compromising the strength of the cloth. An alternative way of achieving this effect is to use ‘slit tapestry’, where discontinuous wefts are not looped around the same warp thread, but around adjacent warp threads instead (for example, 2005.7.1, 2005.7.4 and 2005.7.12). This technique can also be used to build up curves, by staggering the looping of the wefts around successive warp threads. Other methods of ornamentation include ‘inlaying’, where additional weft

Approx. 480mm x 8mm; unbleached linen and dyed linen Description A fragment of narrow braided ribbon in unbleached linen and orange-pink linen. There is a selvedge on both long sides. This piece does not incorporate any areas of tapestry, since all the weft threads pass continuously from one end of the textile to the other. Comparisons There are no known near comparisons in Late Roman, Byzantine or Islamic textiles and it is possible that this fragment does not date from any of these periods. However, analogous pieces executed in wool, rather than linen, were excavated from graves between Abu Simbel and the Sudan frontier (Ancient Nubia) (Thurman and Williams 1979: 68, 76). These have been identified as edging for a sheet or tunic [Grave Q10, number 43 and Grave R12, number 28] or as a belt. Technical details Base fabric

144

Anthea Harris: The Social Life of Museum Textiles

threads, usually in a different colour, are laid at regular intervals between the wefts of the ground weave (for example, 2005.7.12). Whereas the ground weave wefts may pass over and under every alternate warp thread, the inlaid weft may pass over four warp threads, say, before passing under a single warp thread. Some of the textiles have decorative features that resemble fine stitches of embroidery overlying the main tapestry woven design. These ‘stitches’ are usually produced in a single thread of undyed linen against a much darker background colour. They are not actually embroidered, but produced as an integral part of the weaving process by the introduction of a thin weft thread that ‘floats’ diagonally over the rest of the tapestry. The procedure is known as the ‘flying shuttle’ technique (also known as ‘flying thread brocading’ or ‘weft brocading’) and was used to add detail to designs and to highlight patterns and forms.

Czaja-Szewczak, B. (2005): ‘Tunics from Naqlun’, in G. Gabra (ed.), Christianity and Monasticism in the Fayoum Oasis, 133-42. New York. Dauterman Maguire, E. (1999): Weavings from Roman, Byzantine and Islamic Egypt: The Rich Life and the Dance, Urbana-Champaign. De Jönghe, D. (ed) (1999): Ancient tapestries of the R. Pfister collection in the Vatican Library, Vatican City. De Moor, A (ed.), Coptic Textiles from Private Flemish Collections, Zottegem. du Bourguet, P. (1964): Catalogues des étoffes coptes du musée du Louvre, Paris. Durand, M. and Saragoza, F. (2002): Egypte, la trame de l’histoire: textiles pharaoniques, coptes et islamiques, Paris. Forrer, R. (1891): Römische und byzantinische Seidentextilien aus dem Gräberfelde von Achmim-Panopolis, Strasbourg. Ghiggini, F. (2000): Tessuti Copti, la collezione del Museo Storico Didattico della Tappezzeria, Bologna. Granger-Taylor, H. (1993): ‘The Decoration of Coptic Textiles’, in de Moor (ed.), Coptic Textiles from Private Flemish Collections, 15-20. Zottegem. Granger-Taylor, H. (2000): ‘The textiles from Khirbet Qazone (Jordan)’, in Dominique Cardon and Michel Feugère (eds), Archéologie des textiles, des origines au Ve siècle: actes du colloque de Lattes, octobre 1999, 149-62. Montagnac. Griggs, C.W. (2005): ‘Early Christian burials in the Fayoum’, in Gawdat Gabra (ed.), Christianity and Monasticism in the Fayoum Oasis, 185-95. New York. Keenan, J.G. (2007): ‘Byzantine Egyptian Villages’, in R. Bagnall (ed.), Egypt in the Byzantine World, 300-700, 226-243. Cambridge. Kendrick, A. F. (ed.) (1920-1922): Catalogue of Textiles from Burying Grounds in Egypt (Vols. 1-3). London. Kitzinger, E. (1946): ‘The Horse and Lion Tapestry at Dumbarton Oaks: A Study in Coptic and Sasanian Textile Design’, Dumbarton Oaks Papers 3: 2-72 Lafontaine-Dosogne, J. (1988): Textiles coptes des Musées Royaux d’art et d’Histoire de Bruxelles. Bruxelles Linscheid, P (2001): ‘Late Antique to Early Islamic Textiles from Egypt’, Textile History 32 (Special Issue on Medieval Textiles): 75-80. Lorquin, A. (1992): Les tissus coptes au musée national du Moyen Age–Thermes de Cluny. Paris. Martiniani-Reber, M. (1997): Textiles et mode sassanides: les tissus orientaux conservés au département des Antiquités égyptiennes. Paris. Meinardus, O.F.A. (1999): Two Thousand Years of Coptic Christianity, New York. Noever, P. (ed.) with contributions by A. Völker, R. Hofmann-de Keijzer, R. Knaller, V. Mader, and A.S. Weidner (2005): Fragile Remnants: Egyptian Textiles of Late Antiquity and Early Islam. Vienna. Peri, P. (ed.), (1996): Tessuti copti nelle collezioni del Museo del Bargello. Florence. Pritchard, P. (2006): Clothing Culture: Dress in Egypt in the First Millennium AD, Manchester.

Acknowledgements I am grateful for the support of the Aylwin Cotton Foundation and to Dr Amy Smith, Curator of the Ure Museum at The University of Reading for her encouragement and patience, as well as permission to photograph the textiles. Thanks are also due to Vivien Chapman, Gusty Hawkins, Rhianedd Smith, John Peter Wild and Linda Woolley. All errors remain, of course, my own responsibility. Bibliography Azzam, A. and Bruwier, M.C. (eds) (1997): Egyptiennes. Etoffes coptes du Nil, Mariemont. Bender Jørgensen, L. and Mannering, U. (2001): ‘Mons Claudianus: Investigating Roman Textiles in the Desert’, in P. Walton Rogers, L. Bender Jørgensen and A. RastEicher (eds), The Roman Textile Industry: A Birthday Tribute to John Peter Wild, I-II, 1-11. Oxford. Bourgon-Amir, Y. (1993): Les Tapisseries Coptes du Musée Historique des Tissus, Lyon, Montpellier. Bowen, G.E. (2002): ‘Textiles, Basketry and Leather Goods from Ismant El-Kharab’, in C.A. Hope and G.E. Bowen (eds), Dakhleh Oasis Project. Preliminary Reports on the 1994-1995 to 1998-1999 Field Seasons, 87-104. Oxford. Bowen, G.E. (2003): ‘Some Observations on Christian Burial Practices at Kellis,’ in G.E. Bowen and C.A. Hope (eds), The Oasis Papers III: Proceedings of the Third International Conference of the Dakhleh Oasis Project, 167–182. Oxford. Carroll, D.L. (1988): Looms and Textiles of the Copts: First Millennium Egyptian Textiles in the Carl Austin Rietz Collection of the California Academy of Sciences. Seattle. Clarke, M.C., Hiatt, R. P., Kuchar, M.C. J., and Farahnakian, M. H. (2003): ‘Indexing and Cataloging Textiles From the Fag el Gamous Cemetery in Fayum, Egypt to Determine Their Relationship With Known Coptic Textiles’, Clothing and Textiles Research Journal 21(3): 120-29

145

Dress and Identity

Rassart-Debergh, M. (1997): Textiles d’Antioé (Égypte) en Haute-Alsace. Colmar. Renner, D (1985): Die spätantiken und koptischen Textilien im Hessischen Landesmuseum in Darmstadt. Wiesbaden. Renner-Volbach, D (2002): Die sogenannten koptischen Textilien im Museum Andreasstift der Stadt Worms. Wiesbaden. Rizzardi, C. (1993): I Tessuti Copti del Museo Nazionale di Ravenna. Ravenna. Rogers, C. (ed.) (1983): Early Islamic Textiles. Brighton. Schieck, A.P. (2005): Die Koptischen Textilien: Gewebe und Gewänder des ersten Jahrtausends aus Ägypten. Cologne. Schrenk, S. (2004): Textilien des Mittelmeerraumes aus spätantiker bis frühislamischer Zeit. Riggisberg. Seagroat, M. (1961-2): ‘The Coptic Textile Collection’, Liverpool Bulletin 10, n.p. Seagroatt, M. (1965): Coptic Weaves: Notes on the collection of Coptic Textiles in the Merseyside County Museums. Liverpool. Shamir, O. (2001): ‘Byzantine and Early Islamic Textiles Excavated in Israel’, Textile History 32 (Special Issue on Medieval Textiles): 93-105. Smith, A.C. and Harris, A. L. (2008): ‘Recent additions and conservation at the Ure Museum at the University of Reading, 2004-08’, Archaeological Reports 54: 175-85 Thomas, T. (2007): ‘Coptic and Byzantine textiles found in Egypt: Corpora, collections and scholarly perspectives’ in R. Bagnall (ed), Egypt in the Byzantine World, 300700, 137-62. Cambridge.

Thompson, D. (1971): Coptic Textiles in the Brooklyn Museum. New York. Thurman, C.C.M. and Williams, B. (1979): Ancient Textiles from Nubia, Meroitic, X-group and Christian fabrics from Ballana and Qustul. Chicago. Trilling, J. (1982): Roman Heritage: Textiles from Egypt and the Eastern Mediterranean, 300 to 600 AD, Washington D.C. Ure Museum database, A.C. Smith (ed.), The University of Reading. http://lkws1.rdg.ac.uk/cgi-bin/ure/uredb.cgi [accessed July 12th 2010]. van Strydonck, M., K. van der Borg and Ar. de Jong, (1993): ‘The dating of Coptic textiles by radiocarbon analysis’, in A. De Moor (ed.), Coptic Textiles from Private Flemish Collections, 65-71. Zottegem. van Strydonck, M., A. De Moor and D. Bénazeth, (2004): ‘14C Dating Compared to Art Historical Dating of Roman and Coptic Textiles from Egypt’, Radiocarbon 46(1): 231-44. Wild, J.P. and Wild, F.C. (2000): ‘Textiles’, in S.E. Sidebotham and W.Z. Wendrich (eds), Berenike 1998, Report of the 1998 Excavations at Berenike and the Survey of the Egyptian Eastern Desert, including excavations in Wadi Kalalat, 264-273. Leiden (Centre of Non-Western Studies Publications, special series 5). Wild, J.P. (2001): ‘Sails from the Roman Port at Berenike, Egypt’, The International Journal of Nautical Archaeology 30: 211-20.

146